State of Minnesota v. Michael Allan Carbo, Jr.

6 N.W.3d 114

Opinion

STATE OF MINNESOTA

IN SUPREME COURT

A22-1823

Saint Louis County Anderson, J.

Concurring, Procaccini, J.

Concurring in part, dissenting in part, Thissen, J.

Concurring in part, dissenting in part, McKeig, Chutich, Moore, III, JJ.

State of Minnesota, Respondent,

vs. Filed: May 8, 2024 Office of Appellate Courts Michael Allan Carbo, Jr.,

Appellant.

________________________

Keith Ellison, Attorney General, Peter Magnuson, Assistant Attorney General, Saint Paul, Minnesota; and

Kimberly J. Maki, Saint Louis County Attorney, Duluth, Minnesota, for respondent.

Cathryn Middlebrook, Chief Appellate Public Defender, Adam Lozeau, Assistant Public Defender, Saint Paul, Minnesota, for appellant.

Elizabeth G. Bentley, Tyler Blackmon, Certified Student Attorney, Adam Kolb, Certified Student Attorney, Civil Rights Appellate Clinic, University of Minnesota Law School, Minneapolis, Minnesota; and

Teresa Nelson, American Civil Liberties Union of Minnesota, Minneapolis, Minnesota, for amici curiae American Civil Liberties Union and American Civil Liberties Union of Minnesota.

Shauna Faye Kieffer, Kieffer Law LLC, Minneapolis, Minnesota, for amicus curiae Minnesota Association of Criminal Defense Lawyers.

________________________

2

SYLLABUS

1. The district court did not err by denying the defendant’s motion to suppress evidence stemming from a genetic analysis of DNA collected from crime scene materials because the defendant had abandoned his subjective expectation of privacy in that information by leaving his semen at the scene of the crime.

2. The district court did not err by denying the defendant’s motion to suppress evidence stemming from a genetic analysis of DNA collected from his garbage because law enforcement lawfully and independently obtained identical information from a DNA sample the defendant voluntarily provided.

3. The district court abused its discretion by denying the defendant’s motion to present alternative-perpetrator evidence because the defendant’s proffered evidence clearly had an inherent tendency to connect the alternative perpetrator to the commission of the crime and could have been admitted under the ordinary rules of evidence, and the error was not harmless beyond a reasonable doubt.

Reversed and remanded.

OPINION

ANDERSON, Justice.

In this case, we are asked to determine whether the district court erred when it denied appellant Michael Allan Carbo, Jr.’s motion to suppress evidence stemming from analyses of genetic information extracted from semen he left at the crime scene and garbage he left in a communal disposal bin. We must also decide whether the district court abused its discretion when it denied Carbo’s motion to introduce alternative-perpetrator evidence.

3

We decide that the district court did not err in concluding that Carbo had abandoned his subjective privacy interest in the genetic information gathered from the crime scene and that the evidence obtained from Carbo’s garbage need not have been excluded. We nevertheless conclude that the district court abused its discretion in excluding Carbo’s proffered evidence on the grounds that it did not have an inherent tendency to connect the alternative perpetrator to the commission of the crime. Accordingly, we reverse and remand for further proceedings consistent with this opinion.

FACTS

This case arises from the 1986 murder of Nancy Daugherty. That summer, 38-year-old Daugherty was separated from her husband and living alone in her house in Chisholm. B.E., a 31-year-old man, was friends with Daugherty. The two first met through Daugherty’s husband and soon began working together. B.E. was immediately infatuated with Daugherty and pursued a deeper friendship with her. Daugherty and B.E. became sexually involved but ended their relationship due to feeling guilty about Daugherty’s marriage.

After they broke up, B.E. moved away but continued to write letters to Daugherty, telling her how much he still thought about her and asking her to write back. When Daugherty did not do so, B.E. became increasingly more demanding in his desire to maintain a relationship with her. In one letter, B.E. wrote that “[t]here are also times I think about you and I get so mad I could wring your neck.”

Two months before Daugherty died, B.E. visited her in Chisholm. The two went out for dinner and drinks, and although B.E. was “thinking about maybe rekindling some

4

of the intimacy” between them, Daugherty stopped his advances. B.E. then drove Daugherty home, where he noticed a car parked by her house. Daugherty said the car belonged to a man she was dating. This news hurt B.E. He later admitted to investigators that he “suspected all along that [Daugherty] had been seeing somebody.”

B.E. again visited Daugherty on July 15, the day before she died. He had made plans to help Daugherty move some items into storage the next day, as she was in the process of moving away from Chisholm. He arrived at Daugherty’s house around 8 p.m. and had a drink with her. Daugherty and B.E. then went out for a few more drinks at a local bar. Around midnight, they left the bar and went for a drive around a nearby lake. B.E. thought about making advances towards Daugherty but ended up driving back to her house. The two talked in the car for a while until Daugherty said she wanted to go to bed. B.E. then asked to come inside and use the bathroom; Daugherty agreed. Daugherty was sitting at the kitchen table as B.E. left the bathroom, and he sat down and talked with her more. A short while later, Daugherty told B.E. he should go, and he left.

Before he had driven more than a block away, B.E. noticed it was misting and returned to Daugherty’s house to ask her if her moving plans would change if it rained the next morning. He “stuck [his] head in [the door]” and noticed that Daugherty was putting a different shirt on, which indicated to him that she had some other plans that night. He asked Daugherty about how the weather might affect their plans, she responded that they could discuss it tomorrow, and he left a second time. B.E. drove to his parents’ house, where he was staying, and spoke to his younger sister, who was awake and watching television. He then stated that he “had a glass of milk and went to bed.”

5

The next morning, about 9 hours after B.E. had last seen Daugherty, B.E. returned to Daugherty’s house to help her move. She did not answer her door, which was locked, and her shades were pulled down. He left and returned several times before eventually speaking with a neighbor. The neighbor then called the police. B.E., the neighbor, and a police officer went into the house and discovered Daugherty’s naked body under a comforter on her bed.

A medical examination revealed abrasions on Daugherty’s neck and clavicle but no other external injuries. A vaginal swab detected sperm that had been deposited within the previous 48 hours. Daugherty’s death certificate listed her cause of death as asphyxiation due to manual strangulation. Police found grass on Daugherty’s body and on her damp clothes. There was a large patch of matted grass outside Daugherty’s back door where a pile of vomit was found.

The police interviewed the neighbor’s daughter who heard yelling coming from Daugherty’s house. She and her friend heard screams around 3:15 a.m., which sounded “like someone was being choked and shaken, and they were trying to scream through it.” They went outside and walked along the street next to Daugherty’s house, where they saw a dark-colored truck that was “shaped more like a van” or “a truck with a topper” 1 parked in the driveway of the home. They heard more screams and “a man and woman arguing.” When the noises stopped, they headed back inside and went to sleep.

1 At the time Daugherty was killed, B.E. owned a grey Ford Bronco II with a topper.

Carbo’s father had a green truck and a red truck; Carbo denies ever driving either vehicle.

6

The police focused on B.E. as their main suspect. In an interview several years after the murder, B.E. admitted that he had a “bad temper” and said, “[y]ou know, the human mind is a strange thing, and I’ve often wondered jeez, did I wake up in the middle of the night, drive over here and kill her, go back to bed and not know it?” The Bureau of Criminal Apprehension (BCA) tested pubic hairs that were found in Daugherty’s bedroom and found that B.E. qualified as a potential source. Yet the police were never able to find conclusive proof that B.E. committed the murder.

In 2019, the police contracted with Parabon Nanolabs, Inc. (Parabon) to conduct a genetic analysis of DNA preserved from the crime scene. Parabon examined a comprehensive genetic profile generated from crime scene semen and scrapings taken from under Daugherty’s fingernails. It then utilized commercial genealogical databases to determine the likely source of the profile. It generated a report that also gave information about the suspect’s physical traits (including skin, eye, and hair color) and ancestral origin. Parabon concluded it was “highly confident” that Carbo was the likely source for the genetic profile. At the time of Daugherty’s death, Carbo was 18 years old and attending high school in Chisholm. He had no apparent direct connection to Daugherty but went to school concurrently with her two children. Carbo had not previously been investigated by police in connection with Daugherty’s death.

After receiving Parabon’s report, the police began investigating Carbo.

Investigators surveilled Carbo’s apartment and watched him put a bag of garbage in a shared collection bin by the curb. They took the bag from the bin and sent it to the BCA for analysis. The BCA generated a basic DNA profile from items in the garbage bag and

7

determined that the profile matched several of the 1986 crime scene samples, including the semen, the fingernail scrapings, and the vomit in the yard. The police then approached Carbo—without mentioning that they were investigating him—and asked for a voluntary DNA sample. Carbo agreed, and the voluntary sample he provided matched both the crime scene profile and the profile created from the garbage.

In April 2022, a grand jury indicted Carbo with two counts of first-degree murder “while committing or attempting to commit criminal sexual conduct in the first or second degree with force or violence” in violation of Minn. Stat. § 609.185(a)(2) (2022). Carbo moved to admit evidence that B.E. was an alternative perpetrator and to suppress the evidence of his DNA analysis. The district court denied both motions, concluding that Carbo’s proffer did not have an inherent tendency to connect B.E. to the commission of the crime and that Carbo had abandoned any privacy interest in his genetic information. At trial, the State called B.E. to testify about the events leading up to Daugherty’s death and the discovery of her body, but Carbo was not permitted to introduce evidence that might suggest B.E. was an alternative perpetrator. Carbo’s counsel argued that Carbo was likely “blacked out” on the night Daugherty was killed, as he was using drugs and drinking at the time. Carbo’s counsel denied that Carbo killed Daugherty and instead argued that Carbo had consensual sex with her that night, after which, someone else entered her home and killed her. He did not dispute at trial that the crime scene DNA samples came from Carbo. The jury found Carbo guilty of two counts of first-degree murder. He now appeals to this court.

8

ANALYSIS

I.

We must first decide whether the district court correctly admitted evidence obtained from a warrantless and revealing examination of Carbo’s genetic profile. “When reviewing the legality of a search or seizure, an appellate court will not reverse the district court’s factual findings unless clearly erroneous or contrary to law.” State v. Wiegand, 645 N.W.2d 125, 129 (Minn. 2002). We review a district court’s conclusions on constitutional questions involving searches and seizures de novo. State v. Anderson, 733 N.W.2d 128, 136 (Minn. 2007).

The Fourth Amendment guarantees people the right “to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures.” U.S. Const. amend. IV. The Minnesota Constitution offers similar protection. See Minn. Const. art. I, § 10. A search occurs when the government seeks to gain information by infringing upon a person’s “reasonable expectation of privacy.” Katz v. United States, 389 U.S. 347, 360 (1967) (Harlan, J., concurring). Under the Katz approach, a defendant must show that he had an “actual subjective expectation of privacy” in the object searched and that his “expectation [was] reasonable.” State v. Gail, 713 N.W.2d 851, 860 (Minn. 2006) (citations omitted) (internal quotation marks omitted).

A subjective expectation of privacy is usually lost when an object is abandoned. See City of St. Paul v. Vaughn, 237 N.W.2d 365, 370–71 (Minn. 1975). “Abandonment is primarily a question of intent, and intent may be inferred from words spoken, acts done, and other objective facts.” Id. (internal quotation marks omitted) (quoting United States v.

9

Colbert, 474 F.2d 174, 176 (5th Cir. 1973)). We have thus held that, when property is left in a public place where the defendant can no longer expect privacy and the police may lawfully be present, it may be abandoned for purposes of search and seizure. Id. at 371.

Carbo argues that we should distinguish between his expectation of privacy in the semen he left behind at the crime scene and the vast troves of genetic information his semen contained. The abandonment of a privacy interest in the former, he argues, does not equate to abandonment of a privacy interest in the latter. Even accepting this distinction for sake of argument, though, Carbo’s actions unambiguously signaled his abandonment of any privacy interest in his genetic information.

By Carbo’s own account, he got so drunk that he blacked out, had unprotected sex with Daugherty in her yard, then cleaned up in her bathroom and left. These actions do not lead to an inference that Carbo intended to keep his genetic information—which he knew his semen contained—private. See Colbert, 474 F.2d at 176. We do not agree with the concurrence of Justice Procaccini that the abandonment doctrine is an “ill fit” when genetic information is involved. Infra at C-10. Although surely circumstances exist in which an attempt to conceal information may be futile, Carbo’s situation does not fit this description. Carbo’s DNA was not collected from a shed skin cell or a flake of dandruff—it came from semen he voluntarily left inside Daugherty and skin he voluntarily left under her fingernails. 2 Our precedent provides sufficient guidance for us to determine whether these

2 The voluntary nature of Carbo’s actions distinguishes his situation from those cited by the concurrence. Nothing about Carbo’s behavior on the night Daugherty was killed is comparable to the sharing of private hotel registry information we held protected in State

10

actions signaled an abandonment of Carbo’s privacy interest in what those materials could reveal about him, and we conclude that they did.

Because Carbo abandoned his subjective expectation of privacy in his genetic information, 3 a Fourth Amendment search did not occur when the police had Parabon perform a genetic analysis of the biological material Carbo left behind at the crime scene. Accordingly, the district court did not err in denying Carbo’s motion to suppress information gathered from this analysis.

II.

Carbo also argues that the genetic testing performed on items taken from his discarded garbage constituted a search in violation of his Fourth Amendment rights. 4 We

v. Leonard, 943 N.W.2d 149 (Minn. 2020). Nor was the evidence Carbo left behind a mere consequence of his “venturing into the public sphere.” Carpenter v. United States, 585 U.S. 296, 310 (2018). Our conclusion that Carbo abandoned his subjective privacy interests through his actions is necessarily limited to his actions. It thus does not “open[] the door to the indiscriminate collection of deeply sensitive and personal genetic information from the general public,” infra at C-12, insofar as the general public does not undertake the specific combination of actions Carbo did. 3 Because Carbo abandoned his subjective expectation of privacy, we need not consider whether, under the United States or Minnesota Constitutions, the deeply personal and revealing information contained in a genetic profile falls within an expectation of privacy that society is willing to recognize as reasonable. 4 Carbo makes another argument—that the State violated Minnesota law when it performed genetic testing on samples taken from his garbage. He claims that his genetic information is protected under Minn. Stat. § 13.386 (2022), which prohibits the collection, storage, and analysis of genetic information without the source’s consent, unless “expressly provided by law.” This argument fails, however, because Minn. Stat. § 299C.155 (2022) allows the BCA to perform DNA analysis on crime scene specimens. This statute thus provides an express exception to the consent requirement of section 13.386, and Carbo does not otherwise show that the police violated Minnesota law when they analyzed the genetic information in his garbage.

11

need not address whether such a search occurred, as the information gained from this testing was independently gathered through a voluntary sample that Carbo provided to police. “The independent source doctrine allows admission of evidence that has been discovered by means wholly independent of any constitutional violation.” Nix v. Williams, 467 U.S. 431, 443 (1984); see also State v. Diede, 795 N.W.2d 836, 849 (Minn. 2011). So long as “that later acquisition was not the result of the earlier [constitutional violation], there is no reason why the independent source doctrine should not apply.” Murray v. United States, 487 U.S. 533, 541 (1988).

Here, Carbo voluntarily provided a DNA sample shortly after the police generated a genetic profile from his garbage that matched him to the crime scene sample. Carbo was unaware of the testing already performed on his garbage, so his decision to provide a sample could not have been improperly induced by the pressure of knowing he was under investigation. Likewise, because Parabon’s testing had already placed Carbo under individualized suspicion, the confirmation that he was a genetic match to the crime scene sample did nothing to narrow the scope of an investigation already centered on him. As there is “no reason why the independent source doctrine should not apply,” we hold that the district court did not err in denying Carbo’s motion to suppress information gathered from the genetic analysis of his garbage. 5 Id.

5 Carbo’s claim that the State forfeited its independent source argument lacks merit.

While the State did not name the independent source doctrine explicitly in its argument before the district court, it clearly laid out the exact reasons why it should apply.

12

III.

Next, we must decide whether the district court abused its discretion when it denied Carbo’s motion to introduce alternative-perpetrator evidence. We review the evidentiary rulings of the district court for an abuse of discretion, “even when it is claimed that the exclusion of evidence deprived the defendant of his constitutional right to present a complete defense.” State v. Penkaty, 708 N.W.2d 185, 201 (Minn. 2006). Moreover, “[i]f a trial court’s evidentiary ruling is determined to be erroneous, and the error reaches the level of a constitutional error, such as denying the defendant the right to present a defense, our standard of review is whether the exclusion of evidence was harmless beyond a reasonable doubt.” State v. Smith, 876 N.W.2d 310, 331 (Minn. 2016) (citation omitted) (internal quotation marks omitted).

The Due Process Clauses of the Minnesota Constitution and the United States Constitution guarantee a criminal defendant’s right to present a complete defense. Minn. Const. art. I, § 6; U.S. Const. amend. XIV; see also State v. Atkinson, 774 N.W.2d 584, 589 (Minn. 2009) (“A defendant has the constitutional right to present a complete defense.”). Included in the right to present a complete defense is the right to introduce evidence showing that an alternative perpetrator committed the crime. State v. Ferguson, 804 N.W.2d 586, 591 (Minn. 2011). The purpose of this evidence is not to establish the alternative perpetrator’s guilt, but to “create a reasonable doubt as to the defendant’s guilt.” Atkinson, 774 N.W.2d at 590.

Yet the right to a complete defense, we have held, “is not absolute.” State v. Jenkins, 782 N.W.2d 211, 224 (Minn. 2010). Under State v. Hawkins, a two-step process is used to

13

determine the admissibility of alternative-perpetrator evidence. 260 N.W.2d 150, 159 (Minn. 1977). First, the defendant must proffer foundational evidence that has “an inherent tendency to connect [a third party] with the actual commission of the crime.” Id. If the first step is satisfied, the defendant may then introduce “evidence of a motive of the third person to commit the crime, threats by the third person, or other miscellaneous facts which would tend to prove the third person committed the act” if the evidence satisfies the ordinary rules of evidence. Id.

Carbo argues that the district court abused its discretion when it excluded his proffered evidence on the grounds that it did not satisfy the two-step Hawkins test. We agree, and we further conclude that the exclusion of this evidence was not harmless beyond a reasonable doubt.

A.

1.

We first turn to the two-step Hawkins test. Carbo’s proffer included the following evidence: B.E. was with Daugherty the night she was killed, and B.E. drove a vehicle that was physically consistent with an eyewitness’s description of the vehicle parked in Daugherty’s driveway when the eyewitness heard screaming. A BCA report stated that B.E. qualified as a possible source for pubic hairs found in Daugherty’s bedroom after the murder. When interviewed because of his suspected involvement with the crime, B.E. told police that he often wondered whether he woke up in the middle of the night and killed Daugherty. Viewing Carbo’s proffer as a whole, we conclude that this evidence clearly

14

has “an inherent tendency to connect [B.E.] with the actual commission of the crime” under the first step. Hawkins, 260 N.W.2d at 159.

This conclusion is evident in light of our case law. The cumulative weight of Carbo’s proffer is analogous to the scenario we analyzed in State v. Ferguson, in which we reversed the district court and concluded that a proffer containing “multiple pieces of information that may have incriminated [the alleged alternative perpetrator]” had an inherent tendency to connect the alleged alternative perpetrator to the commission of the crime. 804 N.W.2d at 592. There, the alleged alternative perpetrator owned a car that “matched some of the descriptions” of the getaway car as “a brown sedan.” Id. at 591, 588. He also shared physical similarities with the perpetrator, who was described as a “black male, about five-nine, thin build, medium to dark complected” with a nose “a little wider” and “medium sized lips.” Id. at 588 (internal quotation marks omitted). Finally, the alleged alternative perpetrator had his name in the victim’s phone contacts, and “was not in police custody on the date of the shooting.” Id. at 591.

Here, the State offered no evidence of the perpetrator’s appearance, so any physical similarities between Carbo and B.E. would have been irrelevant. But Carbo’s proffered evidence has just as much tendency to connect B.E. to the commission of the crime as the proffered evidence in Ferguson, if not more. B.E.’s admittedly close connection to the victim herself, when added to the evidence suggesting that he was at the scene of the crime, easily places this case in line with Ferguson and in contradistinction with cases where we found alternative-perpetrator evidence lacking.

15

By contrast, in Troxel v. State, we found lacking a proffer that failed to provide any evidence whatsoever that the alleged alternative perpetrator had been at the location where the crime was committed. 875 N.W.2d 302, 309 (Minn. 2016). That proffer differs significantly from Carbo’s, which not only established that B.E. was at the scene of the crime within hours of the murder, but also that a vehicle consistent in appearance with the truck owned by B.E. was parked in the driveway of Daugherty’s house as screaming was heard coming from inside.

Troxel, a case in which the proffer contained no evidence regarding the alternative perpetrator’s presence at the scene of the crime, does not support the conclusion reached by the district court: that Carbo’s proffer fails simply because it did not “establish that it was in fact B.E.’s vehicle, that he was in fact the driver of the vehicle, and that he was in fact in the home.” Essentially, this reads the word “tendency” out of the Hawkins “inherent tendency to connect” requirement, demanding instead a definite and unequivocal connection to the commission of the crime. If this were the standard, though, criminal defendants would effectively need to present evidence sufficient to convict the alleged alternative perpetrator to avail themselves of their constitutional rights to present complete defenses. 6 This is not what Hawkins requires, and we decline here to raise the evidentiary standard required for criminal defendants to present alternative-perpetrator evidence as part of a complete defense.

6 Presumably, if the incriminating facts required by the district court were proven, B.E. would not merely be an “alternative” perpetrator, but the alleged perpetrator.

16

The concern motivating the adoption of the Hawkins test—to avoid ensnaring the trial in a quagmire of matters wholly collateral to the charged offense—is conspicuously absent here. B.E. was not just one of many in a long list of witnesses called by the defense in a desperate hope that some witness, any witness, would spark uncertainty regarding Carbo’s guilt. B.E. was an early suspect in Daugherty’s murder and the sole person with Daugherty in the hours leading up to her death; he speculated to law enforcement about his possible participation in the murder and was also a key witness at Carbo’s trial. Although no conclusive proof was developed establishing B.E. as the murderer, pubic hair evidence, according to the BCA, qualified B.E. as a source for physical evidence at the scene of the crime. It does not require a “significant speculative leap” to take one look at the facts here and conclude that those facts had an inherent tendency to connect B.E. to the actual commission of the crime. Infra at C/D-10. If the alternative-perpetrator evidence here is not sufficient to pass the Hawkins test, it is reasonable to ask whether any alternative-perpetrator evidence is ever admissible under the Hawkins precedent.

Carbo’s entire theory of the case rested on the alleged involvement of B.E. in the crime. “By not allowing the defense to question” B.E. about the facts connecting him to Daugherty’s murder, B.E. “was not exposed to the truth-revealing pressures of the sort of cross-examination which is the heart of our adversary system.” Hawkins, 260 N.W.2d at 160. When the district court concluded that Carbo’s proffer did not satisfy the foundational requirement of Hawkins, it denied Carbo the opportunity to present a complete defense and abused its discretion. We conclude instead that Carbo’s proffer laid a sufficient evidentiary foundation for his alternative-perpetrator defense.

17

2.

Because Carbo’s proffer laid a sufficient foundation that his alternative-perpetrator evidence had an inherent tendency to connect B.E. to the crime, then under the second step, evidence of B.E.’s motive to commit the crime, threats B.E. made against Daugherty, or other miscellaneous facts that would tend to prove that B.E. committed the murder would have been admissible if the ordinary rules of evidence were otherwise satisfied.

The evidence of motive, threats, and miscellaneous facts was substantial in this case.

B.E.’s long and frustrating history of affection for Daugherty could have motivated him to kill her. B.E. wrote to Daugherty repeatedly, expressing his feelings for her despite her lack of reciprocation. He continuously pursued opportunities to rekindle their romantic relationship and was hurt whenever he discovered evidence of Daugherty’s interest in others. B.E.’s potential motivation as a “scorned lover” was in fact so apparent at the time of Daugherty’s death that the police considered B.E. as the main suspect. In addition, B.E. admitted that he had anger issues and that he told Daugherty that he sometimes wanted to “wring [her] neck.” Lastly, B.E.’s purported alibi left open an opportunity for him to travel to Daugherty’s house in the middle of the night and choke Daugherty to death.

The question under the second Hawkins step is whether this evidence is admissible under the ordinary evidentiary rules. We conclude that it is. The State does not meaningfully dispute that this evidence, if it passed the first Hawkins step, would be otherwise admissible. Here, the foremost hurdle to admissibility relates to evidence of B.E.’s motives—specifically, the previous sexual relationship between B.E. and Daugherty and his knowledge of her other romantic relationships which upset him. In the district

18

court, Carbo argued that such evidence should not be precluded under Minn. R. Evid. 412, 7 because the strict application of the evidentiary rule would interfere with his constitutional right to present a complete defense and the probative value of the evidence was not substantially outweighed by unfair prejudice. See State v. Friend, 493 N.W.2d 540, 545 (Minn. 1992) (“In certain cases the due process clause, the right to confront accusers, or the right to present evidence will require admission of evidence otherwise excluded by the rape shield law.”). We agree with Carbo that, here, the probative value of the evidence establishing a motive for B.E.’s purported actions was not substantially outweighed by the State’s interest in “guarding [the victim’s] privacy and in protecting her from harassment.” State v. Caswell, 320 N.W.2d 417, 419 (Minn. 1982). Indeed, the purpose of introducing this evidence was not to prove anything about Daugherty whatsoever, but rather to prove the facts known to B.E. that allegedly motivated him to commit the crime.

Accordingly, we conclude that the facts contained in Carbo’s proffer satisfied both steps of the Hawkins test, and the district court abused its discretion when it excluded this evidence for failure to satisfy the Hawkins test.

B.

Lastly, we turn to the question of whether the district court’s abuse of discretion in excluding this evidence was harmless beyond a reasonable doubt. We conclude it was not. If the jury had heard the evidence that had an inherent tendency to connect B.E. to the actual commission of the murder, along with the evidence of motive and threats, and the

7 Minn. R. Evid. 412 limits the admissibility of evidence of a victim’s previous sexual conduct when the defendant is being prosecuted for an act of criminal sexual conduct.

19

damaging potential of the evidence had been fully realized, we cannot say beyond a reasonable doubt that a reasonable jury would have reached the same verdict. See Smith, 876 N.W.2d at 333. This is not inconsistent with our decision in State v. Westrom, ___ N.W.3d ___, No. A22-1679 (Minn. 2024). There, we concluded that any error in excluding alternative-perpetrator evidence was harmless beyond a reasonable doubt because “there was no evidence that any of [the] alleged alternative perpetrators stood barefoot in [the victim]’s blood when she died, as was the case for Westrom.” Id. at ___, slip op. at 13. By contrast, the evidence here was not so fundamentally incompatible with Carbo’s alternative-perpetrator theory: B.E. could in fact have killed Daugherty after Carbo had sex with her and left. 8

* * *

The question presented here is not whether the damning forensic evidence presented by the State was sufficient to convict Carbo of Daugherty’s murder. Nor is it whether the proffer presented by Carbo conclusively established B.E.’s participation in the crime. Rather, we must decide whether the district court abused its discretion by holding Carbo’s evidentiary proffer to an unobtainable legal standard in violation of his constitutional right to present a complete defense. We conclude that it did, that the evidence was otherwise admissible, and that the error was not harmless. Consequently, we reverse the judgment

8 Justice McKeig’s dissent is skeptical of Carbo’s theory of the case, and perhaps properly so. But the issue here is whether Carbo received a fair trial in the first instance, not whether his defense theory was more probable than not. Carbo need only show, as he has done here, that a reasonable jury could have reached a verdict more favorable to him had it heard this evidence.

20

of conviction and remand to the district court for further proceedings consistent with this opinion.

CONCLUSION

For the foregoing reasons, we reverse the judgment of conviction and remand to the district court for further proceedings consistent with this opinion.

Reversed and remanded.

CONCURRENCE

PROCACCINI, Justice (concurring).

I join the decision to reverse Carbo’s conviction on the grounds that he was improperly precluded from presenting alternative-perpetrator evidence. I also agree that the district court did not err by admitting evidence stemming from the genetic analysis of Carbo’s garbage. I therefore join Parts II and III of the majority opinion. I write separately because I respectfully disagree with the majority’s analysis in Part I. I would instead conclude that the analysis of Carbo’s sensitive and personal information revealed by a comprehensive DNA profile required a warrant but that the evidence stemming from the analysis was nevertheless admissible under the good-faith exception to the exclusionary rule.

In the decades since DNA analysis became a commonplace tool in forensic investigation, its ability to reveal deeply sensitive and personal information has increased dramatically. Our phenotypical makeup, which encompasses many aspects of our observable appearance, can be read from no more than a stray hair. The color of our skin, our likelihood of developing diseases like diabetes, and even the identity of our biological ancestors stretching back centuries is available to anyone with access to the equipment and expertise necessary to unlock our genetic code. We can and should use these new technological capabilities to further the aims of justice by solving cold cases and exonerating the wrongly convicted. At the same time, this court must also continue to uphold Minnesotans’ privacy rights as technology evolves. Accordingly, I would conclude that there is a reasonable expectation of privacy in our most sensitive and personal genetic

C-1

information and that Carbo did not abandon that expectation automatically when he left behind physical evidence that contained his genetic code.

I also recognize that technological leaps like those involved here present challenges to law enforcement as they work in good faith toward justice for crime victims and seek to apply established law to novel circumstances. I would therefore hold that the good-faith efforts of law enforcement do not call for the exclusion of the DNA evidence in this case.

A.

The protections against warrantless searches found in the United States and Minnesota Constitutions are rooted in individuals’ expectations of privacy. Carpenter v. United States, 585 U.S. 296, 303 (2018); State v. Leonard, 943 N.W.2d 149, 156 (Minn. 2020).

Confronted with scientific advances, we should reject a “mechanical interpretation”

of these individual privacy rights. Kyllo v. United States, 533 U.S. 27, 35 (2001). Instead, we should ensure that Minnesotans are not left “at the mercy of advancing technology” that allows law enforcement to intrude into private affairs in ways previously unimaginable. Id. This court has repeatedly recommitted itself to protecting Minnesotans’ constitutional rights—including their right to be free from unreasonable searches—because “a free society will not remain free if police may use . . . crime detection device[s] at random and without reason.” Leonard, 943 N.W.2d at 155 (quoting State v. Carter, 697 N.W.2d 199, 211 (Minn. 2005)) (internal quotation marks omitted); see also Carpenter, 585 U.S. at 305 (“[A] central aim of the Framers was to place obstacles in the way of a too permeating police surveillance.” (citation omitted) (internal quotation marks omitted)). Accordingly,

C-2

“[w]hen law enforcement seeks to conduct a search to uncover evidence of criminal wrongdoing, reasonableness typically requires law enforcement to obtain a judicial warrant before conducting the search.” State v. Thompson, 886 N.W.2d 224, 228 (Minn. 2016).

Because the novel DNA analysis at issue in this case reveals deeply sensitive and personal information about the subject, I would conclude that Minnesotans have a reasonable expectation of privacy in that information and that the United States and Minnesota Constitutions therefore require law enforcement to obtain a warrant before conducting such an analysis.

The DNA profile analyzed in this case revealed significantly more information than the DNA profiles previously considered by courts. Forensic investigators have traditionally relied on short tandem repeat (“STR”) DNA profiles to compare crime scene DNA against a database that catalogues the DNA profiles of convicted offenders at the local, state, and national level. Erin Murphy, Law and Policy Oversight of Familial Searches in Recreational Genealogy Databases, 292 Forensic Sci. Int’l. e5, e5–e6 (2018). STR profiles focus on up to 20 segments of so-called “junk” DNA. Id. These DNA segments reveal nothing about a person’s physical or ancestral traits, but they differ widely among individuals, allowing for easy identification. Id. A decade ago, in Maryland v. King, the United States Supreme Court ruled that collecting and analyzing arrestees’ STR profiles for identification purposes and without a warrant did not violate the Fourth Amendment. 569 U.S. 435, 465 (2013). It analogized an STR profile to a fingerprint, concluding that an STR profile was merely “another metric of identification used to connect the arrestee with his or her public persona.” Id. at 451.

C-3

Forensic DNA technology has come a long way since King. The investigators in this case relied on a new analytical method that extracts much more information from a sample to produce a single nucleotide polymorphism profile, otherwise known as a “SNP” (pronounced “snip”) profile. Murphy, Law and Policy Oversight, supra, at e5–e6. SNP profiles contain several hundred thousand informationally rich segments of DNA, which can be decoded to predict distant genealogical relationships, disease carrier status, and aspects of physical appearance. Id. Indeed, the report delivered to law enforcement in this case used Carbo’s SNP profile to describe his physical traits (including skin, eye, and hair color), as well as to predict his family tree and ancestral origins dating to the late eighteenth century and stretching across continents. The report also contained a “phenotype prediction” that provided a digitally constructed portrait of Carbo’s appearance at age 25 based on the genetic characteristics in his SNP profile.

The scope of personal information that can now be unlocked using modern genetic analytical methods eclipses past capabilities and case law. In 1986, when police officers first began investigating Nancy Daugherty’s murder using an STR profile, the Human Genome Project had not yet begun. See Robert Kanigel, The Genome Project, N.Y. Times Mag., Dec. 13, 1987, at 44 (stating that “today, only about 500 human genes (less than 1 percent of the total) have been decoded” and claiming that the forthcoming project would “grant insight into human biology previously held only by God”). Yet by 2019, when law enforcement sought to analyze a SNP profile from the Daugherty cold case, the Human Genome Project had been completed for 16 years, and the human genome was almost entirely sequenced. Howard A. Zucker & Richard Thomas, A Truly Modern Prometheus:

C-4

Law and Policy at the Brink of a Genetic Revolution, 31 Alb. L.J. Sci. & Tech. 1, 6 (2021). DNA testing had become so widespread that around 60 percent of Americans of Northern European descent could be identified through commercial genealogical databases, whether or not they had participated directly in DNA testing. Yaniv Erlich, Tal Shor, Itsik Pe’er & Shai Carmi, Identity Inference of Genomic Data Using Long-Range Familial Searches, 362 Science 690, 690 (2018). 1 “[T]he Fourth Amendment protects people, not places.” Katz v. United States, 389 U.S. 347, 351 (1967). The sensitive genetic information revealed by a SNP analysis is quintessentially personal, and it is comparable to other categories of information in which courts have found a reasonable expectation of privacy. As technology has advanced, courts have recognized privacy interests in the contents of our cellphones, 2 the physical movements of our daily lives, 3 and the medical information that can be gleaned from our

1 Northern Europeans are statistically overrepresented in commercial genetic databases when compared to people of other ancestries, an inversion of the representation seen in law enforcement DNA databases. Id. The predictive power of this statistic would augur Carbo’s eventual identification through just such a commercial database—as a person of Northern European ancestry who had never uploaded his genetic information to a consumer genealogy service. 2 See Riley v. California, 573 U.S. 373, 396–97 (2014) (recognizing that cellphones include “many sensitive records previously found in the home” and a “broad array of private information”); State v. Barajas, 817 N.W.2d 204, 216 (Minn. App. 2012) (noting that cellphones are “capable of storing substantial amounts of private data, including address books and photographs”). 3 See Carpenter, 585 U.S. at 311 (recognizing that time-stamped location data can reveal not only one’s “particular movements,” but also one’s “familial, political, professional, religious, and sexual associations” (quoting United States v. Jones, 565 U.S. 400, 415 (2012) (Sotomayor, J., concurring)) (internal quotation marks omitted)); State v.

C-5

urine. 4 The private and personal information revealed by a SNP analysis is arguably more private and “previously unknowable” than the information in the cases cited above. Some of the information gleaned from SNP profiles may not be known even by the subjects themselves. While an individual certainly may know their eye, skin, and hair color, they are unlikely to be aware of the precise genetic markers embedded in their DNA that account for those characteristics.

A SNP analysis not only reveals sensitive and personal information about its individual subject—it also reveals similarly sensitive and personal information about others. For example, in conducting their genealogical research based on Carbo’s SNP profile, investigators suspected that a member of Carbo’s extended genetic family was fathered by someone other than that member’s father of record, who was outside of Carbo’s family tree. To support their hypothesis, the investigators researched the individual and found that they were part of a social media group that assists people with identifying unknown parents. While Carbo’s family member may not yet have known the identity of their biological father, the investigators were able to identify the father’s likely identity by analyzing the genetic relationships between Carbo’s other relatives. Clearly, the previously unknown identity of one’s biological parents is “deeply private” information with the

Harvey, 932 N.W.2d 792, 805 (Minn. 2019) (recognizing that cell tower location data is “subject to the Fourth Amendment’s warrant requirement”). 4 See Skinner v. Ry. Lab. Execs.’ Ass’n, 489 U.S. 602, 617 (1989) (recognizing that “chemical analysis of urine, like that of blood, can reveal a host of private medical facts about an employee, including whether he or she is epileptic, pregnant, or diabetic”); Thompson, 886 N.W.2d at 231 (concluding that urine samples “raise[] the same privacy concerns” as blood samples due to the wealth of information they contain).

C-6

potential to disrupt familial relationships if handled improperly. Leonard, 943 N.W.2d at 157 (noting that allowing sensitive information about personal associations or medical diagnoses to “be accessible to the government through a fishing expedition . . . offends our core constitutional principles”).

We have also ruled that the use of specialized technology to extract information from an individual’s home that would “previously have been unknowable without physical intrusion” implicates a protected privacy interest. Carter, 697 N.W.2d at 208 (quoting Kyllo, 533 U.S. at 34). Constitutional protection therefore extends to information that, before the digital age, “law enforcement agents and others would not—and indeed, in the main, simply could not—secretly monitor and catalogue.” Carpenter, 585 U.S. at 310 (quoting United States v. Jones, 565 U.S. 400, 430 (2012) (Alito, J., concurring)) (internal quotation marks omitted). With DNA, the investigatory intrusion goes deeper than just collecting a piece of one’s skin or hair, and it reveals much more than could be gleaned by a simple physical intrusion. Although collecting a DNA sample may be relatively straightforward, obtaining information about ancestry and familial relationships can be achieved only by sending the sample to a sophisticated lab for an analysis that unlocks much of a person’s genetic code.

To ensure adequate constitutional protection, we must consider all the information available to law enforcement from the search, not just the subset of information useful to the investigation. See Birchfield v. North Dakota, 579 U.S. 438, 464 (2016) (finding an enhanced expectation of privacy in a blood sample, which “can be preserved and from which it is possible to extract information beyond a simple [blood alcohol content]

C-7

reading”). Accordingly, the private nature of the information revealed by a SNP analysis, the potential harm that can be wrought by its misuse, and the historical difficulty in accessing this kind of information counsel us to recognize that society would find an expectation of privacy reasonable here. Or, as we put it in Leonard, “most Minnesotans would be surprised and alarmed” to learn that the deeply sensitive and personal information contained in our SNP profiles is accessible to law enforcement without a warrant. 943 N.W.2d at 158.

Moreover, requiring police officers to obtain a warrant before analyzing SNP profiles would not, as a practical matter, hinder their ability to investigate crimes. The power of modern genetics—to solve unsolved crimes and also to exonerate the wrongly convicted—should be neither shunned nor feared. But this power necessitates that we establish safeguards to preserve individual privacy rights. In investigations for serious crimes, such as this one, there will assuredly be sufficient probable cause to justify the SNP analysis of crime scene DNA, and the State’s obligation “is accordingly simple—get a warrant.” 5 Riley v. California, 573 U.S. 373, 403 (2014).

5 To perhaps state the obvious, while it may often be appropriate for crime scene DNA to be subject to a SNP analysis with a warrant, individuals whose DNA is found at a crime scene do not lose the presumption of innocence guaranteed to all criminal defendants. See State v. Portillo, 998 N.W.2d 242, 251 (Minn. 2023) (“Our court has made clear that the presumption of innocence is a . . . bedrock axiomatic and elementary principle whose enforcement lies at the foundation of the administration of our criminal law.” (citations omitted) (internal quotation marks omitted)). There are many reasons why an individual’s DNA may be found at a crime scene, as made clear by the ease with which our DNA is left behind wherever we go.

C-8

A warrant requirement here would result in a process akin to the constitutional treatment of cellphones. Id. (requiring a warrant to search a cellphone’s digital contents even after a lawful post-arrest seizure). Given the significant time needed to compile a SNP profile and complete the subsequent genealogical investigation, requiring a warrant does not pose significant exigency concerns. A warrant requirement would allow for appropriate judicial oversight of the deeply sensitive information uncovered by SNP profiles and also give defendants a remedy should officers exceed the scope of the warrant or misuse the information gleaned from the profile—all without sacrificing the investigatory utility of DNA.

B.

In addition to finding a reasonable expectation of privacy in the information revealed by a SNP analysis, I would conclude that Carbo did not abandon his subjective expectation of privacy in this information. To be clear, I agree that Carbo abandoned his expectation of privacy in the physical evidence he left behind at Daugherty’s house and that we do not recognize a reasonable expectation of privacy in the identifying information that can be drawn from this evidence using standard forensic procedures. State v. Westrom, ___ N.W.3d ___, No. A22-1679, slip op. at 10 (Minn. 2024). But Carbo’s abandonment of this physical evidence did not extend to his abandonment of his expectation of privacy in the deeply sensitive and personal information subsequently revealed by the analysis of his SNP profile. Concluding otherwise is inconsistent with the approach to abandonment taken by this court and the United States Supreme Court.

C-9

In cases where an attempt to conceal information would be futile, the doctrine of abandonment—which turns on whether an individual’s conduct indicates an intent to conceal—is an ill fit. See City of St. Paul v. Vaughn, 237 N.W.2d 365, 370–71 (Minn. 1975). When we venture into public spaces, we simply cannot help but shed our DNA. As amicus curiae American Civil Liberties Union of Minnesota explained in its brief:

Wherever we go, we constantly shed staggering numbers of skin cells, which include our DNA. The average person loses between 40 and 100 hairs per day, a single sneeze spews about 3,000 cell-containing droplets into the world, and merely touching a surface with one’s fingertip causes DNA to be deposited there. Indeed, a person involuntarily sheds roughly 50 million DNA-containing cells each day. With every discarded coffee cup, crumpled tissue, plastic straw, cigarette butt, soda can, piece of gum, and drifting flake of dandruff, people unavoidably and involuntarily leave a copy—and often many thousands of copies—of their genetic blueprint.

Given the virtual impossibility of preventing our deeply personal genetic information from being deposited in public places, “[t]he volition that is implied in abandonment is simply unrealistic” in this context. 6 Elizabeth E. Joh, Reclaiming “Abandoned” DNA: The Fourth Amendment and Genetic Privacy, 100 Nw. U. L. Rev. 857, 867 (2006).

6 This conclusion would not change how we analyze abandonment in other cases. For example, although a cellphone also contains deeply personal information, it is capable of being concealed, and its contents are accordingly susceptible to abandonment when it is voluntarily discarded. Recognizing the unique volitional problems posed by DNA does not require an adjustment to our general abandonment principles. See Vaughn, 237 N.W.2d at 370 (“The issue is not abandonment in the strict property-right sense, but whether the person prejudiced by the search had voluntarily discarded, left behind, or otherwise relinquished his [privacy] interest in the property in question . . . .” (emphasis added) (quoting United States v. Colbert, 474 F.2d 174, 176 (5th Cir. 1973)) (internal quotation marks omitted)).

C-10

In determining individual rights to privacy under the Minnesota Constitution we have previously rejected mechanistic applications of the abandonment doctrine. In Leonard, we concluded that “sharing private information in [certain] spaces does not destroy someone’s reasonable expectation of privacy. . . .” 943 N.W.2d at 159. And the United States Supreme Court has reasoned similarly that the public disclosure of sensitive location information to third-party cell service providers does not defeat a subjective privacy interest in the information because such disclosure is virtually necessary to participate in modern life. Carpenter, 585 U.S. at 310–11 (noting that a cellphone is “almost a ‘feature of human anatomy’ ” and that “[a] person does not surrender all Fourth Amendment protection by venturing into the public sphere”). The case against abandonment is all the more clear when this same logic is applied to DNA. As ubiquitous and essential as cellphones have become in our daily lives, we do still have the ability to turn them off or leave them at home. We have no such options with our DNA and the troves of information it contains. See Iowa v. Burns, 988 N.W.2d 352, 394 (Iowa 2023) (McDermott, J., dissenting) (“A human can turn off a cell phone; a human cannot turn off one’s own automatic, continuous life cycle of creating and casting off DNA.”).

The assertion that the abandonment of all privacy interest in one’s genetic information can be cabined to Carbo’s specific behavior in this case (implying that the result may have been different if Carbo’s DNA was gathered from a less conspicuous source) provides cold comfort. The problem is that the availability of Carbo’s genetic information to law enforcement, as a general matter and regardless of his specific conduct, was out of his control—just as it is for each of us in our daily lives.

C-11

The majority’s conclusion that the total abandonment of one’s expectation of privacy in this information is possible at all—absent perhaps an affirmative waiver—opens the door to the indiscriminate collection of deeply sensitive and personal genetic information from the general public. In State v. McMurray, we concluded that individuals abandon their privacy interest in their garbage when they set it out for collection. 860 N.W.2d 686, 695 (Minn. 2015) (“Any member of the public could have accessed McMurray’s garbage without trespassing on his property, and police do not need a warrant to search items that are exposed to the public.”). If abandonment of one’s privacy interest in physical evidence equates to abandonment of one’s privacy interest in the genetic information contained within that evidence, then McMurray allows police to trawl curbside garbage to genetically profile entire communities. Id. at 702 (Lillehaug, J., dissenting) (“If Minnesotans’ household privacy ends at their sidewalks, there would be nothing unconstitutional about a government program to seize and perform a forensic inspection of the waste from every Minnesota household, all without the barest suspicion.”). Combined with the leeway given to law enforcement to catalogue genetic information, this could create a “backdoor to population-wide data banking.” Joh, Reclaiming “Abandoned” DNA, supra, at 874. So long as the source DNA is collected from an object that lacks independent constitutional protection, the majority holds that police officers may use it in any way they see fit, with no judicial oversight.

* * *

For the above reasons, I would conclude that the analysis of the sensitive and personal information revealed by a SNP profile is a search requiring a warrant under both

C-12

the United States and Minnesota Constitutions. Because the United States Supreme Court has yet to determine for itself whether a SNP analysis is a search under the Fourth Amendment, it is particularly important to consider the scope of the protections provided by Article I, Section 10, of the Minnesota Constitution. Although the language is similar, we have repeatedly held that the Minnesota Constitution provides stronger privacy protections than the United States Constitution. See Leonard, 943 N.W.2d at 159 (protection from unreasonable hotel registry searches); In re Welfare of B.R.K., 658 N.W.2d 565, 578 (Minn. 2003) (protecting the reasonable expectation of privacy of short-term social guests); Carter, 697 N.W.2d at 211 (protection from unreasonable dog sniffs of storage units); State v. Larsen, 650 N.W.2d 144, 154 (Minn. 2002) (protection from warrantless searches of fish houses); Ascher v. Comm’r of Pub. Safety, 519 N.W.2d 183, 187 (Minn. 1994) (protection from suspicionless seizures at DUI checkpoints); State v. Askerooth, 681 N.W.2d 353, 371 (Minn. 2004) (protection from unreasonable expansion of scope and extension of duration of seizures).

When interpreting the Minnesota Constitution, this court has an “obligation to independently safeguard the rights of our citizens.” Women of State of Minn. by Doe v. Gomez, 542 N.W.2d 17, 31 (Minn. 1995). Under the Minnesota Constitution, “a search occurs when law enforcement intrudes upon an individual’s subjective expectation of privacy that society is prepared to recognize as reasonable.” Leonard, 943 N.W.2d at 156. For nearly two decades, Minnesotans have made clear that our society is prepared to recognize a reasonable expectation of privacy in sensitive genetic information. Since 2006, Minnesota law has safeguarded Minnesotans’ genetic information, requiring an

C-13

individual’s written informed consent before that information is disclosed by the government. Minn. Stat. § 13.386 (2022) (enacted in 2006 as the “Genetic Privacy Act”). We have upheld this policy and further interpreted it to cover the collection of blood samples that contain our genetic information. Bearder v. State, 806 N.W.2d 766, 774 (Minn. 2011).

More recently, the Legislature extended protections of Minnesotans’ privacy interests by restricting the use and dissemination of genetic information by commercial genealogical companies. Minn. Stat. § 325F.995, as amended by, Act of May 23, 2023, ch. 57, art. 4, § 18 (the “Genetic Information Privacy Act”). This law not only requires companies to obtain consumers’ consent to the collection and testing of their genetic information, but it also prohibits companies from disclosing genetic information to law enforcement without first obtaining express written consent, a search warrant, or a court order. See id., subd. 2(a)(3). Minnesotans—acting through their elected representatives— have repeatedly affirmed their belief that they have a reasonable expectation of privacy in the sensitive personal information contained in their DNA. A faithful interpretation of the Minnesota Constitution would not allow law enforcement to have unimpeded access to this information.

For these reasons, I would hold that the Minnesota Constitution protects Minnesotans from warrantless searches of the sensitive and personal genetic information revealed by their SNP profiles, independent of any protection also provided by the Fourth Amendment.

C-14

C.

Although I would conclude that the search of Carbo’s genetic information revealed by a SNP analysis required a warrant, I would nonetheless hold that evidence stemming from the search was admissible under the good-faith exception to the exclusionary rule.

The exclusionary rule is a prudential doctrine that compels respect for the guaranty against unreasonable search and seizure. See State v. Lindquist, 869 N.W.2d 863, 868 (Minn. 2015) (citing Davis v. United States, 564 U.S. 229, 236 (2011)). The rule, recognized in Weeks v. United States, 232 U.S. 383, 398 (1914), and applied to the states through the Fourteenth Amendment in Mapp v. Ohio, 367 U.S. 643, 655 (1961), 7 generally requires the suppression of evidence obtained in violation of the Fourth Amendment. Weeks, 232 U.S. at 392; see also State v. Horst, 880 N.W.2d 24, 36 (Minn. 2016) (“In a criminal case, the remedy for an illegal search or seizure is generally limited to the suppression of illegally obtained evidence.”). Although exclusion was initially available

7 Before Mapp, illegally obtained evidence was admissible in Minnesota. See Elkins v. United States, 364 U.S. 206, 224, 227 (1960) (appending a 50-state survey indicating that illegally obtained evidence remained admissible in Minnesota courts, despite the exclusionary rule’s adoption in other states); State v. Pluth, 195 N.W. 789, 792–93 (Minn. 1923) (discussing state precedent on admissibility of evidence obtained pursuant to unlawful search or seizure); State v. Strait, 102 N.W. 913, 914 (Minn. 1905) (declining to entertain an objection to the admission of evidence illegally obtained when the evidence is “pertinent to the issue”). Two years after Mapp, the Minnesota Legislature enacted a provision that mirrors the federal exclusionary rule. Minn. Stat. § 626.21 (1963) (“A person aggrieved by an unlawful search and seizure may move the district court . . . to suppress the use, as evidence, of anything so obtained . . . . If the motion is granted . . . it shall not be admissible in evidence at any hearing or trial.”). Relevant to the analysis in this concurrence, this court has determined that Minnesota Statutes section 626.21 does not prevent recognition of exceptions to the exclusionary rule. See Lindquist, 869 N.W.2d at 875–76.

C-15

as a remedy for any Fourth Amendment violation, see Mapp, 367 U.S. at 655, the United States Supreme Court has limited the remedy’s application to circumstances that most effectively serve its goal of deterring unlawful police conduct. See Arizona v. Evans, 514 U.S. 1, 14 (1995) (declining to suppress evidence obtained pursuant to a mistake of court administration, as “the exclusionary rule was historically designed as a means of deterring police misconduct”); see also Lindquist, 869 N.W.2d at 871 (declining to exclude evidence when suppression would not serve the rule’s central purpose of deterring police misconduct).

Under the good-faith exception, the exclusionary rule does not apply unless suppression “result[s] in appreciable deterrence.” United States v. Leon, 468 U.S. 897, 909 (1984) (citation omitted) (internal quotation marks omitted). The United States Supreme Court first adopted the exception in Leon, declining to apply the exclusionary rule where police obtain evidence in reasonable reliance on a facially valid warrant, later determined to lack probable cause. Id. at 913. Following Leon, the Court has extended the exception to other cases in which exclusion would not deter unlawful police conduct. See Illinois v. Krull, 480 U.S. 340, 352 (1987) (declining to exclude evidence obtained in reasonable reliance on statutes later deemed unconstitutional); Herring v. United States, 555 U.S. 135, 147–48 (2009) (declining to exclude evidence obtained in reasonable reliance on an arrest warrant database managed by police); Evans, 514 U.S. at 14 (declining to exclude evidence obtained in reasonable reliance on an arrest warrant database managed by the judiciary).

We recognized the good-faith exception in Lindquist, holding that it applies when law enforcement officers act in reasonable reliance on binding appellate precedent that

C-16

specifically authorizes the conduct at issue. 869 N.W.2d at 876. In Lindquist, the police arrested Lindquist for driving while impaired and performed a warrantless blood draw. Id. at 865. At the time of the arrest, our decisions in State v. Shriner, 751 N.W.2d 538 (Minn. 2008), and State v. Netland, 762 N.W.2d 202 (Minn. 2009), authorized the warrantless blood draw as a single-factor exigency. While Lindquist’s appeal was pending, however, the United States Supreme Court decided Missouri v. McNeely, 569 U.S. 141 (2013), which overturned the single-factor exigency analysis in Shriner and Netland. Lindquist, 869 N.W.2d at 866.

In Lindquist, we decided to adopt the good-faith exception to the exclusionary rule set out in Davis v. United States, 564 U.S. 229 (2011). In Davis, the United States Supreme Court concluded that the exclusionary rule does not apply to evidence obtained in reasonable reliance on binding appellate precedent, reasoning that there is no police misconduct to deter in those circumstances, as any error rests with judges. Id. at 241.

Faced with an analogous issue in Lindquist, we reviewed our precedent, concluded that “deterrence of police misconduct” is “the central purpose of the exclusionary rule” under Minnesota law, and adopted the good-faith exception articulated in Davis. Lindquist, 869 N.W.2d at 871, 876. In doing so, we noted that our holding was narrow and “should [not] be construed as authorizing the application of exceptions we have not explicitly adopted.” Id. at 876; see also State v. Malecha, 3 N.W.3d 566, 579 (Minn. 2024) (declining to apply the good-faith exception to evidence obtained due to a clerical error by court administration).

C-17

Based on the deterrence rationale articulated in Lindquist, I would conclude that evidence from the analysis of Carbo’s DNA profile was properly admitted, as an existing state statute and extant appellate decisions permitted the challenged search.

To start, Minnesota Statutes section 299C.155 authorizes the Minnesota Bureau of Criminal Apprehension (“BCA”) to “perform DNA analysis and make data obtained available to law enforcement officials in connection with criminal investigations in which human biological specimens have been recovered.” Minn. Stat. § 299C.155, subd. 4 (2022). Because this section does not distinguish between different types of permissible “DNA testing,” officers had no reason to doubt the legality of more advanced protocols.

Moreover, our own precedent has generically endorsed “DNA testing” for identification purposes pursuant to section 299C.155. Minnesota Statutes section 609.117 (2022) incorporates the definition of “DNA analysis,” and the testing procedures and protocols, set out in section 299C.155. See State v. Johnson, 813 N.W.2d 1, 4–5 (Minn. 2012). We have repeatedly upheld the constitutionality of section 609.117, and our case law has not distinguished between STR and SNP analyses.

We first addressed the constitutionality of section 609.117 in State v. Bartylla.

755 N.W.2d 8 (Minn. 2008). There, the issue was the state’s collection of DNA from a person convicted of a felony. Id. at 11. Bartylla was convicted of first-degree murder in a 3-year-old cold case after police collected his DNA and matched it to a sample from the

C-18

original investigation. Id. at 12. 8 We reasoned that because Bartylla was incarcerated, he had a diminished expectation of privacy. Id. at 17. In contrast, the State had a strong interest in solving the cold case. Id. at 18. We therefore held that the DNA collection authorized by section 609.117 did not violate the prohibition against unreasonable searches and seizures. Id. at 18–19.

In Johnson, we again addressed the constitutionality of section 609.117, this time as applied to misdemeanants. 813 N.W.2d at 3. We characterized the issue before us as “whether the collection of a biological specimen to develop a DNA profile for criminal identification purposes is a reasonable search.” Id. at 4 n.2. We also emphasized, in response to the dissent’s concern that a DNA search reveals substantially more about an individual than identity, that section 299C.155 “limits the use of the DNA profile to criminal identification.” Id. at 7 n.4. As in Bartylla, we concluded that an individual convicted of a misdemeanor and subject to probationary conditions has a reduced

8 Although the opinion does not specify the type of DNA analysis performed, the recitation of facts indicates that it was STR testing. To obtain the match, DNA typing was performed on biological specimens left at the crime scene and entered into the BCA-run, state-mandated database. Bartylla, 755 N.W.2d at 12. The BCA ran that profile against those of known contributors. Id. A DNA analyst later testified that “all 13 of the DNA loci tested plus the sex chromosomes from the DNA sample” at the crime scene matched Bartylla. Id. Although not specified, the reference to the 13 loci, or segments of DNA, indicates that the analyst conducted an STR test on the biological specimen. See Karen Norrgard, Forensics, DNA Fingerprinting, and CODIS, 1 Nature Educ. 35 (2008), https://www.nature.com/scitable/topicpage/forensics-dna-fingerprinting-and-codis-736/ (discussing the widespread use in the United States of a 13-loci STR profile to identify and match suspects through crime scene samples).

Modern STR testing analyzes a profile of up to 20 DNA loci. Murphy, Law and Policy Oversight, supra, at e5–e6. SNP testing, by comparison, analyzes hundreds of thousands of DNA segments. Id.

C-19

expectation of privacy in their identity, and the State’s interest in collecting DNA for identification purposes outweighs that limited privacy interest. Id. at 11. We again held that the statute, as applied, did not violate the search-and-seizure provisions of the United States and Minnesota Constitutions. Id.; see also In re Welfare of M.L.M., 813 N.W.2d 26, 38 (Minn. 2012) (upholding the constitutionality of DNA collection authorized by section 609.117 as applied to juveniles adjudicated delinquent).

The United States Supreme Court has also upheld warrantless DNA analysis without distinguishing between STR and SNP analyses. In Maryland v. King, the Court held that taking and analyzing an arrestee’s DNA sample for identification purposes is reasonable under the Fourth Amendment. 569 U.S. at 465–66. There, the defendant was arrested for assault. Id. at 441. During the booking process, law enforcement used a cheek swab to obtain a DNA sample pursuant to state statute. Id. Although the Court acknowledged that DNA can be used for purposes beyond identification, including purposes that would raise greater constitutional concerns like determining “an arrestee’s predisposition for a particular disease or other hereditary factors not relevant to identity,” those purposes were not implicated in that case. Id. at 464–65. For this reason, the challenged DNA analysis was reasonable insofar as it revealed only information relevant to identification. Id.

Like the law enforcement conduct at issue in Lindquist, the investigating officers here complied with both statute and precedent in collecting and analyzing Carbo’s biological sample and seeking to identify a perpetrator. Until now, our case law, and that of the United States Supreme Court, did not distinguish between older STR “junk” DNA testing and newer SNP testing. Here, law enforcement had no reason to believe that they

C-20

needed a warrant to conduct a SNP test. Instead, the police followed the law in good faith, and exclusion would deter no misconduct. “We have refused to suppress evidence in circumstances in which the police have acted in good-faith reliance on a judicial determination or when suppression would not deter police misconduct,” Lindquist, 869 N.W.2d at 871, and we should continue that approach here.

* * *

In a situation such as this, when technological developments leap ahead of existing jurisprudence, the good-faith exception should apply. As a result, although I would hold that the analysis of the sensitive and personal information contained in a SNP profile is a search that requires a warrant, I would conclude that the evidence obtained pursuant to the warrantless search here is admissible under the good-faith exception.

C-21

CONCURRENCE & DISSENT

THISSEN, Justice (concurring in part, dissenting in part).

I agree with the court that Michael Allan Carbo’s conviction must be reversed because the district court improperly excluded alternative-perpetrator evidence. Therefore, I join Part III of the majority opinion.

I dissent from the court’s holding in Part I of its opinion and would hold instead that the evidence stemming from the search of Carbo’s genetic material left in Daugherty’s home should have been suppressed. For the reasons that Justice Procaccini sets forth in Parts A and B of his concurrence, I conclude that Carbo had a reasonable expectation of privacy in the information revealed by a SNP analysis and that Carbo did not abandon his subjective expectation of privacy in his genetic information. I do not join Part C of Justice Procaccini’s concurrence, however. We have no controlling appellate precedent telling police officers that it is permissible to generate a SNP profile without a search warrant. State v. Lindquist, 869 N.W.2d 863, 868 (Minn. 2015), does not control and I would not expand the application of the good-faith exception to Article I, Section 10, of the Minnesota Constitution in this case.

Because I conclude that the evidence stemming from the search of Carbo’s genetic material left at Daugherty’s home should have been suppressed, I need not reach the questions addressed in Part II of the court’s opinion. That said, I do not disagree with the court’s analysis.

C/D-1

CONCURRENCE & DISSENT

MCKEIG, Justice (concurring in part, dissenting in part).

I concur with Parts I and II of the majority opinion, concluding that a defendant abandons an expectation of privacy in genetic information when his semen is left at the scene of a crime, and that the independent source doctrine precluded suppression of information from a genetic analysis of defendant’s garbage. However, I dissent as to Part III and the way the court interprets Minnesota’s test for whether to admit alternative- perpetrator evidence at trial. The court reverses the conviction of a defendant who was found guilty of first-degree murder while committing or attempting to commit criminal sexual conduct. Minn. Stat. § 609.185(a)(2) (2022). The court’s reversal here not only ignores the deference a reviewing court generally gives to a district court’s evidentiary review, it essentially abrogates the alternative-perpetrator test used by Minnesota courts since 1977. Further, even if the district court did err by not admitting the alternative- perpetrator evidence, the facts of this case are similar to those in a recent alternative- perpetrator decision by this court which found such error to be harmless. Because I disagree with how the court interprets its own case law when applying it to the facts of this case, and because today’s decision is arguably inconsistent with another recent alternative- perpetrator decision issued by this court, I dissent.

A.

This court has long stated that it “ ‘largely defer[s] to the trial court’s exercise of discretion in evidentiary matters and will not lightly overturn a trial court’s evidentiary ruling.’ ” State v. Robertson, 884 N.W.2d 864, 872 (Minn. 2016) (quoting State v. Kelly,

C/D-1

435 N.W.2d 807, 813 (Minn. 1989)). Stated plainly, absent a clear abuse of discretion, this court will not reverse a district court’s evidentiary ruling. Dolo v. State, 942 N.W.2d 357, 362 (Minn. 2020). Here, the trial court made an evidentiary ruling to which the court gives no such deference.

Defendant Michael Allan Carbo moved to admit evidence that a third person, B.E., was an alternative perpetrator in a murder committed over 35 years ago. In its order responding to the motion, the district court analyzed each piece of Carbo’s proffered evidence in considerable detail, finding “at best, a possible motive” for the killing, but finding no evidence that had an inherent tendency to connect B.E. to the commission of the crime. Throughout its order, the district court appropriately utilized this court’s previous alternative-perpetrator case law at length, including recent cases State v. Woodard, 942 N.W.2d 137, 141 (Minn. 2020), and Troxel v. State, 875 N.W.2d 302, 307 (Minn. 2016). The district court ultimately denied Carbo’s motion because his proffer did not satisfy the threshold step of the two-step Hawkins test. See State v. Hawkins, 260 N.W.2d 150, 159 (Minn. 1977) (holding that evidence of a third person’s motive, threats, or other facts that tend to prove the third person committed the crime “is inadmissible in the absence of proof of facts to connect that person with the crime.”).

Under Hawkins, a defendant must first lay a proper foundation with evidence that has “an inherent tendency to connect [a third] person with the actual commission of the crime.” Id. Only then is it “permissible to introduce evidence of a motive of the third person to commit the crime, threats by the third person, or other miscellaneous facts which would tend to prove the third person committed the act.” Id. The Hawkins court explained

C/D-2

that “[t]his requirement avoids the use of bare suspicion and safeguards the third person from indiscriminate use of past differences with the deceased.” Id. Put differently, the Hawkins test prevents murder trials from becoming spectacles of finger pointing, resulting in an alternative-perpetrator parade populated by anyone with whom the victim may have had beef or prior intimate contact.

“In considering whether the defendant has established the required foundation, a court must focus on ‘the evidence, not the assertions, contained in the proffer.’ ” Woodard, 942 N.W.2d at 142 (quoting State v. Nissalke, 801 N.W.2d 82, 102 (Minn. 2011)). The court has determined that Carbo “clearly” proffered sufficient evidence, that, when viewed as a whole, has “an inherent tendency to connect [B.E.] with the actual commission of the crime.” But here the court conflates assertions with actual evidence of facts that tend to connect B.E. to the commission of the crime when finding that Carbo had established the necessary foundation.

“Mere speculation is insufficient” and cannot serve as foundational evidence, yet the facts the court found to be evidence connecting B.E. to the commission of the crime are—both by themselves and viewed cumulatively—little more than speculation. Troxel, 875 N.W.2d at 309 (citing Nissalke, 801 N.W.2d at 102 (stating that “bare assertions as to what could have happened are not evidence”)).

One fact that the court highlights is that B.E. admitted to having been with the victim at her house several hours before her murder. This is not, however, suggestive that B.E. was actually at the scene of the crime. A gas station that is robbed in the evening is not the “scene of the crime” earlier in the afternoon prior to the robbery—it is simply a gas station.

C/D-3

Likewise, the victim’s house was not the scene of any crime when B.E. was known to be present as a guest—and it did not become the “scene of the crime” until hours later, when the victim was murdered. Some temporal connection between the commission of the crime and the location of the crime must be required to place a proposed alternative perpetrator at the scene when attempting to establish foundation. To underscore the axiomatic nature of this assumption, neither Carbo’s memorandum in support of his alternative-perpetrator proffer nor the district court discussed the idea that B.E.’s earlier presence at the house was suggestive that he was at the scene of the crime. So, outside of bare assertions and speculation, B.E.’s admitted presence as a guest in the victim’s house several hours before her murder does not have an “inherent tendency” to connect B.E. with the crime, nor does it place him at the scene.

Next, the court finds compelling an eyewitness’ description of a vehicle parked in the victim’s driveway at the time the victim was heard screaming, which was described as a dark-colored vehicle that was “shaped more like a van . . . or a truck with a topper.” Crucially, the eyewitness did not see B.E., did not see the driver of the vehicle, and did not identify the vehicle as one belonging to B.E., yet the court finds the fact that B.E. drove a gray Ford Bronco II at the time to be evidence that has an inherent tendency to connect B.E. to the commission of the crime. I agree with the district court, which found the presence of the vehicle established, at most, that a vehicle similar to B.E.’s was in the victim’s driveway at the time she was murdered and had “not established [B.E.’s] presence at the scene of the crime, let alone a connection to his actual commission of the charged offense[s].” Troxel, 875 N.W.2d at 309.

C/D-4

The same can be said for Carbo’s proffer of pubic hair evidence. A Bureau of Criminal Apprehension (BCA) report indicated that B.E. “qualifies” as a source for four pubic hairs found at the crime scene. 1 Importantly, the BCA report does not state that B.E. was the source of the pubic hairs, but only that his pubic hairs were macroscopically and microscopically like those found at the scene. Ignoring the considerable issue of the veracity and usefulness of non-DNA hair analysis, 2 the only thing the BCA report proves is that a person with pubic hair matching the characteristics of B.E.’s pubic hair was in the victim’s home at some point prior to the murder. Granted, if these hairs had been conclusively determined to belong to B.E. using DNA analysis, the location in which they were found could show an inherent tendency to connect B.E. to the commission of the

1 Seventeen different hairs were collected and tested by the BCA for the report in question. These hairs were collected from several blankets and a pillow in the victim’s bedroom, the victim’s pubic area, her panties, pants, and the area around where her body was found, various items found in a second bedroom, a sexual assault kit, and a sheet used to transport the victim’s body. As part of the investigation, pubic hairs were voluntarily submitted for analysis by six male individuals, including B.E., but not including Carbo. The pubic hairs found to be consistent with the sample provided by B.E. were located: 1) on the victim’s panties from the floor of the bedroom in which she was murdered; 2) on pants from the same bedroom floor; 3) on a bedsheet from the same bedroom; and 4) on a rug from a second bedroom. 2 See, e.g., M. Chris Fabricant & Tucker Carrington, The Shifted Paradigm: Forensic Science’s Overdue Evolution from Magic to Law, 4 Va. J. Crim. L. 1, 91 (2016) (discussing a 2009 National Academies of Sciences report that “concluded that hair microscopy could not uniquely identify one person as the source of a hair; instead, at best, a ‘match’ between two hair samples ‘mean[t] only that the hair could have come from any person whose hair exhibited - within some levels of measurement uncertainties - the same microscopic characteristics’ ”); Daniel A. Klein, Admissibility and Weight, in Criminal Case, of Expert or Scientific Evidence Respecting Microscopic Hair Analysis of Human Hair, 43 A.L.R.7th Art. 9 (Originally published in 2019) (discussing the inconsistent legal landscape of “expert or scientific evidence respecting microscopic analysis of human hair”).

C/D-5

crime. But absent speculation, a BCA report that states pubic hairs could have come from a third person simply does not establish that person’s presence at the scene nor their connection to the actual commission of the crime—much like the presence of a dark, van- like vehicle does not establish a gray Bronco II owner’s presence at or participation in a murder.

Finally, the court also finds compelling B.E.’s statement to police that he often wondered whether he woke up in the middle of the night and killed the victim. Taken out of context, as it was here, this could be seen as evidence that connects B.E. to the commission of the crime. However, I agree with the district court again, which found that the context of B.E.’s statement does not create such an inherent tendency to connect him to the commission of crime. The concerning statement was made over 40 minutes into an interview that B.E. voluntarily gave to police around 12 years after the murder. In it, the interviewer floated the theory that B.E. became jealous of the victim being intimate with another man and then inquired about B.E.’s temper. B.E. admitted to having a temper and said that he had been working on it. The interviewer then shared a theory that B.E. had lost his temper in 1986 to the point that he killed the victim and did not even realize it. This prompted no affirmative responses from B.E. The interviewer then posed this hypothetical:

Q: Um, is it possible that you could’ve lost your temper or lost control at any given time, and been responsible for this in any way?

A: You know, the human mind is a strange thing, and I’ve often wondered jeez, did I wake up in the middle of the night, drive over there and kill her, go back to bed and not know it?

C/D-6

This speculative statement, 12 years after an unsolved murder, is far from an admission of guilt and does nothing to inherently connect B.E. to the commission of the crime. After all, the purported motive of a third party is unconvincing absent “any overt indication of violence, threats, anger, jealousy, or frustration,” and Carbo’s proffer contains none of these. Troxel, 875 N.W.2d at 309; see also Woodard, 942 N.W.2d at 143. Moreover— and for the sake of argument—even if B.E.’s anger issues and this statement could be loosely construed as motive evidence, “[e]vidence of motive alone does not have the inherent tendency to connect a third party to the commission of the crime.” State v. Larson, 787 N.W.2d 592, 598 (Minn. 2010).

The court holds that this collection of evidence—which I would categorize as a collection of assertions—when viewed as a whole has an inherent tendency to connect B.E. with the commission of the crime. Accordingly, the court found that the district court abused its discretion when finding the first prong of the Hawkins test unfulfilled. But this court has never held that the Hawkins foundational requirement may be satisfied by accumulating individual pieces of proffered evidence that, by themselves, do not have an inherent tendency to connect a third party to the commission of the crime.

Granted, an inference may be made to support the idea that cumulative evidence may create such an inherent tendency when looking at the court’s language in other alternative-perpetrator cases. For example, in Troxel, the court stated, “The totality of [the defendant’s] proffered evidence . . . is insufficient to satisfy the foundational requirement establishing [the third party’s] inherent connection to the victim’s murder.” 875 N.W.2d at 309-10 (emphasis added). In Woodard, the court determined that “the evidence

C/D-7

proffered by [the defendant], even when considered cumulatively, did not have the inherent tendency to connect [the third party] to the actual commission of the murder.” 942 N.W.2d at 144 (emphasis added). In fact, the Woodard court even added a footnote of dicta that stated: “There may be cases where overwhelming circumstantial evidence of motive, opportunity, and method of murder has an inherent tendency to connect an alternative perpetrator to the scene of the crime even in the absence of direct evidence of presence at or near the scene.” Id. at 144 n.2.

But the real takeaway from Troxel and Woodard is the similarity between the assertions and circumstantial evidence proffered by the defendants in those cases, and those proffered by Carbo here. See Troxel, 875 N.W.2d at 309-10 (stating that the third party had a potential motive as a “spurned potential lover,” was parked for an hour in the driveway of a house where the victim was present on the night of her murder, and had no confirmed alibi for the time the crime was committed); Woodard, 942 N.W.2d at 142 (stating that the third party shared significant physical characteristics with the murderer, lived within blocks of the crime scene, may have had a motive of revenge, and was not seen at his home following the murder amidst rumors that the killer fled). In those cases, the court upheld the district court’s denial of alternative-perpetrator evidence, yet here, the court found that the proffered evidence “clearly” had the required inherent tendency to connect B.E. with the commission of the crime.

To support its finding, the court claims that Carbo’s proffer is more analogous to that of the defendant in State v. Ferguson, 804 N.W.2d 586 (Minn. 2011), a case in which the court reversed based on a district court’s decision to exclude alternative-perpetrator

C/D-8

evidence. The proffered evidence in Ferguson, however, has an actual tendency to connect a third person to the commission of that crime, rather than the speculative leaps required of the evidence proffered by Carbo. For instance, multiple witnesses saw the shooter in Ferguson and gave a physical description that “was in several ways similar” to the physical characteristics of the proposed alternative perpetrator. Id. at 591. Here, there are no eyewitness accounts of the killer at all, much less one giving a description that resembles B.E. In Ferguson, the investigating officer was told by the victim’s mother that that shooter went by the name C.J. or B.J. Id. Although the defendant had the initials C.F., the proposed alternative perpetrator actually had the initials C.J., had those initials tattooed on his arm, and had his phone number stored in the victim’s cell phone contacts as C.J. Id. Here, no one gave investigating officers any information regarding the name of the killer, much less a name that matched a third person better than it did the defendant.

The only similarity between these cases is that the physical description of the shooter’s car given by witnesses in Ferguson matched that of the third party’s car. Id. But even that is dissimilar to Carbo’s proffer because here, a single witness merely saw a dark, van-like vehicle in the victim’s driveway at the time screaming was heard. In Ferguson, multiple witnesses saw a person flee the scene of the shooting in a car that matched the physical description of the proposed alternative perpetrator’s car. Id. The Ferguson court found that 1) multiple witness descriptions; 2) from the precise time and location of the murder; 3) that matched the physical characteristics of the proposed alternative perpetrator and his car, suggested that the third party was present at the scene of the crime. Id. It was only after connecting the proposed alternative perpetrator to the scene of the crime—with

C/D-9

substantial eyewitness evidence—that the court then considered other proffered evidence cumulatively and found an inherent tendency to connect the third party to the commission of the crime. Id. at 592.

I agree with the decision in Ferguson, but do not find the same similarities to Carbo’s proffer that the court does here. The fact that B.E. had seen the victim on the night of her death and that his pubic hair was similar to hairs found near the victim do not suggest—absent a significant speculative leap—that B.E. was present at the scene of the crime. Nor does a single witness having seen a dark, van-like vehicle parked in the victim’s driveway when screaming was heard suggest—without speculation—that B.E. was present at the scene of the crime. Carbo’s proffer is simply not the same as multiple witnesses describing matching physical characteristics of both a third party and the third party’s vehicle fleeing the scene of a shooting. The court says that the district court’s analysis reads the word “tendency” out of the Hawkins test, instead requiring a “definite and unequivocal” connection. Yet the court itself seems to have neglected the phrase “to the commission of the crime,” because Carbo’s proffered evidence has no inherent tendency to connect B.E. with the actual commission of this murder.

“Generally, a reviewing court defers to the district court’s evidentiary rulings and will not overturn the rulings absent a clear abuse of discretion.” State. v. Dobbins, 725 N.W.2d 492, 505 (Minn. 2006). That deference forms the foundation for this dissent because I fail to see any “clear abuse of discretion” on this issue on the part of the district court that would allow us to overturn Carbo’s conviction. Id. In making the decision to exclude the alternative-perpetrator evidence, the district court took a careful and measured

C/D-10

look at the proffer and offered reasoned analysis—using our case law—explaining why the evidence was to be excluded. This issue was certainly a close one because of B.E.’s past relationship with the victim and the suspicious picture painted of him by Carbo’s assertions. But in close cases, this court has deferred to the district court’s discretion in evidentiary matters, even if another district court may have ruled differently on the admission of evidence. See State v. Blom, 682 N.W.2d 578, 613 (Minn. 2004) (“While we acknowledge that another district court, in the proper exercise of its discretion, may have excluded this . . . evidence, we hold that here the district court did not abuse its discretion . . . .”). I acknowledge that B.E. was an early suspect in this murder, and do not insinuate that Carbo threw darts at a list of potential alternative perpetrators in the hopes of muddying the waters of his guilt. But this case is simply not one that presents the “overwhelming circumstantial evidence” that was pondered by the Woodard court, 942 N.W.2d at 144 n.2, and it is still anything but clear that the district court abused its discretion here.

In reversing the district court’s alternative-perpetrator ruling, the court gives lip service to the requirements of the Hawkins test, and then substitutes its judgment for that of the district court to hold that the district court abused its discretion. In the process, the court’s decision arguably alters our alternative-perpetrator test, opens the door to the “use of bare suspicion” regarding third parties that we condemned in Hawkins, and creates confusion for district courts in the future examining evidentiary proffers like those in this case. Hawkins, 260 N.W.2d at 159. Because the court deferred to the district courts’ discretion in Woodard and Troxel but found it abused its discretion here, district courts will now be left to wonder whether evidence that previously should have only been considered

C/D-11

in step two of the Hawkins test should now be considered when looking for sufficient foundation in step one of the test. Nothing in Carbo’s proffer had the inherent tendency to connect B.E. with the victim’s murder, but now district courts will be forced to make speculative inferences and consider evidence that has no inherent tendency to connect a third party to the actual commission of a crime when making an initial alternative- perpetrator analysis. This decision will invariably diminish the “safeguards” created by Hawkins, allowing any past differences or intimate contact a third person may have had with a murder victim to be aired before the jury in every murder trial in Minnesota. Id.

B.

Although the court determined that the district court clearly erred here, that should not end its alternative-perpetrator analysis. Even if the district court did err in excluding the evidence, this court should only reverse if there is a reasonable possibility that the verdict may have been more favorable to Carbo without the error. State v. Miller, 754 N.W.2d 686, 703 (Minn. 2008). In other words, a district court’s erroneous exclusion of alternative-perpetrator evidence “is harmless beyond a reasonable doubt if the jury’s verdict is surely unattributable to that error.” State v. Vance, 714 N.W.2d 428, 439 (Minn. 2006); see also State v. Smith, 876 N.W.2d 310, 331 (Minn. 2016) (“Under this standard, we must be satisfied beyond a reasonable doubt that if the evidence had been admitted and the damaging potential of the evidence fully realized, an average jury (i.e., a reasonable jury) would have reached the same verdict.”) (internal quotation marks omitted) (citation omitted).

C/D-12

This court need look no further than State v. Westrom, ___N.W.3d ___, No. A22-

1679 (Minn. 2024), published today, to find an analogue for why reversal is improper here. There, the court found unanimously and without lengthy discussion that even if the district court had erred in excluding certain alternative-perpetrator evidence, that the strength of the evidence against the defendant left the court convinced that admitting the alternative- perpetrator evidence would have resulted in the same verdict. Id. at ___, slip op. at 13. The evidence in Westrom included the defendant’s semen found on a comforter in the victim’s home, his DNA found on several items inside the victim’s apartment, and bloody footprints found throughout the victim’s home that were forensically matched to the defendant. I agree that this evidence was strong and that admission of the proposed alternative-perpetrator evidence would not create a reasonable possibility of a more favorable verdict.

But the evidence convicting Carbo is no less strong. The victim here was discovered inside her locked house, naked, and beneath a comforter on her bed. She had grass on her body and there was dirt and grass on her wet clothes, which were also found inside her bedroom. Outside her back door there was a patch of grass that had been obviously matted down by recent activity, which the chief of police at the time described in testimony as an area where a “struggle” took place. Next to that patch of grass were the victim’s keys and a pile of vomit—which was DNA-matched to Carbo. If this were the only evidence connecting Carbo to the killing, then any error by the district court may have been harmful. However, Carbo’s was the only DNA found underneath the victim’s fingernails, and his DNA-matched semen was found on a deep vaginal swab of the victim, on the bedsheets

C/D-13

directly underneath the victim’s groin where she was found, and on a washcloth in the bedroom.

Additionally, in Carbo’s initial interview with police in 2020, he claimed he had “no idea who [the victim] was,” that he never had sex with her, and that he “wasn’t much with girls back then.” But then after his DNA matched that of semen found at the scene, Carbo told police that in those days, he was using drugs and drinking, and that he must have blacked out. In a recorded conversation with his daughter two years later, Carbo said that the only thing that his semen at the scene proves is that he was there before the victim was murdered, and then mused, “All the one night stands I had back then after the bars . . . .” Carbo’s defense attorney claimed that the DNA under the victim’s fingernails proves that the sex was consensual. But to assume that alternative-perpetrator evidence here would have changed the jury’s verdict would also require the assumption that the victim used her fingernails during sex, but not while she was being strangled to death by a person different than Carbo. This assumption is simply not reasonable.

While I continue to believe that there was no clear error in the district court’s denial of alternative-perpetrator evidence involving B.E., 3 the jury’s verdict would surely be “unattributable to that error” based on the strength of the evidence against Carbo. Vance, 714 N.W.2d at 439.

3 B.E. testified at Carbo’s trial for the State. The frequency with which B.E. saw the victim was established through his trial testimony as well as his presence at her house on the evening of her death and his ownership of a gray Bronco.

C/D-14

Carbo’s conviction for first-degree murder should stand. This court’s decisions in Hawkins and its progeny dictate that the district court did not clearly abuse its discretion here in finding that Carbo’s proffer did not have the inherent tendency to connect a third party with the commission of the crime. And even if it did, that error was unattributable to the jury’s verdict in the face of the overwhelming evidence against Carbo. For those reasons, I respectfully dissent on this issue.

CHUTICH, Justice (concurring in part, dissenting in part).

I join Parts I, II, and III.A of the majority opinion, and join in the dissent of Justice McKeig as to Part B, but not as to Part A.

MOORE, III, Justice (concurring in part, dissenting in part).

I join Parts I and II of the majority opinion and join in the dissent of Justice McKeig.

C/D-15