State of Minnesota v. Christopher James Colgrove

• Court: Supreme Court of Minnesota
• Decided: October 4, 2023
• Docket: A220718
• Status: Published

Opinion

STATE OF MINNESOTA

IN SUPREME COURT

A22-0718

Clearwater County Moore, III, J. Dissenting, Thissen, J., Hudson, C.J. Took no part, Procaccini, J.

State of Minnesota,

Respondent,

vs. Filed: October 4, 2023 Office of Appellate Courts Christopher James Colgrove,

Appellant.

________________________

Keith Ellison, Attorney General, Lydia Villalva Lijó, Assistant Attorney General, Saint Paul, Minnesota; and

Kathryn Lorsbach, Clearwater County Attorney, Bagley, Minnesota, for respondent.

Cathryn Middlebrook, Chief Appellate Public Defender, Jennifer Workman Jesness, Assistant Public Defender, Saint Paul, Minnesota, for appellant.

SYLLABUS

The State presented sufficient circumstantial evidence to support defendant’s

conviction for first-degree intentional felony murder while committing burglary.

Affirmed.

1 OPINION

MOORE, III, Justice.

Following a jury trial, appellant Christopher James Colgrove was convicted of

first-degree intentional felony murder while committing burglary, Minn. Stat.

§ 609.185(a)(3) (2022), in the stabbing death of Dawn Swenson. On appeal, Colgrove

argues that the conviction must be reversed because the State presented insufficient

evidence to prove his intent to kill. Because we conclude that the State presented sufficient

evidence to sustain Colgrove’s conviction, we affirm. 1

FACTS

This case arises from the fatal stabbing of Swenson in her home in rural Clearwater

County on September 7, 2020. Colgrove also lived in Clearwater County on a property

near Swenson’s home. Most of Colgrove’s family, including his girlfriend, their children,

his parents, and his sister, also lived on this property.

Colgrove and his girlfriend were intravenous methamphetamine users, and for

several days leading up to the fatal stabbing of Swenson, Colgrove had been injecting

1 The jury also found Colgrove guilty of second-degree intentional murder, Minn. Stat. § 609.19, subd. 1(1) (2022), and second-degree felony murder while committing a felony assault with force or violence, Minn. Stat. § 609.19, subd. 2(1) (2022). The district court did not enter judgment of convictions for these crimes because they are lesser- included offenses. Because we affirm Colgrove’s conviction and sentence for first-degree felony murder while committing a burglary, we need not consider Colgrove’s arguments regarding the jury’s verdict for second-degree intentional murder. See State v. Ortega, 798 N.W.2d 59, 62 n.1 (Minn. 2011). 2 methamphetamine. 2 In the early morning of September 7, 2020, Colgrove purchased more

methamphetamine. His girlfriend saw him use methamphetamine twice that day. That

afternoon, Colgrove’s girlfriend left the house, taking their children with her, because she

could no longer handle the erratic behavior Colgrove had been exhibiting for several days.

Colgrove had been emotional, crying, jealous, paranoid, and scared, and he had barely

eaten and had not slept for almost a week. According to Colgrove’s sister, on the day his

girlfriend left, Colgrove believed that people were under the house and in the surrounding

woods and that they were “trying to get him.” Colgrove’s father also noted that Colgrove

was “not acting normal” that day.

At approximately 8:30 p.m. on September 7, Colgrove called 911 to request an

ambulance. During the call, Colgrove identified himself and described his location. He

explained that he needed an ambulance because he was having trouble breathing.

Sometime after this call, Colgrove’s sister saw him walking outside. Colgrove did not

recognize her until she told him who she was. Colgrove told his sister that he had called

an ambulance earlier because he could not breathe. Colgrove repeatedly asked his sister to

look in his throat and told her that he believed people were following him. Colgrove’s

sister could not see anything in his throat and attempted to calm him down.

2 Colgrove’s girlfriend had also been using methamphetamine with him. At some point, she stopped using because the methamphetamine was making her feel sick, which led her to believe that the drugs were not pure. She asked Colgrove to stop using those drugs as well. She was concerned because they had taken bad drugs before that had made them, in her words, “really crazy.” Colgrove stated that he trusted the person who had sold him the methamphetamine and continued to use it. 3 When the ambulance arrived, Colgrove and his sister were located in their parents’

driveway. The EMT driving the ambulance observed Colgrove and his sister engaged in a

verbal altercation outside the house. Two other people were standing nearby. Colgrove’s

sister informed the EMT that Colgrove needed help, was high on methamphetamine, might

hurt someone, and they needed assistance from the police. Despite the EMT’s efforts to

get everyone other than Colgrove to go inside the house, Colgrove’s sister remained with

Colgrove because he was holding onto her. The EMT, who was in uniform and driving a

marked ambulance, identified herself to Colgrove and said she was there to help him.

Colgrove called her a liar and was uncooperative with her instructions for him to sit on the

ground before she would get out of the vehicle. Concerned for her safety, the EMT decided

to wait in the ambulance until a police officer arrived.

At about 8:45 p.m., an officer arrived at the Colgrove residence, wearing a uniform

and driving a marked squad car with the lights activated. The officer noticed that Colgrove

was acting agitated. He identified himself as a police officer, but Colgrove said he did not

believe him. Because of Colgrove’s behavior, the officer decided to detain Colgrove in

handcuffs, but Colgrove pulled away. The officer drew his taser and ordered Colgrove to

the ground. Colgrove followed instructions to get on his knees, but he refused to get on

his stomach and repeated his concern that the officer was not really a police officer. The

taser malfunctioned and turned off. Colgrove got up to his feet ran down the road toward

the west. The nearest house, which belonged to Swenson, was approximately one-quarter

mile west from Colgrove’s location.

4 The officer drove his squad car down the road in search of Colgrove. The EMT

followed in the ambulance. The officer drove into the driveway of Swenson’s residence to

search for Colgrove on the property. When the officer left his car, he heard a commotion

coming from inside the house and ran toward the front door. Looking through a window

above the door, he could see a man and a woman struggling inside. The officer heard the

woman (later identified as Swenson) yell, “Get out of my house, don’t hurt me.” The

officer attempted to open the front door, but it was locked. He forced the door open with

his shoulder and observed Colgrove struggling with Swenson by the door near the kitchen.

Upon entering the house, the officer drew his taser and commanded Colgrove to put

his hands up, let go of Swenson, and stop resisting. Colgrove did not follow these

commands. The officer deployed his taser and hit Colgrove. The taser appeared to

momentarily affect Colgrove and cause him to “lock up.” Colgrove then appeared angry

and screamed. The officer noticed that Colgrove was holding a knife in his right hand. 3

Colgrove raised the knife and took “two swipes” at Swenson before he pushed her out of

the way and fled out the front door. Rather than chase Colgrove, the officer helped provide

medical treatment for Swenson. The officer and EMT placed Swenson into the ambulance

and transported her to the hospital, where she died from her wounds.

Law enforcement searched Swenson’s home and the surrounding area. In the

kitchen, officers observed an open drawer. One investigator found a bloody knife in the

yard, which was similar to other knives in the open drawer in the kitchen. Law enforcement

3 In his earlier interaction with Colgrove, the officer did not observe Colgrove with any weapons, and Colgrove had appeared unarmed while struggling with Swenson. 5 searched for Colgrove throughout the night of September 7 and into the early morning

hours of September 8 to no avail. Sometime between 3 and 5 a.m. on September 8,

Colgrove’s father located Colgrove in a field up the road from Swenson’s house.

Colgrove’s clothes were torn, and he was shivering and was not wearing shoes. Colgrove

told his father that people were trying to kill him. Colgrove’s father drove Colgrove home.

During the drive back to their property, Colgrove stated that he would never hurt anyone

and he “didn’t do it.” His father noticed that taser probes were stuck in Colgrove’s chest

and that Colgrove was still acting strangely. Colgrove told his father that he had run to a

house and that the “girl there was mean.”

An officer arrived at the property to arrest Colgrove. The officer noted that

Colgrove’s behavior was manic. Colgrove stated that he had gone to a residence, but he

did not know who was there. He also said that a “fake cop” tried to shoot him with a

“dart.” 4 Colgrove’s sister placed handcuffs on Colgrove because he would not allow the

officer to touch him. Colgrove spoke in an animated fashion during the drive to the jail.

He repeated his concerns about a fake cop and getting shot.

As part of the autopsy of Swenson, the medical examiner identified two wounds—

one on Swenson’s lower left back and one on her neck. The medical examiner determined

that the wound to Swenson’s back, which was 5 inches deep and had punctured her

diaphragm, stomach, lung, and heart, was the cause of Swenson’s death. The wound to

Swenson’s neck was a contributory cause of her death. The neck wound was 2 inches deep

4 The officer testified that a “dart” means “taser probe” in “law enforcement speak.” 6 and penetrated her vertebral column and contributed to blood loss, but it did not injure the

spinal cord. A forensic examination of the bloody knife found in Swenson’s yard revealed

that the blood matched Swenson’s DNA profile. The blade of the knife was bent.

Two days after the fatal stabbing, the State charged Colgrove with second-degree

unintentional murder and third-degree murder. The district court ordered an examination

of Colgrove’s mental condition. Following this examination, the district court found that

Colgrove was competent to proceed to trial. In doing so, the district court determined that

at the time of the stabbing Colgrove was not laboring under a mental defect of reason that

substantially impaired his capacity to understand the nature or wrongfulness of his acts.

Subsequently, a grand jury indicted Colgrove with first-degree intentional felony murder

while committing burglary in violation of Minn. Stat. § 609.185(a)(3), second-degree

intentional murder in violation of Minn. Stat. § 609.19, subd. 1(1), and second-degree

unintentional felony murder while committing an assault in violation of Minn. Stat.

§ 609.19, subd. 2(1).

At trial, the jury found Colgrove guilty of all three counts. The court convicted

Colgrove of first-degree intentional felony murder and imposed a life sentence with

eligibility for release after 30 years.

ANALYSIS

On appeal, Colgrove argues that the State presented insufficient evidence to support

his conviction for first-degree intentional felony murder. To sustain the conviction for

first-degree intentional felony murder, the State must prove beyond a reasonable doubt that

Colgrove intended to kill Swenson when he stabbed her with the kitchen knife. See Minn.

7 Stat. § 609.185(a)(3) (providing that a person is guilty of first-degree murder if he “causes

the death of a human being with intent to effect the death of the person or another, while

committing or attempting to commit burglary” (emphasis added)). The State relied on

circumstantial evidence to prove Colgrove’s intent to kill. Colgrove argues that the State

failed to meet its burden of proof because the circumstances proved are consistent with a

reasonable hypothesis that he did not intend to kill Swenson—either because he was too

intoxicated to form the specific intent required or because he intended only to assault

Swenson. We disagree.

When the evidence for an element of a conviction is based on circumstantial

evidence, we apply a heightened, two-step standard of review to decide whether the

evidence is sufficient. State v. Al-Naseer, 788 N.W.2d 469, 474–75 (Minn. 2010). But

even when reviewing a conviction based on circumstantial evidence, we recognize that

“the jury determines the credibility and weight given to the testimony of individual

witnesses.” State v. Olhausen, 681 N.W.2d 21, 26 (Minn. 2004). This “stricter

standard . . . ‘still recognizes a jury is in the best position to evaluate the circumstantial

evidence surrounding the crime.’ ” Id. (quoting State v. Race, 383 N.W.3d 656, 662 (Minn.

1986)).

For the first step, “we identify the circumstances proved.” State v. Hassan,

977 N.W.2d 633, 640 (Minn. 2022). This step requires us to “winnow down the evidence

presented at trial by resolving all questions of fact in favor of the jury’s verdict,” which

results in “a subset of facts that constitute ‘the circumstances proved.’ ” State v. Harris,

895 N.W.2d 592, 600 (Minn. 2017). Our analysis at this step “preserv[es] the jury’s

8 credibility findings” by recognizing that “the jury is in a unique position to determine the

credibility of the witnesses and weigh the evidence before it.” Id. Therefore, “[i]n

determining the circumstances proved, we disregard evidence that is inconsistent with the

jury’s verdict.” Id. at 601.

In the second part of our analysis, “we independently examine the reasonableness

of all inferences that might be drawn from the circumstances proved.” State v. Noor,

964 N.W.2d 424, 438 (Minn. 2021). At this step, “we give no deference to the fact finder’s

choice between reasonable inferences.” State v. Andersen, 784 N.W.2d 320, 329–30

(Minn. 2010) (citation omitted) (internal quotation marks omitted). “The State’s

circumstantial evidence is sufficient when the reasonable inferences are consistent with the

hypothesis that the accused is guilty and inconsistent with any rational hypothesis other

than guilt.” Hassan, 977 N.W.2d at 640. In evaluating the inferences that may be drawn

from the circumstances proved, “[w]e review the circumstantial evidence not as isolated

facts, but as a whole.” State v. Sterling, 834 N.W.2d 162, 175 (Minn. 2013). The

circumstantial evidence standard does not allow us to “analyze and parse each fact” in a

“piecemeal” fashion to conclude that a hypothesis is reasonable. State v. Cox, 884 N.W.2d

400, 415 (Minn. 2016). And “[w]e do not set aside verdicts based on speculation.” 5 Al-

Naseer, 788 N.W.2d. at 480.

5 The focus of our circumstantial evidence analysis is on whether the inferences that can be drawn from the circumstances proved are reasonable. We have said that where an alternative inference has no support in the record, it is speculative. Al-Naseer, 788 N.W.2d at 480. It follows that if an inference is speculative in that sense, it is unreasonable. Accordingly, an inference that the knife bent when it struck Swenson’s vertebral column

9 A.

Applying that standard to the record here, the circumstances proved are as follows.

For several days leading up to the fatal stabbing, Colgrove had been using

methamphetamine and his behavior was abnormal. On the day of the stabbing, Colgrove

injected methamphetamine. He exhibited paranoid delusions and erratic behavior and

appeared to have difficulty recognizing his sister. That evening, Colgrove called 911 and

requested an ambulance. When the ambulance arrived, Colgrove did not cooperate with

the EMT. Colgrove also did not cooperate with the officer who later arrived at his

residence. When the officer’s taser malfunctioned, Colgrove ran away.

Colgrove ran to and entered Swenson’s house. Inside the house, Colgrove struggled

with Swenson, who was yelling at Colgrove to leave and not hurt her. The officer arrived

and forced open the locked door to the house. Upon entering the house, the officer

encountered Colgrove struggling with Swenson and directed him to put his hands up, let

go of Swenson, and stop resisting. After Colgrove ignored the officer’s commands and

continued the struggle, the officer tased Colgrove, but Colgrove hardly reacted to the taser

and became angry. Colgrove then stabbed Swenson twice with a knife that was similar to

others located in Swenson’s kitchen and then fled from the house. Swenson died from a

wound to her back that was 5 inches deep and punctured several organs, including her

diaphragm, lung, and heart. Swenson also sustained a neck wound that was 2 inches deep.

with enough force to penetrate the bone is reasonable because it is grounded in the medical examiner’s testimony that the knife penetrated the bony area surrounding Swenson’s spinal cord. In contrast, an inference that the knife was pre-bent is unreasonable because there is nothing in the record from which such an inference can be drawn. 10 A bloody knife with a bent blade was found in the yard outside Swenson’s home. DNA

on the knife matched Swenson’s DNA. Colgrove’s father located Colgrove between 3 and

5 a.m. in a nearby field. During the drive to their home, Colgrove told his father he would

never hurt anyone and “didn’t do it.” He also admitted that he went to a house and a

“mean” girl was there. Colgrove acted manic and paranoid while being transported to jail.

The State contends that we cannot consider evidence that Colgrove was exhibiting

paranoid delusions the day he stabbed Swenson as part of the circumstances proved

because it is inconsistent with the jury’s verdict. We disagree because “the evidence is not

in conflict” with the jury’s verdict. See Al-Naseer, 788 N.W.2d at 475 n.2.

The first step of our circumstantial evidence test requires that we reject only the

evidence that is inconsistent with—or in conflict with—the jury’s verdict. Harris,

895 N.W.2d at 601. For instance, we would not consider a statement made by a defendant

who testified at trial claiming their innocence.

The State’s argument that evidence of Colgrove’s paranoia and delusions is not part

of the circumstances proved appears premised on the assumption that a person

experiencing this type of behavior can never form an intent to kill another person. Under

this logic, the jury could convict Colgrove of first-degree intentional felony murder and

second-degree intentional murder only if it rejected all evidence related to his paranoia and

delusions. But the State presents no case law to support this assertion, and we decline to

adopt this rule for determining intent to kill. Instead, we have consistently held that “[t]he

mere fact of a person’s [using intoxicants] does not create a presumption of intoxication,

and the possibility of intoxication does not create the presumption that a person is rendered

11 incapable of intending to do a certain act.” State v. Lund, 151 N.W.2d 769, 771 (Minn.

1967). Rather, the test is whether the intoxication is so significant that it prevents a person

from forming the requisite intent. State v. Hale, 453 N.W.2d 704, 706 (Minn. 1990). We

thus conclude that, at this first step of our circumstantial evidence test, the evidence that

Colgrove exhibited paranoid delusions the day he stabbed Swenson is not inconsistent with

the jury’s conclusion that he committed first-degree intentional felony murder because the

jury could accept both that Colgrove had paranoid delusions and that he intended to kill

Swenson. Therefore, we consider this evidence as part of the circumstances proved.

B.

Having identified the circumstances proved, the next step is to “determine whether

the reasonable inferences that can be drawn from the circumstances proved are consistent

with guilt and inconsistent with any rational hypothesis other than [Colgrove’s] guilt.”

State v. Andersen, 784 N.W.2d 320, 331 (Minn. 2010).

The circumstances proved are consistent with Colgrove’s guilt under Minn. Stat.

§ 609.185(a)(3) if it is reasonable to infer that Colgrove acted with intent to kill Swenson.

The Legislature has defined the phrase “[w]ith intent to” to mean that “the actor either has

a purpose to do the thing or cause the result specified or believes that the act, if successful,

will cause that result.” Minn. Stat. § 609.02, subd. 9(4) (2022). Therefore, Colgrove had

the intent required by Minn. Stat. § 609.185(a)(3) if he either had a purpose to kill Swenson

or believed that his actions, if successful, would kill Swenson. See State v. Young,

710 N.W.2d 272, 278 (Minn. 2006). “Intent is an inference drawn by the jury from the

totality of circumstances,” State v. Raymond, 440 N.W.2d 425, 426 (Minn. 1989), and the

12 fact-finder may infer that a person intends the natural and probable consequences of that

person’s actions, State v. Cooper, 561 N.W.2d 175, 179 (Minn. 1997). “A jury may infer

a person’s intent to kill from the nature of the killing.” Young, 710 N.W.2d at 278.

Our decision in Wolfe v. State, 293 N.W.2d 41 (Minn. 1980), is instructive

concerning the nature of the killing. The facts of that case were as follows:

[W]hile arguing with an unarmed acquaintance, [the defendant] opened his knife . . . and stabbed the victim in the chest. The knife penetrated nearly 4 inches into the victim’s body and resulted in a fatal wound to the heart. After the stabbing [defendant] fled the scene, saying to his friends who were witnesses, “See you in 40 years.”

Id. at 42. We concluded that this evidence was sufficient to support a finding of intent to

kill for second-degree intentional murder. Id.

Like the defendant in Wolfe, Colgrove was arguing with Swenson, an unarmed

person, in the kitchen of her home before he stabbed her twice, once in the neck and once

in the back. The wound to Swenson’s back was 5 inches deep and determined to be fatal.

In addition, the tip of the murder weapon, which the officer did not observe in Colgrove’s

hand until after the tasing and which matched the knives in Swenson’s kitchen drawer, was

bent. Colgrove also fled the scene after he stabbed Swenson. It is reasonable to infer from

these facts, which show the nature of the killing, that Colgrove intended to kill Swenson.

See Raymond, 440 N.W.2d at 426 (holding that the defendant’s intent to kill the victim was

inferable from nature and extent of wounds and defendant’s act of leaving the victim to

bleed to death); State v. Andrews, 388 N.W.2d 723, 728–29 (Minn. 1986) (holding that a

single stab wound to the victim’s back and the defendant’s later attempt to leave the scene

showed an intent to kill); State v. Merrill, 428 N.W.2d 361, 370 (Minn. 1988) (holding

13 that a single fatal stab wound that passed through the breastbone and into the heart showed

sufficient intent to kill).

But this conclusion does not end our analysis. To determine whether the State’s

circumstantial evidence is sufficient to support Colgrove’s conviction, we must also assess

whether the circumstances proved are consistent with a reasonable hypothesis other than

guilt. Colgrove puts forth two hypotheses of innocence: first, that his voluntary

intoxication negated his ability to form the intent to kill, and second, that he intended

merely to assault Swenson. In evaluating Colgrove’s hypotheses of innocence, the key part

of our analysis is determining whether the hypotheses are reasonable. We consider each

in turn.

First, Colgrove contends that the circumstances proved support a reasonable

inference that his voluntary intoxication from methamphetamine precluded him from

forming the specific intent to cause Swenson’s death. Section 609.075 governs voluntary

intoxication. This statute provides:

An act committed while in a state of voluntary intoxication is not less criminal by reason thereof, but when a particular intent or other state of mind is a necessary element to constitute a particular crime, the fact of intoxication may be taken into consideration in determining such intent or state of mind.

Minn. Stat. § 609.075 (2022). We have consistently held that “[t]he mere fact of a person’s

[using intoxicants] does not create a presumption of intoxication, and the possibility of

intoxication does not create the presumption that a person is rendered incapable of

intending to do a certain act.” Lund, 151 N.W.2d at 771. Rather, the test is whether the

intoxication is so significant that it prevents a person from forming the requisite intent.

14 Hale, 453 N.W.2d at 706 (stating that the evidence “d[id] not compel the conclusion that

defendant was incapable of forming the requisite specific intent to kill”); State v. Cole,

542 N.W.2d 43, 49–50 (Minn. 1996) (concluding that there was sufficient evidence to

support the jury’s conclusion that the defendant was capable of forming the intent to kill

despite the defendant’s claims that he “[drank] beer, used methamphetamine, cocaine,

heroin, Dilaudid and smoked marijuana” the day of the murder).

The voluntary intoxication statute applies to specific-intent crimes. State v. Fleck,

810 N.W.2d 303, 308 (Minn. 2012) (interpreting the voluntary intoxication statute and

holding that, based on its plain and ordinary meaning, the phrase “particular intent”

unambiguously referred to specific-intent crimes, not general-intent crimes); see also State

v. Wilson, 830 N.W.2d 849, 853 (Minn. 2013). First-degree intentional felony murder is a

specific-intent crime. The jury was never instructed, however, that it could consider

whether Colgrove could not form a specific intent to cause the death of Swenson due to

methamphetamine intoxication. Colgrove’s trial counsel specifically declined a voluntary

intoxication jury instruction. 6

Nevertheless, the fact that no voluntary intoxication instruction was given here is

not dispositive of whether Colgrove was incapable of forming a specific intent to kill.

There may be a point “at which evidence of a defendant’s intoxication . . . is so

overwhelming as to constitute the effective offer of intoxication as an explanation for the

defendant’s actions.” State v. Torres, 632 N.W.2d 609, 617 (Minn. 2001). But this is not

6 The district court had no obligation to sua sponte instruct the jury on the intoxication defense. State v. Hannon, 703 N.W.2d 498, 512 (Minn. 2005). 15 that case. While the circumstances proved clearly support an inference that Colgrove was

impaired by methamphetamine at the time he stabbed Swenson and that his use of

methamphetamine caused him to experience paranoia and delusions, it is unreasonable to

infer that he was so intoxicated that he was incapable of forming an intent to kill a person

when he stabbed Swenson twice with a knife.

Despite his use of methamphetamine, the circumstances proved show that Colgrove

was capable of making rational decisions that evening, including calling 911 for assistance,

providing the dispatcher with details regarding his location, and requesting that the

ambulance proceed quickly. Colgrove initially followed the officer’s instructions to get on

his knees but when the officer’s taser did not work, Colgrove took the opportunity to stand

up and flee to Swenson’s home. Colgrove entered Swenson’s home, obtained a knife,

stabbed Swenson with it, and fled, discarding the knife and evading law enforcement in the

process. Colgrove later told law enforcement that he entered a residence and told his father

that a “mean” girl had been there. Consequently, when the circumstances proved are

viewed as a whole, they demonstrate that Colgrove could deliberate, communicate, act and

recall events, obtain a dangerous weapon, use it, and flee from the scene of the stabbing.

It is unreasonable to infer from the circumstances proved that Colgrove’s actions

were so completely disconnected from reality as a result of the methamphetamine that he

acted with no purpose or intent to kill when he stabbed Swenson. Despite his intoxication,

Colgrove responded with anger when the officer tased him in Swenson’s home. Colgrove

recognized Swenson as a person whom he believed was “mean.” After being found several

hours after the stabbing, Colgrove told his father that he “didn’t do it.” Based on

16 Colgrove’s conduct and statements, it is unreasonable to conclude that Colgrove did not

understand what he had done. Accordingly, the circumstances proved, when viewed as a

whole, do not support the reasonable inference that Colgrove was unable to form the

requisite intent to kill because of his intoxication, i.e., that he had a purpose to kill Swenson

or believed that his actions, if successful, would kill Swenson. 7

Colgrove next argues that the circumstances proved support an inference that he

only intended to assault—but not kill—Swenson when he attacked her with a knife. He

suggests that his act of swinging the knife could reasonably be seen as an effort to

intimidate the officer and the victim to allow him to escape. He contends that it is

reasonable to infer that he swung the knife in retaliation for the officer tasing him and that

Swenson happened to be between him and the officer. Colgrove also argues that it is

possible that he wanted to assault Swenson to get back at the officer for tasing him.

We disagree that these inferences are reasonable when considering the

circumstances proved “as a whole and not as discrete and isolated facts.” Cox, 884 N.W.2d

at 412. Here, Colgrove was with a victim who was isolated in a locked house. He acquired

a kitchen knife—a weapon capable of causing serious injury—and accosted Swenson.

Colgrove became angry when the officer entered the house and tased him. The nature of

the stab wounds is particularly relevant. Colgrove inflicted a 5-inch fatal stab wound to

Swenson’s back, which punctured several organs, including her diaphragm, lung, and

7 The court is unanimous in this conclusion. The dissent explicitly “agree[s] with the court that the evidence does not support the reasonable inference that Colgrove’s voluntary intoxication precluded him from forming the specific intent to cause the death of Swenson.” 17 heart, causing fatal bleeding. Colgrove not only stabbed Swenson in the back of her torso

but also inflicted a stab wound in the back of her neck. In addition, the blade of the knife

was bent. 8 Given the brutal nature and location of these multiple wounds, there is not a

reasonable inference that Colgrove was only trying to intimidate or assault Swenson or the

officer—or that these stabbings were merely part of a defensive attempt to escape the

house—without a purpose to kill Swenson or belief that his actions, if successful, would

kill her. See Andrews, 388 N.W.2d at 727–29 (concluding that circumstantial evidence did

not support a defendant’s assertion that he did not intend to kill the victim where the victim

was stabbed once “with sufficient force to bend the knife blade and cut into a rib bone”).

The totality of the circumstances of this case, including the nature and extent of the

stab wounds inflicted to the victim’s neck and chest, exclude beyond a reasonable doubt

any inference other than that Colgrove intended to cause the result of Swenson’s death.

Accordingly, it is not reasonable to infer from the circumstances proved, as a whole, that

Colgrove only intended to assault Swenson when he stabbed her. 9

8 The dissent notes that the knife may have been bent before the murder. We respectfully disagree. The medical examiner testified that the knife “penetrated to the vertebral column, which is the bony part that surrounds the spinal cord.” (Emphasis added.) Based on this testimony, it is reasonable to infer that the knife bent when it struck the vertebral column with enough force to penetrate bone. In contrast, it is unreasonable to infer that Swenson kept a bent knife in her kitchen drawer, Colgrove reached into the drawer, grabbed the bent knife, and with two quick swipes used a pre-bent knife to stab Swenson five inches into her back and two inches into her neck, penetrating the bony part that surrounded her spinal cord. See Al-Naseer, 788 N.W.2d at 480 (“[W]e do not set aside verdicts based on speculation.”). 9 The dissent contends that the inference that Colgrove intended to assault—but not kill—Swenson is reasonable. In support of this argument, the dissent notes that individual

18 Because the circumstances proved are consistent with the hypothesis that Colgrove

intended to kill Swenson and inconsistent with any reasonable hypothesis of innocence,

the State’s circumstantial evidence is sufficient to support Colgrove’s conviction for

first-degree intentional felony murder.

CONCLUSION

For the foregoing reasons, we affirm the judgment of conviction.

PROCACCINI, J., not having been a member of this court at the time of

submission, took no part in the consideration or decision of this case.

facts comprising the circumstances proved are not “exclusively associated with a person acting with intent to kill and never associated with [a] person acting with intent to harm but not to kill.” For instance, the dissent notes that knives can be used to assault, an angry person can simply intend to hurt (but not kill) someone, and a deep wound might result from an assault without intent to kill. These principles are all true. But we do not view each fact in isolation to determine whether it supports an inference of innocence. And viewing the circumstances proved as a whole, we do not find Colgrove’s hypothesis of innocence reasonable. See, e.g., Cox, 884 N.W.2d at 415 (rejecting a defendant’s hypothesis of innocence, which “analyze[d] and parse[d] each fact,” as unreasonable because “[o]ur circumstantial evidence standard requires that we examine the circumstances proved as a whole and not piecemeal”). Instead of providing an inference that is reasonable in light of the circumstances proved as a whole, the dissent accuses us of relying on “provocative adjectives”—“brutal” and “fatal”—to, in the dissent’s words, “make [our] and the State’s version of the narrative sound more compelling.” The dissent misconstrues our use of these adjectives. We agree that use of adjectives like “brutal” and “fatal” to describe a victim’s wounds inflicted during a murder does not automatically lead to the conclusion that there is sufficient evidence of intent to kill. But once again, we disagree that the circumstances proved as a whole allow for the reasonable inference that Colgrove acted without intent to kill when he stabbed D.S. twice in her home. We rely not on mere adjectives, but on our circumstantial evidence standard to reach this conclusion. 19 DISSENT

THISSEN, Justice (dissenting).

I respectfully dissent.

Dawn Swenson’s death in September 2020 was tragic. In connection with her death,

a jury found Christopher Colgrove guilty of first-degree intentional felony murder under

Minn. Stat. § 609.185(a)(3) (2022), second-degree intentional murder under Minn. Stat.

§ 609.19, subd. 1 (2022), and second-degree unintentional felony murder while

committing an assault under Minn. Stat. § 609.19, subd. 2(1) (2022).

Colgrove does not dispute that he caused Swenson’s death. Rather, Colgrove is

challenging the sufficiency of the evidence on the first two counts that require the State to

prove he intended to cause the death of Swenson beyond a reasonable doubt. 1 He is not

challenging the guilty verdict for second-degree unintentional felony murder. He will

spend substantial time in prison for killing Swenson regardless of the outcome of this

appeal.

Based on the rigor and limited application of our voluntary intoxication

jurisprudence where the defendant did not request a voluntary intoxication jury instruction,

I agree with the court that the evidence does not overwhelmingly support the reasonable

1 For purposes of both first-degree intentional felony murder and second-degree intentional murder, the State may prove intent by showing beyond a reasonable doubt that Colgrove had a purpose to kill Swenson or believed that his actions, if successful, would kill Swenson. See State v. Young, 710 N.W.2d 272, 278 (Minn. 2006) (citing Minn. Stat. § 609.02, subd. 9(4) (2022)). I conclude that the State did not prove beyond a reasonable doubt that Colgrove acted with the purpose to kill Swenson or believed that his act of stabbing would kill Swenson; my analysis applies equally to both bases for proving intent. I will generally refer to proof of intent.

D-1 inference that Colgrove’s voluntary intoxication precluded him from forming the specific

intent to cause the death of Swenson. Nothing in the record tells me that a reasonable

inference may be drawn that Colgrove was so dissociated from reality so as to

overwhelmingly show that he could not form the requisite intent to kill.

I disagree, however, with the court’s conclusion that the evidence does not support

a reasonable inference that Colgrove intended to assault Swenson but not kill her.

Accordingly, because the State did not demonstrate beyond a reasonable doubt that

Colgrove intended to kill Swenson when he stabbed her, I respectfully dissent. I would

reverse Colgrove’s conviction for first-degree intentional felony murder, vacate the guilty

verdict for second-degree intentional murder, 2 and remand to the district court to enter a

conviction on the second-degree unintentional felony murder count and sentence Colgrove

accordingly.

Our review of the sufficiency of the evidence proceeds against the fundamental

background principle that a person in this country is innocent until proven guilty on all

elements of a crime beyond a reasonable doubt. When there is direct evidence establishing

a particular element of a crime, we conduct “ ‘a painstaking analysis of the record to

determine whether the evidence, when viewed in the light most favorable to the conviction,

was sufficient to permit the jurors to reach the verdict which they did.’ ” State v. Ortega,

2 Based on its resolution of the case, the court does not address the guilty verdict on the charge that Colgrove committed second-degree intentional murder. Because both first-degree intentional felony murder and second-degree intentional murder require proof that Colgrove intended to kill Swenson, Minn. Stat. §§ 609.185(a)(3) and 609.19, subd. 1 (both requiring proof of “intent to effect the death of the person or another”), neither conviction may be sustained under my resolution of the issue presented in this appeal.

D-2 813 N.W.2d 86, 100 (Minn. 2012) (quoting State v. Webb, 440 N.W.2d 426, 430

(Minn. 1989)); see State v. Raymond, 440 N.W.2d 425, 426 (Minn. 1989) (“If the jury

could reasonably conclude a defendant had been proven guilty, giving due regard to the

presumption of innocence and the burden of proof beyond a reasonable doubt, this court

will not disturb the verdict.”). We do not disturb the verdict “if the jury, acting with due

regard for the presumption of innocence and for the necessity of overcoming it by proof

beyond a reasonable doubt, could reasonably conclude that a defendant was proven guilty

of the offense charged.” Bernhardt v. State, 684 N.W.2d 465, 476–77 (Minn. 2004)

(citation omitted) (internal quotation marks omitted).

When the evidence for an element of a conviction is based on circumstantial

evidence, the court uses the heightened two-step standard. State v. Andersen, 784 N.W.2d

320, 329 (Minn. 2010); see State v. Al-Naseer, 788 N.W.2d 469, 474–75 (Minn. 2010)

(stating that the heightened-scrutiny standard applies to any disputed element of the

conviction that is based on circumstantial evidence). Circumstantial evidence in this

context means that there is no direct evidence that a certain element of a crime was proved;

rather, the jury must look at the facts established by the evidence and then make an

inference from those facts to determine if the State proved the element. State v. Silvernail,

831 N.W.2d 594, 604 (Minn. 2013). A mens rea element, like proof of intent, is a classic

example of an element that is often proved by circumstantial evidence. State v. McInnis,

962 N.W.2d 874, 890 (Minn. 2021). A primary difference, then, between the direct

evidence test and the circumstantial evidence test is this: When there is direct evidence to

prove an element of a crime, we defer to the jury on all factual determinations. But when

D-3 the jury must make an inference about whether a fact necessary for a finding of guilt has

been proved, we review the inference independently and without deference to the jury.

Andersen, 784 N.W.2d at 329–30 (quoting State v. Stein, 776 N.W.2d 709, 716

(Minn. 2010), for the proposition that this court “give[s] no deference to the fact finder’s

choice between reasonable inferences”).

In short, when a jury’s verdict relies on direct evidence of a fact, we apply our

traditional test for sufficiency of the evidence assessed against the background rule of proof

beyond a reasonable doubt: Could a jury reasonably conclude that the defendant was

proven guilty of the offense charged? We defer to the jury’s fact determinations. But when

a fact necessary to a determination of guilt is not proved by direct evidence, but rather

requires the jury to make an inferential leap from directly proven facts to a fact for which

there is no definitive direct evidence, then we do not defer to the inference drawn by the

jury and we apply what we have called a “heightened” standard—heightened in the sense

that we do not defer to inferences drawn by the jury but rather make an independent

judgment about what reasonable inferences may be drawn from the facts proved. The

circumstantial evidence test was adopted to protect defendants in criminal trials by

ensuring the fundamental principle of proof beyond a reasonable doubt is honored.

The historic concern of courts in cases where a fact necessary for guilt requires an

inferential leap from the facts directly proved by the evidence to an ultimate factual

determination is that the inferences about guilt made by a jury run a greater risk that

prejudice, partiality, hastiness, and lack of due deliberation influenced the jury. State v.

Tscheu, 758 N.W.2d 849, 869–70 (Minn. 2009) (Meyer, J., concurring) (outlining the

D-4 history of the circumstantial evidence test both as a required jury instruction and as a

standard of appellate review); see State v. Harris, 895 N.W.2d 592, 599 n.4 (Minn. 2017)

(stating that the reason for heightened review in circumstantial evidence cases was concern

over the inferences made by the jury from facts proved). For the past several decades, we

have inserted ourselves as a responsible bulwark to ensure that when an accused person’s

guilt turns on inferential leaps drawn from the evidence, the evidence in the case affords

no other rational conclusion but that the person is guilty. As stated, our mechanism for

doing so is to set aside the deference we typically afford jury determinations when the

determination involves an inference from a directly proven fact. Again, in that case, we

afford the jury determination no deference. 3 In short, whether our concerns about jurors

making inferences from direct facts is well founded or not, 4 our job in circumstantial

3 I agree with the court that “ ‘the jury is in a unique position to determine the credibility of the witnesses and weigh the evidence before it’ ” and so we should defer to the jury’s credibility determinations. Supra at 9 (quoting State v. Harris, 895 N.W.2d 592, 600 (Minn. 2017)). But this deference is limited to determining which facts were proven by the evidence (the first step of the circumstantial evidence test). It does not extend to the inferences made by the jury. Andersen, 784 N.W.2d at 329–30. Indeed, I do not understand what it would mean to say that we defer to the credibility of inferences drawn; the whole reason evidence is considered circumstantial is that no witness is testifying to a fact. What credibility determination is being made? If we were to defer to the inferences made by the jury, the whole point of the circumstantial evidence test would be lost and the test would collapse back into the traditional direct evidence test. To be clear, the court in this case does not disagree with this principle. 4 Minnesota courts historically instructed juries that “to authorize a conviction, the circumstances should not only be consistent with the prisoner’s guilt, but they must be inconsistent with any other rational conclusion.” State v. Johnson, 35 N.W. 373, 376 (Minn. 1887). In State v. Turnipseed, 297 N.W.2d 308, 314 (Minn. 1980), we held that this jury instruction language is not constitutionally required. The language is not currently included in the standard criminal jury instructions. See Minnesota CRIM-JIG 3.03

D-5 evidence cases is to make sure that the beyond-a-reasonable-doubt standard is meaningful

in Minnesota.

Here, the parties do not dispute that the State’s evidence proving Colgrove’s intent

to kill is circumstantial evidence. (For instance, Colgrove did not confess that he intended

to cause Swenson’s death.) Likewise, the parties do not dispute that the circumstantial

evidence standard should apply.

The court accurately describes the general contours of the two-step circumstantial

evidence test. The first step of heightened circumstantial evidence review requires us to

identify the “circumstances proved” by the State, accepting the evidence that is consistent

with the guilty verdict and rejecting any evidence that is inconsistent with the verdict.

Andersen, 784 N.W.2d at 329 (quoting State v. Johnson, 217 N.W. 683, 684 (Minn. 1928)).

To me, it is clearer analytically to replace the phrase “circumstances proved” with the

phrase “facts proved.”

(reasonable doubt), 3.05 (defining direct and circumstantial evidence). Our jurisprudence of differing tests for direct-evidence and circumstantial-evidence cases is in large part a response to our decision to let the historical instruction fade away. Notably, we have not held that it would be error for a district court to provide the historical instruction to jurors or for the standard jury instructions to be updated to include the language. See State v. Fox, 868 N.W.2d 206, 222 (Minn. 2015). It may be wise for district courts to include the language in the instructions provided to jurors. One of the challenges of the circumstantial evidence test is that it places appellate courts in the awkward position of questioning the reasons for a jury’s inferences about guilt and requires us to delve into the minds of jurors and guess at what they actually considered in reaching their verdict. If jurors were given the historical instruction that proof beyond a reasonable doubt requires them to acquit when the direct evidence allows them to draw a rational inference that the accused was not guilty—even if they could also draw the conclusion that the facts proved could also support a guilty verdict, it would make more clear that the jurors relied solely on the evidence when they drew an inference that the accused is guilty. Indeed, this instruction may allow us to discard the circumstantial evidence test entirely.

D-6 Before moving on, it is important to keep clear one critical but often unstated point

about this first step: because the purpose of the heightened standard of review is to conduct

an independent, non-deferential review of the inferences made by the jury from the facts

proved, Harris, 895 N.W.2d at 599, the ultimate determination resulting from an inference

(for instance, an inference that the defendant intended to kill a person when there is no

direct evidence like a confession) cannot itself be a fact proved. As we stated in Harris,

our

[c]ircumstantial-evidence standard of review . . . is based on the fact that unlike direct evidence, [circumstantial evidence] requires an additional inference to establish guilt and therefore we have adopted a two-step process to account for the additional inference that must be made: a finding that an alleged fact (which does not by itself establish the required element) exists and then a conclusion that if the alleged fact exists, one can reasonably infer that the required element also exists.

895 N.W.2d at 599 n.4. In other words, the facts proved include only those facts

established by direct evidence.

The second step in the circumstantial evidence test requires us to consider the facts

proved from two perspectives. First, we consider whether the facts proved are sufficient

for a rational person to draw an inference supporting guilt. Al-Naseer, 788 N.W.2d at 478.

This is not a high bar. It is simply asking if a rational person (not every person or even a

majority or plurality of people) could draw the inference based on the known facts proved

by direct evidence. For instance, in this case, we must ask whether the facts proved are

sufficient to conclude that Colgrove intended to cause the death of Swenson. Of course, if

the facts proved are not sufficient to support an inference supporting guilt, the conviction

must be reversed.

D-7 If the facts proved are sufficient for a rational person to draw an inference supporting

guilt, then we must also consider whether the facts proved would allow a rational person

(again, not every person or even a majority or plurality of people) to draw an inference that

an element of the crime was not established. See id. at 478–79; cf. Ramos v Louisiana,

590 U.S. ___, 140 S.Ct. 1390, 1395–96 (2020) (stating that Sixth Amendment requires

unanimous verdicts). For instance, in this case, we must ask whether the facts proved

preclude the conclusion that Colgrove did not intend to cause the death of Swenson, but

rather that Colgrove intended his acts to harm or otherwise affect Swenson without killing

her or believing his acts would kill her. I do not read the court as disagreeing with any of

these introductory points.

I now turn to the specific facts in this case. I am in general agreement—albeit with

a few important nuances discussed below—with the court about the facts proved under step

one of the test. The facts proved—those that are not in conflict with the jury’s verdict—are

as follows: The officer heard a commotion in the victim’s house and observed a man and

woman struggling. The door to the house was locked and the officer forced his way into

the house. When the officer entered the house, he saw Colgrove with Swenson. At that

point, the officer told Colgrove to put his hands up and to let go of Swenson. Colgrove did

not comply with the officer’s orders and the officer tased Colgrove; the taser momentarily

affected him and he tensed up. Then, Colgrove got angry. Colgrove had a knife that was

similar to knives in Swenson’s kitchen drawer. He “took two swipes” at Swenson—in the

neck and the back—before fleeing the house. The back wound was 5 inches deep,

punctured several internal organs including the heart and lungs, and caused the death of

D-8 Swenson; the neck wound was 2 inches deep. A bloody knife, with a bent blade, was found

in the yard. Colgrove’s father found him between 3:00 and 5:00 a.m. in a nearby field.

Colgrove’s father stated that Colgrove admitted to his family that Colgrove went to the

house and that a “mean” girl was there.

I agree with the court that this set of facts proved is sufficient to support the

inference that Colgrove acted with intent to kill Swenson. 5 But I also conclude that this

same set of facts proved is rationally consistent with a conclusion that Colgrove intended

to harm—but not to kill—Swenson. For instance, the facts that Colgrove and Swenson

were struggling when the officer arrived and that the door was locked are not facts that are

exclusively associated with a person acting with intent to kill and never associated with a

person acting with intent to harm but not to kill. Those facts could be true in both

situations; they do not effectively distinguish between the potential two states of mind. The

same is true about the facts that Colgrove was angry after being tased and that Colgrove

fled. One response of a person who is angry is to assault someone without any intent to

kill. And certainly it is not unusual for a person who has harmed someone with a knife to

flee from the scene, especially when a police officer is present, even if they did not intend

to kill the victim. Again, those facts do not rationally distinguish between a person with

5 To support its rationale that the set of facts proved is sufficient to support the inference that Colgrove acted with intent to kill Swenson, the court relies on two cases—State v. Wolfe, 293 N.W.2d 41, 42 (Minn. 1980), and Raymond, 440 N.W.2d at 426. These cases only speak to the direct evidence standard, and the court appropriately relies on them solely to support its conclusion that one reasonable inference is that Colgrove attacked Swenson with an intent to kill. The cases do not support (and the court properly does not rely on them to support) the conclusion that another reasonable inference is that Colgrove did not have an intent to kill Swenson when he attacked her.

D-9 an intent to kill and a person with an intent to harm but not kill. And, of course, knives are

used to assault a person just like knives are used with intent to kill someone.

That analysis leaves us with the facts concerning the depth of the two stab wounds,

one of which was 5 inches deep and injured vital organs, and the existence of the bent knife

that the police found on the scene. I disagree with the court that these two facts exclude

any rational possibility that Colgrove intended to harm Swenson without intending to kill

her.

Turning first to the depth and impact of the wounds: It is not always the case that a

deep wound means a person intended to kill someone, nor is it always true that deep

wounds never occur when someone intends to harm a victim without an intent to kill the

victim. Indeed, in this case, the medical examiner testified for the State and without

contradiction that the “actual depth [of the wound] doesn’t necessarily tell me anything,”

and “[o]nce the knife breaks the skin, it’ll go in pretty easy.” The court’s conclusion that

only a person wielding a knife with an intent to kill could cause such a deep wound is

misplaced. Certainly, especially in light of the medical examiner’s testimony, the depth of

the wound does not preclude a reasonable hypothesis that Colgrove acted without an intent

to kill Swenson.

State v. Andrews, 388 N.W.2d 723, 728 (Minn. 1986), the case upon which the court

relies to support its argument that only a person intent on killing someone would inflict

two wounds, including a wound 5 inches deep, and that a person who only intended to

harm someone without killing them would never inflict such wounds, is readily

distinguishable. In Andrews, the defendant argued that he killed the victim “accidentally”

D-10 when the victim and he were struggling over the knife that the victim held in her hand.

388 N.W.2d at 728. That is not this case. Colgrove does not need to establish a rational

inference that the infliction of the wounds was accidental to prevail on his sufficiency of

the evidence claim. He can still prevail even if he intentionally stabbed Swenson as long

as it is reasonable to infer that he did so without an intent to kill.

In addition, unlike this case, in Andrews a medical examiner expert testified that the

victim was stabbed once with sufficient force to bend the knife blade and cut into a rib

bone and that the knife was repeatedly thrust in the wound. Id. As discussed above, the

medical examiner’s testimony in this case was materially different.

Finally, the fact that the police found a bent knife after Colgrove fled the scene does

not by itself exclude a rational inference that Colgrove intended to hurt but not kill

Swenson. In closing argument, the prosecutor told the jury that “the amount of force used

also bent the blade of the knife.” But all we know is that the knife was bent when the police

found it. There is no direct evidence that the knife was bent as a result of the force of the

stabbing—that assertion is only speculation and inference. Compare Andrews,

388 N.W.2d at 728 (stating that the medical examiner testified that the defendant used

sufficient force to bend the knife blade and cut into a rib bone). The knife may have been

bent before the stabbing. In other words, because there is no actual evidence that the knife

was bent during the stabbing, the assertion that the knife was bent as a result of the stabbing

is itself an inference and it is not permissible for us to consider as a “fact proved” under

the circumstantial evidence test. See State v. Cox, 884 N.W.2d 400, 412 (Minn. 2016).

Accordingly, it is impermissible to rely on the assertion that the knife was bent as a result

D-11 of the stabbing when assessing the reasonableness of the inference that Colgrove intended,

or did not intend, to kill Swenson, without first assessing whether the act of stabbing is

only reasonable inference as to the cause of the bend in the knife. 6 We must turn to the

second step of the circumstantial evidence test and ask (1) whether the inference that the

knife was bent during and as a result of the stabbing is reasonable and (2) whether the

inference that the knife was bent as a result of something else is also reasonable.

This case is somewhat unique because to reach the ultimate question of the

reasonableness of the inferences about Colgrove’s intent, an intermediate inference must

be made. On the one hand, if we credit the speculation or conjecture that the knife was

bent as a result of the force of the stabbing, then it makes the inference that Colgrove acted

with an intent to kill, considered in the totality of the facts proved, somewhat more

reasonable and the inference that Colgrove acted without intent to kill, again considered in

the totality of the facts proved, somewhat less reasonable. On the other hand, if we credit

the speculation or conjecture that the knife was bent before the stabbing or for some other

reason, then it makes the inference that Colgrove acted with an intent to kill, considered in

the totality of the facts proved, somewhat less reasonable and the inference that Colgrove

acted without intent kill, again considered in the totality of the facts proved, somewhat

6 It is analytically important that we do not conflate the analysis in the first and second steps of the circumstantial evidence test when it comes to inferences and speculation. In the first step of the analysis, the lack of direct evidence of when and how the knife was bent goes to the question of whether the cause of the bent knife is a fact proved. The lack of direct evidence concerning the cause of the bend in the knife—the need to “speculate, “conjecture,” “draw inferences” about that—means that we cannot credit speculation or inferences about the reason the knife was bent as a fact proved.

D-12 more reasonable. In other words, this is a situation of analyzing the reasonableness of an

ultimate inference as to a necessary element of a crime based on the reasonableness of an

intermediate inference of fact. Therefore, we must assess both the reasonableness of

competing inferences about how the knife was bent and also the reasonableness of the

competing inferences about whether Colgrove intended to kill Swenson or, alternatively,

Colgrove intended to harm Swenson without intent to kill. And because we assess the

reasonableness of the inferences based on the totality of the facts proved, in neither case

are the alternative inferences that the knife was bent as a result of the stabbing or that the

knife was bent some other way dispositive on the reasonableness of the ultimate alternative

inferences as to intent. If anything, the need to assess the ultimate inferences regarding an

element of a crime based upon the reasonableness of intermediate inference—the longer

string of inferences—makes it harder to exclude one of the ultimate inferences as

unreasonable and conclude that the State proved its case beyond a reasonable doubt.

I agree with the court that the first inference that the knife was bent as a result of the

stabbing is reasonable. The medical examiner did testify that the knife struck the vertebrae.

I do not agree with the court that the second inference—that the knife was bent as a

result of something other than the force of the stabbing—is unreasonable. The medical

examiner did not offer testimony that the knife bent as a result the stabbing. The medical

examiner also testified that “[o]nce the knife breaks the skin, it’ll go in pretty easy.” And

the officer described Colgrove’s actions as “swiping” at Swenson. There is nothing in the

record about the amount of force required to bend a knife through impact with vertebrae.

Further, photographs of the knife (particularly in the context of photographs of the wounds)

D-13 show that the bend in the knife is not such that it would prevent the piercing of the skin or

further penetration. 7 And if the tip of the knife was sharp enough to penetrate the vertebrae,

it could penetrate the vertebrae whether bent or not. Consequently, a rational person could

conclude that the bend in the knife may be the result of something other than the stabbing.

Relying on Al-Naseer, 788 N.W.2d. at 480, the court suggests that we should not

reach the question of the reasonableness of the inference that the knife was bent by some

other cause than the stabbing because that alternative inference is “speculation.” The court

reasons that there is one fact in the record to support the inference that the knife was bent

as a result of the stabbing (the medical examiner’s testimony that the knife hit Swenson’s

vertebrae) and there is no similar indirect evidence of some other cause for the bend in the

knife.

This argument is not convincing. While it is true that Colgrove does not point to

direct evidence explaining how the knife was bent, it is equally true that the medical

examiner’s testimony about the impact of the knife on the vertebrae is not direct evidence

that the knife was bent on impact with the bone. The court must continue to rely on its

own assumptions and inferences about what caused the bend in the knife. If we accept the

fundamental principle underlying the circumstantial evidence test—that we do not defer to

the jury regarding inferences drawn from direct evidence—we cannot logically or fairly

7 The court also suggests that no one would keep a knife with a bent tip in their kitchen drawer. I am not sure I agree or that we can assume that someone would necessarily throw away a kitchen knife with a bent tip. There is nothing in the record to support that assertion.

D-14 reject as “speculation” inferences drawn by the defendant and then turn around and accept

and rely upon parallel inferences drawn by the State from the same set of facts proved.

Zooming out again, then, to view the facts proved as a whole as the court properly

insists I must, State v. Hassan, 977 N.W.2d 633, 640 (Minn. 2022), and supra at 18 n.9

(citing Cox, 884 N.W.2d at 415), 8 the facts proved are as follows: When the officer arrived

on the scene, he saw Colgrove and Swenson struggling in the house and he forced his way

into the locked house. Upon entry, the officer saw Colgrove with Swenson, told Colgrove

to put his hands up and to let go of Swenson, and tased Colgrove when Colgrove did not

comply with his orders. Colgrove got angry. Colgrove had a knife from the kitchen

drawer. He “swiped” at Swenson twice and fled. Colgrove caused a 2-inch-deep wound

in Swenson’s neck, a 5-inch-deep wound in her back and punctured several internal organs.

But no facts proved establish the force with which Colgrove stabbed Swenson. Indeed, the

State’s expert testified without contradiction that the “actual depth [of the wound] doesn’t

necessarily tell me anything,” and “[o]nce the knife breaks the skin, it’ll go in pretty easy.”

Finally, a bent knife was later found near the house. But there are no facts proved as to

8 I disagree with the court’s contention that I am considering the facts piecemeal rather than as a whole. Perhaps the court is expressing concern that I closely examine two of the facts proved—whether the depth of the wounds and the fact that the knife was found bent in fact tell us anything definitive concerning the force or “brutal nature” of the stabbing—before returning to review the facts as a whole. But my analysis of undisputed testimony and evidence as part of the “painstaking analysis of the record” is consistent with our precedent. Ortega, 813 N.W.2d at 100. The court can disagree (in my view incorrectly) with my conclusion that looking at all of the facts proved, it is at least rational to infer that Colgrove did not act with the intent to kill Swenson. But it is factually wrong to say that I am not considering the facts as a whole. The court certainly does not disagree with my recitation of the proved facts.

D-15 how the blade was bent or if it was bent before the events in question. More specifically,

we do not know as a proven fact if the blade was bent as a result of the force of the attack.

Even if we accept that the blade was bent as a result of the force of the stabbing, in

light of the totality of the facts proved, that inferred fact does not foreclose the

reasonableness of a further inference that when Colgrove stabbed Swenson he intended to

hurt her but not kill her. A person who wants to hurt someone but without intent to kill or

belief that his actions, if successful, would kill Swenson, may stab a person with significant

force.

For the same reasons, I am not convinced by the court’s argument that the location

of the wounds and “brutal” and “fatal” nature of the wounds forecloses any rational

possibility that Colgrove did not intend to kill Swenson. Those two adjectives—brutal and

fatal—are doing a lot of analytical work in an attempt to make the court’s and the State’s

version of the narrative sound more compelling and Colgrove’s alternative explanation

sound irrational. But simply adding provocative adjectives to its analysis does not change

the facts proved or the conclusion that a rational possibility exists that Colgrove caused the

injuries to Swenson’s neck and back without intending to kill her. Notably, the wounds

inflicted on assault victims (when there is no intent to kill) are sometimes serious and

brutal. See Minn. Stat. §§ 609.221, subd. 1, 609.02, subd. 8 (2022) (together defining

first-degree assault as assaulting another person while inflicting “great bodily

harm”—bodily injury that “creates a high probability of death,” or that “causes serious

permanent disfigurement,” or that “causes a permanent or protracted loss or impairment of

the function of any bodily member or organ or other serious bodily harm”). And, indeed,

D-16 based on the undisputed testimony of the State’s medical examiner expert witness (the

“actual depth [of the wound] doesn’t necessarily tell me anything”) as well as the

description of the events by the officer (that Colgrove “swiped” at Swenson), the inference

that Colgrove acted brutally when he stabbed Swenson is itself a mere inference and

adjectival speculation.

I conclude that the story the proved facts tell supports both a rational inference that

Colgrove stabbed Swenson with an intent to kill her and a rational inference that Colgrove

stabbed Swenson with an intent to harm her but not an intent to kill her or belief that his

actions, if successful, would kill her. Stated more plainly, the same proved facts in the

record that support a reasonable inference that Colgrove intended to kill Swenson are also

proved facts in the record that support a reasonable inference that Colgrove did not intend

to kill Swenson. Under those circumstances, our case law requires us to hold—and more

importantly, our deep commitment to the principle that a person is innocent until the State

proves him guilty beyond a reasonable doubt compels us to hold—that the State did not

produce sufficient evidence to show that Colgrove committed first-degree intentional

felony murder or second-degree intentional murder beyond a reasonable doubt.

Certainly, Colgrove murdered Swenson through an act of violence. And my

conclusion in this case is that judgment should be entered on his conviction for, and

Colgrove sent to prison for, second-degree unintentional felony murder. But I disagree

that the State proved first-degree intentional felony murder or second-degree intentional

murder in this case.

D-17 The question before us in this case is not the reasonableness of the jury’s verdict but

whether reasonable doubt existed. We cannot lose sight of that constitutional dimension

of our circumstantial evidence case law. The answer does not turn on whether the State’s

narrative is more reasonable—more likely to be true—than that offered by the defendant.

Rather, the question is whether there is a rational possibility that the defendant’s version is

true—giving no preference to the jury’s presumptions about inferences. If the answer to

that question is “Yes,” then an acquittal is required even if the State’s narrative is more

believable and likely to be true. Under that test and consistent with our precedent, we must

reverse in this case. That is what we would demand from a jury and, because we have for

the last several decades thrust ourselves into that role, it is what we must demand from

ourselves even when it is uncomfortable.

For the reasons stated, I would reverse Colgrove’s conviction for first-degree

intentional felony murder, vacate the guilty verdict for second-degree intentional murder,

and remand to the district court to enter conviction on the charge of second-degree

unintentional felony murder and resentence Colgrove.

HUDSON, Chief Justice (dissenting).

I join in the dissent of Justice Thissen.

D-18