Robert John Kaiser v. State of Minnesota

STATE OF MINNESOTA

IN SUPREME COURT

A22-0749

Court of Appeals Moore, III, J.
Took no part, Procaccini, J.
Robert John Kaiser,

Respondent,

vs. Filed: March 13, 2024
Office of Appellate Courts
State of Minnesota,

Appellant.

________________________

Mark R. Bradford, Bradford Andresen Norrie & Camarotto, Bloomington, Minnesota; and

James R. Mayer, Great North Innocence Project, Minneapolis, Minnesota, for respondent.

Keith Ellison, Attorney General, Saint Paul, Minnesota; and

Janelle P. Kendall, Stearns County Attorney, Michael J. Lieberg, Chief Deputy County
Attorney, Saint Cloud, Minnesota, for appellant.

Mary Heath, Fredrikson & Byron, P.A., Minneapolis, Minnesota, for amicus curiae The
Innocence Network.
________________________

SYLLABUS

1. The test articulated in Larrison v. United States, 24 F.2d 82 (7th Cir. 1928),

applies when the State mistakenly presented false expert testimony about a medical fact that

its own expert later admitted could not have been true.
 2. The district court did not abuse its discretion when it granted the respondent

a new trial on the grounds of false testimony because the respondent met his burden under

the Larrison test.

Affirmed.

OPINION

MOORE, III, Justice.

Respondent Robert John Kaiser was indicted for, tried for, and convicted by a jury

of two counts of second-degree felony murder for the death of his 2-month-old son,

William. Kaiser sought postconviction relief, which the district court granted, based in

part on the fact that two expert witnesses testifying for the State of Minnesota presented

false evidence at Kaiser’s trial. The State argues that the district court abused its discretion

when it ordered a new trial for Kaiser. The court of appeals rejected that argument. We

likewise conclude that the district court correctly applied the governing Larrison test to

Kaiser’s false evidence claim and did not abuse its discretion in finding that the expert

testimony at issue was false and that the jury might have reached a different conclusion

without this testimony. Accordingly, we affirm.

FACTS

Kaiser was home alone caring for William on August 27, 2014, when the child

began experiencing medical distress. Later that day, Kaiser and William’s mother brought

William to the hospital. By the time William was admitted, he had begun experiencing

seizures. The cause of William’s condition was not immediately diagnosed. Doctors saw

2
a small bruise on William’s right jaw but no other external signs of trauma. A scan revealed

pools of blood between William’s brain and skull, but his skull was not fractured.

After examining William more closely, the doctors diagnosed him with traumatic

brain injury secondary to non-accidental trauma. Over the next several days, William’s

condition deteriorated. Doctors discovered that William had some healing rib fractures.

They also found extensive retinal hemorrhages in William’s eyes and macular schisis 1 in

his left eye. His abdomen became distended, and when doctors performed exploratory

surgery, they discovered serious gastrointestinal problems.

After a new scan revealed that William’s physical condition was rapidly

deteriorating, the doctors concluded that he was moving towards brain death. Soon after,

doctors discovered that half of his remaining small intestine was dead, and there was likely

not enough living material to absorb nutrients through feeding. Because of the “complete

condition” of William’s brain and intestinal injuries, doctors recommended that he be

removed from life support and allowed to die. After the ventilator was removed, William

died. An autopsy conducted by the medical examiner found that William’s death was

caused by traumatic brain injury and determined that the manner of death was homicide.

Following an investigation and an initial charge, the State submitted the case to a

grand jury. The grand jury indicted Kaiser for murder in the first degree while committing

child abuse with a past pattern of child abuse in violation of Minn. Stat. § 609.185(a)(5)

(2022) and two counts of second-degree felony murder, with third-degree assault and

1
Testimony given at Kaiser’s trial defined macular schisis as a “fold within the retina
which is filled with blood,” in other words, “a blood blister.”

3
malicious punishment of a child as the underlying felonies, in violation of Minn. Stat.

§ 609.19, subd. 2(1) (2022). 2 See also Minn. Stat. §§ 609.223 (2022), 609.377 (2022).

Kaiser pleaded not guilty, and the case proceeded to trial.

At Kaiser’s 4-week jury trial, two medical experts provided critical testimony about

the cause of William’s injuries. Dr. Jeffrey Lynch, an ophthalmologist who examined

William’s eyes while he was critically ill, testified for the State that he observed macular

schisis in the child’s left eye. He then explained the link between macular schisis and

abusive head trauma:

Q: With the macular schisis, . . . do you have an opinion as to whether
that is more associated with abusive head trauma?
....
A: When I see something like a macular schisis cavity, it’s—in seeing a
lot of kids’ eyes, there really isn’t any other type of cause of that, that
I’ve seen in my career.

The State also called Dr. Carl Schmidt, an anatomical, clinical, and forensic

pathologist, to testify about the connection between macular schisis and abusive head

trauma:

Q: Doctor, a macular schisis cavity was found in William’s left eye.
What does this mean to you?
A: That is a fold within the retina which is filled with blood. . . . It’s kind
of like a blister, a blood blister, within the retina.
....
Q: What does that mean to you in regards to William’s case?
A: Well, it’s the kind of thing you see with abusive head trauma.
Q: Do you see it in other cases?
A: I have.
Q: In other cases, other than abusive head trauma?

2
Second-degree felony murder is a form of unintentional murder where the defendant
causes the death of a person while committing or attempting to commit certain felony
offenses. Minn. Stat. § 609.19, subd. 2(1).

4
 A: No. I don’t think it’s been described outside of abusive head trauma.

Defense counsel did not ask Dr. Lynch or Dr. Schmidt whether macular schisis can

be caused by anything besides abusive head trauma. In its closing argument, the State

remarked that “[a] macular schisis cavity is almost diagnostic of abusive injury.”

The jury found Kaiser guilty of two counts of second-degree felony murder but

acquitted him on the first-degree murder charge. The district court convicted Kaiser on

one of the second-degree felony-murder counts and sentenced Kaiser to 240 months in

prison. Kaiser filed a direct appeal. The court of appeals affirmed his conviction and

sentence. State v. Kaiser, No. A17-0571, 2018 WL 2407187 (Minn. App. May 29, 2018).

We denied Kaiser’s petition for review.

Kaiser next petitioned for postconviction relief, requesting a new trial on the

grounds of newly discovered evidence, false evidence, and ineffective assistance of

counsel. The district court held a 9-day evidentiary hearing on Kaiser’s petition. During

the hearing, the following expert testimony was given by Dr. Lynch, the ophthalmologist

who testified at trial:

Q: You talked a little bit about macular schisis. You would agree that
there are numerous causes of schisis?
A: Yep.
Q: Do you recall telling the jury in Mr. Kaiser's case that [abusive head
trauma] is the only cause of macular schisis?
A: I don’t recall.
Q: If you did, would that be incorrect?
A: Yeah, that would be incorrect.
....
Q: Just a couple of more questions, Doctor. Is a macular schisis on it’s
[sic] own diagnostic of abusive head trauma?
A: No.

5
 Q: Is it almost diagnosis—pardon me, diagnostic of abusive head trauma,
a macular schisis?
A: Alone?
Q: Yes.
A: No.

Evidence was also presented that cerebral venous thrombosis appeared on William’s

brain scans. At Kaiser’s postconviction hearing, cerebral venous thrombosis was described

as the formation of blood clots in the veins surrounding the brain, which can lead to seizures

and brain injuries. Kaiser’s experts presented medical literature finding that the condition

can “mimic” the symptoms of abusive head trauma, whether or not any trauma occurred.

In his petition for postconviction relief, Kaiser argued that he was subjected to ineffective

assistance of counsel, in part, because his trial counsel did not investigate how the presence

of this condition could have served as a defense to allegations of abuse.

The district court granted Kaiser’s petition for postconviction relief in part, denied

it in part, and ordered a new trial. It found that the trial testimony of Dr. Lynch and Dr.

Schmidt went beyond mere opinion and stated a false medical fact: that macular schisis is

not caused by anything other than abusive head trauma. It also found that Kaiser’s trial

counsel unreasonably failed to investigate evidence of William’s cerebral venous

thrombosis that would have contested a critical part of the prosecution’s argument. In a

90-page order and memorandum, the district court denied the claims in Kaiser’s petition

asserting newly discovered evidence, but granted in part his petition and ordered a new

trial on the grounds of false testimony and ineffective assistance of counsel. The State

appealed. The court of appeals affirmed. Kaiser v. State, No. A22-0749, 2023 WL

1945553 (Minn. App. Feb. 13, 2023). We granted the State’s petition for review.

6
 ANALYSIS

In this appeal we are tasked with determining whether the district court abused its

discretion by granting Kaiser a new trial because of false expert testimony and ineffective

assistance of counsel at trial. In reviewing the district court’s grant of postconviction relief,

we will affirm the district court unless “it has exercised its discretion in an arbitrary or

capricious manner, based its ruling on an erroneous view of the law, or made clearly

erroneous factual findings.” Martin v. State, 969 N.W.2d 361, 363 (Minn. 2022) (quoting

Pearson v. State, 891 N.W.2d 590, 596 (Minn. 2017)) (internal quotation marks omitted).

I.

We first address the legal framework that guides our analysis here. The threshold

question is whether the district court, in its consideration of Kaiser’s claims of false expert

testimony, appropriately applied the Larrison test for false or recanted testimony as

opposed to the Rainer test for newly discovered evidence. We apply the Larrison test to

determine whether to grant a new trial based on claims of alleged witness recantation or

false trial testimony. See Andersen v. State, 940 N.W.2d 172, 178 (Minn. 2020) (citing

Larrison v. United States, 24 F.2d 82, 87–88 (7th Cir. 1928), overruled by United States v.

Mitrione, 357 F.3d 712, 718 (7th Cir. 2004)). 3 The test has three prongs: “(1) the court

must be reasonably well-satisfied that the testimony in question was false; (2) without that

testimony the jury might have reached a different conclusion; and (3) the petitioner was

taken by surprise at trial or did not know of the falsity until after trial.” Id. “The first two

3
Although Larrison was overruled in Mitrione, Minnesota courts continue to apply
its test for false testimony. See Ortega v. State, 856 N.W.2d 98, 103 n.6 (Minn. 2014).

7
prongs are compulsory, but the third prong is not required in order to grant a new trial.”

State v. Turnage, 729 N.W.2d 593, 597 (Minn. 2007).

In Rainer, we established a separate test for determining whether to grant a new trial

based on a claim of newly discovered evidence. Rainer v. State, 566 N.W.2d 692, 695

(Minn. 1997). To prevail on such a claim, a postconviction petitioner must show that the

evidence (1) was not known to the defendant or defense counsel at the time of the trial;

(2) could not have been discovered through due diligence before trial; (3) is not cumulative,

impeaching, or doubtful; and (4) would probably produce an acquittal or a more favorable

result. Id. A petitioner must establish all four prongs of this test to be entitled to

relief. Andersen, 940 N.W.2d at 178.

Virtually all our cases applying Larrison to claims of alleged witness recantation

involve fact witnesses. See, e.g., Whelan v. State, 214 N.W.2d 344, 344 (Minn. 1974)

(victim later claimed she lied at trial); Sutherlin v. State, 574 N.W.2d 428, 433 (Minn.

1998) (witnesses to shooting made inconsistent statements); Opsahl v. State, 677 N.W.2d

414, 423 (Minn. 2004) (several State fact witnesses described as a “chorus of liars” because

of their recantations); Andersen, 940 N.W.2d at 172 (defendant’s mother gave trial

testimony that was “self-contradictory” to her postconviction affidavit). Because we

generally look “with disfavor on motions for a new trial founded on alleged recantations,”

the first prong of the Larrison test is not met “by a simple statement contradicting earlier

testimony” or a determination that “a witness is generally unreliable.” Pippitt v. State, 737

N.W.2d 221, 227 (Minn. 2007) (citations omitted) (internal quotation marks omitted).

8
Rather, the court must be “reasonably certain that the recantation is genuine before the first

prong is met.” Id. (citation omitted) (internal quotation marks omitted).

In contrast to our cases dealing with the recantations of fact witnesses, in State v.

Caldwell, we applied the Larrison test for the first time in determining that the trial

testimony of an expert witness was false even though the expert did not recant. 322

N.W.2d 574, 587 (Minn. 1982). There, the State’s fingerprint expert introduced

uncontroverted testimony that Caldwell had left his fingerprint on an envelope, which was

the only evidence tending to establish that he was in Duluth when two murders were

committed there. Id. at 581. In the subsequent trial of a co-conspirator, three new experts

testified that the print could not have been Caldwell’s, and the prosecution asked the jury

to disregard the latent print as a misidentified fingerprint. Id. at 581–82. Caldwell later

challenged his conviction, which rested, in part, on the testimony that the fingerprint

belonged to him. Id. at 581, 587. Noting that the prejudicial effect of false testimony does

not depend on the witness’s state of mind, we evaluated the expert’s testimony in

Caldwell’s trial under the Larrison test even though the expert did not intend to make a

false statement but was simply mistaken. Id. at 586–87. Thus, “under the unusual

circumstances of th[at] case, where the uncontroverted testimony of the state’s expert

subsequently prove[d] to be incorrect and that testimony was the basis of the only

circumstantial evidence tending to establish” that the defendant was in town on the day of

the murder, we held that the defendant was entitled to a new trial. Id. at 587.

In the 4 decades since deciding Caldwell, we have not had occasion to discuss the

applicability of Larrison to allegedly false trial testimony from an expert witness. Here,

9
however, the district court relied on Caldwell and applied the Larrison test to determine

whether the expert testimony at Kaiser’s trial about the cause of macular schisis was false.

Consistent with Caldwell, the district court recognized that the statements of Dr. Lynch

and Dr. Schmidt need not have been perjured or fraudulent to be found false; in fact,

nothing in the record suggested that the State’s witnesses intended to present false

testimony at trial. Nevertheless, the district court characterized the testimony as “medical

fact” based on the totality of the State’s presentation, which suggested to the jury that

“macular schisis is only found where there is abuse.” Although the State argues that these

statements were mere opinions, which could not be false, and thus the Rainer test for newly

discovered evidence should control, we agree with the district court and the court of appeals

that the Larrison test applies here.

The record supports the district court’s finding that the disputed testimony given at

Kaiser’s trial was fact, not opinion. Indeed, the State itself treated Dr. Lynch’s testimony

that “there really isn’t any other type of cause” for macular schisis other than abusive head

trauma as factual in its closing argument, telling the jury that macular schisis is “almost

diagnostic of abusive injury.” This was not a so-called “battle of the experts,” in which the

jury was asked to believe the testimony of one expert over another, as there was no actual

scientific support for the proposition that macular schisis is exclusively caused by abuse.

Nor did Dr. Lynch’s qualifying language, “that I’ve seen in my career,” on its own, convert

his statement into an opinion, especially since it followed the phrase “in seeing a lot of

kids’ eyes.” There may be times when the line between fact and opinion is difficult to

discern, but the portrayal of this information as a representation of medical consensus gives

10
us reason to treat it as a statement of fact. See Obiago v. Merrell-Nat’l Lab’y, Inc., 560

So.2d 625, 628 (La. Ct. App. 1990) (“[A]t some point what was formerly merely an opinion

acquires such acceptance by the scientific community . . . that it can be considered as fact

. . . . For example, today we accept as fact what yesterday was only opinion—the existence

of atoms, DNA, and other now unquestioned truths.”). Thus, when viewed in the context

of the State’s full presentation of evidence, the testimony given by the experts here was

factual.

Further, Dr. Lynch in effect recanted his trial testimony. At trial, he testified that

“there really isn’t any other type of cause” for macular schisis other than abusive head

trauma. But at the postconviction hearing, Dr. Lynch admitted that there were many causes

of macular schisis and that he would have been incorrect if he had told the jury that abusive

head trauma is its only cause. This admission not only directly contradicts his trial

testimony but also amounts to a concession that “he was mistaken in his [trial] testimony”

that there is no other cause of macular schisis. See Caldwell, 322 N.W.2d at 587 (emphasis

omitted) (quoting Martin, 17 F.2d at 976) (internal quotation marks omitted). This

admission of mistake further shows that Larrison is the proper test here.

In the end, we conclude that the same confluence of “unusual circumstances” that

led us to apply the Larrison test to the expert testimony in Caldwell is present here. Here,

as there, the factual testimony of the experts “was damning—and it was false.” Caldwell,

322 N.W.2d at 586. This uncontroverted testimony was “subsequently prove[n] to be

incorrect” when the State’s expert in effect recanted his trial testimony at the

postconviction hearing. See id. at 587. Because the challenged testimony here was

11
demonstrably both factual and false, the Larrison test applies. The Rainer test, on the other

hand, is appropriate when new discoveries impugn testimony that was given as an opinion

or was factually accurate at the time of the trial. We therefore hold that the district court

did not err when it applied the Larrison test to the false statements of medical fact made at

Kaiser’s trial.

II.

Having decided that the Larrison test is the legal standard for this case, we now turn

to whether the district court abused its discretion in finding that Kaiser satisfied the

Larrison test and was entitled to a new trial. Larrison’s first prong requires that the court

be “reasonably well-satisfied” that the testimony given at trial was false. Opsahl, 677

N.W.2d at 423. After conducting a 9-day evidentiary hearing, the district court found

credible evidence that there were causes of macular schisis other than abuse. It supported

its findings with testimony from the State’s own experts at the postconviction hearing. 4 It

also relied on clinical reports, studies, and position statements from the scientific

community, which it documented extensively in its 90-page order. Nor does the State

dispute that macular schisis has other causes. Because of the postconviction testimony and

evidence the district court credited, the trial testimony of the State’s experts—that macular

schisis was only caused by abuse—could not have been true. The district court, therefore,

4
Although some of the evidence supporting Kaiser’s Larrison claim was obtained
through testimony given at the postconviction evidentiary hearing, the State had earlier
conceded in a letter that Kaiser met the burden of justifying an evidentiary hearing. We
thus express no opinion on the evidentiary adequacy of Kaiser’s petition for postconviction
relief.

12
did not abuse its discretion when it found itself reasonably well-satisfied that the trial

testimony was false.

The second Larrison prong requires Kaiser to demonstrate that without the false

testimony “the jury might have reached a different conclusion.” Opsahl, 677 N.W.2d at

423. By making no argument before us about this finding, the State has likely forfeited

any challenge to the district court’s conclusion on this prong. See State v. Jorgenson, 946

N.W.2d 596, 606 (Minn. 2020). Regardless, we find no abuse of discretion. The jury could

have found reasonable doubt in the State’s theory of guilt had it not been told that a crucial

piece of circumstantial evidence was “almost diagnostic” of Kaiser’s guilt. This testimony

from two medical experts had a powerful inculpatory effect in a case proven through

circumstantial evidence. The district court thus did not abuse its discretion in finding that

Kaiser established the second Larrison prong.

While not required, the third Larrison prong is satisfied when “the petitioner was

taken by surprise at trial or did not know of the falsity until after trial.” Anderson, 940

N.W.2d at 178. The district court found that Kaiser did not know at the time of trial that

the testimony was false because he had not consulted the kinds of specialists necessary to

explain the causes of macular schisis. 5 And although Kaiser likely knew the State would

offer evidence about the presence of macular schisis in William’s brain scans, nothing in

the record suggests Kaiser had reason to expect that the State would claim macular schisis

can be caused only by abuse or that he would know that this statement was false. Therefore,

5
As with the second prong, the State has made no argument before us about this
finding, and has likely forfeited this issue as well.

13
although this prong is not dispositive, we conclude that the district court did not abuse its

discretion in finding that Kaiser was surprised by the false testimony.

In summary, we hold that because Kaiser met his burden under the Larrison test,

the district court did not abuse its discretion in granting him a new trial. 6

* * *

Consistent with our decision in Caldwell, our holding is narrowly limited to the

combination of circumstances present here, and we do not extend Larrison to apply broadly

to all disputes over expert trial testimony. Here, the State’s experts made statements of

medical fact to the jury that proved crucial in establishing Kaiser’s guilt. The district court

conducted a thorough 9-day evidentiary hearing and gathered the facts necessary to

conclude that the trial testimony was false. And the State’s own witness in effect recanted

his trial testimony by saying later that it was incorrect. The unique cumulative effect of

these facts shows that Kaiser’s trial was prejudiced by false testimony, and the district court

therefore did not abuse its discretion by granting his petition for postconviction relief.

We recognize the tragedy that comes with the death of a child and the pain a

surviving family and community endure in experiencing such an unthinkable loss. We do

not affirm the reversal of a murder conviction lightly, and we make our decision realizing

a new trial will cause renewed pain for William’s family. On this record, however, we

6
The State also challenges the district court’s grant of postconviction relief on the
grounds of ineffective assistance of counsel. Like the court of appeals, we find that the
district court properly granted Kaiser a new trial on the grounds that false testimony was
presented at his trial, so we need not decide whether Kaiser received ineffective assistance
of counsel.

14
agree with the district court and the court of appeals that the right of a defendant to receive

a fair trial requires postconviction relief in this difficult case.

CONCLUSION

For the foregoing reasons, we affirm the decision of the court of appeals.

Affirmed.

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.

15