Jennifer Lynn Baldwin, petitioner, Appellant,

This opinion is nonprecedential except as provided by
Minn. R. Civ. App. P. 136.01, subd. 1(c).

STATE OF MINNESOTA
IN COURT OF APPEALS
A25-0555

Jennifer Lynn Baldwin, petitioner,
Appellant,

vs.

State of Minnesota,
Respondent.

Filed February 2, 2026
Affirmed in part, reversed in part, and remanded
Cochran, Judge

Hennepin County District Court
File No. 27-CR-18-23849

Michael J. Kane, Stuart A. Nelson , Kiersten M. Batzli, Alexander Rafferty, Fish &
Richardson P.C., Minneapolis, Minnesota; and

Grayson P. Sundermeier (pro hac vice), Fish & Richardson P.C., Wilmington, Delaware;
and

James Mayer, Anna McGinn, Great North Innocence Project, Minneapolis, Minnesota; and

Mark R. Bradford, Bradford, Andresen, Norrie, & Camarotto, Bloomington, Minnesota
(for appellant)

Keith Ellison, Attorney General, St. Paul, Minnesota; and

Mary F. Moriarty, Hennepin County Attorney, Adam E. Petras, Senior Assistant County
Attorney, Minneapolis, Minnesota (for respondent)

Mary E. Heath, Margaret G. Severson, Fredrikson & Byron, P.A., Minneapolis, Minnesota
(for amicus curiae The Innocence Network)

Considered and decided by Frisch, Chief Judge; Worke, Judge; and Cochran, Judge.
2
NONPRECEDENTIAL OPINION
COCHRAN, Judge
This is the third appeal arising from appellant’s conviction of two counts of felony
second-degree murder in connection with the death of a nine-month-old child. This appeal
concerns appellant’s second petition for postconviction relief, which seeks to have her
conviction vacated based on claims of false testimony. The petition asserts two
claims: (1) the respondent’s medical witnesses falsely testified that the child died from
abusive head trauma, previously referred to as shaken baby syndrome, and could not have
died from natural causes (false-medical-evidence claim), and (2) a detective falsely
testified on behalf of the respondent that appellant exhibited certain behavioral indicators
during her 911 call that are known indicators of guilt (911-call-analysis claim).
The postconviction court determined that appellant’s false-medical-evidence claim
is procedurally barred under Knaffla1 and that no exception to Knaffla applies. On this
basis, the court concluded that appellant is not entitled to an evidentiary hearing on her
false-medical-evidence claim. Regarding appellant’s 911-call -analysis claim, the
postconviction court determined, after an evidentiary hearing, that this claim is also
Knaffla-barred and, in the alternative, fails on the merits.
Appellant argues that the postconviction court abused its discretion by denying both
claims. We agree with appellant as to her false-medical-evidence claim because, under the
unique circumstances of this case, the record reflects that the interests-of-justice exception

1 See State v. Knaffla , 243 N.W.2d 737, 741 (Minn. 1976) (precluding review of
postconviction claims that were known but not raised on direct appeal).
3
to Knaffla applies and that appellant is entitled to an evidentiary hearing on her false-
medical-evidence claim. But, with regard to appellant’s second claim, we conclude that
the postconviction court acted within its discretion when it determined that appellant’s 911-
call-analysis claim does not merit any relief. We therefore affirm in part, reverse in part,
and remand for an evidentiary hearing on appellant’s false-medical-evidence claim.
FACTS
The following summarizes the lengthy procedural history of this case. After the
death of nine- month-old C.S. in March 2018, respondent State of Minnesota charged
appellant Jennifer Lynn Baldwin with one count of second-degree murder while
committing third-degree assault causing substantial bodily harm and one count of second-
degree murder while committing third-degree assault of a victim under the age of four. See
Minn. Stat. §§ 609.19, subd. 2(1), .223, subds. 1, 3 (2016). At trial, the state presented
testimony of medical experts to support its theory of the case that the cause of death was
blunt force trauma to C.S.’s head inflicted while C.S. was in Baldwin’s care. The state’s
medical experts also testified that there was no possible explanation for C.S.’s death based
on natural, nontraumatic causes. Baldwin did not call any experts at trial because her
retained expert fell ill prior to trial and the district court denied her request for an open-
ended continuance. In 2019, a jury found Baldwin guilty of both counts. The district court
imposed an executed sentence of 255 months’ imprisonment, an upward durational
departure, on count two—second- degree murder while committing third-degree assault of
a victim under four. See Minn. Stat §§ 609.19, subd. 2(1), .223, subd. 3. Baldwin filed a
direct appeal which was stayed to allow her to petition for postconviction relief.
4
First Postconviction Proceeding and Prior Two Appeals
Baldwin petitioned for postconviction relief on the grounds of ineffective assistance
of counsel. She argued that “her attorneys’ failure to call an expert witness to refute key
aspects of the state’s theory” constituted deficient performance because “[t]he case against
[her] relied almost entirely on expert testimony.” She argued that this deficient
performance prejudiced her because a defense expert could have effectively refuted the
state’s experts’ testimony. Baldwin asserted that a defense expert could have testified to
“extensive scientific literature” supporting her theory that C.S. might not have been
immediately symptomatic, suggesting that his injuries could have occurred before he
arrived at Baldwin’s home, and supporting other possible explanations that might have
accounted for C.S.’s injuries including medical conditions or an accident. The state
responded that Baldwin’s petition failed on the merits because Baldwin’s trial counsel
acted in an objectively reasonable manner, as counsel contracted with the National Autopsy
Assay Group (NAA Group) for an expert (although the expert did not testify) and because
Baldwin’s arguments pertained to trial strategy. The state emphasized that the difficulty
in obtaining the expert’s presence at trial was due to the lead expert falling ill just before
trial and the expert-consulting group’s inability to complete the report in time—“two things
outside of trial counsel’s control.” The state also maintained that Baldwin was not
prejudiced by her trial counsel’s performance because Baldwin was able to raise her
defense at trial—that someone else injured C.S. —and did not establish that an expert
making the same argument would have affected the outcome.
5
Prior to filing her petition, Baldwin filed an application for funding under Minnesota
Statutes section 611.21 (2020). The funding was intended to allow Baldwin to obtain and
consult with a forensic pathologist with expert knowledge on interpretation of autopsy
results and medical records to support her postconviction petition. Before requesting the
funding, Baldwin’s postconviction counsel spoke to an expert, who “mentioned . . . that
C.S.’s subdural hematoma could have been caused by factors other than blunt force
trauma,” but the expert could not provide an opinion without reviewing C.S.’s entire
medical file. The postconviction court denied the request for funding.
And, following an evidentiary hearing, the postconviction court also denied
Baldwin’s first petition for postconviction relief, concluding that Baldwin failed to meet
both prongs of the Strickland
2 test to demonstrate ineffective assistance of counsel. T he
court first concluded that, in totality, Baldwin’s trial counsel’s performance “was far above
an objective standard of reasonableness.” As to the second Strickland prong, the
postconviction court concluded that it was “unlikely that the trial outcome would have been
different even if [Baldwin] had secured an expert witness to testify on her behalf at trial.”
After Baldwin was denied postconviction relief, we reinstated her appeal . State v.
Baldwin, No. A19-1677, 2022 WL 1132049, at *1 (Minn. App. Apr. 18, 2022) (Baldwin
I), rev. denied (Minn. July 19, 2022). In her direct appeal, Baldwin asserted that she was

2 Under Strickland, a defendant must prove that (1) “counsel’s representation fell below an
objective standard of reasonableness” and (2) “there was a reasonable probability that, but
for counsel’s errors, the result of the proceeding would have been different.” State v. Nicks,
831 N.W.2d 493, 504 (Minn. 2013) (citing Strickland v. Washington, 466 U.S. 668, 687-
96 (1984)).
6
entitled to a new trial because (1) the district court abused its discretion when it denied her
request for a continuance of trial after her expert fell ill; (2) the postconviction court abused
its discretion when it concluded that she failed to meet the Strickland test for ineffective
assistance of counsel; and (3) the postconviction court erred by denying her postconviction
request for funds to retain an expert witness to support her claim of ineffective assistance
of counsel. Baldwin I, 2022 WL 1132049, at *1-8. Under our deferential standard of
review, we first concluded that Baldwin did not meet her burden to demonstrate that the
district court’s decision to deny her request for an open-ended continuance was a clear
abuse of its discretion. Id. at *4-5. We explained that the district court thoughtfully
considered and weighed the circumstances surrounding her request for an open-ended
continuance, noting that Baldwin had not adequately explained why another expert from
the NAA Group could not testify. Id. at *5. We also noted that Baldwin was able to make
her central argument at trial—that another caretaker was responsible for C.S.’s injuries—
and that “Baldwin has presented no evidence that the NAA Group expert, who was
hospitalized, would have provided testimony that would have materially affected the
outcome of the trial.” Id.
Next, in analyzing her argument regarding the postconviction petition, we
concluded that the postconviction court did not abuse its discretion by denying Baldwin’s
ineffective-assistance-of-trial-counsel claim. Id. Baldwin argued that she received
ineffective assistance because her attorneys did not call an expert at trial. Id. at *6. Giving
deference to the postconviction court’s credibility determinations, we agreed with the
postconviction court’s conclusion that Baldwin did not meet her burden to prove that her
7
attorneys’ conduct fell below an objective standard of reasonableness. Id. at *5-6. W e
concluded that “the record demonstrates that Baldwin’s attorneys undertook reasonable
efforts to retain and present an expert witness at trial,” even though a defense expert
ultimately did not testify. Id. at *7. Because Baldwin did not establish ineffective
assistance of counsel, we did not decide whether the postconviction court abused its
discretion when it denied Baldwin’s request for funds to retain an expert in postconviction
proceedings. Id. at *7-8.
For these reasons, we affirmed Baldwin’s convictions. But we remanded for
resentencing because the district court abused its discretion when it imposed an upward
sentencing departure based on the victim’s particular vulnerability due to age, as the
victim’s age was an element of the underlying offense. Id. After resentencing, Baldwin
again appealed her sentence, arguing that the district court abused its discretion by still
imposing an upward durational departure but on a different basis; we reversed and
remanded for imposition of the presumptive sentence. See State v. Baldwin, No. A22-
1838, 2023 WL 5198738, at *1-2 (Minn. App. Aug. 14, 2023) (Baldwin II).
Second Petition for Postconviction Relief (This Appeal)
In October 2023, following our decision in Baldwin I, Baldwin filed a second
petition for postconviction relief. Baldwin asserted that (1) the state’s medical witnesses
falsely testified that the child could not have died from natural causes and (2) a detective
falsely testified on behalf of the state about what is known about the ability to determine
the guilt of a 911 caller based on certain behavioral indicators. With respect to her false-
medical-evidence claim, Baldwin referenced statements from two state medical examiners
8
and a child-abuse pediatrician who testified at trial for the state. Balwin argued that these
medical witnesses falsely testified that “C.S. died from abusive, blunt force head trauma
and dismissed the possibility that C.S. died from nontraumatic causes of death, including
disease.” Baldwin submitted six expert reports to support her claim that “C.S. very likely
died from natural causes.”
As to her 911-call-analysis claim, Baldwin asserted that “the [s]tate falsely told the
jury that Ms. Baldwin was guilty because of various verbal and behavioral indicators
displayed during her call to 911.” For instance, Baldwin pointed to the detective’s trial
testimony where he explained that he confronted Baldwin about her 911 call when
interviewing her because he “attended training and [he] certainly [has] a great deal of
experience interpreting 911 calls and it is known that when a caller is concerned for
somebody’s condition first and foremost they ask for help for them.” He added that
“[c]allers who contributed to that poor condition offer explanations which may draw
attention elsewhere, distracters, things that may explain what happened before they actually
ask for any help or assistance for the injured person.” Baldwin submitted an expert report
from a psychologist demonstrating that the detective’s testimony related to 911 call
analysis was false. Based on these claims of false testimony, Baldwin requested that her
conviction and sentence be vacated.
The postconviction court originally granted Baldwin an evidentiary hearing on both
claims, but then, after receiving correspondence from the state asserting the false-medical-
evidence claim is subject to the Knaffla procedural bar, denied that claim on Knaffla
9
grounds.3 The postconviction court concluded that Baldwin’s false-medical-evidence
claim is procedurally barred under Knaffla and neither exception to Knaffla applies to
permit an evidentiary hearing. But the postconviction court reaffirmed its grant of an
evidentiary hearing as to Baldwin’s 911-call-analysis claim.
At the evidentiary hearing, Baldwin called an expert witness who explained the
research surrounding 911 call analysis and demonstrated that the technique is not reliable
or valid. The postconviction court also received 11 exhibits, including nine scientific
papers about 911 call analysis.
After the hearing and post-hearing briefing, the court denied Baldwin’s second
petition for postconviction relief. Although no party argued that Knaffla barred the 911-
call-analysis claim, the postconviction court concluded that Baldwin’s 911-call-analysis
claim is procedurally barred under Knaffla and does not fall under any exception. The

3 The circumstances under which the postconviction court originally granted a hearing on
both claims and then later denied the false-medical -evidence claim as Knaffla barred are
as follows. On June 26, 2024, the postconviction court contacted the parties and requested
that they file letter briefs by the next morning clarifying whether the state was asserting
that Baldwin’s claims are Knaffla-barred because the state previously reserved arguments
regarding Knaffla. On June 27, before the state responded, the postconviction court filed
an order granting an evidentiary hearing on both claims and declin ing to address the
Knaffla bar. Regarding Baldwin’s false-medical-evidence claim, the postconviction court
determined that the Larrison test applies, and that Baldwin is entitled to an evidentiary
hearing because the record does not conclusively show that Baldwin is entitled to no relief.
Regarding the 911 -call-analysis claim, the court granted a hearing on the basis of newly
discovered evidence. Later the same day, the state filed correspondence, arguing that
Baldwin’s false-medical-evidence claim is Knaffla-barred. In July, after giving Baldwin
an opportunity to respond to the state’s letter and taking the state’s argument under
advisement, the postconviction court filed an order concluding that Baldwin’s false-
medical-evidence claim is Knaffla-barred and continuing the hearing on the other claim to
a later date.
10
postconviction court also determined in the alternative that “Baldwin’s claim fails under
the Larrison test” for false evidence.
This appeal follows.
DECISION
Baldwin challenges the postconviction court’s denial of both the false-medical-
evidence claim and the 911-call -analysis claim. A person convicted of a crime may seek
postconviction relief by filing a petition alleging that the conviction “violated the person’s
rights under the Constitution or laws of the United States or of the state.” Minn. Stat.
§ 590.01, subd. 1(1) (2022). “Unless the petition and the files and records of the proceeding
conclusively show that the petitioner is entitled to no relief, the court shall promptly set an
early hearing on the petition.” Minn. Stat. § 590.04, subd. 1 (2022). We review the denial
of a petition for postconviction relief, including the denial of an evidentiary hearing, for an
abuse of discretion, reviewing the postconviction court’s factual findings for clear error
and its legal conclusions de novo. Tichich v. State, 4 N.W.3d 114, 119 (Minn. 2024).
I. The postconviction court abused its discretion by denying Baldwin a hearing
on her false-medical-evidence claim.

Baldwin argues that the postconviction court abused its discretion by determining
that her false-medical -evidence claim is procedurally barred under Knaffla and the
interests-of-justice exception to Knaffla does not apply. She further contends that because
there is no procedural bar, she is entitled to an evidentiary hearing on her false-medical -
evidence testimony. We agree.
11
A. The false-medical-evidence claim is not procedurally barred.
Knaffla precludes review of “all claims raised in the direct appeal and all claims that
were known or should have been known but were not raised in the direct appeal.”
Colbert v. State, 870 N.W.2d 616, 626 (Minn. 2015); see also Minn. Stat. § 590.01, subd. 1
(2022) (“A petition for postconviction relief after a direct appeal has been completed may
not be based on grounds that could have been raised on direct appeal of the conviction or
sentence.”). There are two exceptions to the Knaffla procedural bar: “(1) if a novel legal
issue is presented; or (2) if the interests of justice require review.” Gilbert v. State,
2 N.W.3d 483, 487 (Minn. 2024).
Here, the postconviction court determined that Baldwin’s false-medical-evidence
claim was Knaffla-barred because “when testimony was provided by the [s]tate’s medical
experts at trial, Baldwin knew or should have known that those opinions were disputed by
other experts within the field and those claims should have been brought in the previous
postconviction and appellate proceedings.” The postconviction court further concluded
Baldwin’s claim did not fall under either of the two exceptions to Knaffla.
Baldwin disagrees with the postconviction court’s determination that her false-
medical-evidence claim is Knaffla-barred and the determination that the interests-of-justice
exception does not apply. She argues that her false-medical-evidence claim is not
Knaffla-barred because she “could not have known that the [s]tate’s testimony was false at
the time of her direct appeal because she was denied the services of an expert during trial
and on appeal.” Alternatively, Baldwin argues that the postconviction court abused its
discretion in determining that she does not meet the interests-of-justice exception, and she
12
maintains that “fairness requires that the [false-medical-evidence claim] be reviewed in the
interests of justice.” At its core, Baldwin’s argument is that fairness requires that her false-
medical-evidence claim be reviewed in the interests of justice because, until now, she has
been without an expert to counter the state’s evidence.
Baldwin’s alternative argument has merit. Assuming without deciding that the
postconviction court correctly determined that Baldwin’s false-medical-evidence claim is
Knaffla-barred, we conclude that the postconviction court abused its discretion in its
analysis of the interests-of-justice exception and that the interests -of-justice exception
applies to her false-medical-evidence claim.
“To qualify under the interests-of- justice exception, a petition for postconviction
relief must have substantive merit and [the] petitioner must not have deliberately and
inexcusably delayed in bringing the claim.” Id. at 488. The petitioner has the burden of
demonstrating that this exception to Knaffla applies. Id. at 489. “[T]he claim is
procedurally barred unless [the petitioner] can show why fairness requires its
consideration.” Sanders v. State, 628 N.W.2d 597, 601 (Minn. 2001).
Here, the postconviction court determined that the interests-of-justice exception did
not apply to Baldwin’s false-medical-evidence claim because “the postconviction court and
the appellate court have already addressed these claims” and “determined that there was
no error by the district court in denying Baldwin’s request for [a] continuance to retain
expert witnesses and that Baldwin failed to establish that having defense experts at trial
would have affected the outcome.” The postconviction court relied on our decision in
13
Baldwin I to support its conclusion. But the postconviction court’s analysis is based on a
misreading of Baldwin I and reflects an abuse of discretion.
i. Baldwin I does not foreclose application of the interests-of-justice
exception.

Our decision in Baldwin I was limited to the issues raised on appeal and did not
consider whether the state presented false testimony requiring a new trial. Relevant here,
we considered only (1) whether the district court abused its discretion by denying
Baldwin’s pretrial motion for a continuance and (2) whether the postconviction court
abused its discretion by denying Baldwin’s ineffective-assistance- of-trial-counsel claim.
Baldwin I, 2022 WL 1132049, at *4-7. With regard to the first issue, as noted above, w e
concluded that the district court did not abuse its broad discretion by denying Baldwin’s
motion for a continuance after her expert witness with the NAA Group was hospitalized
because Baldwin’s request for a continuance was open-ended , and Baldwin had not
adequately explained why another expert from the NAA Group could not testify. Id. at *5.
Additionally, we emphasized that “Baldwin has presented no evidence that the NAA Group
expert, who was hospitalized, would have provided testimony that would have materially
affected the outcome of the trial if he had been available to testify.” Id. Our analysis was
based on Baldwin’s defense at trial—that a different careta ker was responsible for C.S.’s
injuries. Id.
In reviewing Baldwin’s ineffective-assistance-of-counsel argument, we concluded
that Baldwin did not meet her burden under the two- prong test set forth in Strickland,
466 U.S. at 687. Id. at *5-8. We limited our analysis to the first prong and concluded that
14
Baldwin failed to demonstrate that counsel’s performance fell below an objective standard
of reasonableness. Id. at *7. Consequently, we affirmed the postconviction court’s
decision rejecting Baldwin’s ineffective-assistance-of-counsel claim on the first prong and
did not reach the second prong, that considers whether “a reasonable probability exists that
the outcome would have been different but for counsel’ s errors.” Id. at *6. We also did
not reach the issue of whether the postconviction court abused its discretion by denying
Baldwin’s request for funds to retain an expert in postconviction proceedings. Id. at *8.
Our conclusions in Baldwin I regarding the denial of the continuance and the
ineffective-assistance-of-counsel claim were limited by the applicable standards of review
and Baldwin’s defense that a different caretaker who was with C.S. earlier in the day was
responsible for C.S.’s injuries. Id. at *5. While the issues in Baldwin I concerning the
unavailability of Baldwin’s expert witness may be related to Baldwin’s interests-of-justice
exception argument, there was no assertion on appeal that C.S. died from natural causes.
Accordingly, in Baldwin I, we did not consider whether expert testimony that C.S.’s
injuries could have been due to natural causes would have changed the outcome at trial.
The postconviction court’s reading of Baldwin I to suggest the contrary is inconsistent with
our opinion. Because a close reading of Baldwin I reflects that this court has not already
addressed her claim that the state’s experts presented false testimony and that the outcome
would have been different without that testimony, we conclude that the postconviction
court erred when it relied on our decision in Baldwin I to deny Baldwin a hearing on her
false-medical-evidence claim under the interests-of-justice exception.
15
ii. The record reflects that the interests -of-justice exception applies
and the postconviction court abused its discretion by concluding
otherwise.

As noted above, “[t]o qualify under the interests-of-justice exception, a petition for
postconviction relief must have substantive merit and [the] petitioner must not have
deliberately and inexcusably delayed in bringing the claim.” Gilbert, 2 N.W.3d at 488. In
considering this exception, the postconviction court did not address whether Baldwin’s
false-medical-evidence claim has substantive merit, instead erroneously focusing on
Baldwin I. But the postconviction court did, in effect, conclude that Baldwin deliberately
and inexcusably delayed in bringing her false-medical-evidence claim. Because the record
reflects the opposite— that Baldwin did not deliberately or inexcusably delay in bringing
her claim and her claim has substantive merit—we conclude that the postconviction court
abused its discretion when it determined that the interests -of-justice exception does not
apply.
Question of Delay
In its order, the postconviction court found that “Baldwin’s counsel chose not to
raise the claim of false testimony at the first postconviction proceedings or on direct
appeal” and on that basis concluded that “Baldwin has not shown extenuating or unforeseen
circumstances that make her failure to bring this claim at an earlier stage excusable.” But
our review of the record shows that Baldwin’s failure to raise her false-medical-evidence
claim in her direct appeal or previous postconviction proceeding cannot be characterized
as “deliberat[e] and inexcusabl[e].” Id.
16
The supreme court’s decision in Gilbert is instructive. In Gilbert, the supreme court
concluded that a postconviction petition did not meet the interests-of-justice exception
because “[t]he scientific basis for Gilbert’s false testimony claim —a study from 2015
referenced in a preexisting BCA training manual—was discoverable at the time the direct
appeal was filed.” Id. at 489. The supreme court reasoned that “Gilbert’s appellate counsel
was aware of the issue of potential false testimony and chose not to raise the issue, although
counsel could have with reasonable due diligence investigated and brought such a claim in
a postconviction proceeding filed in connection with his direct appeal.” Id.
In contrast here, it was impossible for Baldwin’s counsel to be aware of the scientific
basis for the false-medical-evidence claim during the previous postconviction proceeding
or direct appeal without the opinion of an expert. The district court approved funding for
an expert witness and Baldwin sought an expert witness from the NAA Group, but Baldwin
was ultimately without that expert at trial because the expert was hospitalized a few weeks
before trial. The expert had not yet completed his analysis and produced a report because
samples from the medical examiner’s office were not ready to send to him until around the
time he was hospitalized. And although we previously concluded that Baldwin did not
adequately explain why another expert from the NAA Group could not testify instead, this
does not change the fact that she was without a forensic pathologist to counter the state’s
medical evidence at trial. The record also reflects that, while her trial counsel was generally
aware of the evidence related to C.S.’s respiratory illness and had knowledge of general
disputes about abusive head trauma (leading counsel to move for a Frye/Mack hearing),
17
her trial counsel did not receive an expert report or an initial assessment offering an opinion
about the cause of C.S.’s death prior to trial.
Similarly, Baldwin’s postconviction counsel submitted an affidavit explaining that
she could not have been aware of the false-medical -evidence claim at the time of the first
postconviction proceeding. After consulting with a forensic pathologist, Baldwin’s
postconviction counsel was made generally aware that C.S.’s subdural hematoma could
have been caused by other factors, but her request for funding to obtain an expert on this
topic was denied. Because the forensic pathologist could not provide an opinion,
postconviction counsel could not make any meaningful arguments to assert a claim about
non-traumatic medical causes for C.S.’s death. Therefore, the postconviction court abused
its discretion when it found that Baldwin’s counsel deliberately and inexcusably failed to
raise the false-medical-evidence claim.
Substantive Merit of Her Claim
When analyzing whether the interests-of-justice exception applies, the
postconviction court did not consider whether Baldwin’s false-medical-evidence claim has
“substantive merit.” Id. at 488. Presumably, the postconviction court failed to do so
because it erroneously concluded that this court had previously addressed the merits of the
claim in Baldwin I. For the reasons discussed below in section II, we conclude that the
allegations in Baldwin’s petition, including reports from multiple medical experts, when
viewed in the light most favorable to Baldwin have substantive merit. The postconviction
court previously found as much. Accordingly, Baldwin’s false-medical-evidence claim
has substantive merit.
18
Conclusion
In sum, Baldwin demonstrated to the postconviction court that fairness requires
consideration of her false-medical- evidence claim set forth in her second postconviction
petition. This case presents a unique set of circumstances where the district court’s
previous decision denying Baldwin a continuance and the postconviction court’s decision
denying Baldwin funds for an expert at the postconviction stage, although not reversed on
appeal, likely resulted in her not being able to bring this claim any earlier. And because
Baldwin’s false-medical-evidence claim has substantive merit and Baldwin’s failure to
raise the claim earlier was not deliberate and inexcusable, we conclude that the
postconviction court abused its discretion by not applying the interests-of-justice exception
to the Knaffla procedural bar for this claim.
B. Baldwin’s petition meets the requirements for an evidentiary hearing on
the false-medical-evidence claim.

Having concluded that Baldwin’s false-medical-evidence claim is not procedurally
barred by Knaffla because the interests-of-justice exception applies, we turn to whether
Baldwin is entitled to an evidentiary hearing. The postconviction statute provides that
“[u]nless the petition and the files and records of the proceeding conclusively show that the
petitioner is entitled to no relief, the court shall promptly set an early hearing on the petition.”
Minn. Stat. § 590.04, subd. 1.
Baldwin asserts that she is entitled to an evidentiary hearing on her false-medical-
evidence claim because, before denying the claim as Knaffla-barred and concluding that
no exception applies, the postconviction court had already determined that “[t]he record
19
does not conclusively show that Baldwin is entitled to no relief .” Specifically, the
postconviction court concluded that Baldwin was entitled to a hearing because the petition
alleged facts that, if true, would satisfy the Larrison test for a new trial based on false
testimony. In response, the state argues that the Larrison test is not applicable. Instead,
the state contends that the Rainer test for newly discovered evidence applies, and Baldwin
is not entitled to an evidentiary hearing under that test. We agree with Baldwin and the
postconviction court that the Larrison test applies, and that Baldwin’s petition satisfies the
test such that she is entitled to an evidentiary hearing.
i. The postconviction court properly determined that the Larrison
test for false testimony applies.

To decide whether Baldwin is entitled to an evidentiary hearing, we must first
determine which legal standard governs the false-medical-evidence claim. Tichich,
4 N.W.3d at 121. We apply the Larrison test to claims of false or recanted testimony when
the challenged testimony is “demonstrably both factual and false.” Kaiser v. State,
4 N.W.3d 95, 103 (Minn. 2024); see 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).4 But we
apply the Rainer test to claims of newly discovered evidence “when new discoveries
impugn testimony that was given as an opinion or was factually accurate at the time of the
trial.” Kaiser, 4 N.W.3d at 103; see Rainer v. State, 566 N.W.2d 692, 695 (Minn. 1997).

4 “Although Larrison was overruled in Mitrione, Minnesota courts continue to apply the
Larrison test in cases involving witness- recantation and false -testimony claims.”
State v. Duol, 25 N.W.3d 135, 139 n.4 (Minn. 2025) (quotation omitted).
20
The supreme court’s recent decision in Kaiser guides our analysis. In Kaiser, the
supreme court concluded that the postconviction court properly applied the Larrison test
to the petitioner’s false-testimony claim because the disputed testimony—that “there really
isn’t any other type of cause” for macular schisis other than abusive head trauma and “I
don’t think [macular schisis has] been described outside of abusive head trauma”— was
treated as fact, not opinion. Kaiser, 4 N.W.3d at 99, 102. When concludi ng that the
testimony in Kaiser was fact, not opinion, the supreme court reasoned that the prosecutor
adopted the statement as factual in the state’s closing argument, and emphasized “[t]his
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.” Id. at 102. The court also
noted that the witness’s “qualifying language, ‘that I’ve seen in my career,’ on its own,
[did not] convert his statement into an opinion, especially since it followed the phrase ‘in
seeing a lot of kids’ eyes.’” Id. The supreme court ultimately stated that “[t]here 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 us reason to treat it as a
statement of fact.” Id. at 102-03.
Like the testimony at issue in Kaiser, the state portrayed the medical testimony at
issue here as fact and medical consensus, rather than opinion. At trial, one of the medical
examiners testified that C.S.’s brain had a hemorrhage, and the cause was a blunt force
injury. He also agreed that there was “no explanation, based on [his] examination of the
brain for the hemorrhage, . . . that would have been naturally occurring or a disease
21
process.” He further testified that, “there was not a competing non- injurious explanation
for the hemorrhage. That in a 9-month-old is very concerning and rendered the manner of
death as homicide.” Similarly, another medical examiner testified that he knew “of no
medical intervention” that would leave a bruise on the left side of C.S.’s head. And a child-
abuse pediatrician testified that C.S.’s injuries were caused by abusive head trauma. He
testified that C.S. “clearly had trauma, severe trauma,” and that C.S.’s injury was not
consistent with any accidental mechanism as “there is no history of any trauma in this child
who would not be able to inflict this amount of injury on himself.” The state treated these
statements as factual in its closing argument, telling the jury: “[t]hey said that this injury
was the result of a significant violent trauma.” And “[w]hat the medicine can tell you is
that this was violent. This head injury was severe, it was catastrophic, and it was almost
immediately fatal.”
The state argues that Rainer, not Larrison, is the proper standard because Baldwin’s
new expert opinions set forth in the postconviction petition merely offer a different view
of the same evidence presented at trial, and “there is no suggestion in [Baldwin’s] case that
any of the [s]tate’s . . . medical experts would testify any differently now than they testified
at trial.” We disagree.
When viewed in the light most favorable to Baldwin, the new expert opinions
demonstrate that medical testimony by the state’s witnesses at Baldwin’s trial was false
because Baldwin’s experts’ opinions support a conclusion that natural causes may have
caused C.S.’s medical condition and death. For instance, one of Baldwin’s experts
concluded “that there is a very high probability that all the medical findings in [C.S.] were
22
caused by his sever e chronic lung inflammation causing respiratory failure and
subsequently heart arrest and, as a consequence of that, a hypoxic ischaemic brain injury
(HII).” The same expert concluded that, “[t]here are no medical findings that can be taken
as proof of violence.” Another expert concluded that “[C.S.] would have bruised easily
given his condition,” and “that the bruising on [C.S.’s] body documented in the autopsy
report cannot be attributed to trauma.” Baldwin’s experts do not merely offer a different
view of the evidence at trial. Rather, they call into question the veracity of the state
witnesses’ testimony that traumatic injury was the only possible cause of death and there
was no natural cause. Moreover, there was no “battle of the experts” at Baldwin’s trial;
the state adopted the testimony as fact and portrayed the conclusions as a medical
consensus. See id. at 102.
We also note that recantation by the state’s witnesses is not required for Baldwin to
prevail, and any recantation by the state’s witnesses would likely not happen until an
evidentiary hearing. See id. at 103 (stating trial testimony of expert witness can be false
even if witness did not recant and noting that witness recanted trial testimony at
postconviction evidentiary hearing). We therefore conclude that the postconviction court
properly determined that the Larrison test for false testimony applies to Baldwin’s false-
medical-evidence claim.
ii. Viewing the facts alleged in the light most favorable to Baldwin,
the false-medical-evidence claim meets the Larrison test.

Under the Larrison test to be entitled to a new trial: “(1) the court must be
reasonably well-satisfied that the testimony in question was false; (2) without that
23
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. at 101 (quotation
omitted). “The first two prongs are compulsory, but the third prong is not required in order
to grant a new trial.” Id. (quoting State v. Turnage, 729 N.W.2d 593, 597 (Minn. 2007)).
“When applying the Larrison factors to determine whether to grant an evidentiary
hearing, the postconviction court must assume the truth of the allegations in the petition.”
Ortega v. State, 856 N.W.2d 98, 103 (Minn. 2014); see also Minn. Stat. § 590.04, subd. 1.
Baldwin’s “burden of proof for a postconviction evidentiary hearing is lower than [her]
burden for a new trial.” Ferguson v. State, 779 N.W.2d 555, 561 (Minn. 2010) (quotation
omitted). And any “doubts about whether to conduct an evidentiary hearing should be
resolved in favor of granting the hearing.” Id. (quotation omitted). The supreme court has
“noted that evidentiary hearings are particularly appropriate when the petition attacks
important evidence in a circumstantial case.” Opsahl v. State, 677 N.W.2d 414, 423 (Minn.
2004).
In analyzing the Larrison factors in its July 2024 order, the postconviction court
determined that all three factors were met. The state asserts that Baldwin cannot satisfy
the first Larrison prong, but it does not make any argument regarding the second prong.
Baldwin maintains, consistent with the postconviction court’s analysis, that she is entitled
to an evidentiary hearing because the Larrison test is satisfied. We agree.
First Prong: False Testimony
The state argues that the trial testimony is not false because it is inconsistent only
with the new expert testimony. This argument is unavailing as discussed above. Other
24
experts’ later contrary testimony can establish falsity if it goes beyond mere difference of
opinion to demonstrate that the original testimony “could not have been true.” Kaiser,
4 N.W.3d at 103- 04. As the postconviction court determined, Baldwin submitted expert
reports, not merely disagreeing with the state’s evidence, but claiming that the evidence
offered by the state’s experts at trial was false. And, as noted above, the state’s evidence
was portrayed as fact and medical consensus. See id. at 102-03.
Second Prong: Impact
The second Larrison prong requires Baldwin to show that without the false medical
testimony, “the jury might have reached a different conclusion.” Id. at 104 (quotation
omitted). This prong does not involve a “sufficiency of the evidence inquiry,” although it
includes a consideration of whether other evidence was “overwhelming.” Martin v. State,
825 N.W.2d 734, 743 (Minn. 2013). “‘[M]ight’ means something more than an outside
chance although much less than ‘would probably.’” Ortega, 856 N.W.2d at 104 (quotation
omitted). The postconviction court concluded in its July 2024 order that the jury might
have reached a different conclusion, reasoning, “The testimony from the [s]tate’s medical
experts had a powerful inculpatory effect in this case proven through circumstantial
evidence.” The finding is supported by the record, as the record demonstrates that the
medical evidence related to C.S.’s injury was critical to the state’s case and the state’s
medical evidence allowed it to argue to the jury that there was only one explanation—
abusive head trauma— for C.S.’s death.
25
Third Prong: Surprise
The third Larrison prong is not required for Baldwin to be entitled to an evidentiary
hearing but is merely “relevant” to a false- evidence claim. Martin , 825 N.W.2d at 740.
The postconviction court explained that “it is likely that Baldwin did not know at the time
of trial that the testimony was false as she was not able to consult with an expert.” This
finding is also supported by the record. As set forth above, Baldwin was likely aware that
the state would present general testimony about abusive head trauma. But that does not
mean that Baldwin was aware that the state would present testimony that C.S.’s injuries
could not have been from natural causes or disease, or more importantly, that she was aware
such testimony was false, as she did not have the benefit of an expert opinion or report.
In short, because Baldwin alleged facts, that when taken as true, meet each prong of
the Larrison test and any “doubts about whether to conduct an evidentiary hearing should
be resolved in favor of granting the hearing,” we conclude that Baldwin is entitled to an
evidentiary hearing on her false-medical-evidence claim. Ferguson, 779 N.W.2d at 561
(quotation omitted). The postconviction court abused its discretion by denying this claim
without an evidentiary hearing. We therefore conclude that reversal and remand for an
evidentiary hearing on Baldwin’s false-medical-evidence claim is required.
II. The postconviction court acted within its discretion when it concluded that
Baldwin is not entitled to a new trial based on her 911-call-analysis claim.

Baldwin next argues that the postconviction court abused its discretion by
determining that her 911-call-analysis claim is Knaffla-barred and, in the alternative, fails
on the merits. The state concedes that it did not assert that Knaffla bars Baldwin’s 911-
26
call-analysis claim and also concedes that it agreed to a contested evidentiary hearing.
Accordingly, the state contends that we should address Baldwin’s 911-call-analysis claim
on the merits only and further argues the postconviction court did not abuse its discretion
in denying a new trial on that claim because Baldwin’s claim fails under the Larrison test.
For the reasons explained below, we agree with the state that the postconviction court
properly denied the 911-call -analysis claim on the merits under the Larrison test and
therefore do not reach the question of whether Baldwin’s claim is also procedurally barred
by Knaffla.
We review the postconviction court’s application of the Larrison test for an abuse
of discretion. See Kaiser, 4 N.W.3d at 103 (determining whether postconviction court
abused its discretion in finding appellant satisfied the Larrison test and was entitled to a
new trial). As explained above, under the Larrison test for a new trial based on false
testimony: “(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. at 101 (quotation omitted). The postconviction court determined that the first
prong was met but the second and third prongs were not met. Consequently, the
postconviction court denied Baldwin’s request for a new trial based on this claim.
As to the first Larrison prong, the state argues the postconviction court abused its
discretion when it determined this prong was met. The state argues that the detective’s
testimony was not false because the state did not present scientific 911 call analysis at
Baldwin’s trial but only asked the detective “to describe many of the different statements
27
that [Baldwin] made and how those statements . . . demonstrated [Baldwin’s] overall
consciousness of guilt.” “ A respondent can raise alternative arguments on appeal in
defense of the underlying decision when there are sufficient facts in the record for the
appellate court to consider the alternative theories, there is legal support for the arguments,
and the alternative grounds would not expand the relief previously granted.”
State v. Grunig, 660 N.W.2d 134, 137 (Minn. 2003). W hile the state may not have
intended to present false testimony and used the detective’s testimony for other purposes,
such as to support its consciousness- of-guilt argument, we disagree with the state’s
assertion that it did not present false testimony about scientific 911 call analysis at
Baldwin’s trial. See Kaiser, 4 N.W.3d at 102 (stating that state’s witnesses did not need to
intend to present false testimony and testimony does not have to be “perjured or fraudulent
to be found false”).
The detective testified that “it is known when a caller is concerned for somebody’s
condition first and foremost, they ask for help for them.” (Emphasis added.) Th e state
presented this uncontroverted statement as factual, but it is now demonstrably false. At the
postconviction evidentiary hearing, Baldwin’s expert testified that the statement was false
because “all of the studies show that there’s no relationship between the timing of a plea
for help and guilt.” We therefore conclude that the postconviction court properly found
that the first Larrison prong was satisfied.
5

5 Baldwin contends that the postconviction court erred when it found that her trial featured
only one false statement related to the timing of the plea for help. The postconviction court
did not explicitly find that the second portion of the detective’s statement—“Callers who

28
We next consider the postconviction court’s determination on the second Larrison
prong—that it “[was] not convinced that without the false testimony, or any of the
challenged portions of the trial, the jury might have reached a different conclusion.”
Baldwin contends that the postconviction court clearly erred in its determination because
“the false [911-call-analysis] testimony was unequivocal, incriminating, and bore directly
on the statements and evidence that followed.” The postconviction court found that the
second Larrison prong was not satisfied because the challenged testimony was a small
portion of the detective’s trial testimony, several witnesses testified about when C.S .’s
injury could have occurred, Baldwin made various statements to law enforcement over the
course of the investigation, and the state’s case relied primarily on medical expert
testimony. The postconviction court also noted that the detective conceded on
cross-examination that Baldwin did request an ambulance at the beginning of the 911 call,
which conflicted with the detective’s suggestion that Baldwin was guilty because she did
not immediately make a plea for help. These findings are reasonably supported by the
record and are not clearly erroneous.

contributed to that poor condition offer explanations which may draw attention elsewhere,
distracters, things that may explain what happened before they actually ask for any help or
assistance for the injured person”—was also false. We conclude that this was error, as it
is undisputed this statement is also false. Baldwin’s expert testified that “subsequent
studies have, again, shown that the relationship between providing extraneous information
and guilt is not significant, or unreliable.” But we are not persuaded that the postconviction
court clearly erred by not finding more of the challenged statements alluding to 911 call
analysis false and finding that “[m]uch of the testimony challenged by Baldwin is not about
the 911 call but rather about statements Baldwin made during other interrogations and
interactions with law enforcement.”
29
Lastly, regarding the third Larrison prong, the postconviction court reasoned that
Baldwin was unlikely taken by surprise because her trial counsel did not object to the
detective’s testimony, effectively cross-examined the detective, and had access to the
interrogation video when the 911 call was mentioned. We note that, while these findings
are supported by the record, Baldwin may not have known of the falsity until after trial as
only one article questioning the validity of the 911 call analysis existed at the time of trial
and the theory has only more recently been conclusively debunked. But this is not a basis
to reverse, as the third Larrison prong is not required, but is merely “relevant” to a
false-evidence claim. Martin, 825 N.W.2d at 740.
We therefore conclude that the postconviction court did not abuse its discretion by
determining that, when the false 911-call-analysis testimony is viewed in isolation, the
second Larrison prong was not met because “significant additional evidence” of Baldwin’s
guilt was presented at trial. Ortega, 856 N.W.2d at 104. Consequently, the postconviction
court properly exercised its discretion by denying Baldwin’s 911-call-analysis claim and
determining that she was not entitled to a new trial based on false 911-call -analysis
testimony.
We observe that if the postconviction court, on remand, concludes after an
evidentiary hearing that the first Larrison prong is met for Baldwin’s false- medical-
evidence claim, that conclusion might affect the postconviction court’s earlier
determination that the second Larrison prong was not satisfied for the 911-call-analysis
claim. In that event, t he postconviction court must analyze the combined impact of both
instances of false testimony in analyzing the second Larrison prong on remand. See
30
Caldwell v. State, 853 N.W.2d 766, 776 (Minn. 2014) (stating that “the second [Larrison]
prong examines the combined impact on the verdict of all of the allegedly false trial
testimony”). Accordingly, if the postconviction court concludes on remand that Baldwin’s
false-medical-evidence claim meets Larrison’s first prong, we instruct the postconviction
court to reconsider whether without both the false medical evidence and the false 911 call
testimony, the jury might have reached a different result, in considering whether to vacate
the conviction and order a new trial.
Affirmed in part, reversed in part, and remanded.