If this opinion indicates that it is “FOR PUBLICATION,” it is subject to
revision until final publication in the Michigan Appeals Reports.
STATE OF MICHIGAN
COURT OF APPEALS
PEOPLE OF THE STATE OF MICHIGAN, UNPUBLISHED
October 29, 2020
Plaintiff-Appellee,
v No. 340931
Van Buren Circuit Court
JASON BRENT KEISTER, LC No. 2016-020783-FC
Defendant-Appellant.
ON REMAND
Before: GLEICHER, P.J., and RONAYNE KRAUSE and O’BRIEN, JJ.
GLEICHER, J. (dissenting).
A jury convicted defendant Jason Keister of criminal sexual conduct in the first and second
degree. The complainant, age 11 at trial, testified that Keister repeatedly sexually assaulted her
two years earlier. Keister vehemently denied any sexual misconduct.
The trial was a classic credibility contest. Two expert witnesses testified on the
prosecution’s behalf: Dr. Angela May, a pediatrician specializing in child abuse and neglect
medicine, and Gloria Gillespie, a therapist specializing in counseling sexually abused children.
Dr. May examined the complainant, found no physical signs of abuse, and opined that based solely
on the child’s disclosures, “there was a high likelihood of abuse.” Dr. Gillespie described several
concepts related to child sexual abuse.
We affirmed Keister’s convictions on direct review. People v Keister, unpublished opinion
per curiam opinion of the Court of Appeals, issued May 16, 2019 (Docket No. 340931). The
Supreme Court vacated the part of our opinion addressing the admissibility of Dr. May’s expert
opinion regarding the likelihood of sexual abuse. People v Keister, __ Mich __; 940 NW2d 117
(2020), mod on recon 944 NW2d 721 (2020). The sole issue presented on remand is whether Dr.
May’s erroneously admitted opinion affected Keister’s substantial rights, warranting a new trial.
-1-
The majority affirms once again, proffering four reasons for rejecting that the improper
testimony tainted Keister’s trial: Dr. May was impeached to some extent, “the jury was powerfully
moved by the victim’s own testimony,” the complainant’s recounting of the events was consistent,
and others “corroborated” what the complainant said. I respectfully dissent.
I
Our Supreme Court has spoken in no uncertain terms: the admission of Dr. May’s opinion
constituted plain error. Id. The Supreme Court declared that May’s testimony obviously
contravened People v Smith, 425 Mich 98; 387 NW2d 814 (1986), People v Peterson, 450 Mich
349; 537 NW2d 857 (1995), and People v Thorpe, 504 Mich 230, 934 NW2d 693 (2019). Keister,
___ Mich at ___. This trio of cases generally stands for the proposition that experts may not bolster
the credibility of a complaining witness by vouching for his or her veracity. Smith, 425 Mich at
109, specifically prohibits an examining physician from offering an opinion regarding whether a
complainant had been sexually assaulted if that “conclusion [is] nothing more than the doctor’s
opinion that the victim had told the truth.”
Our task on remand is to examine whether the prejudice prong of plain-error review has
been satisfied. The question is easily answered in the affirmative. In People v Harbison, the
companion case to Thorpe, the Supreme Court unanimously held that the defendant had
established a virtually identical plain error and concluded that the error affected his substantial
rights, requiring a new trial. Thorpe, 504 Mich at 260. I am unable to distinguish this case from
Harbison.
This case, like Harbison, involved alleged sexual misconduct unwitnessed by anyone other
than the complainant and the defendant. Here, as in Harbison, a pediatric expert testified to a
likelihood of sexual abuse.1 Here, as in Harbison, no physical evidence supported the doctor’s
“diagnosis.” Here, as in Harbison, the expert relied on the child’s account of what had occurred
in reaching her ultimate conclusion of abuse. The expert witness in Harbison, as here, “reinforced
this plain error by claiming that her diagnosis derived from a ‘national [consensus] of
pediatricians[.]”2 Id. at 247. And here, as in Harbison, the prosecutor established the expert’s
credentials and experience in her field.3
1
In Harbison, the expert used the term “probable pediatric sexual abuse,” Thorpe, 504 Mich at
244, while Dr. May testified to “a high likelihood of abuse” despite that her notes referred to a
diagnosis of “probable” abuse. This is a distinction without a difference. Dr. May additionally
stressed that there was only “[a] small possibility” of “non-abuse.”
2
On redirect examination, the prosecutor questioned Dr. Mays extensively regarding the
“guidelines” she followed in reaching her conclusion of a “high likelihood of abuse.” Dr. May
offered that “[e]xperts in our field came up with these guidelines,” and agreed that she “was just
following them at the time” she examined the complainant.
3
Dr. May testified that she had been qualified as an expert in child abuse and neglect medicine
“maybe 20 times by now.”
-2-
The Supreme Court held in Harbison that the expert’s sexual abuse opinion contravened
Smith because it “was based solely on her own opinion that [the child’s] account of the assaults
was ‘clear, consistent, detailed and descriptive.’ ” Thorpe, 504 Mich at 262. The Court highlighted
that “[a]n examining physician’s opinion is objectionable when it is solely based ‘on what the
victim . . . told’ the physician. Such testimony is not permissible because a ‘jury [is] in just as good
a position to evaluate the victim’s testimony as’ the doctor.” Id. (alterations in original). Dr.
May’s testimony mirrored that of the expert in Harbison. Indeed, on redirect examination, Dr.
May explained that her “diagnosis” of “probable abuse” rested on the expert guidelines’
designation of that term when a complainant is “able to provide a clear and descriptive and detailed
account of what happened[.]”
The Supreme Court had no difficulty concluding that the plainly erroneous admission of
the expert’s opinion testimony affected Harbison’s substantial rights, necessitating a new trial. Id.
The Court reasoned as follows:
[T]he instant case is largely a credibility contest. The only evidence against
Harbison was [the child’s] uncorroborated testimony. To bolster its case, the
prosecution presented testimony from a pediatrician who is board-certified in child
abuse pediatrics; who is currently a medical director at the Safe Harbor Children’s
Advocacy Center; who has examined, in her estimate, thousands of children that
have been sexually abused; and who has testified as an expert witness in 32
counties. Given the lack of compelling testimony that forms the basis for the
verdict and the plainly erroneous testimony that [the child] suffered “probable
pediatric sexual abuse,” we conclude that the plain error affected Harbison’s
substantial rights. [Id. at 264.]
Dr. May does not have precisely the same credentials as did the expert in Harbison, but otherwise
the above summary fits this case to a tee.
In Harbison, the Court underscored the impact of the improperly admitted testimony by
observing that the “error strikes at the heart of several important principles underlying our rules of
evidence” because it “not only had the effect of vouching for [the child’s] credibility, but” also
invaded the jury’s province “to determine the only issue in the case.” Id. at 264-265. The Court
summarized, “This improperly admitted testimony very likely bolstered [the child’s] credibility
and affected the verdict. We conclude that the gravity of this significant error seriously affected
the integrity of Harbison’s trial.” Id. at 265-266.
The crux of the Supreme Court’s analyses in Harbison and Thorpe applies with equal force
here: in the absence of corroborating medical evidence, an expert’s opinion that sexual abuse
occurred is overwhelmingly prejudicial. Dr. May’s “diagnoses” rested entirely on her assessment
of the child’s credibility. When a highly qualified and unchallenged expert aligns her opinion on
one side of a pure credibility contest, that opinion likely carries great weight with a jury, verging
on the indisputable establishment of a medical fact. Guided by Thorpe and Harbison, I would
hold that reversal is warranted.
-3-
II
The majority’s attempts to distinguish this case from Thorpe/Harbison are legally
unpersuasive and factually inconsistent with the record.
The majority begins by attempting to portray Dr. May’s testimony as inconsequential
because she was “impeached.” I read the record differently. The majority admits that Dr. May
“had no basis for her diagnosis other than what the victim said.” Yet Dr. May never retreated from
her conclusion; she augmented its validity on redirect, reiterating that the complainant’s story was
credible and sufficient to support the “diagnosis.”
As the majority correctly observes, Dr. May admitted during cross-examination that she
did not personally take a history from the complainant; rather, she interviewed the complainant’s
mother. A social worker on Dr. May’s team obtained information from the child and conveyed it
to Dr. May. The majority finds this evidence impeaching, but I believe that it actually enhanced
the power of the bolstering. Dr. May testified that she did not need to personally interview the
child because as an expert, she believed that the child’s story rang true. The message plainly
telegraphed to the jury was that as an experienced and trained physician relying on “expert
guidelines,” Dr. May knew that the complainant was credible. She did not need to judge the child’s
demeanor, to hear Keister’s side, or to test the evidence in any other way. Rather, Dr. May told
the jury that despite the absence of any corroborating physical evidence, as an expert in child
sexual abuse she could readily determine that the complainant was telling the truth.
The majority rejects the notion that Dr. May’s testimony could have had any meaningful
force in this case, expressing “faith” in “the jury’s ability to think for itself[.]” “Faith” in a jury
regularly gives way to gatekeeping under MRE 105, MRE 403, MRE 404, MRE 407, MRE 410,
MRE 411, and MRE 702—to name but a few rules restricting the admission of otherwise relevant
evidence. Despite our abiding “faith” in a jury’s ability to weigh the evidence and to judge
credibility, for policy reasons the rules of evidence remove from the jury’s consideration certain
facts because they may “weigh too much” and “overpersuade.” Michelson v United States, 335
US 469, 476; 69 S Ct 213; 93 L Ed 168 (1948). Courts uniformly recognize that expert testimony
in particular may have an outsized impact on a jury, filling in compelling gaps in the evidence or
doubts about credibility with the authoritative say-so of a highly educated and experienced
scientist.
The reason parties call experts is that experts know things that lay people don’t. Experts
offer opinions based on deep experience that go beyond first-hand observation—the backbone of
most other testimony. Lay people don’t interview child victims of sex abuse. Experts like Dr.
May do. Lay people lack knowledge of the psychological characteristics of abused children.
Experts like Dr. May are trusted repositories of such knowledge, especially when officially deemed
“expert” by a court. “To a jury recognizing the awesome dilemma of whom to believe, an expert
will often represent the only seemingly objective source, offering it a much sought-after hook on
which to hang its hat.” People v Peterson, 450 Mich 349, 374; 537 NW2d 857 (1995) (quotation
marks, citation, and emphasis omitted). As a matter of policy, Michigan courts protect the integrity
of the trial process in cases involving child sexual abuse by withholding from a jury’s consideration
relevant credibility evidence that may factor too powerfully in the jury’s judgment. This rule
-4-
safeguards faith in the system’s fairness—that the jury will do its job in a way that fully protects
the rights of all participants.
The majority’s next determination, that some jurors’ alleged reaction to the complainant’s
testimony is of “great importance” and mitigates against prejudice, is equally meritless. During
his closing argument, defense counsel said:
I want to touch on the idea of this reasonable doubt. I talked about no second
thoughts and it is hard, I know, some of you were very emotional when [the
complainant] testified. I hope you would be. I am not saying it is easy for her, I
am not saying she is a manipulative little person who is just out to get somebody, I
am saying she is a child.
This observation does not erase or even erode the prejudicial impact of Dr. May’s testimony. That
some jurors displayed sympathy or concern for the complainant is normal, expected, anticipated,
and appropriate. An expression of human feeling on the part of a few jurors is not a marker of
whether improperly admitted evidence affected the fairness, integrity, or public reputation of the
proceedings.4
According to the majority, reversal is additionally unwarranted because the complainant’s
recounting of what happened was consistent and “corroborated” by several witnesses other than
Dr. May. The record does not support the majority’s view. The complainant was impeached with
evidence that she had described what had happened quite differently to her mother than she had to
her friend. And no one “corroborated” the complainant’s story. Other witnesses repeated what
the complainant told them; most were impeached to some extent by contradictions in their
recollections when measured against the complainant’s testimony. The majority’s insinuation that
a witness “corroborated” the fact of the abuse is flatly wrong and misleading.
The trial court plainly erred by admitting expert testimony that improperly vouched for the
complainant’s credibility. As in Thorpe/Harbison, this was a pure credibility contest in which
vouching by an expert witness likely tilted the scales in the prosecution’s favor. I would hold that
the inadmissible evidence seriously prejudiced Keister and mandates a new trial.
/s/ Elizabeth L. Gleicher
4
Inexplicably, the majority characterizes counsel’s closing argument as “evidence.” There is no
evidence regarding the “effect” of the complainant’s testimony on the jury. Counsel’s statement
is an expression of counsel’s opinion, nothing more.
-5-