NOT RECOMMENDED FOR PUBLICATION
File Name: 21a0398n.06
No. 20-4222
UNITED STATES COURT OF APPEALS
FOR THE SIXTH CIRCUIT FILED
Aug 24, 2021
MELISSA DOVALA, ) DEBORAH S. HUNT, Clerk
)
Petitioner-Appellee,
)
) ON APPEAL FROM THE
v.
) UNITED STATES DISTRICT
) COURT FOR THE NORTHERN
TERI BALDAUF, Warden,
) DISTRICT OF OHIO
Respondent-Appellant. )
)
BEFORE: GRIFFIN, WHITE, and READLER, Circuit Judges.
GRIFFIN, Circuit Judge.
On the morning of February 6, 2004, Eileen Callahan-Smath dropped off her five-month-
old son Riley at petitioner Melissa Dovala’s home daycare. When picked up in the afternoon,
Riley was “lifeless” and “blue,” so Callahan-Smath rushed him to a nearby hospital. He died
shortly upon arrival. The coroner determined the cause of Riley’s death was blunt impact trauma
to the head that occurred within just a few hours of death.
An Ohio jury convicted Dovala of felony murder, felonious assault, endangering children,
and involuntary manslaughter, and she was sentenced to an indefinite prison term of fifteen years
to life. She contends her trial counsel was ineffective because he elected not to pursue an expert
who would have challenged the coroner’s cause-of-death and time-of-injury conclusions. The
Ohio Court of Appeals disagreed, reasoning her attorney’s decision to aggressively cross-examine
the prosecution’s medical witnesses instead of engaging in a battle-of-the-experts defense fell
No. 20-4222
Dovala v. Baldauf
within the ambit of Strickland v. Washington’s deference to counsel’s trial strategy. 466 U.S. 668,
687–91 (1984). Over a magistrate judge’s recommendation to the contrary, the district court found
this to be an unreasonable application of Strickland and granted a conditional writ of habeas
corpus. We reverse.
I.
The facts as recited by the Ohio Court of Appeals are presumed correct on habeas review
pursuant to 28 U.S.C. § 2254(e)(1), which are as follows:
Riley Smath was born on August 26, 2003. His mother, Eileen Callahan-Smath,
originally planned to deliver Riley naturally, but doctors had to perform a cesarean
section when she failed to progress after many hours of labor. After the cesarean,
doctors examined Riley and declared him to be a healthy baby. The only problem
Riley ever exhibited was his spitting up during and after feeding.
Mrs. Callahan-Smath . . . arranged to bring Riley to [Dovala]’s house for day care
beginning January 22, 2004.
On February 6, 2004, Mrs. Callahan-Smath called [Dovala] after work to tell
[Dovala] that she was running late to pick up Riley. [Dovala] informed her that
something was wrong with Riley, that she could not wake him, and that he needed
to go to the emergency room. After Mrs. Callahan-Smath arrived at [Dovala]’s
home [around 5:00 p.m., he looked lifeless and appeared blue; so] she rushed Riley
to the hospital. Riley was pronounced dead shortly thereafter, and doctors later
determined his death to be the result of blunt impact trauma to the head [and that
the time between the injury and death was three to five hours].
After Riley’s death, Detective Dan Jasinski interviewed [Dovala] at her home and
recorded the interview on videotape. [Dovala] answered questions about her day
with Riley, but denied that either she or one of the day care children hurt Riley in
any way.
State v. Dovala, 2007 WL 2752395, at *1 (Ohio Ct. App. Sept. 24, 2007); see also State v. Dovala,
2009 WL 806847, at *1 (Ohio Ct. App. Mar. 30, 2009). After unsuccessfully appealing her
convictions, Dovala sought collateral relief in the Ohio state courts. Two opinions from the Ohio
Court of Appeals are relevant, Dovala III and Dovala V. We discuss each in turn.
-2-
No. 20-4222
Dovala v. Baldauf
A.
Dovala III adjudicated on the merits petitioner’s claim that her trial counsel, James Burge,
provided ineffective assistance. State v. Dovala, 2011 WL 2533915 (Ohio Ct. App. June 27, 2011)
(Dovala III). That opinion sets forth extensive facts and reasoning that are at the heart of this case.
1.
The Ohio Court of Appeals initially resolved Dovala’s objections to the trial court’s
findings of fact. Two are of import here.
First, the Ohio Court of Appeals addressed the trial court’s finding that Burge “consulted
with a neurologist throughout the case”:
[Dovala] argues that the “consultation” attested to by Burge was nothing more than
an informal discussion between Burge and his co-counsel’s husband, Dr. Tom
Watson, a neurologist. [Dovala] argues that it is misleading and inaccurate to
consider Burge’s discussion with Dr. Watson a “consultation” because Dr. Watson
was not compensated for his services and did not provide a written report of his
findings. Given the lack of a report, [Dovala] asserts that it is unclear what
materials he reviewed and relied on in arriving at his conclusion that Smath’s death
was the result of an inflicted injury that would have been accompanied by the quick
onset of symptoms. [Dovala] also asserts that the informal nature of the
“consultation” with Dr. Watson resulted in several questions critical to her defense
being left unanswered, specifically, whether the injuries to Smath could have been
inflicted by a younger child at an earlier point in the day. Dovala considers this
possibility one of “huge importance” because there was testimony at trial that
Smath had started crying earlier in the day while Dovala was out of the room, but
her four year old son remained in the room with him. Dovala further argues that
the neurological consultation should have included an assessment of the
inconsistencies between the severe internal tissue and skull damage and the lack of
any external bruising or lacerations.
In his deposition testimony, Burge testified that he has practiced criminal law for
thirty years. Throughout that time, he has defended four infant homicides before
representing Dovala, taking three of those cases to trial. Based on his experience,
he understood that the trauma to the head generally occurred close in time to the
infant’s death. Further, at the point he was retained by Dovala, she had made
several statements to the police, one of which was video recorded and admitted into
evidence at trial. In those statements and in her testimony at trial, Dovala stated
that it was “absolutely not” possible that one of the other children could have done
-3-
No. 20-4222
Dovala v. Baldauf
anything to Smath that day, and further, that there were no accidents or other adults
present in the house that could have caused his injuries. Burge explained that these
statements made it very difficult to mount a defense that another party had caused
the fatal injuries to Smath. He did, however, discuss the case with Dr. Francis
Bartek, a physician specializing in obstetrics and gynecology (“OB/GYN”) who
had testified in the past for one of Burge’s clients, and Dr. Tom Watson, a
neurologist, who was also the spouse of Burge’s co-counsel. Though neither
physician prepared a report on the case, Dr. Bartek, who was compensated $2,000
in December 2004 for his analysis, suggested there might be a congenital weakness
in the skull and reviewed Smath’s medical records from birth. Dr. Bartek gave
Burge the name of two other physicians who could conduct genetic testing to
determine if there was such a weakness in Smath’s skull, one of whom was Dr.
Brian Clark. Burge testified that Dr. Clark was able and willing to perform the
testing on Smath’s parents and that the Smaths had agreed to submit DNA samples
for testing. Burge discussed this strategy with Dovala and informed her of the
ramifications of the test results on her defense, particularly if the test demonstrated
that there was not any congenital weakness. Burge stated he left the decision up to
Dovala, and that she instructed him not to pursue the testing. We note that Dovala
disputes Burge’s testimony, testifying in her deposition that she left the decision up
to Burge and was under the impression that the testing had been done, until she
learned just weeks before her trial that it had not, in fact, been done.
According to Burge, Dr. Watson, who was not compensated for his assessment of
the case, reviewed both the medical records and the autopsy report and concluded
that Smath’s trauma was the result of an “inflicted injury” and that “the onset of
symptoms from the injury would have occurred very quickly.” Dr. Watson further
indicated to Burge that “whoever inflicted it knew they did it.” Burge stated that
he consulted with Dr. Watson approximately three months before the matter was
set for trial and that Dr. Watson attended portions of the trial as well. Burge
admitted that he did not pursue further questioning with Dr. Watson as to the
discrepancies between the internal and external damage to Smath’s skull, in part
because of Dr. Watson’s assessment, but also based on Burge’s own experience
and research with the nature of the head injuries in these cases. For these reasons,
he considered it “unrealistic” to consult with another physician in an attempt to
develop an alternative cause of death. Based on the foregoing, we conclude that
there was some competent, credible evidence to support the trial court’s conclusion
that Burge consulted with two different physicians throughout Dovala’s case.
Id. at *3–4.
Second, it considered one aspect of the testimony of petitioner’s post-conviction medical
expert, Dr. Audrius Plioplys, which was critical of the blunt-force-trauma conclusion drawn by the
prosecution’s experts:
-4-
No. 20-4222
Dovala v. Baldauf
Next, we consider Dovala’s challenge to the trial court’s finding that Dr. Audrius
Plioplys was unable to provide “a precise opinion on the amount of compressive
force needed to cause the fractures” to Smath’s skull. Dr. Plioplys testified at the
post-conviction hearing that Smath’s injuries were the result of “compressive-based
force,” not blunt trauma, as was the opinion of the State’s experts at trial. In
addressing the amount of force necessary to inflict a compressive force injury, Dr.
Plioplys’ testimony ranged from a statement that the injury could have resulted
from “some pressure” from a thumb or a knuckle being applied for a brief period
of time, to a later statement that, although the force “could have been very brief, it
was considerable force.” Though he opined in his affidavit and at the hearing as
well that the force could have been inflicted “by the normal physical strength and
body weight of a child four years of age or older,” he was unable to specify the
amount of weight or calculation of force that was necessary to inflict the fatal
fracture. His only testimony in this regard was that a finger or knuckle placed on
the developing skull of a person Smath’s age could have produced such an injury
“with just enough weight being placed on it.” When explicitly asked if he could
determine the amount of weight or force necessary to cause the type of fracture
suffered by Smath, however, Dr. Plioplys stated he could not. He further agreed
that he was not familiar with the literature or field of research on the amounts of
pressure necessary to cause certain types of fractures. Based on this testimony, we
conclude that there was some competent, credible evidence adduced at the hearing
to support the trial court’s finding as to Dr. Plioplys’ opinion on the amount of force
necessary to cause Smath’s injuries.
Id. at *4 (alterations omitted).
2.
After resolving these factual disputes, the Ohio Court of Appeals then rejected Dovala’s
claim “that Burge was ineffective because he failed to investigate her case and prepare an adequate
defense,” id. at *5, focusing both on Burge’s testimony as to why he pursued aggressive cross-
examination over a battle of the experts and the deference we give to counsel’s strategic decisions:
At his deposition, Burge stated that he was limited in his options for asserting a
defense in light of Dovala’s numerous statements to police that no other adults or
children had been near Smath that day, nor had there been any accidents in the
home. Burge reasoned that it would be inconsistent at trial to blame another person
in the house that day for the injuries based on Dovala’s statements. Additionally,
Burge testified that the coroner’s estimate that the injury had occurred between
three to five hours before Smath’s death, coupled with his own experience in similar
cases, precluded any attempt to suggest that another party, such as Smath’s parents,
had caused the injury. Burge explained that based on his experience in similar
-5-
No. 20-4222
Dovala v. Baldauf
cases, his discussions with Dr. Bartek and Dr. Watson, and the statements Dovala
made to police before retaining him as counsel, he felt the best strategy to pursue
in Dovala’s defense would be to rely on a vigorous cross-examination of Dr.
Daniels and Dr. Matus. The record reflects Burge did so, offering several medical
journals and studies challenging the conclusion of Dr. Matus, so much so that Dr.
Matus agreed there was a possibility, though remote, that there were factors that
could lead to increased intracranial pressure and result in a fracture similar to the
one suffered by Smath.
Consistent with the Supreme Court’s directive, this Court has rejected the assertion
that trial counsel is ineffective if he elects not to put forth an expert on defendant’s
behalf, and instead, relies on a rigorous cross-examination of the State’s witnesses.
Here, the record reflects Burge’s deliberate decision, based on a variety of
considerations including discussions with medical professionals, not to seek an
expert on Dovala’s behalf, but to instead attack the credibility and findings of the
State’s experts. Giving due deference to counsel’s decision, we cannot conclude
that the failure to obtain an expert under these circumstances constitutes ineffective
assistance of counsel. Moreover, “decisions regarding the calling of witnesses are
within the purview of defense counsel’s trial tactics.” In that same regard, we agree
with other courts that have held that a post-conviction relief petitioner cannot
demonstrate ineffective assistance of counsel by presenting “a new expert opinion
that is different from the theory used at trial.” “Ohio case law clearly shows that
alternate or supplementary theories from expert witnesses, which are presented in
post-conviction proceedings, are not sufficient to establish ineffective assistance of
counsel.” Despite Devan’s claims that Burge’s cross-examination was not
effective, the record reveals that Dr. Matus agreed that it was possible that increased
cranial pressure could have caused an existing fracture to expand in size, resulting
in a fracture similar to the one that occurred here.
To the extent Dovala takes issue with Burge’s failure to properly investigate her
claims, Burge testified that he sought the expertise of both an OB/GYN and a
neurologist as well as conducting supplemental research on alternative causes of
infant death in such circumstances. Further, Burge’s initial and supplemental
requests for discovery provide evidence that he pursued relevant information
necessary to prepare a defense in this case. This Court will not employ “hindsight
to distort the assessment of what was reasonable in light of counsel’s perspective
at the time.” Therefore, the trial court did not abuse its discretion in denying
Dovala’s petition for post-conviction relief because there was no evidence that
Burge was deficient in performing his duties as counsel. Having concluded that
Burge was not deficient in acting as counsel, we need not address whether she
suffered any prejudice.
Id. at *7–8 (alterations and internal citations omitted).
-6-
No. 20-4222
Dovala v. Baldauf
B.
Petitioner later moved for relief from judgment after learning “that the neurologist Burge
testified to consulting was Dr. Swanson, not a Dr. Tom Watson” and that “Dr. Swanson . . . never
discussed Dovala’s case with Burge and never rendered a medical opinion.” State v. Dovala, 2014
WL 2522022, at *1 (Ohio Ct. App. June 2, 2014). In response, Burge admitted “that he did not
formally consult with Dr. Swanson, but that Dr. Swanson did provide an unofficial opinion to his
co-counsel, Laura Perkovic, who was at that time married to Dr. Swanson.” Id. The trial court
denied petitioner’s motion, which leads us to the last state-court appeal.
In Dovala V, the Ohio Court of Appeals again detailed Burge’s strategy and provided some
more facts relevant to his consultation with Dr. Swanson:
In support of her petition, Dovala presented the testimony of Dr. Audrius Plioplys,
a retired pediatric neurologist who opined that the victim’s injuries were not caused
by blunt force trauma, but by compressive force applied directly to a single point
on the skull. In his affidavit and testimony in support of the petition, Dr. Plioplys
also disagreed with the State’s trial experts regarding the timeframe during which
the victim’s injuries could have occurred. Specifically, he concluded that he could
not narrow the timeframe further than the twenty-four hour window before the
victim’s death.
The videotaped deposition of Attorney Burge was admitted into evidence during
the hearing on Dovala’s petition. Attorney Burge explained that according to his
understanding of the timeline and potential testimony by the State’s experts, the
onset of the victim’s symptoms happened too late in the day to attribute them to an
injury that occurred before he was in Dovala’s care. He also explained that Dovala
had taken the position, as memorialized in a recorded interview with police, that no
one else in her home had injured the victim. In response to questions about whether
another child in the home could have injured the victim—the theory espoused by
Dr. Plioplys—Attorney Burge reasoned that although the argument could be made
despite Dovala’s own prior statements, it was not a defense that would have proved
successful. Attorney Burge testified that he believed that the only avenue open to
him in light of the substantial limitations imposed by Dovala’s prior statements was
a defense that the injury was a preexisting or congenital condition.
Attorney Burge then described how this defense unfolded. He testified that he
consulted Dr. Thomas Swanson, whose name he recalled incorrectly at the time,
-7-
No. 20-4222
Dovala v. Baldauf
between 60–90 days before trial. Attorney Burge acknowledged that Dr. Swanson
was married to his co-counsel at that time, which provided him the opportunity to
review records. Attorney Burge stated that the case was one of interest to Dr.
Swanson, and that he was not paid a fee. According to Attorney Burge, he did not
pursue a formal consultation with Dr. Swanson because he believed him unlikely
to opine that the victim suffered anything other than an inflicted injury. Attorney
Mark Devan, an expert who testified regarding Attorney Burge’s performance
during the hearing on Dovala’s petition, agreed that Attorney Burge’s interaction
with Dr. Swanson could be characterized as an “unpaid, informal consult” by the
“ex-husband of co-counsel.”
Dovala’s motion for relief from the trial court’s judgment denying postconviction
relief was based on an affidavit obtained from Dr. Swanson. In that affidavit, Dr.
Swanson affirmed that he was married to Attorney Burge’s co-counsel at the time
of Dovala’s trial and that he discussed the case with her “in a casual manner at our
home.” He denied that he discussed the case directly with Attorney Burge, and
stated that he did not “officially opine anything regarding the mechanism of the
death” involved in this case. He also recalled that he “did not have access to or
review the records in any detail that would have allowed a learned opinion in this
case.” Attorney Burge provided an affidavit in response to Dovala’s motion. In
that affidavit, he clarified that any consultations with Dr. Swanson were informally
conducted through co-counsel, to whom Dr. Swanson was married at the time.
Attorney Burge adhered to his recollection that Dr. Swanson believed that the injury
involved in this case was inflicted, and Attorney Burge reiterated that because the
theory of Dovala’s defense had already been developed at that time, he did not
believe that a formal consultation with Dr. Swanson would have been helpful.
While Dovala’s motion was pending, counsel deposed trial co-counsel, Attorney
Laura Perkovic. Attorney Perkovic’s recollection of details surrounding the
defense was hampered by the passage of time. Although she disagreed with parts
of Attorney Burge’s affidavit, she also disagreed with elements of Dr. Swanson’s
affidavit. The substance of her testimony, however, confirmed that Attorney Burge
did ask her to speak to Dr. Swanson about Dovala’s case, and that she did so,
providing access at least to the autopsy photographs and report. According to
Perkovic, Dr. Swanson would not provide an opinion without being formally
retained, but that he did tell her that because the case involved “brain trauma,” he
would be qualified to formally consult on the case.
State v. Dovala, 2016 WL 1295954, at *2–3 (Ohio Ct. App. Mar. 31, 2016) (Dovala V) (emphasis
omitted). And it then summarily concluded Dovala was not entitled to relief from judgment:
Although the witnesses’ recollections of the details differ in light of the passage of
time, a witness during the postconviction proceedings summarized Attorney
Burge’s consultation with Dr. Swanson in a manner consistent with the new
-8-
No. 20-4222
Dovala v. Baldauf
testimony: it was informal, Dr. Swanson was not retained, and it occurred because
Dr. Swanson was married to co-counsel. This much was clear as a result of
Attorney Burge’s deposition during the postconviction proceedings, and it was not
an abuse of discretion for the trial court to conclude that the information set forth
in Dr. Swanson’s affidavit did not undermine the accuracy and credibility of the
judgment.
Id. at *4.
C.
Having exhausted all available state-court remedial options, Dovala filed a petition for a
writ of habeas corpus under 28 U.S.C. § 2254, raising again (and only) her ineffective-assistance-
of-counsel claim. The district court rejected a magistrate judge’s report and recommendation and
instead granted a conditional writ that required the State of Ohio to release Dovala unless it
scheduled a new trial within 120 days. It initially stayed that order pending appeal, which the State
timely commenced. However, following briefing completion here and our scheduling of oral
argument for June 2021, the district court, on April 29, 2021, vacated the stay, ordered that
petitioner be released on bond, and began the 120-day clock anew. We promptly stayed both the
district court’s order granting Dovala’s release and its order conditionally granting Dovala’s
habeas petition pending our resolution of the appeal.
II.
Placed in petitioner’s path to relief are several imposing hurdles.
First, the Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA) “recognizes a
foundational principle of our federal system: State courts are adequate forums for the vindication
of federal rights.” Burt v. Titlow, 571 U.S. 12, 19 (2013). Accordingly, “habeas corpus is a guard
against extreme malfunctions in the state criminal justice systems, not a substitute for ordinary
error correction through appeal.” Harrington v. Richter, 562 U.S. 86, 102–03 (2011) (internal
-9-
No. 20-4222
Dovala v. Baldauf
quotation marks omitted). AEDPA provides that federal courts can overturn a state conviction for
an issue adjudicated on the merits, like Dovala’s claim here, only if the relevant state-court
decision was “contrary to, or involved an unreasonable application of, clearly established Federal
law, as determined by the Supreme Court of the United States.” 28 U.S.C. § 2254(d)(1). This is
a “formidable barrier,” Burt, 571 U.S. at 19; “a state prisoner must show that the state court’s
ruling on the claim being presented in federal court was so lacking in justification that there was
an error . . . beyond any possibility for fairminded disagreement,” Harrington, 562 U.S. at 103.
Being wrong or even clearly erroneous “will not suffice.” White v. Woodall, 572 U.S. 415, 419
(2014) (citation omitted). “This is a difficult to meet and highly deferential standard for evaluating
state-court rulings, which demands that state-court decisions be given the benefit of the doubt.”
Cullen v. Pinholster, 563 U.S. 170, 181 (2011) (internal quotation marks and citations omitted).
Second, a prisoner asserting an ineffective-assistance-of-counsel claim must demonstrate
both deficient performance and prejudice. Strickland, 466 U.S. at 687. The first component
requires a showing that the lawyer’s performance “fell below an objective standard of
reasonableness” as measured by “prevailing professional norms.” Id. at 687–88. Given the “strong
presumption that counsel’s conduct falls within the wide range of reasonable professional
assistance,” the Supreme Court has cautioned that “every effort be made to eliminate the distorting
effects of hindsight, to reconstruct the circumstances of counsel’s challenged conduct, and to
evaluate the conduct from counsel’s perspective at the time.” Id. at 689. “Even under de novo
review, the standard for judging counsel’s representation is a most deferential one,” Harrington,
562 U.S. at 105, and “a court still may not grant relief if the record does not reveal that counsel
took an approach that no competent lawyer would have chosen” even in cases where “there is
reason to think that counsel’s conduct was far from exemplary,” Dunn v. Reeves, 141 S. Ct. 2405,
-10-
No. 20-4222
Dovala v. Baldauf
2410 (2021) (per curiam) (brackets and internal quotation marks omitted). And upon a prisoner’s
satisfaction of a deficient-performance showing, he must also establish “that there is a reasonable
probability that, but for counsel’s unprofessional errors, the result of the proceeding would have
been different.” Strickland, 466 U.S. at 694. “The likelihood of a different result must be
substantial, not just conceivable.” Harrington, 562 U.S. at 112.
The combined effect of these two “highly deferential” legal prisms means that we must
defer to the state court’s adjudication of an ineffective-assistance-of-counsel claim “doubly so.”
Id. at 105 (internal quotation marks omitted). Accordingly, “the question is not whether counsel’s
actions were reasonable. The question is whether there is any reasonable argument that counsel
satisfied Strickland’s deferential standard.” Id. With this overlay, Harrington instructs that we
“must guard against the danger of equating unreasonableness under Strickland with
unreasonableness under § 2254(d).” Id. This necessarily means that under double-deference
review, “even a strong case for relief does not mean the state court’s contrary conclusion was
unreasonable.” Id. at 102. So, as the Supreme Court reminded us earlier this year, “the critical
question [i]s not whether the Sixth Circuit itself c[an] see [a Strickland violation.] All that matter[s
i]s whether the [state] court, notwithstanding its substantial latitude to reasonably determine that
a defendant has not shown [a Strickland violation], still managed to blunder so badly that every
fairminded jurist would disagree.” Mays v. Hines, 141 S. Ct. 1145, 1149 (2021) (per curiam)
(alterations and internal quotation marks omitted).
III.
Dovala must first establish that the Ohio Court of Appeals unreasonably concluded Burge
did not provide ineffective assistance when he elected an aggressive cross-examination strategy
over pursuing experts of his own as part of Dovala’s defense. We review de novo the district
-11-
No. 20-4222
Dovala v. Baldauf
court’s legal conclusion that Dovala overcame double-deference review. See, e.g., Johnson v.
Genovese, 924 F.3d 929, 933 (6th Cir. 2019). Harrington compels us to disagree with the district
court.
A.
Strickland’s performance-standard “rule is as general as they come.” Davis v. Carpenter,
798 F.3d 468, 473 (6th Cir. 2015). A lawyer provides constitutionally adequate representation
when his efforts fall “within ‘the wide range of reasonable professional assistance.’” Id. (quoting
Strickland, 466 U.S. at 689). With respect to an attorney’s decision not to investigate, that choice
“must be directly assessed for reasonableness in all the circumstances, applying a heavy measure
of deference to counsel’s judgments”; moreover, “strategic choices made after less than complete
investigation are reasonable precisely to the extent that reasonable professional judgments support
the limitations on investigation.” Strickland, 466 U.S. at 690–91.
The Supreme Court has not had the occasion to examine a case with similar contours to
Dovala’s. No case has set a floor for the steps an attorney must take when determining whether to
employ a battle-of-the-experts defense, let alone whether those steps must constitute formal
engagement (as Dovala suggests) over informal and limited fact-gathering (as Burge did). Cf.
Davis, 798 F.3d at 473–74. But what the Supreme Court has instructed—doing so about five
months before the Ohio Court of Appeals adjudicated Burge’s performance in Dovala III—is that
Strickland does not require a defense attorney to present an expert whenever a prosecutor does so.
Harrington, 562 U.S. at 111 (“Strickland does not enact Newton’s third law for the presentation
of evidence, requiring for every prosecution expert an equal and opposite expert from the
defense.”). Furthermore, the Court endorsed the very strategy Burge utilized as an alternative to
expert-testimony: cross-examination, “the greatest legal engine ever invented for the discovery of
-12-
No. 20-4222
Dovala v. Baldauf
truth.” California v. Green, 399 U.S. 149, 158 (1970) (internal quotation marks omitted). “In
many instances,” the Harrington Court noted, “cross-examination will be sufficient to expose
defects in an expert’s presentation. When defense counsel does not have a solid case, the best
strategy can be to say that there is too much doubt about the State’s theory for a jury to convict.”
562 U.S. at 111. These directives, the generality of Strickland’s minimum-performance standard,
and AEDPA deference, mean we owe the Ohio Court of Appeals considerable latitude. See
Knowles v. Mirzayance, 556 U.S. 111, 123 (2009). In our view and consistent with the magistrate
judge’s opinion, Dovala v. Tim, 2019 WL 10890248, at *17–21 (N.D. Ohio Oct. 30, 2019), we
cannot find that “every ‘fairminded juris[t]’ would agree that every reasonable lawyer would have
made a different decision,” Dunn, 141 S. Ct. at 2411 (quoting Harrington, 562 U.S. at 101),
particularly given significant factual developments that limited Burge’s options.
Begin with the restrictions placed on Burge by Dovala’s statements to law enforcement
made shortly after Riley’s death. She told police that there were no accidents at the house
involving Riley and disclaimed any possibility that another child (or anyone else) inflicted the
injury on him. Those admissions limited Burge’s options for undermining the prosecution’s
charge that the injury occurred while in her care—e.g., that someone else in her home was
responsible, by accident or purpose.
Consider too what Burge learned from his consultations with medical professionals about
the case. One (Dr. Bartek) told him of the possibility that Riley’s skull was congenitally weak,
and that it would be possible to use his parents’ DNA profile to account for that possibility. Burge
testified that he raised this possible line of defense with Dovala, but that she worried what his
-13-
No. 20-4222
Dovala v. Baldauf
strategy would be if the test showed no congenital abnormalities.1 And he elicited informal advice
from a neurologist (Dr. Swanson) who Burge believed gave him no reason to consider challenging
the coroner’s conclusions regarding the cause or timing of death.
Finally, considering what was already known about the case, there was little to put Burge
on notice that a further investigation and formal consultation with an expert witness might be
fruitful to Dovala. See Strickland, 466 U.S. at 691; see also Wiggins v. Smith, 539 U.S. 510, 527
(2003) (“In assessing the reasonableness of an attorney’s investigation, however, a court must
consider not only the quantum of evidence already known to counsel, but also whether the known
evidence would lead a reasonable attorney to investigate further.”); Caudill v. Conover, 881 F.3d
454, 463 (6th Cir. 2018) (noting the difference between “a complete failure to investigate” and
“failing to dig deeper” (internal quotation marks omitted)). No information before Burge, other
than Dovala’s telling of the events, indicated that there was reason to be skeptical of the coroner’s
causation or timing conclusions. So although Burge’s decision to pursue cross-examination rather
than calling an expert witness may have constituted less than exemplary lawyering, it cannot be
said that he had no possible reason for his decision. See Pinholster, 563 U.S. at 196. As the
Supreme Court recently explained in Dunn v. Reeves, a state court is “entitled to reject [a
defendant’s] claim if trial counsel had any ‘possible reason for proceeding as they did.’” 141 S.
Ct. at 2412 (quoting Pinholster, 563 U.S. at 196 (alterations omitted)). Because Burge did not
have a particularly “solid case” to defend, we cannot find that “every ‘fairminded juris[t]’ would
agree that every reasonable lawyer would have made a different decision.” Id. at 2411 (quoting
1
Dovala disagrees with Burge’s testimony here, Dovala III, 2011 WL 2533915, at *3, but
she does not anchor any part of her ineffective-assistance claim in this aspect of Burge’s
representation. Indeed, it was perhaps good trial strategy not to pursue a skull test because it could
have done more harm than good if Dovala’s worry came to fruition.
-14-
No. 20-4222
Dovala v. Baldauf
Harrington, 562 U.S. at 101). Thus, established Supreme Court precedent precludes habeas relief
in this case.2
B.
The district court saw this case differently. It faulted Burge for both not consulting a doctor
on the timing of Riley’s injuries and for his “blanket acceptance of Dr. Swanson’s alleged opinion
regarding the cause of Riley’s injuries.” However fair that criticism may be, the district court’s
decision to grant habeas relief reflects one that viewed Dovala’s claim through Strickland’s
familiar standard without adding Harrington’s constraining lens. Under that controlling precedent,
“the question is not whether counsel’s actions were reasonable. The question is whether there is
any reasonable argument that counsel satisfied Strickland’s deferential standard.” Harrington,
562 U.S. at 105. True, the district court cited and briefly discussed Harrington. It did not,
however, apply the question Harrington decrees, let alone the Court’s on-point language about
pursuing cross-examination over expert witnesses, especially “[w]hen defense counsel does not
have a solid case” as examined above. Id. at 111. Admittedly, our decision is guided in part by
the Supreme Court’s recent application of the Harrington standard in Dunn, which provided
helpful analysis regarding the limited instances in which federal courts may grant habeas relief on
ineffective assistance of counsel claims—guidance that was not available at the time of the district
court’s decision in this case. Regardless, the Harrington standard as elaborated in Dunn, precludes
the relief granted by the district court.
2
Because fairminded jurists could debate the Ohio Court of Appeals’ holding that Burge’s
performance did not fall below constitutionally permissible levels, we need not address the district
court’s separate holding that Dovala established a “reasonable probability that, but for counsel’s
unprofessional errors, the result of the proceeding would have been different.” Strickland, 466
U.S. at 694.
-15-
No. 20-4222
Dovala v. Baldauf
Finally, we are not persuaded that any of the authority relied upon by the district court and
petitioner mandate a different outcome. Richey v. Bradshaw is the chief one. 498 F.3d 344 (6th
Cir. 2007). There we similarly considered an attorney’s decision to pursue cross-examination
instead of hiring an expert to testify on his client’s behalf, and concluded the attorney performed
deficiently because he did not make a determination as to why the consulted expert (who agreed
with the findings of the state’s experts and thus would not have provided helpful testimony)
reasoned as he did before declining to pursue a battle-of-the-experts defense. Id. at 362–63. Given
the factual similarities, we can see the allure of that case. But we have subsequently characterized
Richey as a decision that applied an “outdated,” pre-Harrington-style-of-review that “barely
referenced AEDPA’s deferential standards when granting relief.” Kendrick v. Parris, 989 F.3d
459, 476 (6th Cir. 2021).3
IV.
For these reasons, we reverse the judgment of the district court.
3
Dovala’s other main cases similarly miss the mark. Stermer v. Warren, 959 F.3d 704 (6th
Cir. 2020), also dealt with the failure to retain an expert. But, as Kendrick sets forth, “our
discussion of counsel’s duties with respect to rebutting expert testimony was dicta” because the
Stermer panel “expressly declined to decide the merits of the IAC claim based on counsel’s failure
to retain an arson expert.” 989 F.3d at 476 (discussing Stermer, 959 F.3d at 738). Frazier v.
Huffman, 343 F.3d 780 (6th Cir. 2003), and Wiggins, 539 U.S at 527, are also distinguishable—
they too are pre-Harrington cases and involved counsel, who unlike Burge, had known evidence
at their ready that would have caused a reasonable attorney to investigate further.
-16-