CourtListener 10679758

Full text

315 Ga. 287
FINAL COPY

S22A1051. SMITH v. THE STATE.

PETERSON, Presiding Justice.

Nineteen years ago, Danyel Smith was convicted of the murder

of his infant son based on a theory of “shaken baby syndrome”

(“SBS”). Smith now argues that the science regarding diagnosis of

brain injuries in infants has changed so much since his trial that he

is entitled to a new trial based on a new expert affidavit ruling out

battery or shaking as the cause of the baby’s death. The trial court

rejected that argument and denied Smith’s extraordinary motion for

new trial without a hearing. Because Smith’s extraordinary motion

alleged facts that, if proven, may warrant relief, the trial court was

not authorized to deny the motion without a hearing. We therefore

vacate the trial court’s ruling on the motion and remand for further

proceedings.

1. Background
(a) Trial, conviction, and appeal

Smith was convicted of felony murder and aggravated battery

in connection with the death of his infant son. This Court affirmed

his convictions in 2008. See Smith v. State, 283 Ga. 237 (657 SE2d

523) (2008). In that decision, in the light most favorable to the

verdicts, we summarized the evidence presented at Smith’s

November 2003 trial as follows:

[O]n April 29, 2002, two-month-old Chandler was taken
by his parents, Marsha Collins and Smith, to his
pediatrician for a checkup. Following the checkup,
Chandler was declared to be in good health. Chandler was
then released from the doctor’s office and into his parents’
care. On the way home, the family stopped at a QuikTrip
gas station. That afternoon, Collins left Chandler with
Smith while she attended an appointment to apply for
WIC public assistance. While Collins was at her
appointment, she called Smith and told him to bring the
baby to the WIC office. While Smith was en route to
Collins’ location with Chandler, Collins called him again,
and Smith told Collins that Chandler was not breathing.
Smith arrived at the WIC office with Chandler, who was
limp and cold and had blood running from his nose.
Collins called 911, and emergency responders rushed
Chandler to the emergency room. At the emergency room,
Chandler’s heartbeat was restored, but he remained
comatose, unresponsive, and unable to breathe on his
own. A CT scan of Chandler’s brain revealed a skull
fracture, a hematoma, and swelling of the brain. Based on

2
the CT scan as well as Chandler’s broken wrists, retinal
hemorrhages, and the quick onset of his symptoms,
Chandler was diagnosed as a “shaken baby,” who had
been subjected to vigorous shaking that was probably
coupled with impact. Smith was the only person with
Chandler during and immediately prior to the onset of his
symptoms. The nature of Chandler’s injuries did not
indicate that they were self-inflicted, and the injuries
were inconsistent with a simple fall or accidental trauma.
After seven days in the hospital without any evidence of
brain function, Chandler was removed from life support
and died on May 6, 2002. A physical examination of
Chandler’s body after his death revealed abdominal
bruising that was consistent with the spacing of adult
knuckles.

Id. at 237-238 (1).

Dr. Anne Frankel, the pediatrician who saw Chandler for a

check-up hours before he became nonresponsive, testified at trial

that there were complications during Chandler’s mother’s

pregnancy and that Chandler was born prematurely via Caesarean

section (“C-section”). She testified that Chandler could not have had

a life-threatening brain injury when she saw him on April 29, 2002.

Dr. William Boydston, a pediatric neurosurgeon who treated

Chandler after he arrived at the hospital, testified that he had

concluded that Chandler had experienced a brain injury due to

3
shaking, based on Chandler’s retinal subdural hematomas, blood

clots of various ages, skull fracture, abdominal bruising, wrist

fractures, and lack of a documented history of physical trauma or

other medical history to explain his injuries. The State also

presented the testimony of Dr. Steven Dunton, who both examined

Chandler at the hospital in his role as a forensic pediatrician for the

hospital and performed Chandler’s autopsy as the county medical

examiner. At trial, Dr. Dunton testified that Chandler died by

homicide as a result of blunt-force head trauma. He acknowledged

that some of the things he observed in Chandler could be explained

by something other than abuse. But Dr. Dunton ultimately made “a

collection of findings . . . that are classic and in some cases virtually

exclusive for violent shaking.”

There was evidence presented at trial that, several weeks

before Chandler became unresponsive, his mother had summoned

emergency medical assistance for Chandler due to a concern about

breathing problems or a seizure. Dr. Boydston testified that the

medical history of Chandler that was provided to him did not

4
reference seizures, but knowing Chandler had such a history would

not change his conclusion. Dr. Dunton acknowledged on cross-

examination that, prior to completing his autopsy report, he was not

made aware of the prior request for emergency medical assistance

for Chandler due to a possible seizure.

In his trial testimony, Smith was adamant that he did not

shake, punch, beat, or kick Chandler, and that Chandler never

experienced a fall in his presence. Smith’s trial counsel emphasized

in his closing argument that Chandler’s mother also had access to

the baby on the day he became nonresponsive and raised questions

about her credibility. He did not challenge the expert medical

testimony directly, although he noted Dr. Dunton’s testimony that

he had been unaware of the prior call for medical care due to a

possible seizure, argued that Dr. Dunton had a conflict of interest

given his two roles in the matter, and suggested that unskillful CPR

performed on Chandler may have caused the baby’s various injuries.

Smith was convicted of felony murder (predicated on first-

degree cruelty to children) and aggravated battery, and moved for a

5
new trial. He raised several claims of ineffective assistance of

counsel, including that trial counsel had failed to investigate

competently the medical evidence prior to trial — relying on an

inadequate review by the defense expert — and failed to present

additional medical evidence at trial. Smith argued that a “competent

expert” would have concluded that Chandler suffered a head injury

shortly before a seizure that he experienced in March 2002 — when

Collins and other people besides Smith had access to the child.

Smith also suggested that vaccinations may have been a cause of

Chandler’s death, given that he was born prematurely and was

predisposed to seizures. At the 2007 hearing on the motion for new

trial, Smith presented the expert testimony of a forensic pathologist

who opined that Chandler’s skull fracture was the result of a birth

injury or an injury that occurred after he was discharged from the

hospital after birth but before paramedics were summoned to his

home due to a possible seizure. This injury led to another, fatal

seizure on April 29, 2002, he said. The pathologist testified at the

motion-for-new-trial hearing that the SBS theory was “very

6
controversial,” saying that “there are a number of [academic] papers

denying its existence.”

The trial court denied the motion for new trial. In our 2008

opinion affirming Smith’s convictions, we said that evidence

supported the conclusion that counsel made a reasonable decision

not to pursue additional medical investigations after consulting with

his expert, who believed that Chandler’s injuries were consistent

with physical abuse that occurred on the day alleged by the State.

See Smith, 283 Ga. at 238-239 (2) (a).

(b) Extraordinary motion for new trial

In March 2021, Smith filed an extraordinary motion for new

trial. Relying on an expert affidavit, various academic journal

articles, and position papers by the American Academy of Pediatrics

(“AAP”), the motion described a major shift in how the medical

community thinks about infant head trauma, from generally

presuming child abuse when an infant presents with head injuries,

to instead requiring a full examination of the child’s medical record,

including the circumstances of the child’s birth. As the motion

7
framed the issue, a hypothesis was “well-entrenched in the medical

and legal communities” at the time of Smith’s trial that violent

shaking was presumptively to blame when an infant presented with

“the triad” of subdural hemorrhage, retinal hemorrhages, and

cerebral edema, with long falls (as from a multi-story building) and

car crashes the only other possible explanations for the combination

of those three symptoms. Smith cited a 2001 AAP position paper,

which stated that “[t]he constellation of” injuries associated with

SBS “does not occur with short falls, seizures, or as a consequence of

vaccination.” Amer. Acad. of Pediatrics, Comm. on Child Abuse and

Neglect, Shaken Baby Syndrome: Rotational Cranial Injuries —

Technical Report, 108 Pediatrics 206 (2001).

Smith’s extraordinary motion said that a major shift in the

medical community’s thinking began in 2006 when the National

Association of Medical Examiners withdrew a position paper

endorsing the “triad” as diagnostic of SBS. The medical community

increasingly began to accept the idea that the “triad” of symptoms

once considered diagnostic of SBS may also be caused by birth

8
injuries, short falls, or other diseases, the motion posited. The

motion cited a 2009 position paper by the AAP indicating that

“[m]edical diseases” can “mimic” the presentation of abusive head

trauma (“AHT”), a broader term that includes head injury due to

shaking. See Amer. Acad. of Pediatrics, Comm. on Child Abuse and

Neglect, Abusive Head Trauma in Infants and Children, 123

Pediatrics 1409 (2009).

Smith emphasized a 2018 position paper by the AAP and other

professional organizations (“2018 Consensus Statement”). See A.K.

Choudhary et al., Consensus Statement on Abusive Head Trauma

in Infants and Young Children, 48 Pediatric Radiology 1048 (2018).

The 2018 Consensus Statement, which framed itself as “intended to

help courts improve the scientific accuracy of their decisions,”

decried “denialism of child abuse” and contentions by defense

attorneys and their expert witnesses proffering “speculative

causation theories” — including birth-related injuries — as

alternative diagnoses in child abuse cases. The statement called the

notion of a “triad” of symptoms as diagnostic of AHT a “straw man”

9
“fallacy” that is “a legal argument and not a medically valid term.”

Smith framed the 2018 Consensus Statement as “mandat[ing]” for

the first time “that pediatricians presented with patients whose

diagnosis previously would have defaulted to AHT must now

thoroughly investigate the possibility of alternative causes.” Smith

claimed that the 2018 Consensus Statement represented the AAP’s

first recognition of birth trauma as an “alternative diagnosis” for the

sort of symptoms presented by Chandler. Citing a 2020 journal

article and his expert’s affidavit, Smith also posited that “[t]oday it

is known that vaccinations, including Hepatitis B, can cause

seizures and encephalopathy even in healthy infants” and that “the

modern medical literature recognizes that prematurity and other

health conditions must be accounted for in vaccine administration.”

The expert affidavit attached to Smith’s motion was provided

by the chair of neurosurgery at Mount Sinai West and Mount Sinai

Morningside, Dr. Saadi Ghatan, opining that the cause of Chandler’s

death was pre-existing conditions resulting from birth injury and

other events, and not from SBS. In his affidavit, Dr. Ghatan cited

10
several ways in which the medical understanding of infant head

injuries has changed since the time of Smith’s trial. Regarding

Chandler’s death in particular, Dr. Ghatan explained how various

events — including acute fetal distress prior to Chandler’s birth,

prolonged labor by his mother, premature delivery via C-section and

vacuum extraction, and prior seizures — led to a brain injury that

was exacerbated by two things that happened shortly before

Chandler became nonresponsive and stopped breathing: a seven-

hour car ride the night before (that would have left a young infant

dehydrated) and vaccinations received the same day.1 Dr. Ghatan

also explained how the medical evidence, in the light of current

medical understanding, ruled out conclusions that Chandler’s death

was a result of battery or shaking: Chandler did not present with

the sort of injuries that one would expect to see in a “battered” or

1 In explaining how routine infant vaccinations could have been so
problematic for Chandler, Dr. Ghatan noted Chandler’s low birth weight of four
pounds, seven ounces, and that “he received more vaccinations than
customary, and at an accelerated pace”; the extraordinary motion averred that
Chandler had accidentally been given two doses of the Hepatitis B vaccine
during his initial neonatal hospitalization, such that the shot he received on
April 29, 2002, was an “overdose.”
11
“shaken” baby. Dr. Ghatan also posited that Chandler’s abdominal

bruising was caused by CPR performed on him by untrained

persons. Dr. Ghatan added that he did not intend to suggest that the

doctors who “handled” Chandler’s case did anything improper under

the standard of care at the time, but were working with a now-

outdated framework.

Without holding an evidentiary hearing, the trial court denied

the extraordinary motion for new trial.2 Smith filed a discretionary

application, which we granted. The case was orally argued before

this Court on October 4, 2022.

2. Analysis

In denying Smith’s extraordinary motion, the trial court

concluded that the sort of expert opinion he offered could never

constitute newly discovered evidence requiring a new trial. The

court also concluded that Smith had failed to satisfy two of the

requirements for obtaining a new trial based on newly discovered

2 The trial court initially dismissed the motion, but then entered a new

order denying it.
12
evidence. Because we conclude that the trial court was wrong to

deny the motion on these bases without first affording Smith an

evidentiary hearing, we vacate the trial court’s order and remand

for further proceedings.

“All motions for new trial, except in extraordinary cases, shall

be made within 30 days of the entry of the judgment on the

verdict[.]” OCGA § 5-5-40 (a). “When a motion for a new trial is made

after the expiration of a 30 day period from the entry of judgment,

some good reason must be shown why the motion was not made

during such period, which reason shall be judged by the court.”

OCGA § 5-5-41 (a). Under some circumstances, newly discovered

evidence may authorize the grant of an extraordinary motion for

new trial:

A new trial may be granted in any case where any
material evidence, not merely cumulative or impeaching
in its character but relating to new and material facts, is
discovered by the applicant after the rendition of a verdict
against him and is brought to the notice of the court
within the time allowed by law for entertaining a motion
for a new trial.

OCGA § 5-5-23; see also Mitchum v. State, 306 Ga. 878, 880-882 (1)

13
(a) (834 SE2d 65) (2019).

As we framed the requirements in Timberlake v. State, 246 Ga.

488 (271 SE2d 792) (1980):

It is incumbent on a party who asks for a new trial on the
ground of newly discovered evidence to satisfy the court:
(1) that the evidence has come to his knowledge since the
trial;
(2) that it was not owing to the want of due diligence that
he did not acquire it sooner;
(3) that it is so material that it would probably produce a
different verdict;
(4) that it is not cumulative only;
(5) that the affidavit of the witness himself should be
procured or its absence accounted for; and
(6) that a new trial will not be granted if the only effect of
the evidence will be to impeach the credit of a witness.

Id. at 491 (1) (citation and punctuation omitted). “Failure to show

one requirement is sufficient to deny a motion for a new trial.” State

v. Gates, 308 Ga. 238, 250 (3) (840 SE2d 437) (2020) (citation and

punctuation omitted). “Extraordinary motions for new trial are not

favored, and a stricter rule is applied to an extraordinary motion for

. . . new trial based on the ground of newly available evidence than

to an ordinary motion on that ground.” Id. (citation and punctuation

omitted). But “[w]here a defendant pleads [facts sufficient to

14
authorize that the motion be granted if the facts developed at the

hearing warrant such relief] in his extraordinary motion and

submits supporting affidavits, a trial court errs by ruling on the

motion without first holding an evidentiary hearing.” Stinchcomb v.

State, 308 Ga. 870, 875 (2) (843 SE2d 847) (2020). “Whether a

defendant has pleaded sufficient facts to entitle him to a hearing is

a question of law that we review de novo.” Id.

Before the trial court, the State contested Smith’s

extraordinary motion on only two of the Timberlake requirements,

arguing that Smith had not shown that the motion was in fact based

on evidence that had come to his knowledge since trial or that he

had acted with due diligence. In its March 2022 order denying

Smith’s extraordinary motion for new trial, the trial court found that

Smith had failed to meet his burden as to at least those two

requirements. As to the first requirement, the court found that the

sort of evidence offered as new — a different expert interpretation

of the medical records that were used at trial — categorically did not

meet the definition of newly discovered evidence. Moreover, the trial

15
court determined, the substance of the expert opinion on which

Smith relies — in particular, “expert opinion to challenge the

scientific basis of shaken baby diagnosis” — “has been available

since the 1990s and was available at the time of the Defendant’s

trial[,]” and so was not “newly discovered.” The trial court concluded

that, for similar reasons, Smith also had failed to satisfy the second

Timberlake requirement because he had not shown that he could not

have obtained through due diligence a medical expert to challenge

at trial the medical conclusions of the State’s experts.

We begin our analysis by explaining why the trial court was

not correct to conclude that the sort of evidence at issue here cannot

qualify as newly discovered evidence.

(a) The trial court erred by categorically rejecting Smith’s
evidence as a basis for a new trial on the ground that it was opinion
evidence, without holding a hearing.

In reaching his conclusion that Smith could not satisfy the first

Timberlake requirement, the trial court concluded that the sort of

expert opinion evidence offered could never qualify as newly-

discovered evidence:

16
In this case, the Defendant has not shown newly
discovered evidence at all. Instead, he offers a different
interpretation of the medical records used at trial through
a new expert witness. In his Affidavit, Dr. Ghatan relies
exclusively on the same medical records that were always
available to the Defendant at trial. . . . Expert opinion
does not constitute “new and material facts” but merely
“opinion evidence [which] fails to constitute newly
discovered evidence.” Wesleyan [College] v. Weber, 238 Ga.
App. 90, 97 [(517 SE2d 813)] (1999).

The trial court erred.

The text of the relevant statute does not exclude expert opinion

evidence from the sort of evidence that may provide the basis for an

extraordinary motion for new trial. OCGA § 5-5-23 provides that the

new evidence supporting such a motion may include “any material

evidence, not merely cumulative or impeaching in its character but

relating to new and material facts,” that is discovered by the

applicant after trial and presented to the court with the requisite

diligence. Of course, expert opinion testimony is “evidence.” See, e.g.,

OCGA § 24-7-701 (b) (“Direct testimony as to market value is in the

nature of opinion evidence.” (emphasis supplied)); OCGA § 24-7-702

(f) (“It is the intent of the legislature that, in all proceedings, the

courts of the State of Georgia not be viewed as open to expert
17
evidence that would not be admissible in other states.” (emphasis

supplied)); OCGA § 24-7-703 (“If [particular facts or data are] of a

type reasonably relied upon by experts in the particular field in

forming opinions or inferences upon the subject, such facts or data

need not be admissible in evidence in order for the opinion or

inference to be admitted.” (emphasis supplied)). It may be that expert

opinion evidence does not fall within the term “facts,” but the text of

OCGA § 5-5-23 does not provide that the evidence in question must

itself be “new and material facts.” It requires only that the evidence

must be “relating to new and material facts.” (Emphasis supplied.)

Expert opinion testimony may, indeed, relate to new and material

facts.

The case law relied on by the trial court here does not demand

a conclusion that expert opinion testimony can never support an

extraordinary motion for new trial, either. Although the Court of

Appeals opinion in Wesleyan College, cited by the trial court,

included language suggesting that such evidence could not

constitute new evidence under OCGA § 5-5-23 because it was expert

18
testimony, the opinion also described the proposed testimony as

“cumulative” and “impeaching,” 238 Ga. App. at 96-97 — aspects of

the evidence that, under Timberlake, likely would have doomed it as

a basis for a motion for new trial. Moreover, there was no suggestion

in Wesleyan College that the evidence — an opinion by the plaintiff’s

former expert that the landowner-defendant could not have foreseen

that a particular tree would have fallen — was based on new

scientific developments or other facts that had become known to the

defendant after trial. See 238 Ga. App. at 95-96.3 The decision of this

Court on which Wesleyan College relied also involved evidence that

was merely impeaching, and, with little description of the evidence

at issue, the opinion of this Court in that case contains no indication

that the evidence was related to any new facts. See Allen v. State,

187 Ga. 178, 180 (1) (200 SE 109) (1938) (“Aside from the fact that

3 Indeed, the expert who offered the “new” opinion inspected the tree

prior to trial. See Wesleyan College, 238 Ga. App. at 96. The dissent insisted
that because the expert in question examined the tree well before the experts
who testified for the plaintiff at trial, that expert’s testimony “established facts
which he had a unique opportunity to observe; he was not merely an opinion
witness.” Id. at 99 (Smith, J., dissenting).
19
[the evidence in question] is largely or entirely opinionative, newly

discovered evidence that is merely impeaching in character affords

no legal reason for the grant of a new trial.”).

Other decisions of this Court rejecting expert opinion evidence

as a basis for granting a motion for new trial also did so on the

grounds that the evidence was merely impeaching, without

indicating that those opinions were related to new facts. See Ruger

v. State, 263 Ga. 548, 551 (2) (c) (436 SE2d 485) (1993) (affirming

denial of motion for new trial based on expert opinion that method

employed by State’s expert in conducting experiment about

bloodprints was “scientifically unsound”; new opinion “tendered to

disprove the facts on which the [testimony of the State’s expert] was

founded”); Wright v. State, 184 Ga. 62, 71 (9) (190 SE 663) (1937)

(rejecting expert affidavit challenging State’s testimony about

substance found on pipes near defendant’s workplace as basis for

new trial, as it was “impeaching”).4 Rejecting a motion purportedly

4 In Rogers v. State, 257 Ga. 590 (361 SE2d 814) (1987), relied on in

Ruger, we concluded that the affidavit of a pathologist who challenged various

20
based on newly-discovered evidence on the ground that the expert

opinion evidence at issue was “impeaching” — a problem that is at

least potentially fatal under the Timberlake standard — is not the

same as saying that expert opinion evidence can never support an

extraordinary motion for new trial. Nothing in our case law

categorically excludes expert opinion evidence from serving as the

basis for an extraordinary motion for new trial premised on newly

discovered evidence.

And the conclusion that new expert analysis of existing

physical evidence may constitute new evidence justifying the grant

of an extraordinary motion for new trial accords with a recent

decision of this Court. In State v. Gates, we affirmed the grant of an

extraordinary motion for new trial based on analysis of DNA

evidence through the TrueAllele software, which was not available

to the parties at the time of the trial in that case. See 308 Ga. at 264

(3) (b). We rejected various challenges by the State to the reliance

aspects of the State’s expert testimony failed to satisfy the Timberlake
standard, but we didn’t say why. See id. at 591 (2) (a).
21
on this analysis as a basis for granting a new trial, including that

the defendant showed insufficient diligence:

The State first argues that Gates should have brought his
extraordinary motion much earlier, given the prevalence
of DNA evidence in criminal proceedings since at least the
1990s. . . . As the State implicitly concedes by that
argument, however, the “newly discovered evidence” in
this case is not simply the DNA found on the belt and tie,
or even the GBI’s initial inconclusive test results for them.
Those items, that DNA, and those results, have little
value to Gates’ case because the GBI’s human
interpretation of the DNA results was inconclusive. It was
instead the TrueAllele analysis of those results that
yielded Gates newly discovered evidence on which he
could stake a claim to a new trial. Because the record
established that the TrueAllele software had the ability
to provide probative analysis of complex and degraded
DNA mixtures in a way that traditional human methods
could not (and apparently, to this day, cannot), it was not
necessary under Timberlake for Gates to have sought
TrueAllele analysis of the DNA located on the belt and tie
at any point prior to 2005 when TrueAllele was first used.

Id. at 257 (3) (a) (iii). We used the shorthand “TrueAllele analysis”

in describing the evidence at issue, but the evidence ultimately came

in the form of expert testimony by the creator of the TrueAllele

software: namely, “that the TrueAllele software determined that

Gates is excluded as a contributor to the DNA mixture on” the

physical evidence at issue. Id. at 250 (3). Although our opinion in
22
Gates did not address specifically the question of whether expert

opinion testimony was categorically in the realm of “new” evidence

that supports an extraordinary motion for new trial — the parties

and the Court appear to have assumed that it was — Gates is

consistent with the notion that new expert analysis of existing

physical evidence may constitute newly discovered evidence

supporting a grant of an extraordinary motion for new trial.

The trial court thus erred by denying Smith’s extraordinary

motion on the basis that “[e]xpert opinion does not constitute ‘new

and material facts’” and “opinion evidence . . . fails to constitute

newly discovered evidence[,]” without considering whether the

expert opinion that is offered as the primary support for Smith’s

motion relates to new and material facts. We will leave that ultimate

determination for the trial court to make in the first instance. But

Smith certainly has offered a pleading sufficient to satisfy that

standard for purposes of obtaining a hearing. On its face, Dr.

Ghatan’s opinion purports to be based on new and material facts. In

particular, his affidavit makes a number of factual assertions about

23
the development of the medical understanding of infant head

injuries since the time of Smith’s trial:

 “Experience documented since 2002 shows that the
obstetrician’s hands, a knife, vacuum, and forceps used during
any C-section can all cause trauma to a baby’s head[.]”

 Although Chandler’s medical team did not scan Chandler’s
head following his birth despite swelling to his head —
“appropriately so in 2002” — “[w]ith the more recent
application of ultrasound technology, where there is no
radiation exposure, we routinely document a much higher
frequency of skull fractures and traumatic brain injuries in
infants due to birth and incidental traumas than was done so
two decades ago.”

 “Since the time of Chandler’s death, significant experience has
been accumulated regarding the risk of seizures with
vaccinations, which would only exacerbate the susceptibility of
the brain of an infant such as Chandler, to experience a
seizure.”

 “Retinal hemorrhages, which . . . were commonly assumed to
be due to non-accidental trauma 20 years ago, today are known
to be associated with myriad causes such as stroke, raised
intracranial pressure, and the nervous system being starved of
oxygen.”

 In the two decades since Chandler’s death, “our perspective on
child abuse and intentional brain injury has evolved” such that
in a case of non-accidental trauma one would expect to see
injuries not observed in Chandler when he presented at the
hospital nonresponsive.

24
 “In 2002, the neurodiagnostic literature was rife with the belief
that chronic subdural hematomas and acute subdural blood,
when seen on the same CT scan, were commonly associated
with abuse. Twenty years later, we know that infants who
undergo scanning in the first four months of life often have
chronic subdural hematomas and other fluid collections related
to birth trauma. Twenty years later, we also know that there
can be components of acute blood within the chronic fluid that
are not necessarily caused by non-accidental trauma, but by
trivial bumps or other metabolic causes.”

 “[T]he standard of care has changed dramatically in the last
twenty years thanks to advances in science and technology. In
2002 and 2003, the standard of care was to diagnose the
symptoms observed in Chandler as the result of abuse, absent
specific diseases or a known, large-scale accident. Today,
unlike in 2002-2003, the diagnostic procedures and attention
to particular details in a child head trauma case is entirely
different.”

These pleadings at the very least allege facts that, if proved at the

hearing, would be sufficient to warrant a conclusion that the opinion

offered in support of the motion relates to new and material facts.

See Stinchcomb, 308 Ga. at 875 (2). The trial court thus erred by

denying the motion without a hearing on the basis that Dr. Ghatan’s

expert opinion was categorically excluded from the statutory

definition of newly discovered evidence.

(b) The trial court erred by denying the extraordinary motion
without a hearing on the basis that Smith had failed to show that his
25
motion was based on evidence that has come to his knowledge since
the trial.

In addition to concluding that opinion evidence could never

constitute newly discovered evidence, the trial court also denied

Smith’s motion on the ground that Smith could not obtain relief

because he had failed to show that the particular evidence on which

his motion was based has come to his knowledge since the trial.

Smith argues on appeal that the trial court erred in making these

findings without an evidentiary hearing. Here as well, we agree.

The trial court broadly concluded based on two articles that

“[t]his type of expert opinion [offered by Dr. Ghatan] has been

available since the 1990s,” such that Smith had failed to show that

his motion was based on evidence that has come to his knowledge

since the time of trial. But, on its face, Dr. Ghatan’s particular

opinion could not have been offered at the time of trial, let alone in

the 1990s. As detailed above, Dr. Ghatan’s opinion as outlined in his

affidavit purports to be based on developments that occurred after

trial. Moreover, the extraordinary motion itself contains several

statements of fact to the effect that the medical community’s
26
approach to infant head trauma has changed since the time of

Smith’s trial. In particular, the motion states that, around the time

of Smith’s trial, medical schools taught that shaking was the

primary or exclusive cause of the so-called “triad” of subdural

hemorrhage, retinal hemorrhages, and cerebral edema — and the

AAP also took the position that this “constellation” of symptoms

gave rise to a presumption of abuse and did not occur with short

falls, seizures, or as a consequence of vaccination. Now, the motion

states, “the medical and pediatric communit[ies] agree that child

abuse is no longer the presumptive diagnosis when an infant

presents with head injuries. Instead, a thorough examination of the

full medical record is necessary, as it may reveal one of numerous

possible alternative and unintentional causes, including birth

trauma.” Many of these factual assertions in the motion and

affidavit involve research, clinical observations, or organizational

changes of position occurring after the time of trial. Assuming for

the sake of our analysis the truth of these statements in the

extraordinary motion and supporting affidavit, Smith could not

27
have discovered prior to trial the same factual and opinion evidence

that he offers now.5

In addition to differences between Dr. Ghatan’s actual expert

opinion as expressed in his affidavit and a hypothetical “type of

expert opinion” that might have been offered 19 years ago, Dr.

Ghatan’s opinion is offered against the backdrop of the post-trial

scientific developments that he references. As pleaded, those

developments may make Dr. Ghatan’s actual opinion more credible

than a hypothetical, similar opinion that might have been offered at

the time of trial, in ways that are, in the parlance of the Timberlake

requirements, “so material that [they] would probably produce a

different verdict.” 246 Ga. at 491 (1) (citation and punctuation

omitted).6

5 Some of us are skeptical that changes in position by professional or

other organizations — as opposed to the scientific studies that may or may not
underlie those positions — are themselves so material that they would
probably produce a different verdict. But this appeal does not require us to
decide that question.
6 Moreover, to say that an expert willing to offer a similar opinion might

have been found in the past is not to say that such an opinion would have been
admissible at that time. Although neither party appears to have argued
explicitly that Dr. Ghatan’s testimony would have been inadmissible had it

28
been offered at the time of trial, Smith does argue that “it is only in the recent
past that enough scientific research has been peer-reviewed and published to
refute the shaken baby hypothesis as it was presented at the trial of this case”
and that “[t]his established medicine did not previously exist” such that he
“could not have presented it” at trial. The State implies that Dr. Ghatan’s
testimony might not be admissible even today. At any rate, we observe that if
an expert opinion would not have been admissible at a particular time in the
past, that expert opinion cannot be said to be “evidence” that could have been
discovered by the defendant at that particular time. Cf. Timberlake, 246 Ga. at
491 (1) (“Implicit in [the] six requirements for [granting a new trial on the basis
of newly discovered evidence] is that the newly discovered evidence must be
admissible as evidence.”). Until recently, the “opinions of experts on any
question of science” were generally admissible in criminal cases. See former
OCGA § 24-7-707 (2013); see also Debelbot v. State, 305 Ga. 534, 542 (2) (826
SE2d 129) (2019) (noting that many of the State’s challenges to medical
evidence proffered by the defendants may “be more properly characterized as
challenges to the qualification of the witnesses as experts or to the
persuasiveness of the experts’ testimony in the light of conflicting testimony
from the State’s experts”). But trial courts still were empowered to exclude
expert testimony based on a particular “procedure or technique” on the ground
that it had not “reached a scientific stage of verifiable certainty[.]” Harper v.
State, 249 Ga. 519, 525 (1) (292 SE2d 389) (1982); see also, e.g., Riley v. State,
278 Ga. 677, 683 (4) (604 SE2d 488) (2004) (trial court did not abuse discretion
in refusing under Harper to allow expert testimony on “false[-]confession
theory”). Under that approach, trial courts were cautioned against “simply
calculating the consensus in the scientific community.” Harper, 249 Ga. at 526
(1). The General Assembly recently has amended the Evidence Code, however,
to extend to criminal cases the federal standard of admissibility of expert
testimony articulated in Daubert v. Merrell Dow Pharmaceuticals, 509 U.S.
579 (113 SCt 2786, 125 LE2d 469) (1993), and its progeny. See Ga. L. 2022, p.
201, § 1 (amending OCGA § 24-7-702). Under that standard, a trial court must
evaluate the reliability of the expert’s proffered testimony; proper
considerations include “whether a theory or technique can be tested, whether
it has been subjected to peer review and publication, the known or potential
rate of error for the theory or technique, the general degree of acceptance in
the relevant scientific or professional community, and the expert’s range of
experience and training.” HNTB Ga. v. Hamilton-King, 287 Ga. 641, 642 (1)
(697 SE2d 770) (2010).
29
Rather than assuming the truth of the statements found in

Smith’s extraordinary motion and supporting affidavit for purposes

of arguing that the motion was properly denied without a hearing,

however, the State insists on challenging the credibility of these

statements. Citing various articles, the State argues that “the

specific theory advanced here — that birth trauma could have been

the cause of the victim’s injuries rather than shaken baby syndrome

— goes back nearly 30 years and was available in the 1990s, years

before the Appellant’s trial.” The State also argues that Dr. Ghatan’s

affidavit “merely offers the same type of expert opinions and theories

— thoroughly discredited by the 2018 AAP Consensus Statement —

that would have been available before his 2003 trial.” To the extent

that the State challenges the credibility of the averments of Smith

and his supporting expert that there in fact have been new

developments in the scientific community’s understanding of infant

trauma, or that Dr. Ghatan’s opinion is in fact based on those

developments, the place for that is an evidentiary hearing. By not

affording Smith a hearing in which he could show that he had

30
evidence about the cause of Chandler’s death materially different

from that which could have been discovered, introduced, and

admitted at the time of his trial, the trial court erred. See

Stinchcomb, 308 Ga. at 875 (2) (trial court errs by denying

extraordinary motion for new trial without a hearing where a

defendant pleads facts sufficient to authorize that the motion be

granted if the facts developed at the hearing warrant such relief).

(c) The trial court erred by denying the extraordinary motion on
the pleadings on the basis that Smith had failed to show that he had
exercised due diligence.

For similar reasons, the denial of Smith’s extraordinary motion

on the due diligence prong without a hearing also was error. The

trial court denied Smith’s extraordinary motion on the alternative

ground that he could not satisfy the due diligence requirement

because he had failed to show that he could not have found an expert

to challenge the State’s expert at trial — again, saying that the “type

of expert opinion” that Smith attempts to present now would have

been available at the time of trial. But, again, assuming the truth of

statements contained in the extraordinary motion and supporting

31
affidavit, Smith could not have discovered prior to trial the same

factual and opinion evidence that he offers now. To the extent that

the State challenges the credibility of the statements made in the

extraordinary motion and supporting affidavit to the effect that

scientific developments have in fact occurred subsequent to trial,

and that Dr. Ghatan’s opinion is in fact based on those

developments, that presents a factual dispute calling for a hearing.

The State notes that “[t]he statutes which control

extraordinary motions for new trial based on newly discovered

evidence require a defendant to act without delay in bringing such

a motion.” Llewellyn v. State, 252 Ga. 426, 428 (2) (314 SE2d 227)

(1984). The State argues that Smith has failed to show why he

waited 18 years after his conviction and 13 years after his direct

appeal to bring his “newly discovered evidence” to the court’s

attention. The State suggests that, taking Smith’s extraordinary

motion on its own terms, the scientific developments supporting that

motion occurred well before a group of organizations issued the 2018

Consensus Statement. In particular, the State noted, the motion

32
cited developments in 2006, 2011, and 2012.

Although the State argued below that Smith could not “show

that he exercised either pre-trial due diligence or subsequent due

diligence[,]” the trial court appears to have limited its analysis of the

due diligence prong to only whether Smith showed pre-trial due

diligence. But even assuming that the issue of Smith’s post-trial

diligence is properly before us, we cannot say that Smith’s pleadings

as to his post-trial diligence were insufficient to require a hearing.

Although the extraordinary motion indicates that a shift in the

medical community’s understanding of abusive head trauma in

infants began in 2006, it also alleges developments after that date.

In addition to the 2018 Consensus Statement, the motion cites a

2020 article about adverse events resulting from vaccination. It is

true that Dr. Ghatan’s affidavit does not make clear when exactly

the developments that he references occurred. But Dr. Ghatan’s

affidavit repeatedly suggests that his analysis is based on fairly

recent developments — repeatedly contrasting medical

understanding of infant head injuries at the time of Chandler’s

33
death with that of “today,” “twenty years later.” Moreover, as

explained above, a defendant who brings an extraordinary motion

for new trial based on new scientific developments cannot prevail

unless those developments are “so material that [they] would

probably produce a different verdict.” Timberlake, 246 Ga. at 491 (1)

(citation and punctuation omitted). And a convicted defendant may

file only one extraordinary motion for new trial, see OCGA § 5-5-41

(b), so a prudent defendant predicating such a motion on scientific

developments would wait until he is confident in the materiality of

those developments. Although the State may have plenty of room to

argue at a hearing that Smith did not act with sufficient diligence

in bringing forth the evidence that is the basis for his motion, we

cannot say as a matter of law that Smith’s pleadings are insufficient

to even obtain a hearing on that point. The trial court erred in

denying the motion without a hearing on the ground that Smith had

failed to show that he acted with due diligence. See Stinchcomb, 308

Ga. at 879 (2) (b) (concluding that trial court erred in denying

extraordinary motion based on due diligence factor without a

34
hearing).7

3. Conclusion

In sum, the trial court erred by denying Smith’s extraordinary

motion for new trial on the ground that it was based on opinion

evidence that could never support such a motion. The trial court also

erred by denying the motion, without a hearing, on the alternative

bases that Smith had not shown that his motion was based on

evidence that has come to his knowledge since the trial or that he

had brought that evidence to the court’s attention with due

diligence. The State has not opposed Smith’s bid for a hearing on

7 In his briefing before this Court, Smith argues that we should grant

him a new trial outright, because he satisfied the first two Timberlake
requirements through his written submissions and the State did not contest
the other requirements below, arguing only in the alternative that this Court
should remand the case for an evidentiary hearing. But Smith cites no
authority for mandating that his extraordinary motion be granted at this
stage. Even if the State has waived its right to contest Smith’s motion based
on certain Timberlake requirements — an issue we need not resolve here —
our conclusion that Smith has satisfied the pleading standard as to the
Timberlake requirements the State contested below means only that he is
entitled to a hearing, not that he is entitled to a new trial. To secure a new
trial, Smith will still need to prove that these requirements have been met. See
Timberlake, 246 Ga. at 491 (1) (“It is incumbent on a party who asks for a new
trial on the ground of newly discovered evidence to satisfy the court” as to the
six requirements. (citation and punctuation omitted; emphasis supplied)).
35
other grounds. We therefore vacate the trial court’s order denying

the motion and remand for the trial court to consider the motion

anew after affording Smith an evidentiary hearing.

Judgment vacated and case remanded with direction. All the
Justices concur.

Decided December 20, 2022.

Motion for new trial. Gwinnett Superior Court. Before Judge

Batchelor.

Mark A. Loudon-Brown, Whitney K. Lee, for appellant.

Patsy Austin-Gatson, District Attorney, Christopher M. DeNeve,

Brenda L. Romero, Assistant District Attorneys; Christopher M.

Carr, Attorney General, Patricia B. Attaway Burton, Deputy Attorney

General, Paula K. Smith, Senior Assistant Attorney General, for

appellee.

NOTICE: This opinion is subject to modification resulting from motions for reconsideration under Supreme Court
Rule 27, the Court’s reconsideration, and editorial revisions by the Reporter of Decisions. The version of the
opinion published in the Advance Sheets for the Georgia Reports, designated as the “Final Copy,” will replace any
prior version on the Court’s website and docket. A bound volume of the Georgia Reports will contain the final and
official text of the opinion.

In the Supreme Court of Georgia

Decided: December 20, 2022

S22A1051. SMITH v. THE STATE.

PETERSON, Presiding Justice.

Nineteen years ago, Danyel Smith was convicted of the murder

of his infant son based on a theory of “shaken baby syndrome” (SBS).

Smith now argues that the science regarding diagnosis of brain

injuries in infants has changed so much since his trial that he is

entitled to a new trial based on a new expert affidavit ruling out

battery or shaking as the cause of the baby’s death. The trial court

rejected that argument and denied Smith’s extraordinary motion for

new trial without a hearing. Because Smith’s extraordinary motion

alleged facts that, if proven, may warrant relief, the trial court was

not authorized to deny the motion without a hearing. We therefore

vacate the trial court’s ruling on the motion and remand for further

proceedings.
1. Background

a. Trial, conviction, and appeal

Smith was convicted of felony murder and aggravated battery

in connection with the death of his infant son. This Court affirmed

his convictions in 2008. See Smith v. State, 283 Ga. 237 (657 SE2d

253) (2008). In that decision, in the light most favorable to the

verdicts, we summarized the evidence presented at Smith’s

November 2003 trial as follows:

2
own. A CT scan of Chandler’s brain revealed a skull
fracture, a hematoma, and swelling of the brain. Based on
the CT scan as well as Chandler’s broken wrists, retinal
hemorrhages, and the quick onset of his symptoms,
Chandler was diagnosed as a “shaken baby,” who had
been subjected to vigorous shaking that was probably
coupled with impact. Smith was the only person with
Chandler during and immediately prior to the onset of his
symptoms. The nature of Chandler’s injuries did not
indicate that they were self-inflicted, and the injuries
were inconsistent with a simple fall or accidental trauma.
After seven days in the hospital without any evidence of
brain function, Chandler was removed from life support
and died on May 6, 2002. A physical examination of
Chandler’s body after his death revealed abdominal
bruising that was consistent with the spacing of adult
knuckles.

Id. at 237-238 (1).

Dr. Anne Frankel, the pediatrician who saw Chandler for a

check-up hours before he became nonresponsive, testified at trial

that there were complications during Chandler’s mother’s

pregnancy and that Chandler was born prematurely via Caesarean

section (“C-section”). She testified that Chandler could not have had

a life-threatening brain injury when she saw him on April 29, 2002.

Dr. William Boydston, a pediatric neurosurgeon who treated

Chandler after he arrived at the hospital, testified that he had

3
concluded that Chandler had experienced a brain injury due to

shaking, based on Chandler’s retinal subdural hematomas, blood