Derek Sivo v. A.T. WALL, Director, Rhode Island Department of Corrections

United States Court of Appeals
For the First Circuit
No. 10-1836
DEREK SIVO,
Petitioner, Appellant,
v.
A.T. WALL,
Director, Rhode Island Department of Corrections,
Respondent, Appellee.
APPEAL FROM THE UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF RHODE ISLAND
[Hon. William E. Smith, U.S. District Judge]
Before
Torruella, Boudin and Lipez,
Circuit Judges.
Richard K. Corley, by appointment of the court, for appellant.
Aaron L. Weisman, Assistant Attorney General, with whom
Patrick C. Lynch, Attorney General, was on brief for appellee.
June 30, 2011

-- 1 of 17 --

The facts are primarily drawn from the Rhode Island Supreme 1
Court's decision affirming Sivo's conviction, State v. Sivo, 925
A.2d 901 (R.I. 2007), and are supplemented when necessary with
facts from the record when they are consistent with the Rhode
Island Supreme Court's findings, see Lynch v. Ficco, 438 F.3d 35,
39 (1st Cir.), cert. denied 549 U.S. 892 (2006).
-2-
BOUDIN, Circuit Judge. Derek Sivo, convicted of first-
degree child abuse in state court, appeals from the federal
district court's denial of his petition for a writ of habeas
corpus. The charge against Sivo was that he inflicted injuries on
John W. Jr. ("J.J."), the then two-and-a-half-year-old son of
Sivo's girlfriend, Kimberly Mathieu ("Kim"). Sivo's central claim
now before us is one of insufficient evidence. We outline briefly
the central facts and prior proceedings.1
On November 1, 2001, Kim and J.J. woke up at around 7:30
or 8:00 a.m. and went out for breakfast. J.J. said that he did not
feel well and wanted to stay home that day. Kim left for work at
around 11:15 a.m., leaving J.J. in the care of her boyfriend, Sivo.
Kim called home between 3:30 and 4:30 p.m. and Sivo said that J.J.
was sick.
Upon Kim's return home from work at around 6:30 p.m.,
Sivo told Kim that J.J. was not feeling well and had earlier that
day fallen down some of the stairs that led to their basement
apartment. Sivo reported that when J.J. fell, he was not crying
and seemed mostly okay. When Kim went in to check on J.J., who was
sleeping, he awoke and said "Hi, Mommy," and then resumed his

-- 2 of 17 --

-3-
sleep; Kim felt that J.J. was running a "little fever," and so she
gave him Motrin and checked on him periodically during the night.
The next morning, on November 2, 2001, J.J. did not wake
up as usual at 7:30 or 8:00 a.m. When Kim awoke at 10:00 a.m. and
realized that J.J. was still sleeping and found him "not
responsive," she called J.J.'s pediatrician, Dr. Cheryl Flynn.
When Kim brought J.J. to Dr. Flynn's office later that day, Dr.
Flynn found that J.J. was barely awake but "arousable"; was having
trouble "hold[ing] himself up"; and "could not bear [his own]
weight or walk." At Dr. Flynn's direction, J.J. was taken to
Hasbro Children's Hospital.
At the hospital, J.J. was found to be in an "altered
level of consciousness" and was "virtually nonresponsive." J.J.
was treated by Dr. John Allison Duncan III, chief of neurosurgery
at Brown University Medical School and an expert in pediatric
neurosurgery, and Dr. Seth Asser. They determined that he had
incurred a subdural hematoma--a collection of blood on the surface
of the brain--and was suffering increased intercranial pressure
that threatened his life.
In an emergency operation, the surgeons removed two blood
clots from J.J.'s brain--a "fresh" one caused by an injury that the
doctors determined occurred sometime between a few days to a few
hours before the surgery, and an older one that was four to eight
weeks old. While the surgery saved J.J.'s life, he suffers

-- 3 of 17 --

-4-
significant complications from his injuries, including weakness on
his right side, lack of peripheral vision, speech problems and a
permanent limp. Hasbro's Child Protection Team contacted the
police.
A detective from the Cranston Police Department arrived
at Hasbro at 7:00 p.m. the same evening, questioned Kim and one of
J.J.'s doctors and then summoned Sivo. Sivo came to the station at
10:40 p.m. and repeated what he had told Kim earlier: that J.J.
fell down the last few stairs on the way down to their basement
apartment; that J.J. was not crying after the fall, and only
indicated that his "butt" and the back of his head hurt; and that
J.J. napped later that day because he was ill. Sivo also gave
similar statements to other investigators.
On February 13, 2002, the Cranston Police Department
filed a criminal information charging Sivo with one count of first-
degree child abuse of a child under the age of five, in violation
of state law. R.I. Gen. Laws § 11-9-5.3 (2002). The offense
occurs when "a person having care of a child . . . knowingly or
intentionally . . . [i]nflicts upon a child serious bodily injury."
Id. § 11-9-5.3(b)(1). Serious bodily injury is defined to include
injury that creates a substantial risk of death, causes protracted
impairment of function or evidences subdural hematoma. Id. § 11-9-
5.3(c).

-- 4 of 17 --

-5-
At trial, the jury heard--in addition to what is
recounted above--testimony from all three doctors who saw J.J. on
November 2. Doctors Duncan and Asser testified that the subdural
hematomas must have been caused by a severe blow such as a fall
from a significant height; that the injury was "non-accidental";
that (according to Dr. Duncan) a fall down the stairs could explain
the injury only if J.J. were thrown down the stairs and landed
primarily on his head; and that (according to Dr. Asser) child
abuse was the likely cause. None of the three doctors identified
signs of abuse other than the subdural hematomas.
Sivo presented as a witness Dr. Thomas Morgan, a
neurologist; Dr. Morgan gave as his medical opinion J.J. was not
the victim of abuse: J.J. had none of the telltale signs of being
abused, including facial bruises, a black eye, a broken jaw, chest
or pelvic injuries, fractured skull, broken ribs or swollen tissue.
Dr. Morgan suggested that the subdural hematomas were likely caused
by an August 2001 seizure that resulted in J.J. becoming weaker on
his right side and more prone to falls.
Dr. Morgan further testified that J.J. was predisposed to
developing subdural hematomas, and so any fall--including a fall
down the stairs--could have triggered one; and that the lack of
external evidence of abuse and J.J.'s behavior on November 1-2--at
first normal but becoming progressively weaker and more lethargic--
was consistent with Dr. Morgan's assessment. Dr. Morgan conceded

-- 5 of 17 --

-6-
that he never examined J.J., but instead based his conclusions on
a review of police reports, medical records and an interview with
Kim.
There was conflicting evidence on yet another matter.
According to Kim, falling was not unusual for J.J.; she claimed
that after two falls in August 2001, J.J. began falling frequently,
a matter that concerned Kim enough to raise it with J.J.'s
pediatrician. J.J.'s maternal grandmother, Roseann Mathieu
("Roseann"), and his biological father, John W. Sr., testified for
the state, asserting that J.J. was not clumsy and fell no more than
an average two-year-old. Roseann also testified that J.J. often
did not want to go to Sivo's home.
At the close of evidence, Sivo moved for a judgment of
acquittal, which was denied. On October 23, 2002, the jury found
Sivo guilty of first-degree child abuse; he was later sentenced to
twenty years imprisonment, twelve to serve and eight suspended with
probation. Although the Rhode Island Supreme Court ordered a
hearing in the trial court on Sivo's new trial motion, that motion
failed and the Rhode Island Supreme Court ultimately affirmed the
conviction and sentence.
Thereafter, a federal district court considered a habeas
petition by Sivo, 28 U.S.C. § 2254 (2006), ultimately denying
relief but granting a certificate of appealability on one issue,
id. § 2253(c)(2), namely, "whether the denial of Sivo's motion for

-- 6 of 17 --

-7-
judgment of acquittal violated his constitutional rights." Our
review of the district court is de novo, Santiago v. O'Brien, 628
F.3d 30, 33 (1st Cir. 2010), but our review of the underlying state
court determination--that the evidence against Sivo was
sufficient--is more restricted.
Specifically, under the habeas statute, factual
determinations by state trial and appellate courts are presumed
correct unless disproved by "clear and convincing evidence." 28
U.S.C. § 2254(e)(1); Clements v. Clarke, 592 F.3d 45, 47 (1st
Cir.), cert. denied, 130 S. Ct. 3475 (2010). As to legal issues,
the state court legal ruling must stand unless it is contrary to
Supreme Court precedent or amounts to "an unreasonable application
of" clearly established such precedent, 28 U.S.C. § 2254(d)(1),
which requires more than mere disagreement by the federal court,
McCambridge v. Hall, 303 F.3d 24, 36 (1st Cir. 2002) (en banc).
The federal constitutional rule invoked by Sivo is based
on the due process clause of the Fourteenth Amendment and set forth
in Jackson v. Virginia, 443 U.S. 307, 313-16 (1979). The standard
under Jackson is "whether, after viewing the evidence in the light
most favorable to the prosecution, any rational trier of fact could
have found the essential elements of the crime beyond a reasonable
doubt." Id. at 319. The Rhode Island court evaluated Sivo's claim
under state law doctrine equivalent to Jackson, see State v. Day,
925 A.2d 962, 974 (R.I. 2007), so the "unreasonable application"

-- 7 of 17 --

-8-
test governs, Foxworth v. St. Amand, 570 F.3d 414, 426 (1st Cir.
2009), cert. denied, 130 S. Ct. 1710 (2010).
In practice habeas review under Jackson, constrained by
doctrines already described, is reserved for unusual cases and its
standard "is rarely met where there is plausible evidence to
support a verdict." Tash v. Roden, 626 F.3d 15, 20 (1st Cir.
2010). Put differently, Jackson applies where there is no
substantial evidence of guilt or where the evidence amounts to
little more than colorable speculation. O'Laughlin v. O'Brien, 568
F.3d 287, 300-01 (1st Cir. 2009), cert. denied, 130 S. Ct. 1142
(2010). In this case the evidence against Sivo was circumstantial
but it was nevertheless substantial.
First, the two doctors who were most closely involved in
treating J.J. for the subdural hematoma testified that it could
only have occurred with an extremely powerful blow or by being
thrown from a considerable height and was not consistent with a
mere stumble down a few stairs. The defense expert disagreed but
he had not examined J.J., and in any event the jury, which heard
the witnesses testify, is entitled to evaluate their credentials
and the persuasiveness of their explanations. See United States v.
Calderon, 77 F.3d 6, 10 (1st Cir. 1996).
Second, Sivo was alone with J.J. for the pertinent
period. The state's medical testimony was that once J.J.'s injury
occurred, he would not have been able to function normally: Dr.

-- 8 of 17 --

-9-
Asser testified that J.J.'s injury would not be "consistent with
any kind of lucid interval," and that J.J. would likely have been
groggy, would not have been able to stand or walk, would not
respond if someone talked to him or moved him, and would have been
moaning. Yet, Kim's testimony indicates that, at breakfast the
same day and before he was entrusted to Sivo, J.J. was functioning
normally even if he did not feel well.
Third, Sivo's own statement to Kim during their phone
call--he did not testify at trial--confirmed that J.J.'s condition
worsened while he was in J.J.'s care and he was asleep and
apparently groggy when his mother arrived home, worse the next
morning and by that afternoon almost unable to function. Combined
with the medical testimony, this fairly indicates that something
occurred while J.J. was in Sivo's care that precipitated the
injury.
Fourth, both Dr. Duncan and Dr. Asser testified that the
injury was consistent with J.J. having been hit hard with a hard
object; there was no evidence that he had fallen from a great
height or been in a car crash; and the only explanation provided
before trial by Sivo (that J.J. had stumbled down a few stairs) was
ruled out by Dr. Duncan as sufficient to cause so serious an
injury. Roseann's testimony that J.J. did not want to be left with
Sivo reinforced, if only slightly, the likelihood that Sivo had
struck the child.

-- 9 of 17 --

E.g., Thomas Pittman, Significance of a Subdural Hematoma in 2
a Child with External Hydrocephalus, 39 Pediatric Neurosurgery 57,
57-58 (2003) (noting that some conditions "predispose a child to
develop a subdural hematoma after even minor injury," and
discussing specifically one condition--external hydrocephalus--that
-10-
On appeal, Sivo points out that no one can show exactly
how the injury was incurred, but that is of little help to his
position. If the jury could rationally conclude that the only
plausible injury was a fierce blow and that only Sivo was around to
administer it, the precise motive, weapon and other details are not
critical to the chain of reasoning. Nor does the lack of obvious
signs of prior abuse absolve Sivo. A single act of great violence
is as much a violation as a string of episodes.
Sivo also says that the state failed to rule out all
other possible scenarios that would be consistent with Sivo's
innocence, but that is not part of the state's burden: "'[b]eyond
a reasonable doubt' does not require the exclusion of every other
hypothesis; it is enough that all 'reasonable' doubts be excluded."
Stewart v. Coalter, 48 F.3d 610, 616 (1st Cir.), cert denied, 516
U.S. 853 (1995). In fact, Dr. Asser testified that there was no
reasonable explanation other than deliberate violence for this
injury.
Sivo also submitted to this court articles published in
medical journals that express the view that subdural hematomas can
arise in some children for reasons other than abuse and severe
trauma. However, one such article explains that in a study of the 2

-- 10 of 17 --

might make children especially prone to developing subdural
hematomas).
-11-
causes of subdural hematomas, only four percent of those caused by
accident were the result of falls, all from a height of "greater
than 10 feet onto unyielding surfaces," Kenneth Feldman et al., The
Cause of Infant and Toddler Subdural Hemorrhage: A Prospective
Study, 108 Pediatrics 636, 643 (2001), and discusses an extensive
body of literature that suggests that "stairway falls are unlikely
to cause serious infant and toddler injury," id. at 637.
What is more, the articles are not evidence presented to
the jury and the question is whether a rational jury could convict
on the evidence before it. Leftwich v. Maloney, 532 F.3d 20, 27
(1st Cir. 2008). For habeas claims generally, the record is that
developed in the state court, Cullen v. Pinholster, 131 S. Ct.
1388, 1398-99 (2011), save in very limited circumstances not
present here--for example, when evidence was not discoverable
through due diligence at the time of the trial, see 28 U.S.C.
§ 2254(e)(2); Leftwich, 532 F.3d at 27 & n.6.
This case is readily distinguishable from O'Laughlin, 568
F.3d at 287. There, the defendant was a highly plausible suspect
in a murder, but there was no evidence that he had in fact been
present--let alone uniquely present--at the time that murder
occurred. J.J. was unquestionably in Sivo's custody, and the

-- 11 of 17 --

-12-
explanation for the injury given in Sivo's statement to the police
was directly contradicted by the testimony of two doctors.
In sum, the Rhode Island Supreme Court did not
unreasonably apply the Jackson standard in upholding the jury's
verdict, and the district court did not err in denying the petition
for habeas.
Affirmed.
--Concurring opinion follows--

-- 12 of 17 --

Dr. Duncan testified that he did not see any marks on J.J.'s 3
body evidencing the cause of his injury and that he could not say
where or how J.J.'s injury occurred, but could pinpoint the
approximate time of J.J.'s two brain injuries. State v. Sivo, 925
A.2d 901, 907 (R.I. 2007).
-13-
TORRUELLA, Circuit Judge, Concurring. I join the
majority because the deferential standard of review compels that we
affirm Derek Sivo's conviction, but write separately to stress how
near to the boundary this case falls.
The evidence, when viewed in the light most favorable to
the prosecution, showed that J.J.'s injury required a high degree
of force -- far in excess of the defendant's suggested cause -- and
that it was suffered while in the care of the defendant. The
prosecution, however, offered no explanation as to how or why
J.J.'s injury took place, other than to generally attribute the 3
injury to the defendant. On this decidedly sparse basis a jury
returned a conviction, which was then in turn upheld by the Rhode
Island Supreme Court. See State v. Sivo, 925 A.2d 901 (R.I. 2007).
Although circumstantial evidence can be sufficient to
sustain a conviction, O'Laughlin v. O'Brien, 568 F.3d 287, 302 (1st
Cir. 2009), cert. denied, 130 S. Ct. 1142 (2010), it must still
meet some constitutional baselines. A conviction will survive
habeas review if "after viewing the evidence in the light most
favorable to the prosecution, any rational trier of fact could have
found the essential elements of the crime beyond a reasonable

-- 13 of 17 --

Dr. Duncan testified that a blow, sufficiently powerful to 4
leave a subdural hematoma, would have left J.J. in such a state
that his mother would have noticed before she left. See Sivo, 925
A.2d at 907.
-14-
doubt." Jackson v. Virginia, 443 U.S. 307, 319 (1979) (emphasis in
original).
We accept, as we must, the assessment (by the
prosecution's experts) of the approximate time as of which J.J.
suffered his injury, and, consequently, that the injury took place
while in the sole care of the defendant. However, this evidence 4
is not sufficient, by itself, to sustain Sivo's conviction, since
one of the elements of the crime in question requires that the
defendant "knowingly or intentionally" inflict serious bodily
injury. R.I. Gen. Laws § 11-9-5.3(b)(1) (2002). Therefore, only
in conjunction with the finding that the injury was knowing or
intentional can the prosecution's evidence suffice. Dr. Duncan and
Dr. Asser testified to this effect, opining that the injury was
"non-accidental." Sivo, 925 A.2d at 907-08. How they were able to
reach such a conclusion from merely observing a subdural hematoma
is at least one of the circumstances that causes me to have
concerns regarding the outcome we must approve. Be that as it may,
this was apparently valid opinion evidence, which was rebutted by
defendant's expert and ultimately believed by the jury. Hence the
question becomes whether the jury can convict a defendant -- of
knowingly or intentionally inflicting serious bodily injury to a

-- 14 of 17 --

-15-
child -- on the basis that the defendant had sole access to the
child when the latter suffered an injury that, according to the
prosecution's expert witnesses, could not have been accidental.
The criminal justice system must allow for some
inferences in order to properly function, but courts must endeavor
to ascertain the point at which a reasonable inference becomes
overly speculative. This case poses such a quandary. Although we
must draw all reasonable inferences in favor of the prosecution, "a
reviewing court must refrain from giving credence to evidentiary
interpretations and illations that are unreasonable, insupportable,
or overly speculative." Foxworth v. St. Amand, 570 F.3d 414, 425
(1st Cir. 2009) (internal quotations omitted). Some level of
intellectual rigor beyond bald acceptance of state court findings
is required. Id. We approached a similar issue in O'Laughlin,
overturning a denial of a habeas petition where the prosecution's
case relied primarily on the defendant's mere presence near the
scene of the crime. O'Laughlin, 568 F.3d at 302. While the
instant case may have a superficial congruency to O'Laughlin, there
remains a crucial distinction. Whereas in O'Laughlin there were
numerous individuals with access to the victim, Sivo had sole
custody of J.J. during the critical period. Hence, Sivo's
conviction resulted not from the rank speculation found in
O'Laughlin, but from a reasonable inference that the only
individual capable of inflicting the injury did in fact do so, a

-- 15 of 17 --

-16-
criminal res ipsa loquitur. The line separating permissible
inferences from groundless speculation defies clarity, but it lies
somewhere between the instant case and O'Laughlin.
In the present case, a rational jury could find Derek
Sivo guilty beyond a reasonable doubt, if it believed the
government's circumstantial case. That said, although I concur
with the majority, I wish to stress how scant the evidence against
Sivo was. The prosecution's medical testimony was barely
sufficient to sustain the conviction. If anything, the
prosecution's doctors disproved the defendant's suggested
explanation for J.J.'s injury, but failed to provide a more
specific cause than "acceleration and deceleration of the brain
like that caused by a car accident or a fall from a height of
several stories." Sivo, 925 A.2d at 908. There was no history of
child abuse or outward signs of the cause of J.J.'s brain injury
and the prosecution could offer no explanation for why or how Sivo
purportedly administered the injury, even though such an injury
allegedly required the force of a car accident, severely limiting
the possible causes. Id. at 907-08. Further, Dr. Duncan testified
that typically there is a visible mark on the scalp of a child who
suffers a subdural hematoma, yet none of the three doctors who
examined J.J. saw one. Id. at 907. In addition, it is debatable
how the prosecution's expert witnesses could have opined with
confidence that J.J.'s injury was "non-accidental," where they

-- 16 of 17 --

-17-
admitted to not being able to ascertain how or where it occurred
and where the possible explanations provided by such experts (e.g.,
a car accident or a fall from height) are situations that may also
ostensibly occur accidentally.
The prosecution's construction of the pertinent time line
was also suspect. J.J.'s mother was able to communicate with him
after the alleged blow (e.g., J.J. responded "Hi, Mommy" when Kim
checked on him after returning from work on November 1), see Sivo
v. Wall, 2010 U.S. Dist. LEXIS 65045, *13-14, 2010 WL 2636024, *4
(D.R.I. June 28, 2010), in direct contradiction to Dr. Asser's
testimony that, immediately following the injury, J.J. would not
have responded if someone talked to him or moved him, and would
have been moaning, see Sivo, 925 A.2d at 908.
The aforementioned inconsistencies, compounded by the
generally meager offering of proof, should have given the jury
pause. Instead, the jury convicted and put a man in jail on a
showing of evidence that flirts with the fringes of permissibility,
even under Jackson's limited review.
Though I believe this case to be exceedingly close, given
the limited review which is available to this court, I have no
choice but to concur in affirming this outcome. I do so, however,
with extreme reluctance. I would not have convicted Derek Sivo,
but the jury was within its rights, if ever so barely.

-- 17 of 17 --