UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF MISSISSIPPI NORTHERN DIVISION
DEVIN BENNETT PETITIONER
V. CIVIL ACTION NO. 3:24-CV-613-KHJ
BURL CAIN, et al. RESPONDENTS
ORDER
This is a capital habeas case. Petitioner, Devin Bennett (“Bennett”), moved to stay and abate proceedings while he exhausts claims in a successive state-court post-conviction proceeding. Mot. to Stay [25]. For the reasons below, the Court grants the motion. I. Background1 On August 25, 2000, Bennett took his infant son, Brandon, to the emergency room at River Oaks Hospital in Flowood, Mississippi. When they arrived, Brandon was not breathing and had no pulse, but doctors revived him. They transferred Brandon to the pediatric unit at the nearby University of Mississippi Medical Center (“UMMC”), where the infant spent two days in a coma before he died. During this time, doctors at UMMC discovered that Brandon had bruising on his
1 The state-court record has not yet been filed, and the Court’s account of the case’s background relies heavily on the prior opinions of the Mississippi Supreme Court. , 933 So. 2d 930 (Miss. 2006); , 990 So. 2d 155 (Miss. 2008); , 383 So. 3d 1184 (Miss. 2023). scalp, shoulder, back, and forearm. He also had hemorrhaging in his eyes, a fractured skull, and blood pooling on his brain. Throughout Brandon’s medical treatment and in response to police
questioning after his death, Bennett provided several explanations for his son’s injuries, most often claiming that the infant had fallen out of his carrier. Bennett eventually admitted that he had shaken his son, but he maintained that he was trying to wake Brandon, not hurt him. On November 7, 2000, Bennett was indicted for capital murder with the underlying crime of felonious child abuse. Bennett’s trial began on February 18, 2003. The State introduced expert
testimony from Brandon’s treating physician at UMMC, a pediatric neurosurgeon from UMMC, and the state pathologist who performed Brandon’s autopsy. The experts testified that (1) a ten-month-old infant could not have lifted himself out of a carrier, (2) Brandon could not have sustained such serious injuries by simply falling out of his carrier, and (3) the injuries tracked with being shaken and thrown down on a hard surface. The jury returned a guilty verdict and death sentence on February 28, 2003.
Bennett appealed, raising fifteen assignments of error, but the Mississippi Supreme Court affirmed the conviction and sentence on May 11, 2006. , 933 So. 2d at 938–39, 956. Bennett petitioned for post-conviction relief, and on August 28, 2008, the Mississippi Supreme Court granted him leave to proceed in the trial court on a post-conviction claim of ineffective assistance of penalty-phase counsel. , 990 So. 2d at 162. The trial court held an evidentiary hearing on March 25, 2021, and denied the petition on April 1, 2021. Pet. [7] at 4. The Mississippi Supreme Court affirmed on November 16, 2023. , 383 So. 3d at 1199–1200.
Bennett petitioned this Court for a Writ of Habeas Corpus in October 2024. Pets. [6, 7]. He asserts 21 grounds for relief, and the parties agree that Grounds 17, 18, 19, and 20 have never been presented to the Mississippi Supreme Court. Resp. [21] at 4; Rebuttal [24] at 8. In Ground 17, Bennett claims that he is innocent of capital murder, citing new scientific research and evidence that was previously unavailable. [7] at 95. In Ground 18, he claims that the State suppressed
information that could have been used to impeach the testimony of the State’s expert, Dr. Steven Hayne, and knowingly presented Dr. Hayne’s false testimony at trial. at 106. In Ground 19, Bennett alleges prosecutorial misconduct, citing alleged improper comments and questioning by the prosecutors throughout the trial. at 108. Finally, in Ground 20, Bennett claims that his trial counsel provided ineffective assistance by failing to object to the prosecutor’s alleged misconduct, failing to properly cross-examine or challenge the State’s experts, and
trying another case in Chancery Court during a break in Bennett’s capital murder trial. at 113. In September 2025, Bennett moved to stay this case pending the resolution of a successive post-conviction petition which he filed in the Mississippi Supreme Court on the same day. [25]. The Court now addresses Bennett’s [25] Motion. II. Standard The Antiterrorism and Effective Death Penalty Act of 1996 (“AEDPA”) governs this case. 28 U.S.C. § 2254(a). AEDPA limits district courts’ ability to
grant habeas relief. , 28 U.S.C. § 2254(d). It was designed to “curb the abuse of the statutory writ of habeas corpus, . . . to address the acute problems of unnecessary delay and abuse in capital cases,” , 168 F.3d 762, 764 (5th Cir. 1999), and to “further the principles of comity, finality, and federalism,” , 529 U.S. 420, 436 (2000). Accordingly, a federal habeas petitioner “must exhaust all claims in state court prior to requesting federal
collateral relief.” , 515 F.3d 392, 400 (5th Cir. 2008); 28 U.S.C. § 2254(b)(1)(A). “The exhaustion requirement is satisfied when the substance of the federal habeas claim has been fairly presented to the highest state court.” , 515 F.3d at 400 (citation modified). Generally, district courts should dismiss mixed petitions—habeas petitions that include both exhausted and unexhausted claims. , 701 F.3d 171, 174 (5th Cir. 2012). Yet “because exhaustion is based on comity rather than
jurisdiction, there is no absolute bar to federal consideration of unexhausted habeas applications.” (citation modified). A district court may deny an unexhausted claim on the merits, “notwithstanding the failure of the applicant to exhaust the remedies available in the courts of the State.” (citation modified). District courts also have discretion to stay a mixed-petitioner case to allow the petitioner to present his unexhausted claims to the state court and later return to federal court for review of his perfected petition. , 544 U.S. 269, 274–79 (2005). Such stays should only be available in “limited circumstances,” where “the district court determines there was good cause for the petitioner’s failure
to exhaust his claims first in state court.” at 277. “[E]ven if a petitioner had good cause for that failure, the district court would abuse its discretion if it were to grant him a stay when his unexhausted claims are plainly meritless.” Finally, mixed petitions “should not be stayed indefinitely.” The Court should impose “reasonable time limits” on the trip back to state court and “condition the stay” on the petitioner diligently pursuing the unexhausted claims in state court. at 278.
In summary, the Court should grant a stay “if the petitioner had good cause for his failure to exhaust, his unexhausted claims are potentially meritorious, and there is no indication that the petitioner engaged in intentionally dilatory litigation tactics.” If the Court determines that a stay is not merited, it “should allow the petitioner to delete the unexhausted claims and to proceed with the exhausted claims if dismissal of the entire petition would unreasonably impair the petitioner’s right to obtain federal relief.”
III. Analysis The Court first analyzes whether Bennett has good cause for failing to exhaust his claims. Then, the Court considers whether his unexhausted claims are potentially meritorious. A. Good Cause “There is little authority on what constitutes good cause to excuse a petitioner’s failure to exhaust” under . , 745 F.3d 977, 980 (9th
Cir. 2014). The Fifth Circuit has described “good cause” as an “equitable” standard in this context. , 504 F.3d 523, 529 n.17 (5th Cir. 2007). Similarly, the Supreme Court has stated “[a] petitioner’s reasonable confusion about whether a state filing would be timely will ordinarily constitute ‘good cause’ for him to file in federal court.” , 544 U.S. 408, 416 (2005). But ineffective assistance of a habeas petitioner’s post-conviction counsel “cannot serve
as ‘good cause’ for a stay.” , 90 F.4th 857, 863 (5th Cir. 2024); , 504 F. App’x 269, 284 (5th Cir. 2012) (in context of motion for stay, post-conviction counsel’s “ignorance or inadvertence” cannot provide good cause). District courts in the Fifth Circuit have recognized “good cause” for failing to exhaust a habeas claim in these circumstances: (1) a pro se petitioner did not know about his obligation to exhaust claims, and a dismissal would have prevented him
from returning to federal court because of AEDPA’s statute of limitations, , No. H-20-3501, 2022 WL 2239851, at *7 (S.D. Tex. June 22, 2022); (2) a pro se petitioner alleged difficulty in hiring an attorney, and AEDPA’s statute of limitations would preclude refiling, , No. 3:18-CV-174-NBB-DAS, 2021 WL 2907889, at *4 (N.D. Miss. July 9, 2021); (3) a petitioner’s unexhausted claim arose from a change in applicable law, , No. CV 18-121-BAJ- SDJ, 2020 WL 7753075, at *1–2 (M.D. La. Dec. 29, 2020); (4) the “legal landscape” surrounding a petitioner’s claim was “confusing and ambiguous” when the petitioner pursued his initial state post-conviction relief, , No.
3:17-CV-3396-G-BT, 2020 WL 4877586, at *1 (N.D. Tex. Aug. 19, 2020); (5) the state appellate court effectively instructed a petitioner to raise claims of ineffective assistance of trial counsel on post-conviction, , No. CV 18-9393, 2019 WL 2469678, at *4 (E.D. La. June 13, 2019); and (6) the state appellate court would not consider claims of ineffective assistance of trial counsel on direct appeal, , No. 17-11674, 2018 WL 2013105, at *4 (E.D. La. Apr. 30,
2018). Other jurisdictions have found “good cause” in the following circumstances: (1) the petitioner provided evidence of ineffective assistance of state post-conviction counsel, , 745 F.3d at 983; (2) the petitioner had no counsel in his state post- conviction proceeding, , 847 F.3d 714, 720–21 (9th Cir. 2017); (3) a dismissal would jeopardize the timeliness of the petitioner’s habeas petition as to the unexhausted claims, , 835 F. App’x
675, 677 (3d Cir. 2020); (4) the petitioner’s counsel discovered undisclosed material after the petitioner had filed his federal habeas petition, , 592 F. App’x 422, 431 (6th Cir. 2014); (5) the petitioner did not learn of the factual basis of a claim until after his state post-conviction proceeding concluded, and he had exercised due diligence in developing the pertinent facts, , 756 F.3d 477, 486 (6th Cir. 2014); and (6) the prosecution’s wrongful withholding of information caused the petitioner’s failure to exhaust, , 657 F.3d 293, 304–05 (6th Cir. 2011). Bennett argues that there is good cause for a stay because he has discovered
new evidence that was previously unavailable to him. He contends that “the scientific and medical evidence” surrounding shaken baby syndrome (“SBS”) and abusive head trauma (“AHT”) “has drastically changed since his 2003 trial and new opinions from the medical community have surfaced.” [25] at 4. Bennett discusses this issue in Ground 17 of his Petition, an actual innocence claim. [7] at 94–105. In support of the good-cause argument, he presented a variety of evidence, including
reports from three experts. First, Dr. John Galaznik, a pediatrician with experience in cases of alleged physical abuse of infants and toddlers, provided a report on the “changing position” of the American Academy of Pediatrics (“AAP”) as to SBS and AHT, “the recent scientific changes to the understanding of short falls in children,” and “the scientific changes to the understanding of retinal hemorrhages.” Dr. Galaznik Report [32-7] at 1, 10. Dr. Galaznik stated that “[a]t the time Devin Bennett’s case was presented
and tried, pediatricians would have relied on the 2001 AAP position paper entitled ‘Shaken Baby Syndrome: Rotational Cranial Injuries – Technical Report,’” which relied on three criteria to diagnose SBS. at 11. If those criteria were present, the position paper advised doctors “to make a presumption of child abuse, specifically by abusive shaking of that infant or toddler,” and it “further declared” that the criteria “did not occur in short fall injury events or other disease processes.” In 2007, the AAP released another report that “acknowledged that a particular injury pattern is rarely . . . uniquely characteristic . . . of abuse,” and “that there is a developing recognition of other causes of brain injury that can
‘mimic’” the three criteria traditionally associated with SBS. at 11, 30. In 2009, after several years of documented cases in which “short fall impacts and other accidental events” caused the same criteria traditionally associated with SBS, the AAP “acknowledged the expiration of the 2001 Technical Report on SBS and revised its official position regarding SBS in a Policy Statement entitled ‘Abusive Head Trauma in Infants and Children.’” at 11. This publication “recommended
dropping the term ‘shaken baby syndrome’ . . . and adopting the term ‘abusive head trauma,’ which did not imply a specific mechanism of injury.” at 11–12, 45–47. It also “acknowledged other mechanisms of injury,” redefined the criteria suggesting abusive head trauma, and “acknowledged the controversy surrounding the validity of the traditional Abusive Shaking/SBS diagnosis.” at 12, 45–47. This trend continued in the 2010s as more professional literature moved further from the previous criteria used to diagnose SBS. at 12–15. Various
reports acknowledged that retinal hemorrhaging, one of the three previous criteria, can have “numerous other, non-abusive causes.” at 12, 14, 50. In 2018, the AAP revised its official position to hold that although shaking a baby can cause retinal hemorrhaging, there may be other causes, “especially in critically ill children.” at 14. It also published a “Consensus Statement,” asserting that a diagnosis of AHT “must exclude” other potential medical causes, and that an AHT diagnosis “signifies that accidental and disease processes cannot plausibly explain the etiology of the infant/child’s injuries.” at 15, 178–79. In 2020, the AAP published a policy statement titled “Abusive Head Trauma in Infants and Children,” which Dr.
Galaznik summarized as illustrative of the “controversy as to what specific primary injuries might result from an abusive shaking.” at 15. He asserted that the AAP’s 2020 policy statement “recognizes that none of these [criteria] are specific to abuse,” and that “recognized non-abusive alternatives” could cause the same symptoms. Dr. Galaznik summarized his research on the history of SBS diagnoses in the
justice system by stating that “the AAP must now acknowledge that much of what was previously thought to be true regarding the abusive shaking allegation has actually turned out to either be incomplete or has failed to be validated by experimental research.” at 16. This “raises serious doubts about the validity of previous convictions based on shaking as a mechanism for the death of an infant or toddler.” He also stated that the “scientific understanding of short falls has changed
significantly since the time Devin Bennett’s case was presented.” And he opined that “the absence of . . . rib fractures” and “structural ligament injury” in the vertebra shows that Brandon “was not the victim of abusive shaking.” at 18. Further, he observed that retinal hemorrhages were not documented until after Brandon had received “prolonged CPR,” which can cause “extensive retinal hemorrhages.” at 19. And he asserted that there were alternative explanations in the medical documentation for the bruising on Brandon’s head and body, rather than abusive blunt force impact. Dr. Galaznik concluded that “there is no evidence that [Brandon] was injured
by abusive shaking on August 25, 2000,” and that “the accidental short fall event from the bed as reported can no longer be denied as a plausible explanation for this infant’s presentation and findings and outcome.” at 20. In Dr. Galaznik’s opinion, “the evolution of our understanding of the abusive shaking hypothesis since 2003 would not support [the State’s expert] testimony today.” at 21. Bennett also provided a report from Dr. Julie Mack, a radiologist who has
researched infant cranial anatomy and physiology. Dr. Mack Report [32-1] at 19. After reviewing Brandon’s scans, medical records, and a variety of other evidence from Bennett’s trial, Dr. Mack concluded that there “are no radiological indicators of major impact” and that the “radiology is consistent with impact from a fall followed by bleeding or thrombosis and/or loss of autoregulation, possibly in a predisposed child.” at 19–20. Dr. Mack believes that it is possible Brandon was “susceptible to disproportionate effects from minor head injury” because of a “pre-existing skull
fracture” from a forceps delivery and “[d]ifficulty and delay in intubation (prolonged lack of oxygen).” She concluded: “Since there is no radiological evidence of significant impact injury (no soft tissue swelling, no visible hemorrhage in areas of fracture), no evidence of shaking (insufficient blood for bridging vein rupture, no neck injury), the fall described by the father is consistent (rather than inconsistent) with the imaging findings.” at 31–32. Finally, Bennett provided a forensic review report from Dr. Janice Ophoven, a forensic pathologist with specialized training and experience in pediatric pathology. Dr. Ophoven Report [32-6] at 1, 15. Ophoven concluded that Brandon’s
skull fracture “was not a recent injury . . . based on the absence of any documented soft tissue swelling and/or hemorrhage in the subgaleal space and absence of hemorrhage in the tissues on the surface of the skull . . . around the fracture.” at 16. She believed that the skull fracture was probably “a residual complication” of Brandon’s “traumatic birth” by forceps. at 17. In fact, Ophoven questioned whether Brandon’s skull was fractured at all. She noted that Dr. Hayne, who
performed the first autopsy, did not document a skull fracture. Dr. Emily Ward, who conducted the second autopsy, “described a . . . skull defect that she interpreted as a fracture . . . .” Ophoven concluded: Based on my review of the material available to me at this time, I have serious concerns for an unsafe conviction of Mr. Bennett. . . . The autopsy does not confirm a fatal impact to the child’s head – but certainly could have been the result of the fall described by his father. . . .
The diagnosis of the shaken baby is now recognized by many as a highly controversial and unproven theory in evaluating infant head injuries. In this case the presence of the so-called triad of injuries – retinal hemorrhages, subdural hemorrhage and brain swelling is not at all diagnostic of abusive head injury.
at 17–18. requires this Court to determine whether there is “good cause” for a stay “in the equitable sense.” , 504 F.3d at 529 n.17; , 745 F.3d at 982. Thus, requires more than a “good excuse.” , No. 3:20- CV-516-G-BN, 2020 WL 1788714, at *3 (N.D. Tex. Mar. 4, 2020), , 2020 WL 1702822 (N.D. Tex. Apr. 8, 2020). But it does not require “extraordinary circumstances.” , 745 F.3d at 981. “Petitioners
show good cause when they provide a reasonable excuse external to their efforts to comply with the [exhaustion requirement] and one that cannot be rationally placed onto them.” , 2020 WL 1788714, at *3. In assessing good cause for a stay, the question for this federal habeas Court is not whether Bennett’s new evidence existed at the time of his trial. Rather, it is whether the evidence was previously discoverable by Bennett or his counsel at
a time when the claims premised on it could have been exhausted by presentation to the Mississippi Supreme Court. , , 940 F.3d 1175, 1190–91 (5th Cir. 2019); , No. 5:13-CV-241-KHJ, 2023 WL 3574776, at *4 (S.D. Miss. May 19, 2023). Respondents argue that Bennett could have presented his unexhausted claims at any point during the thirteen years his post-conviction proceeding was pending in the trial court. Defs.’ Mem. in Opp’n [27] at 4–5. Bennett disagrees,
contending that he could not have exhausted the new claims during the post- conviction proceeding because the Mississippi Supreme Court “granted [the trial court] leave to proceed only on issues of ineffective assistance of counsel in sentencing,” [25] at 5, so the trial court’s “jurisdiction was limited by the remand order,” Reply [30] at 4. The Court agrees that Bennett could not have pursued the new claims or sought leave to amend his post-conviction petition in the trial court. The Mississippi Supreme Court has held that when it grants leave to proceed in the trial court on a
petition for post-conviction relief, the “trial court [has] no jurisdiction to address any claims outside those specifically addressed in [the Mississippi Supreme Court’s] remand order.” , 163 So. 3d 240, 263 (Miss. 2014) (citing , 897 So. 2d 960, 970–71 (Miss. 2004); , 879 So. 2d 1000, 1003 (Miss. 2004)). Thus, a trial court hearing a claim for post-conviction relief on remand from the Mississippi Supreme Court may not “broaden the scope of the
hearing or . . . allow a proffer on . . . proposed new claims.” , 897 So. 2d at 971. And since most of the research relied on by Bennett’s experts did not exist until after the Mississippi Supreme Court had remanded the case to the trial court,2 Bennett could not have discovered the new evidence in time to include claims premised on it in his original petition for post-conviction relief. Though he could not have brought these claims in his original petition or sought leave to amend his petition in the trial court, he could have filed a request to
pursue amended post-conviction claims in the Mississippi Supreme Court. , 897 So. 2d at 97 (acknowledging that the proper forum to file amended post-
2 The Mississippi Supreme Court remanded the case in 2008. Dr. Galaznik described multiple publications from 2009 or later, and 12 of the 16 exhibits attached to his report were published after 2008. [32-7] at 12–15, 44, 50, 61, 65, 92, 97, 104, 138, 145, 168, 178, 197. And according to Dr. Galaznik, three of the most significant changes occurred a decade or more after the Mississippi Supreme Court remanded the case—the 2018 Consensus Statement, the 2018 AAP COCAN Clinical Report, and the 2020 AAP Policy Statement on “Abusive Head Trauma in Infants and Children.” at 14–15. conviction claims is the Mississippi Supreme Court). The trial court did not hold a hearing on his petition for post-conviction relief until March 2021—after some of the new research on SBS and AHT was discoverable. Despite some of the evidence
being available as early as 2020, he did not seek leave to amend his petition for post-conviction relief or otherwise present the new claims to the Mississippi Supreme Court until after he initiated this federal habeas case.3 Still, the Court finds that Bennett has good cause for his failure to exhaust Grounds 17–20 of the Petition. He arguably could have presented his new claims to the Mississippi Supreme Court in a brief window of time before he began this
federal habeas action, but, in practice, it can be difficult to pinpoint specific dates on which a particular scientific theory gains or loses consensus. Therefore, based on the current record, the Court declines to make a factual finding about the specific date on which the collective body of new scientific research reached a critical mass, triggering Bennett’s obligation to exhaust the new claims.4 The equitable nature of a stay advises against such a hyper-technical analysis. It is a close enough call that the Court concludes Bennett has satisfied his burden.
3 But even if Bennett had moved for leave to amend with the Mississippi Supreme Court, the court could have construed it as a successive petition. , 987 So. 2d at 971–72 (affirming the lower court’s denial of post-conviction relief and acknowledging a pending motion to amend); , No. 2002-CA-28-SCT, Order at 2 (Miss. Apr. 21, 2005) (treating a motion for leave to amend as a successive petition). So his case could have ended up in the same posture as it stands now. 4 To be clear, the Court expresses no opinion on the validity of Bennett’s claims about the purported change in scientific consensus, or whether such change would have made a difference in his trial. As noted below, the Mississippi Supreme Court should have a chance to address that issue before Bennett presents it to this federal habeas court. The Court acknowledges that the Mississippi Supreme is currently tasked with determining whether Bennett’s new evidence was “reasonably discoverable at the time of trial” and “of such nature that it would be practically conclusive that, if
it had been introduced at trial, it would have caused a different result in the conviction or sentence.” Miss. Code Ann. § 99-39-27(9); , No. 2025-DR-1074-SCT, Mot. to File Post-Conviction Relief Pet. at 21–22 (Miss. Sept. 9, 2025). For that reason, principles of federalism and comity demand that, without evidence drawing a bright line, the Court avoid making a specific finding of fact as to the date on which Bennett’s body of new evidence was reasonably
discoverable. B. Merits “[E]ven if a petitioner had good cause for [his failure to exhaust], the district court would abuse its discretion if it were to grant him a stay when his unexhausted claims are plainly meritless.” , 544 U.S. at 277. “In determining whether a claim is ‘plainly meritless,’ principles of comity and federalism demand that the federal court refrain from ruling on the merits of the claim unless it is perfectly
clear that the petitioner has no hope of prevailing.” , 847 F.3d at 722. The Court must not “deprive state courts of the opportunity to address a colorable federal claim in the first instance and grant relief if they believe it is warranted.”
In the spirit of comity and federalism, the Court declines to delve deeply into the factual merits of each of Bennett’s unexhausted claims. Based on the current record, the Court concludes that Bennett has asserted colorable claims for relief. For now, it is not “perfectly clear” to the Court that he has no hope of prevailing on the unexhausted claims. Nor does the record suggest that Bennett has engaged
in intentionally dilatory litigation tactics. IV. Conclusion The Court has considered all arguments. Those not addressed would not have changed the outcome. For the stated reasons, the Court GRANTS Petitioner’s [25] Motion. , 423 F.3d 474, 478–79 (5th Cir. 2005) (actual innocence claim had to be exhausted in state court before assertion in federal
habeas case). This case is stayed until further notice. The Court further orders this case be administratively closed for statistical purposes until Bennett pursues his unexhausted claims in state court. Bennett has already filed a motion in the Mississippi Supreme Court seeking leave to file a successive petition for post-conviction relief in the trial court. The Court orders Bennett to file a status report in this federal habeas case every 120 days until resolution of the state-court proceeding. The parties shall provide notice
to this Court no later than thirty days after resolution of the state-court proceeding. Finally, Bennett must diligently pursue his unexhausted claims in state court, or the Court may lift the stay on this habeas case. SO ORDERED, this 1st day of April, 2026. s/ Kristi H. Johnson UNITED STATES DISTRICT JUDGE