Wilson v. Sellers

LII note: The U.S. Supreme Court has now decided Wilson v. Sellers.


Can a district court look past a state appellate court’s summary denial of a habeas petition and review the lower state court decision to determine whether the appellate court’s decision to deny habeas relief was unreasonable?

Oral argument: 
October 30, 2017

This case asks the Supreme Court to determine whether the Antiterrorism and Effective Death Penalty Act (“AEDPA”) allows a federal court reviewing state habeas decisions to “look through” a summary decision made by a state appellate court and instead examine the last reasoned opinion of a lower state court. Petitioner Marion Wilson, the habeas petitioner, contends that the Supreme Court case Ylst v. Nunnemaker guides district courts to “look through” state appellate court summary decisions to reasoned lower state court decisions when examining a state criminal proceeding’s constitutionality. Respondent Eric Sellers, Warden of the Georgia Diagnostic and Classification Prison, argues that the later Supreme Court case of Harrington v. Richter, along with the text of § 2254, rejects Ylst’s “look through” method. This case has potential implications for federalism concerns and the structure of state habeas proceedings.

Questions as Framed for the Court by the Parties 

Did this Court’s decision in Harrington v. Richter, 562 U.S. 86 (2011), silently abrogate the presumption set forth in Ylst v. Nunnemaker, 501 U.S. 797 (1991) – that a federal court sitting in habeas proceedings should “look through” a summary state court ruling to review the last reasoned decision – as a slim majority of the en banc Eleventh Circuit held in this case, despite the agreement of both parties that the Ylst presumption should continue to apply?


In 1997, Petitioner Marion Wilson was charged with and convicted of murder, armed robbery, hijacking a motor vehicle, and several other violent crimes. During the sentencing portion of the trial, Wilson’s counsel argued that the death penalty was undeserved, in part because Wilson had experienced a violent childhood. After jury deliberations, Wilson was sentenced to death.

Wilson then petitioned for a writ of habeas corpus in state court, contending that his counsel failed to fully investigate mitigating evidence regarding his childhood. The lower state court denied Wilson’s petition, finding that his counsel was not ineffective and, even if they had been, their failure to investigate did not negatively affect the outcome of the trial, and thus did not meet the prejudicial standard. Wilson then applied for a certificate of probable cause to appeal his petition’s denial, which the Supreme Court of Georgia summarily denied without issuing an explanatory opinion.

Wilson then petitioned for a writ of habeas corpus in federal district court under 28 U.S.C. § 2254, the Antiterrorism and Effective Death Penalty Act of 1996 (“AEDPA”). Section 2254 reformed the federal habeas review process by directing federal courts to focus on the decisions of the reviewing state courts. The district court denied Wilson’s petition, finding that the state court’s determination that his counsel had not been ineffective reasonably comported with the law.

The district court’s opinion was affirmed by the Eleventh Circuit Court of Appeals, which explained that the Supreme Court of Georgia’s summary denial was the final decision “on the merits” subject to review under § 2254(d) and thus the only decision that needed to be found reasonable. The Court of Appeals then concluded that the Supreme Court of Georgia could have reasonably denied Wilson’s writ of habeas corpus after finding that additional evidence would not have created a reasonable probability of a different sentence.

The Eleventh Circuit granted rehearing en banc to determine whether the district court should have “looked through” the Supreme Court of Georgia’s summary denial to review the lower state court’s reasoned opinion. The majority concluded that the federal court need not have looked through, as the Supreme Court of Georgia had examined evidence and reached a decision on the merits. The court found that this determination satisfied § 2254(d)’s review of “the last-state court adjudication on the merits.”

The majority also held that the district court was not required to “look through” the Supreme Court of Georgia’s summary denial since the Harrington v. Richter test for reviewing a summary decision only required federal courts to determine whether there were theories consistent with legal precedent that could have supported the summary decision. They refused to extend Ylst v. Nunnemaker—which had implicitly directed courts to look past the appellate court opinions to reasoned lower court opinions—to Wilson’s case, as they did not believe that the Supreme Court of Georgia necessarily adopted the lower court’s reasoning when issuing its summary denial. Five judges dissented on this point, finding that the district court should have presumed that the Supreme Court’s summary judgment implicitly adopted the Superior Court’s reasoning.

Wilson petitioned the Supreme Court for a writ of certiorari in November 2017, which was granted in February 2017.



Wilson argues that when a federal court reviews a state court habeas decision for an unreasonable determination of law or fact under § 2254(d), Supreme Court precedent suggests that the review should focus on what the reviewing state court “knew and did.” Wilson contends that the best way in this case to know what the reviewing state court knew and did is to look through the habeas denial and consider the reasoned decision of the lower court. Under AEDPA, Wilson maintains, a federal court must review the reasonableness of both the outcome and the underlying findings and determinations yielding that outcome. Wilson notes that the Supreme Court has followed the “what the state court knew and did” approach in every § 2254(d) case except one—Harrington v. Richter. Wilson contends that this approach respects principles of federalism and comity, as well as the state court’s resources. Wilson asserts that Richter is a unique case because the state appellate court issued a summary decision without explanation after no other state court issued a full opinion on the merits. According to Wilson, in unique cases like Richter where no reasoned merits decision exists, a federal court may only “imagine . . . what ‘could have supported’ the [state court’s] decision” in a § 2254 analysis. Wilson argues, then, that Richter did not contravene the approach directing courts to evaluate “what the state court knew and did” for federal habeas proceedings.

On the other hand, Respondent Eric Sellers, Warden of the Georgia Diagnostic and Classification Prison, argues that in Richter the Court held that § 2254’s deference applies to the last state court merits decision regardless of whether the decision included the court’s reasoning for denying relief. Thus, according to Sellers, when the last state court’s merits decision includes no explanation, the petitioner must meet the burden of showing that “there was no reasonable basis for the state court to deny relief.” Sellers also asserts that Richter instructs courts to identify any possible arguments that could have yielded the state court’s decision and then determine whether any of those arguments provide a reasonable basis for the decision. Sellers notes that, in the past, federal appeals courts reviewed § 2254 cases and state court merit decisions using this approach and Richter merely confirmed the approach. Sellers argues that the Richter approach is superior because it does not automatically assume that the state appellate court has adopted unreasonable lower court opinions, even if the state appellate court does not issue a supporting opinion or explanation. To hold otherwise, according to Sellers, would reinstate the pre-AEDPA “grading papers” regime in which federal courts essentially reviewed state court merit decisions concerning § 2254 de novo. Sellers further contends that this “reasonable basis” standard protects the “integrity” of state case law and respects principles of federalism and comity by not requiring the state court to issue a written opinion.


Wilson argues that the Supreme Court stated in Ylst v. Nunnemaker that “silence implies consent, not the opposite,” here meaning that when a state appellate court summarily affirms the decision of a lower court to reject a claim, the court does so based upon the same reasoning provided by the lower court. Wilson contends that the federal courts, then, must “look through” to the last reasoned state court decision as the best evidence for deciding cases concerning § 2254. Although the Ylst case dealt with a different issue within federal habeas review, Wilson asserts that federal courts find the Ylst presumption appealing when applying § 2254 because the presumption is easy to apply and is consistent with AEDPA. Wilson notes that historically every Circuit Court of Appeals to consider the issue—with the exception of the court below—has “look[ed] through” to the last reasoned state court decision when applying § 2254. Even the Supreme Court, according to Wilson, adopted the “look through” method when applying § 2254 in post-Richter cases. Thus, Wilson argues that federal courts should apply the Ylst “look through” method to federal habeas review as the appropriate method for evaluating summary state appellate court decisions pursuant to § 2254. Furthermore, adopting Sellers’s and the Eleventh Circuit’s expansive reading of Richter, Wilson maintains, would effectively “abandon[]” the precedent set forth by the Supreme Court in Ylst.

In opposition, Sellers emphasizes that § 2254 mandates only that the federal court afford deference to the state court’s “decision” and not the state court’s “reasoning” or “opinion.” Thus, Sellers contends, a federal court may review only the last state court merits decision, and not a lower court’s explanatory opinion. Sellers asserts that in reviewing an appellate summary denial, the court must find only that the decision is based in reasonable conclusions of law and fact. Furthermore, Sellers argues that although these reasonable conclusions need not be stated in the appellate decision, the reviewing court may look to the lower court’s opinion for potential supporting conclusions. If the federal court finds that a lower state court’s opinion provides the last state court with a reasonable basis that could support its summary merits decision, Sellers maintains that the inquiry is over. If, on the other hand, the federal court finds that the lower state court’s opinion resulted from an unreasonable conclusion, Sellers contends, the court must look beyond the opinion for other possible reasonable supporting conclusions for the last state court merits decision.


Wilson argues that, given Georgia’s two-tiered habeas review system, “look[ing] through” to a reasoned decision by a Georgia state court is the best approach to determine “what the state court knew and did.” Georgia’s legislature has designed and implemented a habeas review system, according to Wilson, that invests extensive resources and responsibility in lower state courts to hear evidence and produce reasoned decisions. Therefore, Wilson contends, applying Richter’s “could have supported” approach in a system like Georgia’s, rather than looking to the reasons laid out in the lower court opinion, makes little sense because it often precludes an examination of “what the state court knew and did.” Wilson asserts that adopting the Eleventh Circuit’s expansive reading of Richter would effectively bar federal courts from considering a Georgia state denial’s underlying legal or factual reasoning. This, Wilson argues, would provide federal courts with a “blank canvas” on which to impose their own “hypothetical reasoning” to state court denials and would reinstate the pre-AEDPA regime.

Sellers argues that a broad application of Richter and § 2254 to the Georgia Supreme Court’s summary decision is consistent with principles of federalism and comity. Sellers contends that federal courts will pay due respect to both tiers of Georgia’s habeas review system under Richter’s “reasonable basis” standard. Sellers asserts that Wilson’s proposed approach, on the other hand, would compel a federal court to discount a state appellate court’s merit decision in favor of a lower state court’s decision. In other words, according to Sellers, Wilson’s approach would incorrectly allow federal courts to disregard the second tier of Georgia’s system and only look to the first tier. Sellers further argues that the Georgia Supreme Court makes a denial decision only after “fully considering the complete record and transcript” and that to discount this decision disrespects the Georgia legislature’s decision to create a two-tiered system. Therefore, Sellers contends that when the Georgia Supreme Court issues an unexplained merits decision, the federal courts must review that decision and apply the “reasonable basis” standard.



A group of former state Supreme Court justices (“former justices”) in support of Wilson contend that Congress intended, when drafting the AEDPA, for federal courts hearing habeas proceedings to review the rationale behind state court judgments. The former justices argue that in cases where a state appellate court has not issued a reasoned opinion, the federal courts should examine the last reasoned lower court opinion, because this process will promote comity and federalism goals by elevating and recognizing the state court’s efforts in habeas adjudications. The justices also posit that the Eleventh Circuit’s en banc decision to expand Richter and abrogate Ylst will allow federal courts to determine habeas cases without affording due deference to the state courts. This, the justices claim, will disincentivize the state courts’ endeavors to deliver well-reasoned and reflective opinions and disregard the substantial state court resources put towards adjudicating state habeas proceedings.

In response, twenty-five states (“States”) in support of Sellers assert that the AEDPA’s language calls for federal courts hearing habeas proceedings to focus solely and exclusively on the final decisions of the state courts, regardless of whether they are summary decisions. The States contend that allowing federal courts to look past state appellate court summary decisions to reasoned lower court decisions will encourage the federal courts to inaccurately presume that state appellate court summary decisions adopt the reasoning of the lower courts. In cases where a lower state court made a constitutional mistake, the States argue, the federal courts would be forced to assume that the state appellate courts made the same mistake. They posit that state appellate courts, when faced with these assumptions, would be forced to write opinions affirmatively outlining which parts of the lower court’s opinion they agreed and disagreed with. This would create inefficiency and undermine principles of comity.


The former justices, in support of Wilson, contend that the Eleventh Circuit’s decision undercuts state court systems designed to dispense efficient and well-reasoned habeas decisions. They point to Georgia’s collateral review system, which requires trial courts to issue reasoned opinions in habeas proceedings. The former justices argue that this system shows the state’s intent that the Georgia Supreme Court rely on trial-level rationales when deciding whether to issue certificates of probable cause to appeal. They assert that, based on the history of this system, Georgia Supreme Court summary merits decisions are not intended to create distinct and reviewable decisions on the merits for habeas cases. The former justices contend that the Eleventh Circuit’s approach unreasonably elevates appellate court summary decisions, like those of the Georgia Supreme Court’s, over the reasoned trial opinions actually at issue. Looking through the appellate courts to the judgments of the trial courts, on the other hand, would reflect the practical realities of the state court systems.

In response, the States contend that Wilson’s understanding of Georgia’s collateral review system is inaccurate. They assert that Georgia merely requires a reasoned decision from the trial court and does not intend for the state’s Supreme Court to constrain itself to the rationale of the lower court’s written opinions. The States contend that systems like Georgia’s intend to give petitioners two merits decisions in the state courts and that, given the language of AEDPA, the decision of the court of last resort should be the decision given due deference. The States note that not all collateral review systems are structured like Georgia’s, with two decisions on the merits. Therefore, the States argue, if a state wanted federal courts to give their trial court decisions AEDPA deference, the state has the power and ability to do so. Thus, the States contend, Wilson’s reading would require federal courts to ignore this aspect of state choice regarding their habeas systems and offend comity concerns.

Written by 

Edited by 


Additional Resources