The Life and Death Stakes of Supreme Court Informal Practices
By Michael Dorf
Yesterday I attended a talk by George Mason Law Professor Ross Davies on the history of solo actions by Supreme Court Justices. To oversimplify, Davies argued that, in general, the Court gives its members a great deal of room to freelance in their extra-judicial activities when those activities do not purport to take the form of adjudication. He made the further point that in the exercise of this discretion, individual Justices have reached widely divergent answers about what is appropriate. As an example, he pointed to the enormous variation in the time that individual Justices have said must be allowed to pass before their papers become public record. E.g., Thurgood Marshall's papers became public immediately upon his death; David Souter has embargoed them for 50 years following his donation of his papers.
Although Prof. Davies did not offer a policy prescription, it struck me that one approach that builds on his observations might be for the Court--or, where appropriate, other institutions, like Congress--to resolve as a matter of formal policy some of the questions that are now governed either by individual discretion or some loose sense of custom. If so, some such policies might be extended to matters that arise in an adjudicatory context. Here I'll focus on the four/five problem for capital cases.
Although nowhere codified in the Supreme Court Rules, it has long been understood that it takes four votes to grant a petition for a writ of certiorari. That customary practice is so well established that it need not be codified. Likewise, it is equally well established that to issue an order, a majority of the voting members of a quorum of the Court must vote in favor of the order. Absent a recusal or a vacancy, that means five. This too is a well-established practice.
Now consider what happens when a petitioner under sentence of death seeks cert. It is possible that four Justices will vote to grant the cert petition, but that no fifth vote can be found to stay the execution. Thus, the Court developed a practice whereby to honor the "Rule of Four" for granting cert, if there were only four votes to grant cert, one of the Justices who voted to deny cert would nonetheless vote to stay the execution so that the Court would not find itself in the awkward circumstance of having granted review in a case, only then to have the case mooted by the fact that the petitioner was executed.
But that informal practice of the courtesy fifth vote was never made official, and it has occasionally broken down. In Herrera v. Collins, four Justices voted to grant cert but there was no courtesy fifth vote for a stay. Fortunately, the Texas Court of Criminal Appeals issued stays that prevented Herrera's case from being mooted by his execution. But in Hamilton v. Texas (1990), there was no such intervention: four votes to grant cert; no fifth vote for a stay; so Hamilton was executed while his case was pending before the Court. To be sure, the merits of Hamilton's case made this result somewhat more complicated: Hamilton himself was not seeking review; he had decided not to raise any further challenges to the death penalty; the cert petition was filed by a "next friend." But still, even in a case like Hamilton, there is something macabre about permitting a state to moot a case by executing the petitioner.
In addition to the inspiration taken from Prof. Davies' talk, I raise this issue now because it may bear on the disposition of the cert petition currently before the Court in Hernandez v. Stephens (and for which my colleague Sheri Johnson is counsel of record). The case presents a number of very important issues regarding how the Texas courts have implemented the rule of Atkins v. Viriginia--perhaps none more important than the question of whether the Fifth Circuit erred when it ruled that Atkins permits a state court to find that a defendant is not intellectually disabled, notwithstanding IQ scores in the disabled range, if that score is (just barely) brought up to the bottom of the non-disabled range when "scaled to Mexican norms." The Fifth Circuit appears to be endorsing the view that whether someone counts as intellectually disabled--and thus death penalty-ineligible--can depend on his national origin. Even the SCOTUS conservatives ought to be interested in this case on equal protection grounds.
The Court should grant cert in Hernandez but if it does not, it should at least hold the petition pending its disposition of Hall v. Florida (which I discussed in an earlier Verdict column). Both Hall and Hernandez present questions about the proper role of state law in Atkins intellectual disability determinations; thus, even if the Court does not regard Hernandez as cert-worthy, its ruling in Hall could lead to what Supreme Court lawyers call a GVR--grant, vacate, and remand--for reconsideration in light of Hall.
What has that all got to do with Herrera and Hamilton? Texas has set an execution date for Hernandez that would have the state killing him before the SCOTUS could hear plenary argument in his case (if the Court grants cert) and likely before it would resolve Hall (which was only argued last week). So, whether the Court grants or holds the petition in Hernandez, it will take either five votes from the Supreme Court or the cooperation of the Texas courts to ensure that Texas does not moot the case by killing Hernandez before his petition is considered.
That sort of thing has also happened. In 1987, in Watson v. Butler, four Justices voted to hold the petition pending the outcome of another case on the docket. But, as Justice Brennan explained: "Three votes suffice to hold a case, but it takes five votes to stay an execution. The Court thus permits Mr. Watson's legal claim to stay alive while condemning Watson himself to die under a sentencing scheme that within a matter of months the Court may conclude is unconstitutional." One hopes that Hernandez will not be the occasion for a repetition of that sort of grisly irony, but more generally, the Court should rationalize its stay rules with its cert grant and hold rules.
Postscript: Speaking of the death penalty, Adam Liptak has a new NY Times story about claims by condemned prisoners that they have a right to know what method of execution the state plans for them. The story quotes an earlier blog post of mine in which I said that the Supreme Court may be skeptical of "opportunistic" attacks on methods of execution by lawyers who oppose the death penalty in all circumstances. The Times story accurately quotes me and links to my post, but a too-casual reader might get the erroneous impression that I think that such claims are problematic because opportunistic. Here I'll just reiterate what I said in the blog post: I was describing how I thought the Justices might react, not offering a normative justification for that reaction.
Yesterday I attended a talk by George Mason Law Professor Ross Davies on the history of solo actions by Supreme Court Justices. To oversimplify, Davies argued that, in general, the Court gives its members a great deal of room to freelance in their extra-judicial activities when those activities do not purport to take the form of adjudication. He made the further point that in the exercise of this discretion, individual Justices have reached widely divergent answers about what is appropriate. As an example, he pointed to the enormous variation in the time that individual Justices have said must be allowed to pass before their papers become public record. E.g., Thurgood Marshall's papers became public immediately upon his death; David Souter has embargoed them for 50 years following his donation of his papers.
Although Prof. Davies did not offer a policy prescription, it struck me that one approach that builds on his observations might be for the Court--or, where appropriate, other institutions, like Congress--to resolve as a matter of formal policy some of the questions that are now governed either by individual discretion or some loose sense of custom. If so, some such policies might be extended to matters that arise in an adjudicatory context. Here I'll focus on the four/five problem for capital cases.
Although nowhere codified in the Supreme Court Rules, it has long been understood that it takes four votes to grant a petition for a writ of certiorari. That customary practice is so well established that it need not be codified. Likewise, it is equally well established that to issue an order, a majority of the voting members of a quorum of the Court must vote in favor of the order. Absent a recusal or a vacancy, that means five. This too is a well-established practice.
Now consider what happens when a petitioner under sentence of death seeks cert. It is possible that four Justices will vote to grant the cert petition, but that no fifth vote can be found to stay the execution. Thus, the Court developed a practice whereby to honor the "Rule of Four" for granting cert, if there were only four votes to grant cert, one of the Justices who voted to deny cert would nonetheless vote to stay the execution so that the Court would not find itself in the awkward circumstance of having granted review in a case, only then to have the case mooted by the fact that the petitioner was executed.
But that informal practice of the courtesy fifth vote was never made official, and it has occasionally broken down. In Herrera v. Collins, four Justices voted to grant cert but there was no courtesy fifth vote for a stay. Fortunately, the Texas Court of Criminal Appeals issued stays that prevented Herrera's case from being mooted by his execution. But in Hamilton v. Texas (1990), there was no such intervention: four votes to grant cert; no fifth vote for a stay; so Hamilton was executed while his case was pending before the Court. To be sure, the merits of Hamilton's case made this result somewhat more complicated: Hamilton himself was not seeking review; he had decided not to raise any further challenges to the death penalty; the cert petition was filed by a "next friend." But still, even in a case like Hamilton, there is something macabre about permitting a state to moot a case by executing the petitioner.
In addition to the inspiration taken from Prof. Davies' talk, I raise this issue now because it may bear on the disposition of the cert petition currently before the Court in Hernandez v. Stephens (and for which my colleague Sheri Johnson is counsel of record). The case presents a number of very important issues regarding how the Texas courts have implemented the rule of Atkins v. Viriginia--perhaps none more important than the question of whether the Fifth Circuit erred when it ruled that Atkins permits a state court to find that a defendant is not intellectually disabled, notwithstanding IQ scores in the disabled range, if that score is (just barely) brought up to the bottom of the non-disabled range when "scaled to Mexican norms." The Fifth Circuit appears to be endorsing the view that whether someone counts as intellectually disabled--and thus death penalty-ineligible--can depend on his national origin. Even the SCOTUS conservatives ought to be interested in this case on equal protection grounds.
The Court should grant cert in Hernandez but if it does not, it should at least hold the petition pending its disposition of Hall v. Florida (which I discussed in an earlier Verdict column). Both Hall and Hernandez present questions about the proper role of state law in Atkins intellectual disability determinations; thus, even if the Court does not regard Hernandez as cert-worthy, its ruling in Hall could lead to what Supreme Court lawyers call a GVR--grant, vacate, and remand--for reconsideration in light of Hall.
What has that all got to do with Herrera and Hamilton? Texas has set an execution date for Hernandez that would have the state killing him before the SCOTUS could hear plenary argument in his case (if the Court grants cert) and likely before it would resolve Hall (which was only argued last week). So, whether the Court grants or holds the petition in Hernandez, it will take either five votes from the Supreme Court or the cooperation of the Texas courts to ensure that Texas does not moot the case by killing Hernandez before his petition is considered.
That sort of thing has also happened. In 1987, in Watson v. Butler, four Justices voted to hold the petition pending the outcome of another case on the docket. But, as Justice Brennan explained: "Three votes suffice to hold a case, but it takes five votes to stay an execution. The Court thus permits Mr. Watson's legal claim to stay alive while condemning Watson himself to die under a sentencing scheme that within a matter of months the Court may conclude is unconstitutional." One hopes that Hernandez will not be the occasion for a repetition of that sort of grisly irony, but more generally, the Court should rationalize its stay rules with its cert grant and hold rules.
Postscript: Speaking of the death penalty, Adam Liptak has a new NY Times story about claims by condemned prisoners that they have a right to know what method of execution the state plans for them. The story quotes an earlier blog post of mine in which I said that the Supreme Court may be skeptical of "opportunistic" attacks on methods of execution by lawyers who oppose the death penalty in all circumstances. The Times story accurately quotes me and links to my post, but a too-casual reader might get the erroneous impression that I think that such claims are problematic because opportunistic. Here I'll just reiterate what I said in the blog post: I was describing how I thought the Justices might react, not offering a normative justification for that reaction.