The DIG Rule and the Rule of Four Reconsidered

By Mike Dorf

Less than a month ago, in a post on the oral arguments in the Same-Sex Marriage Cases, I wrote that
it takes six votes to DIG [that is, dismiss as improvidently granted].  (Since it takes only four votes to grant cert--the so-called "rule of four"--permitting even five votes to accomplish a DIG would undermine the rule of four.)
Then, last week, the Supreme Court proved me wrong.  In Boyer v. Louisiana, the Court DIG'd by a vote of 5-4.  What gives?

Okay, so to begin, I hadn't expressed the general understanding of the DIG rules quite precisely or accurately.  As traditionally stated, what's required is that one of the Justices who voted to grant must agree to DIG.  If there were originally only four votes to grant, then that shift will mean that it takes six votes to DIG: the original five no votes at the cert stage plus the Justice who has had a change of heart.  So I should have said it usually or often takes six votes to DIG.

But if the DIG rule is phrased in the way I've just described, it can still undermine the rule of four.  Let's use Boyer as an example.  The 5 votes to DIG were CJ Roberts plus Justices Scalia, Kennedy, Thomas, and Alito, with Justices Ginsburg, Breyer, Sotomayor, and Kagan wanting to reach the merits.  We don't know who voted for and against cert in the first place but let's suppose that the original votes to grant were by the four merits-reachers (Ginsburg, Breyer, Sotomayor, and Kagan), plus one of the Justices who voted to DIG: I'll guess Justice Alito, as he wrote the concurrence in the DIG.  After oral argument, Justice Alito concludes that he originally voted to grant based on an erroneous assumption, so he now votes to DIG.  The result is a DIG by 5-4, even though four of the Justices who originally voted to grant still think the grant was proper.  By joining the vote to grant and then changing his, Justice Alito (in my hypo) was able to convert an otherwise-unDIGable case into a DIGable one.  So even though the rule appears to be that it takes a total of 5 votes to DIG, so long as one of the five originally voted to grant, I think a better rule would be the one I announced last month: It should always take 6 votes to DIG.

At least that would be a better rule if one were interested in protecting the Rule of Four.  But maybe there's no very good reason to protect the Rule of Four.  The Rule of Four itself is not even codified. The Supreme Court Rules set forth criteria for granting (and denying) certiorari but not the procedures that govern the Court's decision about how to resolve conflicts regarding how to apply those procedures.  The Rule of Four appears to be a longstanding practice that the Court observes, regarding it as binding on itself.

This leads to the question: Why four?  If only four Justices want to hear a case, why not deny?  One answer might be that a rule requiring five votes to grant would merge the merits and the decision to grant--but that's also a risk under the Rule of Four.  Suppose an ideologically divisive case in which four liberal Justices want to grant to reverse a conservative circuit court ruling but they think that the remaining five Justices will affirm the lower court ruling if the case is granted.  If they're worried about the merits, one or more of the four liberals might vote to deny, thinking that at least this way the lower court's ruling will be confined (temporarily) to one circuit.  Cognoscenti call such strategic voting a "defensive deny," and there is some debate over whether it is legitimate and over how often it occurs.

Putting aside strategic behavior and ideological cases, the justification for the Rule of Four is plain enough: If four Justices think a case satisfies the cert criteria, that ought to be enough to place it on the docket.  They're not necessarily saying the lower court got the outcome wrong, just that the issue is important.

But once we recognize that four Justices' say-so is simply a rough measure of importance, there's nothing special about four.  Indeed, I think one could make a good argument for a Rule of Three or maybe even a Rule of Two.  After all, the Rule of Four originated at a time when a large portion of the Supreme Court's docket consisted of mandatory appellate cases.  Congress eliminated nearly all of the mandatory appellate jurisdiction in 1988, but the Rule of Four persists--and this probably plays some role in the shrinkage of the Court's docket.

Hence, my modest proposal would be to go to a Rule of Three or even a Rule of Two.