On Saturday, I criticized Jeb Bush’s stated criteria for choosing judges as what has led to continued disappointment by conservatives in Republican nominated justices. He said: “You focus on people to be Supreme Court justices who have a proven record of judicial restraint.” But what does “judicial restraint” mean?  There are two quite different possibilities:

  1. Constraint: “Judicial restraint” could refer to confining oneself to following the meaning of the text of the Constitution (and of statutes) — by which is meant its original meaning — whether this leads to upholding or invalidating properly enacted statutes; or
  2. Deference: “Judicial restraint” could refer to deferring to the will of the majority as reflected in the acts of the more “democratic” branches — i.e. “unelected unaccountable” judges should avoid wherever possible thwarting the will of the people, by which is meant the political preferences of the majority of the electorate. The emphasis here is not on the correctness of constitutional analysis, but on judicial deference to majority will.

What is of utmost importance is that these are not the same thing.

  • The first of these is about the constraint imposed on judges by the text; the second is about judges deferring to the will of the more “democratic” branches;
  • According to the first definition of “restraint,” you can’t know if a judge is being “restrained” without assessing the correctness of his or her constitutional analysis; according to the second, a judge is being unrestrained — or “activist” — whenever he or she invalidates a popularly enacted law.

So “restraint” could merely refer to being constrained by the Constitution — in which case it adds nothing to the injunction that judges should only invalidate laws that are truly unconstitutional and uphold constitutional ones. Or it could refer to deferring to the will of the people as manifested by their elected representatives, wholly apart from (or in addition to) disobeying its original meaning.

For example, in the gay marriage case, Justice Kennedy is not being “restrained” either because his reasons for concluding that traditional marriage violates the Fourteenth Amendment are insufficient, in which case the fault is lack of constraint, or because he is invalidating laws that were properly-enacted by the democratic branches in which case the fault is lack of deference. What is key is that one can only reach the first of these judgments by addressing his legal reasoning, but the second requires only that he is invalidating a properly-enacted law — any properly-enacted law.

So, to be clear, I favor the first type of “judicial restraint” — which I and others call “constraint” to avoid semantic confusion — while opposing the second, which I associate with the term “judicial deference.” On  Twitter, Ed Whelan and I went around this subject a few times. While our previous online exchange about “judicial supremacy” vs “judicial equality” shows that Ed and I have some fundamental disagreements, here there is less disagreement than meets the eye. And, while he does regularly seem to use “activism” as a pejorative for invalidating popularly-enacted laws and “restraint” for failing to defer to legislatures, he embraces “restraint” as being constrained to follow the text of the Constitution or statutes, whether to uphold or invalidate.

On the fundamental issue of constraint I have no doubt that we clearly agree. We can agree, for example, that whether the outcome of Obergefell is correct or not, Justice Kennedy’s reasoning is entirely unconstrained by the original meaning of the text. I have not read so unconstrained an opinion since the Warren Court made an art form of it (albeit often to reach a correct result). Even Chief Justice Roberts’s opinion in NFIB was more constrained by the text when he felt the need to adopt a “saving construction” of the statute to avoid invalidating the Affordable Care Act because the individual mandate as written exceeded the enumerated powers of Congress. By contrast the opinion of the four liberal justices was entirely unconstrained.

Ed and I may well disagree about whether a particular judicial invalidation is constitutionally sound or not (though I think we rarely do). We may also disagree over who bears the burden of showing that a law is constitutional or unconstitutional.  And even on this issue, we are not as far apart as it appears. Though I favor a “presumption of liberty,” I could live with a “presumption of constitutionality” that is genuinely rebuttable, as the elder Justice Harlan urged in his dissenting opinion in Lochner v. New York. (What I vehemently reject is Justice Holmes’s solo dissent in Lochner, which was praised on Friday by Chief Justice Roberts, that effectively makes the presumption irrebuttable.) And like Justices Thomas and Scalia, Ed and I may also disagree about the degree to which “precedent” — or the rulings of dead justices — should trump original meaning. In my view, an embrace of precedent on questions of textual meaning allows judges too freely to avoid the constraint of the text, but on this many originalists today disagree.

But Ed and I agree that laws that conflict with the original meaning of the Constitution are unconstitutional, however popular or properly-enacted they may be. And I know we both agree that judges should be constrained in their rulings by the original meaning of the text. I also agree with him that the principal problem with Chief Justice Roberts’ approach in King was that it adopted “purposivism,” but I see the Chief Justice justifying this deviation from plain meaning as a way to defer to the intent of the enacting Congress, as he did when he adopted the “saving construction” of the statute in NFIB.

So here is what we should demand of judicial nominees: a “proven record of willingness to be constrained by the original meaning of the text, whether this means upholding or invalidating a properly-enacted law.” Picking “judges who have a proven record of judicial restraint,” might get you that because that is one possible meaning of “restraint.” But it will just as easily –indeed more likely — get you a “deferential” judge who moves heaven and earth to uphold the constitutionality of laws as in NFIB, or defers to the “intent” or purpose of the Congress that enacted the law as in King v. Burwell, or as George Will dubbed it “judicial abdication.”

To repeat, in no way was I suggesting that the majority decision in Obergefell, or Justice Kennedy’s opinion, was an example of “judicial restraint” or “deference.” To the contrary. My point is that what makes it objectionable, if it is, was Justice Kennedy’s justification for invalidating the state laws, not the fact he thwarted the will of the majority. By its lack of constraint, not a failure to defer. And further that Justice Thomas’s dissent more clearly addressed the issue of the decision’s correctness than did the dissent of Chief Justice Roberts, which relied more heavily on judicial role and deference rather than on constitutional error.

In short, because of both the ambiguity of “judicial restraint” and its historical association with deference to legislative majorities, demanding “a proven record of judicial restraint” is very likely to lead to judges who disappoint conservatives and libertarians. Republicans should not be fooled again. And the only way to avoid being fooled again is to demand judges with a proven commitment to the original meaning of the text — by which I mean the whole text, not just the parts that lead to the results they like.

UPDATE: Greg Weiner replies: The Constitutional Morality of Restraint