Why Danforth v. Minnesota Does Not Undermine My View About State Court Decisions To Follow Lower Federal Court Precedent

by Michael Dorf

Last week I asked whether state courts can gratuitously give to federal appeals court precedents more binding force than they have of their own accord--which is to say ANY binding force. My answer was no. The binding effect--or as it turns out here, lack of binding effect--of a federal appeals court precedent on a state court is itself a question of federal law, and states are not generally free as a matter of state law to "over-enforce" federal law.

My post on this subject generated a very interesting discussion, in the comments, in private correspondence, and on other sites. The most potent challenge to my position comes from Steve Vladeck, both in his appearance on Oral Argument the week after mine and in a blog post. Christian Turner favorably cited Vladeck's argument on his blog (in the course of making some broader points). Professors Vladeck and Turner, as well as others, rely on Danforth v. Minnesota for the proposition that states can in fact over-enforce federal law as a matter of state law. As I'll explain, that's a plausible claim, but I think that it ultimately misreads Danforth.

To see why, we need to start with some basics. Under the Orwellian-named Antiterrorism and Effective Death Penalty Act (acronymed AEDPA, which is pronounced "edpa" by the cognoscenti) and an earlier judicial decision called Teague v. Lane, federal courts considering habeas corpus petititions from state prisoners are generally barred from granting relief where the petition relies on a "new rule" of constitutional law--i.e., a rule that was not "dictated by precedent" at the time the petitioner's original case became final. Danforth involved Teague rather than AEDPA but the basic rationale of both the statute and the substantially overlapping judge-made rule is the same: Federal habeas as a collateral remedy exists to ensure that state courts comply with federal limits as they exist at the time they consider the cases that come before them, rather than for the development of the case law. (Many people, including me, think this is a too-narrow conception of habeas, but the Rehnquist/Roberts Court and Congress disagree with us, so we take this as given.)

In Danforth, the question was whether the Minnesota Supreme Court was entitled to disregard Teague--and thus to give a habeas petitioner the benefit of a new rule on state habeas review. The Court said yes. Danforth thus appears to say that a state supreme court can choose to give greater effect to a federal precedent than it is required to give it, and this does indeed appear to support the position of my critics.

But a closer look shows that Danforth is not generalizable in the way that the critics think. Here is how the Danforth majority characterized Teague:
A close reading of the Teague opinion makes clear that the rule it established was tailored to the unique context of federal habeas and therefore had no bearing on whether States could provide broader relief in their own postconviction proceedings than required by that opinion.
In other words, the Teague rule is best understood as either a construction of the federal habeas statute itself or as a freestanding judicial gloss on the limits applicable to a federal court when considering a habeas petition from a state petitioner, in light of principles of federalism. Danforth establishes no state court power to disregard some general principle of federal law stating that new rules are nonapplicable in all (i.e., both federal and state) collateral proceedings because there is no general principle of federal law stating that new rules are are nonapplicable in all collateral proceedings.

By contrast, my interlocutors have been willing to concede (at least for the sake of argument) that there is a general principle stating that lower federal court precedents do not bind state courts. To get from there to the principle that a state court can gratuitously decide to be bound by lower federal court precedent (or anything else that is not binding of its own force), one must posit that state courts can disregard such a principle. Whatever reasons one might have for thinking that, Danforth v. Minnesota, properly understood, should not be among them.

By the same token, Danforth doesn't affirmatively help my argument either. In Danforth the Court repeats the longstanding principle that state law can provide remedies for federal law violations beyond those that are required by federal law itself. That means that there are contexts in which state courts are allowed to over-enforce federal law.

Now if anything turned on it, I would contest that last proposition. I would say that a state court selection of a remedy that goes beyond what federal courts provide is not actually over-enforcement; it's simply a recognition that the law of remedies is not strictly tied to the law of substantive liability. But I don't think that anything turns on this characterization issue, so I'll concede for the sake of argument what I said at the end of the last paragraph: Danforth affirms that there are some contexts in which states can be said to over-enforce federal law.

However, other cases--such as Michigan v. Long--make clear that there are other contexts in which state over-enforcement of federal law is impermissible, and efforts in that direction would simply amount to an erroneous construction of federal law. The question, therefore, is whether the binding effect of lower court precedent falls into the Danforth category or the Long category. That question must be answered by reference to the specifics at issue here, not the general question of whether over-enforcement is ever permissible. Indeed, thinking of this issue as a matter of "over" or "under" enforcement of federal law is at best an analogy, since (as I noted in my last post) a state court that voluntarily decides to be bound by some otherwise non-binding authority on federal law is not systematically opting to give more or less force to federal law; it is instead (at least in some cases) merely opting to give different content to federal law than it would give if making its own best judgment.

To the extent that the Danforth/Long dichotomy does apply, I do think that the current issue falls in the Long category. Under the Madisonian Compromise, Congress didn't even have to create any lower federal courts; prior to 1875 the lower federal courts did not have federal question jurisdiction; and even then and for a century thereafter, there was an amount-in-controversy minimum in federal question cases. Accordingly, for much of U.S. history, state courts could not treat as binding the decisions of lower federal courts because there would be no such decisions. So when did state courts get this power? Did it exist before lower federal courts had federal question jurisdiction? If so, what was it? A power to treat just about any source (such as a treatise or the opinion of wise men or of the House of Lords) as binding with respect to federal law? That seems a reductio of the argument of my interlocutors. Yet if the power sprang into existence with the creation and subsequent expansion of lower federal court authority to decide questions of federal law, that is attributing an awful lot to a conferral of jurisdiction on federal courts that says nothing whatsoever about state courts.

Finally, I want to make clear that I do not have a policy objection to Congress making federal appeals court rulings on federal law binding on the states within their respective coverage, nor do I have a policy objection to Congress authorizing state courts to voluntarily treat such rulings as binding (although I would have a policy and probably a constitutional objection to Congress authorizing state courts to treat other non-SCOTUS sources of federal law as binding). I just don't think Congress has done either of those things or that, absent a plausible claim that Congress did, there is authority anywhere else for state courts taking this action.