« Sherry on the "Kardashian Court" | Main | Interview with Calvin Morrill about the Jurisprudence & Social Policy Ph.D. Program at Berkeley Law »

Thursday, August 01, 2019

Richard Re on the "Law of Clarity" and Constitutional Avoidance:

As numerous scholars (see, e.g., Brian Slocum's fine piece) and even one judge (Justice Kavanaugh in a widely cited book review) have noted, few concepts are less clear than the concept of legal clarity. (The problem can be equivalently restated as legal ambiguity’s being ambiguous). Given that a lot rides on a determination that a statute is plain, clear, unambiguous, etc., one would think that courts would spend more time analyzing the concept(s) in judicial opinions. Alas, no — and perhaps predictably so: Keeping the concept of ambiguity ambiguous gives judges discretion that intelligible analysis of ambiguity/clarity might eliminate. Using an unanalyzed concept of clarity, judges can turn Chevron off and on like a spigot, hedge on how much notice to give to criminals about what the law forbids, or vary their deference to state judges in habeas proceedings — all without having to engage in a lot of pesky, tiresome reason-giving.

Leisure-seeking judges beware (but Leg-Reg profs like myself, rejoice): Richard Re has just posted a razor-sharp analysis of the concept of legal clarity. Richard’s piece is full of insights, but its most basic point is also the most important: Legal clarity, he argues in Part I, “is not an empirical or linguistic fact but rather a legal characterization that ultimately rests on normative premises.” Legal clarity is not a linguistic concept at all: It is a legal conclusion dependent on the purposes of the law. So one needs to know the normative goal that any clarity doctrine (e.g., lenity, Chevron, qualified immunity etc.) seeks to accomplish before one can define “clarity.” Different doctrines define clarity in different ways, relying on different levels of confidence about how different sorts of interpreters would read some allegedly clear phrase. The “Law of Clarity,” therefore, is characterized by “clarity pluralism,” about which Richard has ambivalent thoughts.

There is so much to like about Richard’s article that it would be churlish of me to pick nits about his treatment of constitutional avoidance, a topic to which he devotes four pages. Naturally, after the jump, that’s what I am going to do. In brief, I think that Richard might have overlooked what I call the “Lazy and/or Scared Judge” theory of clarity, viz.: Sometimes the point of a clarity doctrine is neither to make the law more predictable nor to insure that legislative purpose is more faithfully carried out but instead to save judges the politically troublesome, intellectually vexing task of resolving disputes about what the law means. By demanding high levels of semantic clarity in statutes from lawmakers, judges (try to) force those lawmakers to do the work that the judge wants to avoid doing.

The Constitutional Avoidance canon is the best example of such prescriptive clarity rules designed to change the behavior of lawmakers. Take, for instance, Gregory v Ashcroft (a case that Richard does not discuss). The Gregory Court held that, contrary to the implication of noscitur a sociis, the Age Discrimination in Employment Act’s exemption for state officials “at the policymaking level” applied to state judges. According to the Court, the statute was just not plain enough, even though it was canonically unambiguous, because Garcia’s requirement that Congress take seriously its constitutional duty to protect federalism required a plainer statement to extend federal protection from age-based discrimination to the state judiciary. Quiting Larry tribe's treatise, Gregory stated that "'[T]o give the state-displacing weight of federal law to mere congressional ambiguity would evade the very procedure for lawmaking on which Garcia relied to protect states' interests."

Richard suggests that Gregory-style “clear statement rules” “might require an especially high level of confidence about legislative intent before impinging on legally recognized interests or purposes.” But that characterization of Constitutional Avoidance raises a bit of a puzzle. State governments do not actually enjoy any “legally recognized interest” to protection from ADEA in Gregory, if, by “legally recognized interest,” one means a “judicially protected interest.” The Gregory Court was actually pretty clear that it would not strike down the extension of ADEA to judges if Congress explicitly enacted it. (In 1991, Congress quickly eliminated most of the exemptions for state officials " at the policymaking level" not themselves directly elected by voters). Moreover, once Congress, acting as the relevant constitutional decision-maker, decides that extending ADEA to state judges is necessary and proper for executing Article I powers, then the states actually have no “legally protected interest” of any sort whatsoever to be protected from ADEA: The authoritative interpreter — Congress — has clarified that the the burden on state autonomy is indeed “necessary and proper” for carrying out Article I, such that the extension of ADEA “impinges” on nothing at all.

Richard’s account of constitutional avoidance, in other words, does not explain Gregory, at least as Gregory explains itself: According to Gregory, its federalism "clear statement rule" protects only a procedural entitlement to legislative due process, not a substantive interest in state autonomy. This means that Gregory has nothing to do with prediction or certainty about legislative intent/purpose. Even if the Court were 100% “ confident about legislative intent,” it would rule the same way in Gregory, because the point of the doctrine is to prescribe rather than predict Congress’ behavior: The Court wants Congress to speak plainly to save the Court the trouble of judicially deciding a constitutional issue that the Court deems more fit for Congress to decide. In effect, the Gregory Clear Statement Rule is a judicial prod to Congress to force it to do a constitutional job that the Court wants to avoid — namely, making a decision about the exact scope of state autonomy.

Gregory might possibly provide an addition to Richard’s taxonomy of clarity rules. Richard divides up Clarity Rules into rules that promote legal “certainty” and legal “predictability.” The first aim for accuracy in the interpretation of the law; the second aim to enable third parties like businessmen, cops, or criminals to predict how the law will be applied to them. It is not obvious that Gregory-style Clear Statement Rules aim in either direction. Such rules are not trying to “predict” how Congress

There might be a third category of clarity rules that are designed to affect the incentives rather than predict the decisions of lawmakers. Richard notes that clarity rules can be prescriptive: Courts could demand, for instance, that police officers seeking qualified immunity make more accurate predictions about the law that actually (rather than just “clearly”) applies to their actions. Using clarity rules to prescribe proper behavior for lawmakers as opposed to the subjects of laws, however, does not quite seem to fit into his binary division.

So consider this post a friendly amendment to Richard’s article, a clarification that Richard’s taxonomy might not be 100% complete. The article, however, does not purport to offer a comprehensive taxonomy of clarity rules: It provides a foundational analysis to clear the ground. In clarifying that the Law of Clarity is a prescriptive body of normative legally rules rather than an empirical body of linguistic data, Richard's Clarity Doctrines needs no further clarification: It is an unambiguous success.

Posted by Rick Hills on August 1, 2019 at 05:07 PM | Permalink


Richard suggests that maybe Gregory can be understood as an effort by SCOTUS to achieve more certainty about what Congress really intended or to make the law of federalism more predictable. Maybe so: One could understand these sorts of “clear statement rules” as penalty default rules to encourage Congress to speak more plainly about “important” matters. The procedural character of the rule suggests to me that the goal is not to clarify the law but to force Congress to make decisions that the Court does not want to make about the necessity and propriety of certain burdens on state autonomy. (I suppose that this procedural entitlement could be understood as a way to achieve certainty not about the meaning of the ADEA but rather Article I of the Constitution).

Asher writes that Richard’s article is “somewhat disappointing” because its “main insight” is “comically trivial.” Somehow my interpretation of the article “is worse.”

Asher, your comment is characteristically blunt but uncharacteristically confusing, at least to me. I am afraid that I do not understand your complaints about alleged triviality. (Really? Every lawyer already understands that clarity is a legal conclusion and not a matter of linguistic fact? News to me). And I could not tell whether or not you thought my anti-textualist take was “worse” because more “trivial,” more comic, or just more wrong.

Maybe I will just write another post explaining why I think that lawyers who believe that the proper role of extra-textual materials is merely to disambiguate “ambiguous” text are fooling themselves. The extra-textual stuff must come in right away, at the outset, as Ryan Doerfler has argued, regardless of how “plain” text seems to be when read in isolation from statutory purpose, because we cannot tell whether or not text is ambiguous until we think about the aim of the statute, defined pragmatically. Note, by the way, that Richard’s “comically trivial” insight leads to the same place as Ryan’s position — and Richard’s own position, argued well in his Green Bag article — and is at odds with what I believe (maybe mistakenly) are your views about the lexical priority of text read in isolation from purpose over extra-textual evidence of the mischief that the statute was supposed to address.

I suspect that post would clear up our misunderstandings about what sensible textualism requires. In the meantime, I am happy to discuss offline.

Posted by: Rick Hills | Aug 3, 2019 8:21:02 PM


Thank you for the very kind words about my paper and of course also for these helpful comments. If you are right that there is a “third category of clarity rules,” apart from the two I focus on, that would be fine by me. As you note, my two model types aim to provide a helpful framework but don’t purport to exhaust the range of clarity pluralism out there. We need more work on legal clarity!

That said, I’m not sure that we need a third category in order to capture your nice insights about “prescriptive” clarity doctrines. As you kindly note, I also have some thoughts on incentive-influencing doctrines (such as qualified immunity). So it may be the case that you are developing an interesting clarity subcategory within my framework, rather than breaking out of it.

Here is what I have in mind. The choice between certainty and predictability is perhaps most fundamentally a choice of perspective: is clarity assessed from a first- or third-person point of view? Now, I do suggest that those perspectives are paradigmatically associated with certain objectives, such as to maximize correctness or to make allowances. But I also tried to show that more bespoke objectives work their way into the process of designing many clarity doctrines. Again, my qualified immunity discussion, which you note, is an example.

Your discussion of Gregory seems focused on the objectives-level aspect of the clarity doctrine, as indicated by your discussion of the case’s “aim.” And I think you're right that there is something interesting and distinctive going on in Gregory, at least insofar as the Court’s rule was purely procedural in nature. (More on that below.) But I don’t think that that approach can escape the choice between first- and third-person perspectives on assessing clarity. Put another way, when Gregory “wants Congress to speak plainly” or “required a plainer statement,” what perspective does it use to assess plainness? Once we confront that question, it seems we are back to the certainty / predictability choice, and I suspect that the paradigmatic purposes that I focused on will again start to matter.

I also think I disagree on your more specific point about Gregory not fitting into my admittedly brief discussion of constitutional avoidance. Here is my tentative reaction. You note my suggestion that one possible basis for avoidance is to help secure “legally recognized interests or purposes.” And you doubt that states have any such interest as to the federal statute interpreted in Gregory. But you also posit that Congress must “take seriously its constitutional duty to protect federalism” and link that duty to a “procedural entitlement to legislative due process.” So it seems to me that federalism is the relevant interest and that it is being recognized, among other ways, through the procedural entitlement that Gregory established. The fact that the Court “would not strike down” the relevant measure “if Congress explicitly enacted it,” as you point out, may cast light on the nature of the interest or on the Court’s preferred means of safeguarding it. But that point doesn’t seem to make the interest any less “legally recognized.” Rather, it would reveal an especially interesting way of recognizing the interest.

One last thought. There is an aspect of your argument that made me wonder whether you thought Gregory actually or sincerely created a clarity doctrine at all. As you put it, “Even if the Court were 100% confident about legislative intent,’ it would rule the same way in Gregory, because the point of the doctrine is to prescribe rather than predict Congress’ behavior.” You add: “In effect, the Gregory Clear Statement Rule is a judicial prod to Congress to force it to do a constitutional job that the Court wants to avoid . . . .” One way to read these remarks – not the only way – is that the Gregory Court was going to find a lack of clarity no matter what Congress did. The clarity threshold was essentially infinitely high. If that were the case, then Gregory would indeed be able to avoid the choice between first- and third-person perspectives. The mere fact that Congress tried to legislate in the relevant area would in itself generate unclarity and so trigger the federalism-protecting doctrine. But that would only be because the clarity doctrine was really being used as an elaborate cover for what is really a much simpler principle.

Again, many thanks for these thoughts! Again, plenty to ponder here.

Posted by: Richard | Aug 2, 2019 1:40:35 PM

I think Richard's article is somewhat disappointing, as I've told him, but that your characteristically anti-textualist overreading of it is worse. On why it's a disappointment, its main insight is comically trivial and far too belabored. It has probably never occurred to even the most benighted lawyer to imagine that "clarity" means the same thing in Chevron as it does in qualified immunity or habeas (to say nothing of lenity). It probably wouldn't even be possible to make them mean the same thing if courts tried for consistency's sake; in the one case you're interpreting a statute, in the other you're interpreting a body of judicial opinions, in the one case the question is whether the statute is textually ambiguous on what the law is, in the other the question isn't so much whether an opinion is ambiguous on what the law is, but rather whether the legal principle it states, however clearly it states it, has clear or unclear application to a set of facts. So a great deal of the article is a response to a coherentist view of clarity doctrines that no one's ever held. Indeed, the doctrines he discusses are so hopelessly unrelated to each other that his managing to say anything remotely insightful about them as a group is quite a feat; a more focused treatment of the meaning of clarity in statutory interpretation would have been far more useful. That said, he eventually does get down to the useful task of discussing, e.g., how the purposes of a given clarity doctrine bear on the relevance of judicial disagreement to whether something or another is clear or not, and that is helpful.

Another way in which the article is disappointing is the bit you seize on in an exaggerated way (though perhaps I'm being charitable to Richard and the article's as bad on this as you make it out to be), namely the points where he seems to fallaciously suggest that, just because we use "clear" to mean different things in different legal contexts (often when we're talking about the clarity of very different things), clarity in *none* of these contexts is an empirical fact about linguistics. Putting aside that a lot of the doctrines he talks about are not really concerned with the clarity of language at all, it could just be a contingent fact about the law that certain clarity doctrines look to empirical facts about language. So it may well be, for example, that lenity's standard for clarity is much less demanding for legal reasons than that of an empirically linguistic approach, and that clear statement rules like the one you discuss above are more demanding in that they require explicitness and specificity even when it's possible to, in a linguistic sense, clearly cover something through implication or generality. I think that's probably right. But it could nevertheless be the case that (a) there are empirical facts about whether something is linguistically clear, and (b) a lot of ambiguity-triggered canons, like Chevron, or the presumption that recodifications aren't intended to change the law unless they do so clearly, look to such facts and not some legally constructed clarity-plus or clarity-minus.

To be sure, there is a normative legal question of whether Chevron should look to such facts instead of, say, the sort of it's-"clear"-so-long-as-it-isn't-(almost)-unconstitutionally-vague standard that lenity more or less is. But if you think as a matter of positive law that we have, in fact, opted for an empirical/linguistic approach in Chevron and many other ambiguity-triggered canons of interpretation, as I think we have, then it is worth getting clearer, no pun intended, on what that approach entails, instead of fallaciously saying that because the meaning of legal clarity always involves a prior normative choice about how to construct the doctrine, legal clarity is never a linguistic fact.

Other than that, I agree with you that Congress-disciplining canons are a third distinct category from the two in his model, but I think it could be argued that clear-statement rules are concerned with predictability, e.g., ensuring that state governments know with absolute certainty when they're being regulated. I also think that there's an incoherence in your account of Gregory; supposedly the clear-statement rule is intended to save the Court the intellectually vexing of deciding a difficult constitutional question, but also, "the Gregory Court was actually pretty clear" about what it thought about that question, and gratuitously so. How vexing could the question have been then? Off the top of my head, some more plausible theories than "The Lazy and/or Scared Judge" theory are that the Court wanted to discourage a kind of legislation it didn't like by making it more politically costly, to force Congress to have total accountability for its decisions in this sensitive area, or to ensure that states would have a chance to lobby against legislation that would regulate them by forcing Congress to write it so clearly that states would know they would be regulated by proposed legislation.

Posted by: Asher Steinberg | Aug 1, 2019 11:33:54 PM

The comments to this entry are closed.