Constitutional constructions and constitutional decision rules: thoughts on the carving of implementation space.

• Author: Berman, Mitchell N.

INTRODUCTION

Let's start with the obvious: court-announced constitutional doctrine is frequently not identical to the announcing court's understanding of what the text of the Constitution means. Consider, for example, the doctrine that implements the First Amendment's Free Speech Clause. That Clause is terse: "Congress shall make no law ... abridging the freedom of speech." But a comprehensive statement of judicial doctrine effectuating that command would tax even the most expert First Amendment scholar. Here's just a first and partial stab:

A law constitutes an impermissible abridgment of the freedom of speech if: it regulates expression on the basis of its content or viewpoint and is not narrowly tailored to achieve a compelling governmental interest, except that content-based regulation of non-misleading speech that proposes a lawful economic transaction is permitted if the regulation directly advances a substantial government interest that could not be advanced equally well by a less speech-restrictive regulation, and except too that content-based regulation of speech is freely permitted if, inter alia, the regulated speech proposes an unlawful economic transaction or a lawful transaction in a misleading way, or if it is sexually explicit and as a whole appeals to the prurient interest, and depicts or describes sexual conduct in a patently offensive way, and lacks serious artistic, political, or scientific value, or if it includes the sexually explicit depiction of children, or if the speech, by its very utterance inflicts injury or tends to incite an immediate breach of the peace; all subject to the caveat that even when speech may permissibly be regulated, if that regulation takes the form of a prior restraint on its issuance, then the regulation is ordinarily presumptively impermissible; and furthermore, a content-neutral regulation of speech is impermissible unless it is narrowly tailored to achieve a significant government interest and leaves open ample alternative channels of communication. As complicated as is this statement, it captures, at best, only some neighborhoods of constitutional free speech law. I have not yet said anything about those portions of the doctrine that govern defamation, intentional infliction of emotional distress, or invasion of privacy, or limited public fora, or campaign finance expenditures, or the speech of public employees, and so on.

Even with only a pre-reflective untheorized sense of what constitutional meaning is or amounts to, or what are the conceptual bounds of the activity of constitutional interpretation, it seems exceedingly unlikely that, in contributing to the formation of this intricate free-speech doctrine, Supreme Court Justices believed that each building block was a partial statement of what the First Amendment means, or that all they were doing in developing and announcing this doctrine falls within the bounds of the activity properly denominated interpretation, or that the output itself amounts to an interpretation of the First Amendment. In some ways, leading scholars have been drawing attention to just this point for 30-odd years (think of Henry Monaghan's work on "constitutional common law" and Larry Sager's on "underenforced constitutional norms"). (1) But Richard Fallon was particularly helpful in focusing attention on it a decade ago when urging that what federal courts do in the process of constitutional adjudication is more felicitously described under the capacious label "constitutional implementation" than as "constitutional interpretation." (2)

For a constitutional theorist, especially one of a more conceptual orientation, the question raised by this proposed change in perspective and vocabulary--from "interpretation" to "implementation"--is how best to conceptualize what is going on, or what may or should go on, in this implementation space. And by conceptualize, I mean how best to think about and understand any more or less distinct stages of implementation and more or less distinct outputs of the activity. The theorists I have mentioned have all carved the space in two: Monaghan contrasted "Marbury-shielded constitutional exegesis" with "constitutional common law"; Sager distinguished "constitutional norms" from "constitutional constructs"; Fallon differentiated "constitutional meaning" from "constitutional doctrine." In addition, Kim Roosevelt and I, followed now by others, distinguish "constitutional operative propositions" from "constitutional decision rules." (3) All these scholars, then, adhere to what I have called, in previous work, "the two-output thesis": (4) each of these frameworks recognizes two conceptually distinct outputs of constitutional adjudication (both of which lie upstream from the application of law or doctrine to fact that is necessary to reach case-specific holdings), one of which is, in a fairly straightforward sense, logically and perhaps normatively prior to the other.

It is against this background that we can consider the distinction between constitutional interpretation and constitutional construction introduced a decade ago by Keith Whittington (5) and embraced and further developed by such other prominent "new originalists" as Randy Barnett and Larry Solum. (6) As Solum puts the distinction:

Interpretation is "the activity of determining the linguistic meaning--or semantic content--of a legal text"; Construction is "the activity of translating the semantic content of a legal text into legal rules, paradigmatically in cases where the meaning of the text is vague." (7) Constitutional "pragmatists" like Rick Hills who resist the very enterprise of carving implementation space into conceptually distinct pieces on the grounds that "pragmatically speaking, the meaning of a constitutional provision is its implementation," (8) naturally reject this particular distinction too. But as another take on the two-output thesis, the distinction strikes me, at least at first blush, as unobjectionable and potentially valuable. I say the distinction is unobjectionable "at first blush" and "potentially" valuable because, all by itself, it is not very informative or helpful. To distinguish activities occurring in implementation space in these terms is not yet, I think, to make clear just how this framework differs from previous conceptualizations of the two distinct types of general normative or propositional outputs of constitutional adjudication. (9) Sager spoke of "norms" and "constructs." Is interpretation just the name for the process of deriving the norms, with construction being the name for the process that results in, well, the constructs? Fallon spoke of "meaning" and "doctrine." Plausibly, meaning is the outcome of interpretation. Is doctrine just the outcome of construction?

In order to ascertain whether the interpretation/ construction distinction is merely a notational variant of one or more other formulations of the two-output thesis, we need to hear more. In particular, we need to know: (1) what is linguistic meaning or semantic content of the constitutional text; and (2) what are the constraints on the translation of the Constitution's semantic content into legal rules--that is, what is the nature of such constraints, and what is their content.

The new originalists provide a reasonably clear answer to the first question and at least gesture to or intimate an answer to the second, even if what they say on this score is looser than one might reasonably wish. Somewhat simplified, and elaborated on at greatest length by Solum, the new originalists take the semantic content of a legal text to be Gricean sentence meaning, which entails that the semantic content of any constitutional provision is essentially its original public meaning. Furthermore, legal rules, in their view, may permissibly depart from the semantic content of the Constitution (understood, more or less, as the original public meaning) only in very limited ways. For the most part, the activity of construction is proper when the semantic content is unable to resolve concrete legal disputes--because, paradigmatically, it is too vague or otherwise underdeterminate. And when construction is called for, its proper scope is necessarily and severely restricted--to making a vague norm more precise or to choosing between original meanings that conflict, or the like. (10)

If one adopts the interpretation/construction distinction along the foregoing general lines--interpretation is the activity of determining linguistic meaning, and construction is the activity of translating linguistic meaning into legal rules--and one adopts the new originalist views of linguistic meaning and of the strictures that such linguistic meaning puts on the process of construction or translation (whatever, precisely, those strictures turn out to be), then one has tools of consequence. One armed with this whole package is apt to make different moves in the practice of constitutional implementation, and reach different destinations, than one not so armed. The question, accordingly, is whether one should adopt this package of views. Very generally, that is what I mean to explore in this essay.

More precisely, I aim to address three distinct but related questions. First, are the originalist views I have just sketched about linguistic meaning and about the permissible scope of construction true? Second, if originalist views about the linguistic meaning of the constitutional text and the constraints that that meaning imposes on the announcement of law and the generation of legal doctrine are not jointly true, does the interpretation/construction distinction, as glossed by its proponents, still have value? That is, is its utility independent of particular controversial claims about linguistic meaning and the permissible relationship between linguistic meaning and legal doctrine? Third, if the interpretation/construction distinction is not useful when divested of...