Wednesday, August 11, 2010

Polymorphism 3: The New Originalism

By Mike Dorf (Updated/Corrected Post, as Explained Below)

A couple of weeks ago, I posted here and here about constitutional polymorphism--the notion that a single word or phrase in the Constitution could mean one thing in one context and something else in another context.  The posts led to some spirited exchanges in the comments over a number of questions about both statutory and constitutional polymorphism, including: 1) Is it ever justified?  2) How common is it?  3) Were the specific examples I gave actually instances of polymorphism or were they better explained as instances in which a single word or phrase was sufficiently vague that it could have a consistent meaning across contexts, even as the implications of that single meaning varied in those different contexts?

To conclude this discussion of polymorphism, I'd like to connect question 3) to some of the questions raised by the so-called "new originalism."  To summarize, the new originalism--by contrast with the "old" originalism--is essentially a theory about how to read texts rather than a theory about how to constrain judges.  Where the old originalism was often intentionalist and paid substantial attention to the concrete expectations of the framers, the new originalism looks to original public understanding and the concepts captured by the words of the Constitution.  Prominent new originalists include Randy Barnet, Larry Solum, and Keith Whittington.  Jack Balkin professes to be a new originalist as well, although there is debate over whether he should be admitted to the club.  (More on that in a moment.)

Although I regard the move from intentions to public meaning as important, that move was already being made by originalists in the 1980s.  To my mind, the real practical significance of new originalism is its modesty.  Solum describes most contemporary originalists (by which I take him to mean new originalists) as subscribing to what he calls the "moderate contribution thesis": "The semantic content (linguistic meaning of the constitutional text) forms a constraining part of the legal content of constitutional doctrine, but it can be supplemented (by constitutional construction) and may be subject to limited defeasibility conditions (e.g., extraordinary emergencies)."  Likewise for Whittington, much of the contestation in constitutional law is not about constitutional meaning--what he calls "interpretation"--but about "constitutional construction," a process that is not determined by the original meaning of the text.  Solum, Whittington and other new originalists save originalism by shrinking it: Originalism in their hands becomes the claim that public meaning plays an important role in framing constitutional controversies, but much of the heavy lifting is left to "construction."  Thirty years ago Paul Brest called this approach "moderate originalism," which he defined as follows: "The text of the Constitution is authoritative, but many of its provisions are treated as inherently open-textured."  Filling in the open texture is what the new originalists call "construction."

Brest's article (The Misconceived Quest for the Original Understanding in the 1980 BU L Rev) distinguished among three views: strict originalism; moderate originalism; and non-originalism.  After attacking strict originalism (on grounds that have more bite against intentionalist than textualist theories), Brest concluded that moderate originalism is defensible but that non-originalism should generally be preferred.  That is, in principle, a real difference with the new originalism, but I think the discussion of polymorphism provides a nice analogy for showing why in practice the distinction doesn't amount to much.

Consider an example I used in my second post on constitutional polymorphism.  I said that in construing the Establishment Clause as a structural principle not subject to exceptions, the Supreme Court had interpreted the words "no law" differently from how the Court had interpreted the same terms as applied to freedom of speech, where strict scrutiny applies.  Some commenters objected that one could get this pair of results by giving "no law" a consistent meaning, so long as one understood that "freedom of speech" only encompasses protected speech.  (One commenter astutely noted that the Supreme Court once said strict scrutiny applies to denominational preferences under the Est Cl, but I regard that case as amomalous and, in any event, nothing turns on the particular example.)   I agree.  It's usually possible to recharacterize a seemingly polymorphic interpretation of some word as non-polymorphic (monomorphic?), so long as there are some sufficiently vague words lurking around.

Brest makes the same point about moderate originalism.  He says that one can get to the result that the Equal Protection Clause forbids most sex discrimination by purporting to use the original understanding at a sufficiently high level of generality.  However, he says that such decisions are better characterized as really non-originalist.  Or, in the new originalists' terms, the real action takes place in the domain of constitutional construction, after original understanding has run out.

Brest's claim is empirical.  Some of the new originalists contest it, saying that the semantic content will actually rule out some of the results that nonoriginalists would otherwise reach.  But their list is short and contestable.  And Balkin's efforts to fly the new originalist flag while affirming liberal constitutionalism rather seriously undercut them.  Unless the other new originalists have some good ground for kicking Balkin out, his recent work seems to support Brest's claim that results purportedly reached in the name of moderate originalism are not seriously derived from the original understanding.

I'll end with a confession.  I have not read all of the new originalism scholarship as closely as I probably ought to have, mostly because I regard it as a rearguard action to save originalism by sacrificing what made it a distinctive approach.  Hence, it's possible I've missed something important here.  But I doubt it.

Postscript:  Perhaps in confirmation of my concluding confession, the original version of this post mistakenly quoted Professor Solum's definition of the minimal contribution thesis rather than the moderate contribution thesis.  Thanks to him for calling the error to my attention in the comments and on his blog.  I don't think the error undermined the substantive points I make here but I apologize nonetheless.