Questions for Judge Katzmann

by Michael Dorf

Yesterday, a panel of the U.S. Court of Appeals for the Second Circuit--Chief Judge Katzmann, Judge Hall, and Judge Wesley--heard oral arguments here at Cornell Law School rather than in their usual courthouse at Foley Square in lower Manhattan. Continuing the festivities, today I will moderate a discussion with Judge Katzmann. I plan to ask some questions and then some follow-ups depending on what Judge Katzmann says, before opening the discussion up to the audience.

Of course there are standard questions one asks of a judge in this sort of setting, but I'm going to try to focus on questions that arise out of Judge Katzmann's 2014 book Judging Statutes. The book is more or less a defense of purposivism in statutory interpretation--and especially the use of legislative history as a means of construing vague or ambiguous legislative language. Conversely, the book critiques textualism. Judge Katzmann's excellent and highly readable book illustrates his own views with in-depth studies based on three cases his court decided and that were subsequently reviewed by the Supreme Court.

Here is a preview of three of the lines of inquiry I hope to pursue in our conversation later today. I'll pose them here as though I'm speaking directly to Judge Katzmann.

(1) In making the case for the relevance of legislative history, you argue that Congress itself and the administrative agencies consider legislative history essential to how they create, understand, and administer statutes. But does it follow that courts should resort to the same material? Let me suggest that you need some further argument for that proposition.

Here’s a suggestive metaphor for what I regard as the gap in the argument. Editors of textbooks often also write Teacher’s Manuals, which are available to instructors but not students. The teachers use the manuals to get a deeper understanding of the material and to make lesson plans, but the students don’t have access to them. Might legislative history work similarly? Or consider the owner’s manual of a car. There are a great many documents that are relevant to the production and servicing of a car, but the owner—if she is not an amateur mechanic—only looks at the owner’s manual. I’m not suggesting that either of these examples is anything like a perfect analogy. I’m simply giving them to make the point that it is sometimes sensible for one audience to have reference to a narrower range of documents than another audience. Why do you think that judges should have reference to all of the technical production documents for a statute rather than just the owner's manual?

(2)  In your book, you distinguish between purposivism and textualism, expressing a preference for purposivism. When I teach my students about different approaches to statutory construction, I usually include a third approach: Intentionalism. The chief difference between intentionalism and purposivism is that the former asks what the legislature actually intended, whereas the latter asks what purpose or purposes can reasonably be attributed to a statute and then how best to carry them out. A purposivist might, but also might not, be amenable to consulting legislative history. Do you disagree with this schema? If not, do you consider yourself an intentionalist or a purposivist? Why?

(3) You note in your book that some textualists ground their approach in public choice theory, which you criticize on the ground that it inaccurately disregards the sometimes-public-regarding reasons Congress has for enacting laws. For what it's worth, I agree with your critique of public choice.

You also note, however, that there are other arguments for textualism that do not rely on public choice theory. One leading such argument--originally proposed by Prof. John Manning and later adopted by other textualists--treats textualism as a non-delegation doctrine. When courts treat committee reports, floor statements, and other sorts of legislative history as highly probative of the meaning of statutes, textualists argue, they effectively delegate to congressional staff or to particular members of Congress the task of legislating. But cases like Bowsher v. Synar and Metrop Washington Airports Authority v. Citizens for the Abatement of Aircraft Noise forbid such delegations. Article I, Section 7 says that Congress, not a servant or subset of Congress, has the power to legislate.

You say that the non-delegation argument for textualism is mistaken. You write:
The contention that the use of legislative history violates the constitutional proscription against self-delegation . . . is premised on a mistaken view of the legislative process. Legislative history accompanying proposed legislation precedes legislative enactment. When Congress passes a law, it can be said to incorporate the materials that it, or at least the law's principal sponsors (and others who worked to secure enactment), deem useful in interpreting the law. After all, Article I of the Constitution gives each chamber the authority to set its own procedures for the introduction, consideration, and approval of bills. And each chamber has established its own rules and practices governing lawmaking--some favoring certain proceedings over others--establishing [what Professor Brudney calls] "a resultant hierarchy of internal communications." Those rules and procedures give particular legislators, such as committee chairs, floor managers, and party leaders, substantial control over the process by which legislation is enacted. Communications from such members as to the meaning of proposed statutes can provide reliable signals to the whole chamber.
It seems to me that there are potentially three distinct arguments in that passage. One is simply about timing. It says that because the legislative history process takes place before enactment of the law, giving it effect as law is not delegation. If meant to stand alone, the timing point strikes me as not very forceful. Suppose that a law contained the following clause: "In ascertaining the meaning of Section 2j of this Act, the views expressed by the junior Senator from Vermont prior to the law's enactment shall be deemed authoritative." Or worse: "In ascertaining the meaning of Section 2j of this Act, the views expressed by the Duke of Wales prior to the law's enactment shall be deemed authoritative." If there is no evidence that Congress was actually aware of what the junior Senator from Vermont (or the Duke of Wales) said, then we cannot treat Congress as incorporating by reference his expressed views. Rather, this looks like delegation either to a subset of Congress or wholly outside Congress, plain and simple. The timing doesn't appear to matter.

Accordingly, I take you to be saying that the timing interacts with the actual legislative process. That brings me to the second argument: The idea is that the whole of Congress is keenly aware of what the committees and sponsors say and intend, and so when a statute uses unclear language, we can safely assume that such language incorporates by reference the statements found in the legislative history. This argument strikes me as persuasive if the factual assumptions underlying it are correct, but I think they're pretty plainly false. As you note in the book, most members of Congress are not aware of much of the language in most statutes they enact, much less of the glosses given to it by the committee process.

That brings me to the third argument in the passage quoted above: Congress has the power, in virtue of Article I, Sec. 5 ("Each House may determine the Rules of its Proceedings"), to delegate actual lawmaking power to its committees, so long as it follows up by enacting legislation. For this argument to work, one must either think that cases like Bowsher and Metrop. Washington are wrongly decided or think that the absence of a formal delegation to committees makes a big difference.

I hasten to add that I don't think that reliance on legislative history to fill statutory gaps violates the non-delegation doctrine for two reasons. First, the non-delegation doctrine is quite toothless. The authority granted by looking to legislative history is much more minor than in cases like Bowsher and Metrop. Washington (which, technically, are not non-delegation cases so much as Art I, Section 7 formalism cases). Second, as you write elsewhere in the book, looking to legislative history where the statutory meaning is otherwise unclear constrains courts. It would be better if Congress could be clearer (it can't always be), but as between judges making stuff up and pretending to find their preferences in the text versus looking to what will probably be a pretty good indication of legislative intent, the latter makes more sense.

* * *

I'm looking forward to a terrific session.