« Congressional Procedure = Real Constitutional Law | Main | The Times insults every civ pro professor in the country »

Sunday, March 14, 2010

Richard Primus, Obama's Oath, and Our Indifference to Plain Constitutional Text

There is a widespread sense among scholars and laity that, whatever the force of ambiguous constitutional text, the plain text always governs, regardless of other considerations: Two Senators from each state, only 35-year-old Presidents, and so forth. But a new paper by Richard Primus nicely illustrates that this widespread belief understates our willingness to ignore even the plainest of constitutional texts when the departure is within the apparent purpose of the constitutional provision.

Primus gives the example of the Presidential oath, famously muffed by Chief Justice Roberts the first time around but, less famously, muffed by every President-elect since FDR (including Obama), because Presidents-elect have inserted their name into the words of the oath specified by the Constitution. “I, Barack Hussein Obama…” actually departs from the precise words of the text of Article II, section 2, which provides no blank in which President-elects may insert their names. Of course, we do not care: We do not even notice. Why not? Because (to quote McCulloch) we instinctively deduce the "minor ingredients from the Constitution's "important objects," such that even the plainest of text defining the former gives way to the latter. Because Obama's departure from text disturbs no important object of Article II, we allow it.

Primus offers, as another less trivial example, the Tidewater problem familiar from federal jurisdiction. Following Nat’l Mut. Ins. Co. v. Tidewater Transfer Co., 337 U.S. 582 (1949),, federal courts routinely treat suits between citizens from Washington, D.C. and other states as falling within their jurisdiction, even though such a grant cannot fit the text of Article III, which provides for jurisdiction over “Controversies … between Citizens of different States.” By no stretch of 18th century diction is Washington, D.C. a “State[].” The Supreme Court agonized over the basis for such a departure from text, split between denying that any such departure exists (the Rutledge position) and elaborating a convoluted theory that Congress can confer Article I jurisdiction on Article III courts in excess of Article III (the Jackson position). But both are flat contradictions of (respectively) the 18th century semantic meaning of the term “States” and the implicitly exclusive enumeration of jurisdictional categories in Article III, section 2. Primus notes that lawyers have gradually just accepted D.C.-state “diversity” without a murmur despite this insult to textual purity.

But maybe someone should complain? Do those strict semanticists out there – Larry Solum, Kurt Lash, etc – have a beef with this lackadaisical attitude towards semantics? Do they think that President Obama ought to do the oath over one more time? If not, why does not Obama’s departure from “public semantic meaning” of Article II violate some shared sense of fidelity towards law?

Posted by Rick Hills on March 14, 2010 at 10:12 AM in Constitutional thoughts | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference Richard Primus, Obama's Oath, and Our Indifference to Plain Constitutional Text:


I'm not sure it's accurate to say that "[t]here is a widespread sense among scholars and laity that, whatever the force of ambiguous constitutional text, the plain text always governs, regardless of other considerations." Rather, I would say that there's a widespread sense that the best kinds of legal arguments are textual arguments, and that as the text gets clearer, the argument gets stronger. But there are lots of cases that deviate from the plain text, so it can't be that plain text is seen as always governing. Or so it seems to me.

Posted by: Orin Kerr | Mar 15, 2010 1:36:56 AM

The first statute enacted by Congress prescribed the congressional oath as "I, A. B. do solemnly swear or affirm (as the case may be) that I will support the Constitution of the United States." 1 Stat. 23, s. 1 (1789). I don't think anyone took the public meaning of that text to require utterance of the letters, "A. B." Is there any evidence that the failure to include "A. B." in Article II was read to abandon the traditional practice of stating one's own name? Cf. An Act for the further Security of his Majesty's Person and Government, s. 1, 1 Geo. I, c. 13 (1714) ("I, A.B., do swear ..."); cf. also the judicial oaths at 1 Stat. 76.

(This practice, by the way, nicely preserves the meaning of an indexical term. George W. Bush can say "I, George W. Bush," but Barack Hussein Obama can't, because when he says "I" it denotes "Barack Hussein Obama." Even if one insists that it's a departure from the text, it's hard to think of a departure that does less violence to the text's meaning. By contrast, the Tidewater rule actually changes whether actual people have a right to sue.)

Posted by: anon | Mar 14, 2010 11:48:22 AM

"I'm out back" is either synecdoche or metonymy (probably the latter), in which "I" means the same as "my car." Hence, it's not textually false any more than saying metaphorically "the doctor is a butcher" or "Jeff is a space cadet" is textually false. See Steve Winter's A Clearing in the Forest for more on this, but words are motivated, and if we understand and do not object to the motivation, we do not object to the words.

Metaphoric usage is troubling or provocative when it shakes up our usual categories. That's a matter of difference in degree, not difference in kind. And metaphoric usage that at one time was controversial can become conventional (a la Tidewater).

In that sense, I don't see a lot of difference between Tidewater and the oath.

Steve Griffin from Tulane gave an interesting talk last week at Suffolk on this very issue in the context of the development of executive war powers authority since WWII. The basic idea is that there's a kind of symbiotic amendment to the Constitution going on all the time - conventional usage shapes the meaning we attribute to the text, and the text can limit conventional usage. I find that very persuasive not only in the Constitutional debate, but in any tradition in which text and usage evolve, such as, for example, the evolution of new meaning in traditional Jewish law, or Halakha, to changed conventions or circumstances.

It's why my reaction to most of the "semantic originalism" debate is "so what?" I'm willing to accept that such-and-such words meant such-and-such in 1787, but I'm no more persuaded than governs their usage today than the very Orthodox position on the application of the Fourth Commandment ("Remember the Sabbath day...") to the turning on of an electrical light switch.

Posted by: Jeff Lipshaw | Mar 14, 2010 11:23:26 AM

Also, the Oath Clause has to be read against a background of legal practices concerning oaths. If other oaths described in legal documents were known to be administered with the name included, then the framers' failure to write "I, (insert your name here)" would just have been following linguistic conventions of the time. Any similar argument for Tidewater would be fanciful.

Posted by: anon | Mar 14, 2010 11:08:07 AM

I don't see these two cases as raising the same issue. Tidewater fits with Primus's theory; it involves a departure from the text's reasonably understood meaning. But the oath, I'm less sure about. Show that text to 100 people and show them what a President-elect actually says, and I suspect that a substantial fraction of them will not notice that there's a divergence. This isn't a departure justified by purpose, in their heads. As a matter of pragmatics, the text of the oath can include the name, just as it includes either "swear" or "affirm," but not both, even though they both appear within the quotation marks. It's like saying "I'm out back" when asked where you've parked your car: no one bats an eye at the utterance even though it's textually false.

Posted by: James Grimmelmann | Mar 14, 2010 10:44:46 AM

Post a comment