Treaties Can Create Domestic Law of Their Own Force, But It Does Not Follow That Treaties Can Increase The Legislative Power of Congress

Yesterday, the Supreme Court granted certiorari in United States v. Bond, which raises the question of whether a treaty can increase the legislative power of Congress. Guest Blogger Rick Pildes has already noted the cert grant here, and Ilya Somin posted his thoughtful take on the case here. I merely add that I am delighted that the Court has taken the case. Missouri v. Holland addressed this issue in one unreasoned sentence; I believe that it deserves a far more thorough treatment.

As it happens, Rick and I are in the midst of debating this very issue. Rick set the stage with some historical background, and I largely agreed with – but slightly re-characterized – his account. Rick offered some structural or pragmatic reasons to believe that treaties can increase the legislative power of Congress. I contended that these arguments put the cart before the horse.

The first question, I suggested, is whether there is any basis in constitutional text for this proposition. (And, in light of the Tenth Amendment and the enumeration of legislative power, the burden of proof surely lies with anyone claiming that Congress’s legislative power can be expanded, virtually without limit, by treaty.) The conventional view is that the textual basis may be found in a combination of the Treaty Clause and the Necessary and Proper Clause. I have attempted to explain why this is not so.

And the absence of textual support is unsurprising, because the proposition itself is in such deep tension with the basic structural axioms of the Constitution. The Constitution goes to great pains to limit and enumerate the powers of Congress. It emphasizes that the powers of Congress (unlike the powers of the President and the courts) are only those “herein granted.” It creates an elaborate mechanism, really four mechanisms, for its own amendment, by which the legislative power can be — and repeatedly has been — augmented. And for good measure, it underscores that “[t]he Powers not delegated to the United States by the Constitution, nor prohibited by it to the states, are reserved to the states respectively, or to the people.”

Given all this, it is hard to imagine that the Constitution includes a fifth mechanism, unmentioned in the text, by which the legislative power of Congress can be increased, virtually without limit, by treaty. As Justice Scalia says: “I don’t think that powers that Congress does not have under the Constitution can be acquired by simply obtaining the agreement of the Senate, the President and Zimbabwe. I do not think a treaty can expand the powers of the Federal government.” (oral argument, Golan v. Holder (2012)).

Despite all this, Rick insists that that Justice Scalia is wrong, and that treaties can increase the legislative powers of Congress. He has advanced two arguments so far. In this post, I will address his first point, about self-executing treaties. I will address his second point in a subsequent post.

Rick points out that treaties generally can be self-executing; that treaties are supreme law of the land; and that self-executing treaties create domestic law of their own force, perhaps preempting state law in the process. (See the Treaty Clause + the Supremacy Clause.) If all that’s so, he wonders, what’s so bad about a non-self-executing treaty giving Congress new legislative power? Why should we object to the two-step displacement of state law (non-self-executing treaty followed by statute) if the one-step displacement (self-executing treaty) is permissible?

The short answer is that process and structure matter in constitutional law. In the canonical structural cases, like INS v. Chadha (legislative veto) and Clinton v. New York (line item veto), the losing argument generally takes this form: If the government could have achieved something similar by procedure X, then what’s so bad about letting it use procedure Y? The winning side reminds us that functional equivalence does not suffice; there is no substitute for “a single, finely wrought and exhaustively considered procedure” required by the Constitution.

In any case, here we are not talking about functional equivalence. It is one thing for a treaty to create domestic law of its own force — a distinct, well-defined, section of federal law, whose preemptive force would be clear on its face, just like a federal statute. It is quite another matter for a treaty to create an entirely new font of legislative power (like the new fonts of power in various constitutional amendments) — power that Congress may use, at its discretion, to regulate entirely local matters forever after. Or at least until the President of the United States — or the President of, say, Zimbabwe — abrogates the treaty.

If this were permissible, the Constitution would create a doubly perverse incentive — an incentive to enter into new international entanglements precisely to enhance domestic legislative power. The Framers were very wary of foreign entanglements (see, e.g., Washington’s Farewell Address). And they were deeply fearful of the legislature’s tendency to “everywhere extend[] the sphere of its activity, and draw[] all power into its impetuous vortex,” Federalist #48 (Madison). It is, therefore, implausible that they would have created a doubly perverse incentive by which treaty makers (the President and Senate) could undertake new foreign entanglements — and thereby increase the power of lawmakers (the President, Senate, and House). This is not “ambition … made to counteract ambition,” Federalist #51 (Madison); this is ambition handed the keys to power.

Happily, this is not what the Constitution requires. It nowhere suggests that treaties can increase the legislative power of Congress.