The Legal Status of Treaties that Require Violations of the Constitution

Co-blogger Nick Rosenkranz and I agree on most of the practically important issues regarding the constitutional status of treaties. But in his insightful recent post responding to my most recent comment on the subject, Nick does identify one theoretically interesting difference between us. He believes that treaties that require action that violates the Constitution are in some sense legally valid, whereas I do not:

If the President signs a treaty promising that Congress will enact certain legislation, but Congress would ordinarily lack the power to enact that legislation, what happens? Missouri v. Holland seems to say that the treaty automatically gives Congress the legislative power at issue. Ilya and I both disagree.

Ilya would say that, under these circumstances, the treaty itself is void. He would say that the President has no power to make such a promise. In his view, the treaty power only empowers the President to make promises that the federal government knows it can keep.

In my view, the answer is different. I believe that the President can make such a promise, even though Congress lacks present power to keep it. Making such a promise is not generally advisable, to be sure, but it is permissible. To see why, consider that for every person, and every politician, and every government, the capacity to make promises exceeds the capacity to keep them. Many of our promises may turn on circumstances beyond our control, including the actions of third parties.

To be clear, I don’t doubt that the president can make that promise. I just deny that the promise has any legal validity of the kind that would be enjoyed by a treaty that only requires action within the constitutional limits of federal power. It has the same status as any other promise to do something we have no legal right to do. For example, if I sign a contract promising to force a third party blog for the Volokh Conspiracy, I certainly have the right to put my signature to the piece of paper. But it would create no binding legal obligation. The same goes for a treaty committing the federal government to do something it lacks the constitutional authority to do.

Nick correctly points out that we often have a right to contract to do things that we might not ultimately succeed in carrying out, such as promising to build a house within a time-frame that turns out to be impossible. But there is a difference between that kind of promise and a promise to do something that is actually outside the scope of the promisor’s legal authority. It’s the distinction between the contractor who promises to build a house on an unrealistic schedule, and one who promises to, say, commit murder for hire. Because the latter has no right to commit murder in the first place, his promise is legally void.

Nick argues that a presidential commitment to a treaty that requires action beyond the power of the federal government might be seen as a promise to use Article V of the Constitution to pass a constitutional amendment. If the treaty merely requires the president to take action to pass a constitutional amendment, that may be so. But most international agreements go beyond this, stating that the US is actually required to perform Action X, as opposed to merely requiring the president to use persuasion to try to enact a constitutional amendment.

There is a difference, moreover, between a treaty that would require a constitutional amendment to implement, and one that merely requires ordinary legislation. The latter is within the power of the federal government as a whole, even if not that of the president by himself. And the president is himself an official of the federal government. By contrast, a constitutional amendment requires the consent of a supermajority of states, which are not part of the federal government and have their own separate sovereign authority.

Nick worries that my approach might lead to disaster in some circumstances:

Imagine that the United States is defeated in a disastrous war, and the victorious country requires, as a term of a peace treaty, a concession that would violate the Bill of Rights. It proposes, for example, to allow the United States to maintain some military bases abroad, but insists that any crimes committed by people there, including the spouses of soldiers, must be tried by military commission. Can the United States agree to the term and end the war?

Such a treaty cannot be self-executing; if it were, then making it would violate the Bill of Rights. And if such a treaty were non-self-executing, it would not empower Congress to pass legislation executing it. A treaty cannot itself violate the Bill of Rights, and nor can it empower Congress to violate the Bill of Rights. These are the holdings of Reid v. Covert, and Rick [Pildes], Ilya, and I all agree with them.

But does it follow that the President has no power to enter into such a treaty in the first place, even if it is non-self-executing? Ilya would say yes: If Congress has no power to execute such a treaty, then the President has no power to sign such a treaty, and if he does so, the treaty is void. But why? Would we really be obliged to fight to the last man rather than sign such a treaty?

In my view, such a treaty would indeed be legally void. To make it legal, we would have to pass a constitutional amendment. But notice that the practical situation is little different under Nick’s view. In theory, he would say that the treaty is valid. But he also argues that it can’t be enforced either through self-execution or through congressional legislation. Presumably, the president cannot enforce it by executive order. Under Nick’s theory, the treaty would have no real effect until there is a constitutional amendment. From the standpoint of a victorious power that wants to see results in the real world, there is little difference between my view and Nick’s. In practice, both would require us to either pass a constitutional amendment quickly or violate the Constitution if we wanted to appease the enemy and end the fighting.

This is just one of many possible examples of how any constitutional limit on government power could potentially lead to disaster. Any such limit could turn into a suicide pact in some theoretically conceivable situation. But that does not mean that we should simply do away with constitutional restrictions on government. Unconstrained government power also poses grave risks. I wrote about the suicide pact dilemma in greater detail here.