All You Ever Wanted to Know About the Treaty Power and Federalism, But Were Afraid to Ask*

by Julian Ku

I am remiss in not linking earlier to this excellent and absorbing discussion of the upcoming US Supreme Court decision in Bond v. United States at the Volokh Conspiracy.  Rick Pildes at NYU is doing a nice job, but he is single-handedly taking on Professors Nick Rosenkranz, Ilya Somin, and Eugene Kontorovich on various aspects of the treaty power and its limitations under the U.S. Constitution.  So Professor Pildes is a little outgunned, although he is doing a nice job nonetheless.

My own two cents so far:  I find the textual argument for a limited treaty-execution power fairly compelling, especially under the Necessary and Proper Clause.  I also agree that the Bill of Rights limitation on the treaty power, inferred from the Supreme Court’s plurality decision in Reid v. Covert, is not a particularly powerful precedent in favor of the Missouri v. Holland result.  If anything, the Reid v. Covert conclusion that treaties cannot violate the Bill of Rights should limit the impact of the earlier Missouri decision, which may not have agreed with Reid.

But Professor Pildes certainly has a powerful argument on this front (at least to me): If a self-executing treaty can exceed Congress’ Article I powers, than why not a statute implementing that same treaty?  What is the structural logic of this result?

For this reason, I associate myself with Professor Curtis Bradley’s view that it makes sense to read a federalism limitation on the self-executing effects of a treaty as well. That question was the subject of Missouri’s main holding, and that holding is also troubling and suspect. I understand that the arguments for limiting the implementation power are stronger, at least textually, but that doesn’t mean there aren’t good structural arguments for limiting the self-executing effects of treaties as well.

13 Responses

  1. People overlook that Missouri v Holland is presented as a property dispute more than a question of powers. At the founding of the US, natural resources appeared to be inexhaustible. Experience with beaver trapping and buffalo herds showed this was not the case, and around the presidency of Teddy Roosevelt (and the cases in question) conservation became an issue. The prior cases on which Missouri based its arguments noted that when a natural resource was the common property of all people, and when it could be regarded as inexhaustible or else was located exclusively in one state (like clams rather than migrating birds) then the 10th Amendment made the regulation of hunting, fishing, and harvesting a matter for state rather than Federal regulation. That is, if Maine adopts rules on lobster pots that the Federal government regards as too lax, then Maine will suffer from the over-harvesting and will have to realize its mistake and correct it.
    In this case, Missouri cited precedent to assert that on behalf of its citizens, it had an exclusive property right in the “common property” of migrating birds when they were within the state and, therefore, the exclusive right to make rules governing hunting. Had any other state decided to challenge Missouri’s title to the bird population, then the dispute might have ended up before the Supreme Court. Nobody would have asserted that the Supreme Court could not decide matters in a dispute between states if the subject matter of the dispute would not have been a matter for Federal courts if only one state were involved. However, Missouri would argue that Congress could not make law to enforce a treaty that resolved a dispute between a state and a foreign country unless Congress could make the same law absent the international dispute.
    Between 1913 (the earlier case) and 1920 (Missouri v Holland), it became almost universally understood that migratory birds were not inexhaustible, and that overhunting by Missouri would be detrimental to the rights of other states. However, it turned out that only Canada was willing to contest Missouri’s assertion of exclusive “property”ownership. When the property dispute is between a state and a foreign country, it is resolved by treaty negotiation rather than by a Supreme Court case. Then, while the prior precedent still asserts that the 10th Amendment prohibits the Federal government from asserting a Federal property right in the birds in opposition to the Missouri claim, it could contest Missouri’s claim based on the competing Canadian claim which was not blocked by the 10th Amendment and which the treaty had resolved by limiting what Missouri or any other state could do.
    Even the judge that rendered the 1913 ruling in favor of the states had, by 1918 realized that the bird populations had to be protected. In the Missouri v Holland decision, Holmes appears frustrated that we have to rely on Canada because the US could not do the right thing on its own: “But for the treaty and the statute, there soon might be no birds for any powers to deal with. We see nothing in the Constitution that compels the Government to sit by while a food supply is cut off and the protectors of our forests and our crops are destroyed. It is not sufficient to rely upon the States. The reliance is vain, and were it otherwise, the question is whether the United States is forbidden to act.”
    This still means that Missouri v Holland may not stand for anything when there are no competing property rights to resolve. This was not a case about abstract regulation through enumerated powers. If a property belongs to the US, then the Federal government has the power to regulate its use based on its ownership without regard to enumerated powers in Article I. Therefore, even if conservation was not regarded in 1920 as an enumerated power of Congress, the regulation would have been obviously constitutional had the Federal government been able to assert ownership of the birds themselves. Under the 10th Amendment, the USG could not assert such property rights, but Canada did and Wisconsin could have but did not.
    Since none of the Bill of Rights, and very little else in the Constitution is based on an assertion of property rights, there is nothing in Missouri v Holland that plausibly allows a self-executing treaty, or enabling legislation for any other treaty, to expand the power of Congress beyond conservation, water quality, and navigation in rivers and lakes shared with Canada or Mexico. Certainly you cannot find a disputed property right that touches on capital punishment, health insurance, or private gun ownership.

  2. I still don’t see the structural logic in limiting the treaty power to the other enumerated powers. The States ceded their entire sovereign power to make treaties to the federal government. There must be some entity that can make treaties on matters of State concern. Or did powers somehow mysteriously vanish as they were getting ceded? Given that there is no textual limit on the treaty power, for me the functional argument is pretty determinative regarding treaties adopted in the Constitutional way (i.e., Pres + 2/3 Senate). This, of course, doesn’t answer whether the necessary and proper power allows the legislature to adopting implementing legislation to give domestic force to treaties that are not self-executing. 
    //Regarding Professor Pildes’ argument that you find convincing: with all respect, I do not. The treaty power arises in Article II–it is a power of the executive, not of the legislature, and is not subject to the limits placed on the legislature. Self-executing treaties avoid the limitations of Article I simply because they arise from a totally different federal power. (This is not to say that the president isn’t bound by Article I restraints when executing laws under the “take care’ power, but that is because the laws that he is taking care to execute are themselves bound by Article I restraints.)
    //Really, as far as I see it, the sole question is whether the necessary and proper clause gives Congress the power to implement treaties or other international law arrangements constitutionally entered into by the President (which would be required primarily when those agreements are not self-executing). The text here suggests that it is: Congress is given the power “to make all laws which shall be necessary and proper for carrying into execution” not only “the foregoing powers” (that is, the prior enumerated powers), but also “all other powers vested by this Constitution in the government of the United States, or in any department or officer thereof”. As a textual issue, Congress has the power to make all laws which shall be necessary and proper for carrying into execution the President’s treaty power. The question is whether implementing legislation is necessary and proper.
    //As far as the 10th amendment goes, I can’t see how it can fundamentally affect the constitutional structure here. There is no question that we are working with an enumerated power: the treaty power, which was turned over generally by all States to the president. Unlike legislative powers, where the States retained residual power and enumerated certain powers that Congress could exercise, the treaty power given to the President was given plenarily, and with no residual power left in the States. And if one of the enumerated legislative powers is to make laws that are necessary and proper for carrying into execution the President’s treaty power, then Congress is acting within an enumerated power by implementing a treaty, assuming that the implementation is in fact necessary and proper for carrying the treaty into execution.
    //Of course, I don’t know anything of the Federalist Papers, etc., and perhaps I’m all wrong as a matter of original understanding (although I’m not convinced that that matters). But the structure of the Constitution–States ceding their plenary treaty-making powers to the federal government–does suggest that this is an area that States’ rights concerns should not be as strong, so long as the federal government is not entering into treaties as a pretext.
    //Even with all of this, I’m still not sure what the result should be in Bond. Was the implementation legislation–reaching an area of general criminal law so divorced from the core of the treaty–really necessary and proper? I could see myself agreeing with an opinion that interprets the implementing legislation narrowly, as bounded by what is necessary and proper to execute the treaty.

  3. Re…I still don’t see the structural logic in limiting the treaty power to the other enumerated powers. The States ceded their entire sovereign power to make treaties to the federal government.
    In Reid v Covert the Court noted that neither the President nor the Senate had the power to amend the Constitution, whether acting alone or in combination.
    It also noted that statutes and treaties must comply with the terms of the Constitution, since neither were listed among the prescribed methods in Article V for amending that document. So the thinking goes: that a treaty (self-executing or not) can’t preempt the Bill of Rights or the Constitution and broaden the delegated powers granted to any branch of the federal government.

  4. I fail to see how Reid v Covert is relevant. The holding of Reid is that the Constitution applies to treaties: only those treaties which are in accordance with the Constitution are the law of the land. But my argument above was not that the Tenth Amendment does not apply, but simply that it is not breached. The Tenth Amendment reserves to the States any power that they have not delegated to the federal government. Fine, but they have fully delegated all of their power to make treaties–they retain none of it, and the President is simply exercising a Constitutionally delegated power when he makes treaties on behalf of the United States–and moreover one that is enumerated in Article II as among his powers. There is no breach of the 10th Amendment there. And furthermore, so long as the law is necessary and proper, Congress is exercising an enumerated power specifically delegated to it when it makes laws executing a treaty constitutionally entered into by the President. In other words, yes, Reid says that the provisions of the Constitution apply to treaties, but, as I see it, there is no breach of any provision of the Constitution, and the actors are all exercising powers provided for by the Constitution, so its an irrelevant point that the Constitution applies. There is no need to amend a Constitution that already gives you the power you seek.

  5. When the Constitution was written treaties resolved disputes between nations. In the last 60 years, however, treaties are increasingly a form of international legislation. The Constitution did not include an enumerated power for the Supreme Court to invalidate a law passed by Congress if it was contrary to the Constitution. That was added later on by Marbury. Reid extends Marbury to Legislation by Treaty. If DOMA is invalidated, the political branches cannot bypass this by adopting the Defense of Marriage Treaty (DOMT) and then claim that because it is a treaty is falls outside the scope of judicial review.
    However, our Federal system divides domestic legislative power into a State sphere and a Federal sphere. When there is a legitimate international dispute, the Federal government negotiates on behalf of all 50 States and, to the extent that resolution of the dispute requires enabling legislation, the subject matter of the dispute becomes a Federal power even though without the dispute it would have been an exclusive power of the states. However, treaties that simply mandate a code of conduct and do not resolve disputes about legitimate foreign interests do not become a backdoor legislative mechanism completely at variance with what the founders intended.

  6. Howard: your history is off re: actors (and “nation” is even diff. than “state” — see Brierly, etc.) as well as re: treaties only addressing disputes between formal actors in the internaitonal legal process.  The U.S., for example, has had treaties with Indian nations (e.g., the Navajo nation) and tribes, as well as with free cities, and other non-state actors.  See
    Treaties fairly early on had addressed numerous subjects.  They have trumped state powers within the U.S. b/c of the Supremeacy Clause, which expressly and unavoidably mandates that “all” treaties [not merely self-executing ones, or a few] are supreme law  and binding on the states, including the state judiciary.  Several cases, including Supreme Court cases other than Missouri, have affirmed (which was also affirmed per dictum in Reid) that with respect to treaties the States within the U.S. have no 10th Amendment power.  Please see
    With respect to enhancement of congressional power and cases in addition to Missouri, please see our West casebook, Paust, Van Dyke, Malone, International Law and Litigation in the U.S. 251-273 (3 ed. 2009).

  7. Under the Tenth Amendment to the U.S. Constitution, after the formation of the Constitution any remaining state powers or aspects of sovereignty were expressly limited to (1) those “powers not delegated to the United States by the Constitution,” and (2) those powers that are not “prohibited by it to the States.” The treaty power was expressly delegated to the United States in Article II, Section 2 (i.e., the President “shall have Power, by and with the Advice and Consent of the Senate to make Treaties”) and Article III, Section 2 (i.e., “[t]he judicial Power shall extend to all Cases … arising under Treaties”). The treaty power was also expressly prohibited to the States in Article I, Section 10 (i.e., “[n]o State shall enter into any Treaty”). Additionally, the text of the Constitution clearly and unavoidably mandates that “all” treaties are supreme law of the land binding the states and the state judiciary. As Supreme Court cases have long recognized, there is therefore no competing or inhibiting power of the states. Indeed, with respect to treaty law, “the Tenth Amendment is no barrier” and “whatever is within … [the] scope [of the treaty power] is not reserved to the states [and] the Tenth Amendment is not material.” As one federal court recognized in 1880:
    There can be no mistaking the significance or effect of these plain, concise, emphatic provisions. The states have surrendered the treaty-making power to the general government, and vested it in the president and senate; and, when duly exercised…, the treaty resulting is the supreme law of the land, to which not only state laws but state constitutions are in express terms subordinated.
    No nonconforming theoretic construct or revisionist professorial preference can rightly avoid the textual strictures of the Constitution in this regard. The textual strictures are reason enough to preempt invitations to engage in judicial reconstruction of the Constitution in the name of an alleged federalism that was not chosen.
    As the Supreme Court has rightly affirmed, “treaties of course ‘are as binding within the territorial limits of the States as they are elsewhere throughout the dominion of the United States’” and treaties can reach “matters which in … [their] silence … the State may regulate.” With respect to matters that “usually fall within the control of the State,” “a treaty may override its power.” Indeed, a treaty is “obligatory” domestically and “must necessarily control all acts issuing from the inferior authority which might contravene it” (i.e., the state, including unavoidably the state judiciary). As one federal court recognized:
    the treaty furnishes the law, and with that treaty no state or municipal corporation thereof can interfere. Admit the wedge of state interference ever so little, and there is nothing to prevent its being driven home and destroying the treaty-making power altogether.

  8. So Jordan, does this mean that the President and Senate can ratify the Defense of Marriage Treaty with, say, the Mashantucket Pequot Tribal Nation (Foxwoods Casino) and that puts it beyond any challenge. Or on the other side, can it negotiate the Patient Protection and Affordable Care Treaty with them to put the Health Insurance Mandate beyond challenge even if it isn’t a tax. I think the Treaty clause is qualified by an assumption of reasonableness and is subject to challenge in court if used in pretense. Your bold statements are accurate simply because the power has previously been used honestly for the purposes to which it was intended, not as a loophole to achieve a questionable legislative purpose unrelated to any proper treaty.

  9. Don’t forget that a treaty as such must receive consent from 2/3 of the Senators for the President to ratify.  That is a significant political break.  Such does not exist with respect to a sole Exec. Ag. or, what most forms of U.S. int’l ags. take, a congressional-exec. ag.
    We all know that each form is trumped by the Cosnt., but what the Const. requires is, of course, the major issue.  Certainly the 10th Amend. “is no barrier” and treaties can be directly operative as law of the United States.
    If treaties can prohibit torture anywhere in the U.S. and can protect ducks, I cannot see why they cannot protect women or children.
    An “assumption of reasonableness”? surely that would be judicial legislation contrary to the choices of the political branches — activists judges and all that

  10. Response…Daniel said I fail to see how Reid v Covert is relevant.
    I didn’t say that it was relevant to Bond. I simply explained its rationale and why it is relevant to some of the more bizarre hypotheticals that are being discussed. Some of those suggest or imply that a treaty can be used to amend the Constitution without the consent of the states. Article V denies the President and the Senate any such power.

  11. Yes, a treaty would not “amend” the Supremacy Clause (Art. VI, cl.2), which expressly and unavoidably mandates vis a vis the states that “all” treaties (not some of them, some that implicate “foreign affairs” [what treaty does not?]} are supreme law of the land and binding on the states and their judiciary.  This is unavoidable constitutional text, and coupled with the unavoidable constitutional text in the 10th Amend, there is no inhibiting state power vis a vis treaties — see Reid’s dictum re: Missouri and the 90 cases in my study (32 of which are supreme Court cases!). So text, structure, and early decisions of the judiciary (and all the way up to Missouri and Lara) dictate that there is no limitation with respect to the reach of the treaty power vis a vis the states. To decide otherwise wold be to ignore text, structure, etc. and engage in activist revisionism.

  12. hostage: Thanks for the response. I understand the rationale, but I was trying to express that, if there is a federalism-based restriction, it does not come from the 10th amendment or the list of delegated powers in Article I, but rather from an inherent (I used the word “internal”) limitation within the treaty power itself. Reid stands for the proposition that treaties must comply with the constitution and cannot operate so as to amend the constitution. That’s true but not the issue in regard to federalism, since the treaty power is an explicit, enumerated power that states have ceded to the federal government. The issue is rather whether the treaty power has some sort of internal limitation. I can think of three internal limitations. First, is the treaty power in Article II (dealing with the executive) somehow limited by the list of enumerated powers in Article I (dealing with the powers of the legislature)? I argued before that there was a structural disconnect, that the list of enumerated powers comes in the context of legislation, which the states retain their sovereign powers and cede only limited, enumerated powers to the federal government, whereas the treaty power is given completely and with no residual power left in the states. This operates as a counterargument to reading the treaty power in light of Article I, but I don’t even think a counterargument is needed–there is no textual reason to think that the list of enumerated powers in Article I limits the treaty power in Article II. Is there another limitation? I think two other limitations are conceivable, and they are represented above by Howard and Jordan. As Howard notes, an originalist view of treaties could be used to limit the scope of the treaty power–i.e., it extends only to those issues that would have been the subject of a treaty at the time of the founding. Jordan presents a different view, that of political checks and balances. The treaty power may be broad, but the political process needed to adopt one is quite stringent. So perhaps the founders meant to cede the treaty power wholesale to the federal government, knowing quite well that the matters of international concern shift and develop throughout the years, but they checked that expansive power by making it hard to pass a treaty. In neither of these scenarios, however, is the issue in Reid raised.
    I’d also like to add another issue to the debate–one that I know none about, however. On what constitutional basis are extradition treaties passed? I know of no constitutional clause providing a special international extradition power, and it would seem that extradition treaties cover areas of power that are not enumerated in Article I as ceded to Congress. If there is no special constitutional grant, and the power does indeed cover areas of legislation reserved to the states, then it must be concluded that the treaty power is not limited by the list of enumerated powers in Article I. Of course, extradition was a matter of international concern in 1789, and extradition treaties are passed according to Article 2 processes, so both potential internal limitations would be met. This just goes to show that the list of enumerated legislative powers in Article I cannot operate as an external limitation to the treaty power in Article II.

  13. and Daniel, note that Art. VI, cl. 2 expressly applies to “all” treaties and Art. II, Sec. 3 contains no limitation regarding subject matter or anything else. The treaty power is an independent power.  An imaginary implied “federalism” limit is totaly inconsistent with the text of the Const. (esp. the 10th Amend.), and so is an imaginary “subject matter” limitation.

Trackbacks and Pingbacks

  1. There are no trackbacks or pingbacks associated with this post at this time.