A Response to David Sloss by Carlos Vázquez

[Carlos Vázquez, Professor of Law, Georgetown Law Center responds to David Sloss, Executing Foster v. Neilson:The Two-Step Approach to Analyzing Self-Executing Treaties. This post is part of the Third Harvard International Law Journal/Opinio Juris Symposium.]

David Sloss’s article, Executing Foster v. Neilson, is an important contribution to the literature on the judicial enforcement of treaties. I agree with much of it, as I agree with much of Professor Sloss’ other writing on treaties. In particular, I agree that the two-step approach to treaty enforcement that he proposes is generally the right approach, and I agree that the “intent-based” approach to the self-execution issue that he criticizes is highly problematic. But Professor Sloss and I disagree about the source of this problematic approach. I have traced this approach to Chief Justice Marshall’s opinion in Foster v. Neilson. Professor Sloss traces it to courts and scholars (including me) who, in his view, have misread Foster. I shall address our differences on this point below. First, however, I shall explain my general agreement with the two-step approach to treaty enforcement that Professor Sloss defends.

The much-controverted question of treaty self-execution is widely understood to concern whether a treaty may be enforced directly by the courts or must instead await legislative implementation. Professor Sloss proposes a two-step analysis for addressing this question. The first step is to determine what the treaty obligates the United States to do. This is a question of treaty interpretation, to be answered through the application of the international law of treaty interpretation. The second step is to identify which domestic officials have the power and duty to enforce the obligation. This, Professor Sloss argues, is entirely a matter of U.S domestic law, not a matter of treaty interpretation. Courts and commentators have fallen into error, and produced much confusion, by treating the second question as one of treaty interpretation, seeking an answer in the text of the treaty or in the parties’ intent. Professor Sloss notes that treaties seldom address the question of which domestic officials – legislative, executive, or judicial – are responsible for enforcing the treaty. Instead, treaty parties almost always leave that question to the domestic law of the states-parties.

Professor Sloss is entirely correct to note that seeking the answer to this question in the treaty itself is highly problematic. Although there is nothing in international law that prevents states from addressing that question in the treaty itself, the fact is that states almost never do so. Domestic officials take their cues from domestic law, and states have very different constitutional rules concerning the need for legislative implementation of treaties. In the U.K. and most nations of the British Commonwealth, treaties are never enforceable in the courts until they have been implemented by legislation. The constitutional law of other countries permits the direct judicial enforcement of some treaties but not of others. In the United States, for example, treaties that require the criminalization of conduct or the appropriation of money must be legislatively implemented because the Constitution as been interpreted to require a statute for those purposes. Because of the diversity of domestic constitutional rules on the question, states rarely, if ever, address the issue in the treaties they conclude. Seeking an answer to the self-executing question in the treaty itself is thus, in Justice Breyer’s words, like hunting for the Snark. No matter how hard they look, the courts will almost never find an answer there.

States instead leave the question to the domestic law of each state-party. The most relevant provision of our Constitution is the Supremacy Clause, which provides that “all Treaties” of the United States are “the supreme Law of the Land,” and instructs judges to give them effect. This clause, I have argued, was intended to reverse the British rule, which we would otherwise have inherited. It establishes that treaties in the United States do not as a constitutional matter always require implementing legislation; and it appears to establish that treaties are judicially enforceable in the same circumstances as constitutional and statutory provisions of like content. The first task for a court confronted with a treaty should thus be to identify the treaty’s content, which, as Professor Sloss argues, is a matter of treaty interpretation. The next step should to be to ask whether the obligation imposed by the treaty is one that would be judicially enforceable if it were found in a statute. Thus, as with statutes, a treaty would not be judicially enforceable if it were unconstitutional. (This would include treaties that purported to accomplish something for which the Constitution requires a statute, such as criminalization of conduct or appropriation of money.) Similarly, a treaty would not be judicially enforceable if it imposed an obligation requiring the exercise of political judgment. This would include treaties that set forth aspirations or that contemplate the exercise of discretion requiring political judgment. But, as Professor Sloss correctly argues, a treaty imposing a nondiscretionary duty of government to behave in a determinate ways towards individuals should not generally give rise to questions regarding its judicial enforceability at the behest of such individuals.

Regarding all of the above, Professor Sloss and I appear to agree. Our main point of disagreement concerns my claim that the Supreme Court in Foster recognized another category of non-self-executing treaty: treaties that are non-self-executing because the treaty parties intended that the United States’ obligation be subject to legislative implementation. Professor Sloss contends that the Court in Foster did not treat the issue as one of treaty interpretation, but instead held that the treaty involved was non-self-executing because it imposed an obligation that required legislative implementation for constitutional reasons.

Even here, my disagreement with Professor Sloss is narrow: for the reasons that Professor Sloss highlights in his article, and that I have sketched out above, I have always regarded the intent-based category of non-self-execution as highly problematic. Because states rarely address the issue, courts will rarely find evidence of any intent regarding the need for legislative implementation. If they think that have found evidence, they are almost certainly misreading the treaty, attributing to the parties a nonexistent intent. I have accordingly argued that the Supremacy Clause should be understood to establish a strong presumption that a treaty was not intended to require legislative implementation, reversible only through clear evidence that the parties did so intend. If Professor Sloss and I are right that the parties to a treaty rarely, if ever, address the question of implementing legislation, then the clear statement rule that I have proposed should lead the courts rarely, if ever, to find a treaty non-self-executing on the basis of the parties’ intent. Thus, my approach and Professor Sloss’ will almost always lead to the same result.

Professor Sloss claims that this problematic approach to the self-execution question is based on a misreading of Foster. He argues that the Court in Foster determined that the obligation that the treaty imposed on the parties was one that required legislative implementation by virtue of Article IV of the Constitution. Specifically, he argues that the treaty at issue in Foster conferred on the grantee only an inchoate or equitable title in the property at issue. Legislation was required, according to Professor Sloss, because Article IV of the U.S. Constitution assigns exclusively to Congress the power to transform an inchoate title to property otherwise held by the United States into a perfect title.

Would that he were right. Unfortunately, Professor Sloss’ attempt to reinterpret Foster is unsupported by Court’s opinion. Foster involved Article 8 of an 1819 treaty with Spain, which provided that “all grants of land made before the 24th of January 1818 by his catholic majesty, or by his lawful authorities, in the said territories ceded by his majesty to the United States, shall be ratified and confirmed to the persons in possession of the lands, to the same extent that the same grants would be valid if the territories had remained under the dominion of his catholic majesty.” Professor Sloss notes that some of the grants of land made by his catholic majesty conferred perfect titles and some conferred inchoate titles. If the Spanish grant conferred a perfect title, then the treaty, correctly construed, required the United States to recognize a perfect title, enforceable without prior legislative intervention. If the Spanish grant conferred an inchoate title, then the United States was only required to recognize an inchoate title, and, by virtue of Article IV of the Constitution, the power to transform and inchoate title into an actual title belongs to Congress. Professor Sloss points to a line of post-Percheman cases articulating this distinction, requiring legislation with respect to inchoate grants but not with respect to perfect grants.

Foster, however, involved a Spanish grant of land located in territory that the Court in Foster held had ceased to belong to Spain in 1803. Because the grant related to land that was not within the “territories ceded by his catholic majesty to the United States,” a majority of the Court held that Article 8 was simply inapplicable. For these Justices, the analysis ended there. The portion of the Foster opinion that has come to be understood as recognizing the doctrine of non-self-executing treaties was relevant only to Chief Justice Marshall and one other Justice, and that was because these two Justices construed Article 8 to require the United States to recognize these grants as if the land had in fact belonged to Spain between 1803 and 1819. These two Justices were inclined to accept such a construction because a declaration was appended to the treaty specifying that, notwithstanding Article 8, three specified grants did not need to be recognized. One of these three specifically excluded grants lay in territory that, according to the Court’s earlier analysis in Foster, did not belong to Spain between 1803 and 1819. In the view of the Chief Justice, this declaration “made it difficult to resist the construction that the excepted grants . . . would otherwise have been within [Article 8].” Marshall went on, however, to hold that, even if Article 8 were construed to apply to lands located within the disputed territory, the courts could not enforce the Spanish grants because Article 8 was non-self-executing.

Professor Sloss points to Supreme Court decisions, handed down after Marshall’s death, holding that individuals holding Spanish grants within this disputed territory acquired an inchoate or equitable interest in the land binding on the conscience of the sovereign. According to Professor Sloss, Foster is best read, in light of these subsequent cases, as holding that equitable interests of this sort may be transformed into perfect titles only through legislative action.

But the Foster opinion leaves no doubt that the Court understood the need for legislation to arise from what the treaty itself had to say about the need for legislation, which in turn depended on the particular wording of Article 8. According to the Court, the matter turned on the “sound construction of the eighth article.” Legislation was necessary because Article 8 “addresses itself to the political, not the judicial department.” The question, according to the Court, was: “Do these words act directly on the grants, so as to give validity to those not otherwise valid; or do they pledge the faith of the United States to pass acts which shall ratify and confirm them?” Most importantly, the Court emphasized that “[t]he article . . . does not say that those grants are hereby confirmed. Had such been its language, it would have acted directly on the subject, and would have repealed those acts of congress which were repugnant to it; but its language [is different].” The Court did not distinguish between grants of perfect titles and grants of inchoate titles, nor did it point to any particular characteristic of the grants involved in the case.

According to the Court’s reading of Article 8, therefore, even perfect grants would not have been directly enforceable in court. Had Article 8 provided that the grants were “hereby” confirmed, on the other hand, the treaty would have had the effect that Sloss claims it had: perfect grants would have been directly enforceable and inchoate grants would have been subject to implementing legislation. The fact that the Court distinguished the actual wording of Article 8 from a different wording that would have produced the result that Sloss defends shows that the Court was not reading Article 8 as Professor Sloss contends.

The Court’s subsequent analysis in Percheman confirms the conventional understanding of Foster. Had Marshall understood Foster’s non-self-execution holding to be limited to inchoate grants, his analysis in Percheman could have stopped upon concluding that Percheman involved a perfect grant. But the Court found it necessary in Percheman to revisit Foster’s reading of Article 8 as requiring legislative implementation. Again, the Court focused on the words of the treaty. But, this time, with the Spanish text before him, Marshall concluded that the words did not necessarily “stipulate for some future legislative act,” as the Court had mistakenly concluded in Foster.

In short, while Professor Sloss has offered an alternative basis on which the Court in Foster could perhaps have reached the same result, he has not shown that the Court in fact decided the case on such grounds. To the contrary, the conclusion that the Court based its decision on what the treaty itself had to say about whether the United States was required to “pass acts” is irresistible. While much about Foster is unclear, the Court’s understanding of the issue as one of treaty interpretation is not.

Professor Sloss gives three reasons for rejecting this conclusion. First, he notes that “the claim that Article 8 requires legislative implementation . . . has no basis in the treaty text.” I agree that the Court misread the treaty. But that does not show that the Court did not base its decision on its reading of the words of the treaty, any more than the fact that the Medellín Court misunderstood the meaning of the term “undertakes” in the U.N. Charter establishes that the Court in Medellín did not purport to be answering the self-executing question by reference to the words of that treaty. Second, Professor Sloss notes that it was already well established at the time “that international law does not govern the internal processes by which a nation implements its treaty obligations.” That is also true, but that, too, is a basis for criticizing the Court’s approach to the self-execution question, just as Professor Sloss (and Justice Breyer, and I) have criticized the Medellín majority’s focus on treaty text in answering the same question.

Finally, Professor Sloss argues that the conventional interpretation of Foster and Percheman is inconsistent with the Court’s post-Percheman decisions, which “repeatedly affirmed that congressional legislation was necessary to perfect inchoate titles.” But these latter decisions are not inconsistent with the conventional interpretation of Foster. Under the Foster interpretation, because of the wording of article 8, legislative implementation was necessary before the courts could enforce any Spanish grants, even perfect grants such as that involved in Percheman. That is why the Court had to revisit the self-execution question in Percheman. Having corrected its misconstruction of Article 8, the Court held in Percheman that the no legislation was required to authorize the courts to recognize Spanish grants of perfect titles. But, since Article 8 only required recognition of Spanish grants “to the same extent that the same grants would be valid if the territories had remained under the dominion of [Spain],” the U.S. courts were only required to recognize inchoate grants as inchoate. If Professor Sloss is right about the constitutional need for legislation to transform inchoate grants into perfect grants, and if, as Professor Sloss maintains, the grant involved in Foster was indeed an inchoate one, then the Court in Foster might well have reached the same result in that case even if it had not misconstrued Article 8 as requiring legislation across the board. Marshall could thus perhaps have avoided his unfortunate disquisition on treaties that are self-executing because of the parties’ intent.

But, unfortunately, Marshall did not take that course. As a result, we were left with the highly problematic intent-based approach to determining when treaties require implementing legislation. Marshall himself eventually recognized his mistake in reading Article 8 to require legislation. I have argued that, consistent with the Court’s approach and language in Percheman, a treaty should not be read to reflect an intent to require implementing legislation unless it “stipulates for some future legislative act.” As noted, this approach to the intent-based category of non-self-executing treaties should produce the same results as Professor Sloss’ approach. In short, Foster is ripe for criticism and limitation, even outright overruling. But it is not susceptible to the reinterpretation that Professor Sloss offers.