Michigan Law Review

November, 1998

Article

*390 THE TREATY POWER AND AMERICAN FEDERALISM

Curtis A. Bradley [FNa1]

Table of Contents

*391 Introduction

For much of this century, American foreign affairs law has assumed that there is a sharp distinction between what is foreign and what is domestic, between what is external and what is internal. This assumption underlies a dual regime of constitutional law, in which federal regulation of foreign affairs is subject to a different, and generally more relaxed, set of constitutional restraints than federal regulation of domestic affairs. In what is perhaps its most famous endorsement of this proposition, the Supreme Court stated in 1936 that "the federal power over external affairs [is] in origin and essential character different from that over internal affairs." [FN1] For a variety of reasons, however, the distinction between domestic and foreign affairs has been eroding in recent years, and this trend is likely to continue. [FN2] As a result, there will be an increasing need to reexamine the differential treatment of federal foreign affairs powers.

This Article reexamines one example of such differential treatment‑‑the purported immunity of the treaty power from federalism limitations. The Constitution provides that the President "shall have Power, by and with the Advice and Consent of the Senate, to make Treaties, provided two‑thirds of the Senators present concur." [FN3] Since the adoption of the Constitution, the President has exercised this power to commit the United States to hundreds of international obligations. The President also has committed the United States to thousands of additional obligations without going through the Article II process, by means of so‑called "executive agreements." [FN4] The treaties entered into by the President are deemed by the Constitution to be part of the supreme law of the land, [FN5] and the Supreme Court has construed this supremacy to extend *392 to executive agreements as well. [FN6] This means, among other things, that treaties and executive agreements preempt inconsistent state law. [FN7] Because of the supremacy of treaty law over state law, the treaty power implicates important issues concerning this country's federal system of government.

A central principle underlying American federalism, often recited by the Supreme Court, is that the national government is one of limited, enumerated powers. [FN8] A corollary of this principle is that when the federal government makes supreme federal law, it is restrained in what it can do either by inherent limits in the scope of its delegated powers, or by the Tenth Amendment's reservation of powers to the states, or both. [FN9] To be sure, these restraints are not nearly as strong as they once were, and the Supreme Court's willingness to police these restraints has varied throughout U.S. history. [FN10] Nevertheless, neither the Court nor most commentators deny the existence of such restraints. Even in the Garcia decision, the low point of judicial protection of federalism, the Court acknowledged that there are "limitation[s] on federal authority inherent in the delegated nature of Congress' Article I powers" and that *393 "[t]he States unquestionably do 'retai[n] a significant measure of sovereign authority.' " [FN11]

This is not the conventional wisdom, however, with respect to the treaty power. Although the treaty power is understood as being subject to the individual rights protections of the Constitution, [FN12] and perhaps also to separation of powers restrictions, [FN13] treaties and executive agreements are not thought to be limited either by subject matter or by the Tenth Amendment's reservation of powers to the states. As Professor Lori Damrosch has stated, "our constitutional law is clear: the treaty‑makers may make supreme law binding on the states as to any subject, and notions of states' rights should not be asserted as impediments to the full implementation of treaty obligations." [FN14] For the sake of convenience, I will refer to this conventional wisdom as the "nationalist view." The nationalist view has been endorsed by a number of prominent foreign affairs commentators, [FN15] as well as by the influential Restatement (Third) of the Foreign Relations Law of the United States. [FN16]

As suggested by Professor Damrosch, the nationalist view of the treaty power has two components. First, largely on the basis of the Supreme Court's decision in Missouri v. Holland, [FN17] it generally is understood today that "the Tenth Amendment, reserving to the several States the powers not delegated to the United States, does *394 not limit the power to make treaties or other agreements." [FN18] Second, while it "was once widely accepted" that treaties could be made only with respect to matters of "international concern," [FN19] most commentators today either disagree with such a limitation or assume that it is insignificant, given that most matters upon which treaties are likely to be concluded can plausibly be characterized as of international concern. [FN20]

In this Article, I question the nationalist view. As I explain, the two components of the nationalist view have developed in isolation. While either component might seem relatively unproblematic by itself, when considered together they violate the principle of limited, enumerated powers. The treaty power in our Constitution is a power to make supreme federal law. If such law can be made on any subject, without regard to the rights of the states, then the treaty power gives the federal government essentially plenary power vis‑a‑ vis the states. Such plenary power, however, is exactly what American federalism denies. This inconsistency between the nationalist view and American federalism is particularly significant today, in light of the Supreme Court's renewed commitment to protecting federalism and the rapidly expanding nature of this country's treaty commitments. [FN21]

I should make clear at the outset the nature of my argument. I am not defending here the value of federalism, or judicial review of federalism, subjects that have generated enormous literature. [FN22] My argument is simply that if federalism is to be the subject of judicial protection‑‑as the current Supreme Court appears to believe‑‑there is no justification for giving the treaty power special immunity from such protection. My argument is one against treaty power exceptionalism, not necessarily one in favor of federalism. In addition, I am not assuming here the legitimacy of any particular method of constitutional interpretation, such as originalism or textualism. Instead, I consider all the standard interpretive materials, including text, history, structure, and changed circumstances, and I *395 conclude that none of these materials justifies giving the treaty power special immunity from federalism limitations.

This Article proceeds in five parts. Part I describes why the relationship between the treaty power and American federalism is particularly significant today, in light of recent changes in the nature of treaty‑making, as well as the recent federalism jurisprudence of the Supreme Court. Part II examines materials from the Founding period and the nineteenth century and concludes that, contrary to claims by its proponents, the nationalist view lacks substantial support in history. Part III recounts how the nationalist view became orthodoxy, beginning with the Supreme Court's 1920 decision in Holland and followed by the eventual academic repudiation of a subject matter limitation on the treaty power. Part IV then sets forth a critique of the nationalist view. In particular, it questions the three principal justifications for the nationalist view: that the treaty power is immune from federalism restrictions because that power has been exclusively delegated to the federal government; that federalism limitations are unnecessary because the political process is sufficient to protect states' rights; and that imposing federalism limitations on the treaty power would unduly interfere with the ability of the federal government to speak with one voice in foreign affairs. Part V argues that, while it may not be feasible to limit the treaty power by subject matter, this power should at least be subject to the same federalism limitations as Congress's legislative powers. To the extent that this conclusion would require overruling Holland, this Part argues that the justifications for stare decisis are weak in this context, given the substantial changes in both the nature of treaty‑making and the scope of permissible federal legislation.

I. Contemporary Significance of the Federalism Issue

The relationship between the treaty power and American federalism is not a new issue. It has been a matter of controversy since the Founding of the Constitution. It also was the subject of substantial academic and official attention during both the early part of this century and the 1950s. [FN23] During the last several decades, however, it seems largely to have receded from view. In this Part, I explain why this issue is likely to come back into focus, and why it deserves our attention.

*396 A. Changes in Treaty‑Making

To those unfamiliar with international law, it might seem that the treaty power would be an unlikely threat to federalism. Treaties, after all, concern the relations among nations, whereas federalism concerns the relationship between the national government and the constituent states. To put it differently, treaties concern inter‑national relations whereas federalism concerns intra‑national relations.

This dichotomy might have been accurate at one time in American history, when treaties were generally bilateral and regulated matters such as diplomatic immunity, military neutrality, and removal of trade barriers. [FN24] The nature of treaty‑making, however, has undergone a radical transformation, especially in the years since World War II. As the Restatement (Third) of Foreign Relations Law explains, "[u]ntil recently, international law was essentially customary law: agreements made particular arrangements between particular parties, but were not ordinarily used for general law‑making for states." [FN25] During the latter part of this century, however, there has been a proliferation of treaties, such that treaty‑making has now eclipsed custom as the primary mode of international law‑making. [FN26] Moreover, many of these treaties take the form of detailed multilateral instruments negotiated and drafted at international conferences. These treaties resemble and are designed to operate as international "legislation" binding on much of the world. [FN27]

Even more significant than these structural changes is the change in the content of modern treaty‑making. While many treaties continue to concern matters traditionally viewed as inter‑national in nature, numerous others concern matters that in the past countries would have addressed wholly domestically. This change in treaty‑making is most evident in the area of international human rights law, which purports to regulate the relationship between nations and their own citizens. There is now general agreement "that how a state treats individual human beings, including its own citizens, in respect of their human rights, is not the state's own business *397 alone . . . but is a matter of international concern and a proper subject for regulation by international law." [FN28] As a result, there are today a host of multilateral human rights treaties that purport to confer a variety of rights that individuals can assert against their own governments. These treaties address issues such as racial and gender equality, criminal procedure and punishment, and religious freedom. [FN29]

This transformation in treaty‑making is so fundamental that it alters the very essence of international commitments. As Professor Henkin has explained, "[h]uman rights law has shaken the sources of international law, reshaped its character, enlarged its domain." [FN30] Because treaties now regulate matters that countries traditionally have considered internal, there is an increasing likelihood of overlap, and conflict, with domestic law. This is particularly so, given that "in certain important respects, international human rights norms are more rights‑protective than the corresponding domestic law standards." [FN31] In a federalist system like the United States, this means that some of the overlap and conflict is likely to occur at the state level.

*398 Another important development during this period has been the increasing use by the President of executive agreements. Executive agreements are, quite simply, international agreements concluded by the President without the two‑thirds senatorial advice and consent specified in Article II of the Constitution. [FN32] Executive agreements approved in advance or after the fact by a majority of both houses of Congress are referred to as "congressional‑executive agreements." [FN33] Executive agreements concluded by the President alone are referred to as "sole executive agreements." [FN34] The Supreme Court has endorsed the constitutional legitimacy of executive agreements, [FN35] and it has held that even sole executive agreements are supreme federal law and thus supersede inconsistent state law. [FN36] The Court has not addressed the permissible scope of executive agreements, but the prevailing view today is that at least congressional‑ executive agreements are fully interchangeable with treaties and thus may be used any time that a treaty would be proper. [FN37]

While executive agreements were relatively infrequent during the nineteenth and early twentieth centuries, the vast majority of international agreements concluded by the President in the latter half of this century have been in the form of executive agreements rather than Article II treaties. [FN38] Commentators have debated at various times the constitutional legitimacy of such agreements, [FN39] but the issue may be largely academic in light of the widespread and now longstanding nature of the practice and Congress's acquiescence to it. The rise of executive agreements is relevant to the federalism question because the two‑thirds Senate consent requirement in Article II, which these agreements bypass, has long been thought to provide special protection of states' rights. [FN40]

B. Supreme Court's Renewed Commitment to Federalism

As the above changes in treaty‑making began to unfold after World War II, concerns were in fact raised regarding their potential impact on American federalism. These concerns were expressed most prominently in connection with debates in the 1950s over the proposed "Bricker Amendment" to the Constitution. [FN41] This controversy died out, however, in part because of the Supreme Court's expanded reading of Congress's commerce and other powers, beginning in the New Deal era and continuing into the civil rights era. [FN42] Once it became understood that the federal government had almost unlimited domestic lawmaking powers, the particular scope of the treaty power (or executive agreement power) became less relevant. [FN43] Although the Supreme Court did attempt to reinvigorate the Tenth Amendment in its 1976 National League decision, [FN44] this effort was short‑lived, as a majority of the Court overruled that decision just nine years later in the Garcia case. [FN45] Thus, once again, the Court seemed to allow the federal government essentially unlimited lawmaking power vis‑a‑vis the states. [FN46]

*400 During the 1990s, however, the Court has shown a willingness in a number of areas to limit the scope of Congress's domestic powers, and to enforce the rights of the states against federal regulation. For example, in New York v. United States, the Court invalidated a federal statute that in effect compelled state disposal of radioactive waste as "inconsistent with the federal structure of our Government established by the Constitution"; [FN47] in United States v. Lopez, the Court invalidated a federal statute criminalizing the possession of firearms near school zones as exceeding Congress's powers under the Commerce Clause; [FN48] in City of Boerne v. Flores, the Court invalidated the Religious Freedom Restoration Act (RFRA) [FN49] on the ground that it exceeded Congress's powers under the Fourteenth Amendment; [FN50] and in Printz v. United States, the Court invalidated a federal statute requiring state law enforcement officials to conduct background checks on prospective handgun purchasers as "compromis[ing] the structural framework of dual sovereignty." [FN51] The recent Court also has protected federalism in other, less direct ways‑‑for example, by narrowing the situations in which Congress can override the states' Eleventh Amendment immunity from suit. [FN52]

The Supreme Court's renewed commitment to protecting federalism is likely to increase the importance of the scope of the treaty power. If the treaty power is immune from federalism restrictions, as the nationalist view maintains, then it may be a vehicle for the enactment of legislative changes that fall outside of Congress's domestic lawmaking powers. [FN53] Indeed, commentators recently have begun to seize on this possibility.

*401 Professor Gerald Neuman, for example, has argued that even if RFRA is not a valid exercise of Congress's domestic lawmaking powers, it can be justified constitutionally as part of this country's implementation of a treaty‑‑the International Covenant on Civil and Political Rights (ICCPR). [FN54] The ICCPR, which the United States ratified in 1992, contains a long list of individual rights that cannot be infringed by member countries. One of these rights is a right "to freedom of thought, conscience and religion," which the Convention says includes a right "to manifest [one's] religion or belief in worship, observance, practice and teaching." [FN55] The Convention also says that the only limitations that may be imposed on this right are those "necessary to protect public safety, order, health, or morals or the fundamental rights and freedoms of others." [FN56]

Neuman makes the plausible argument that these provisions authorize Congress to grant religious freedom protection such as is reflected in the RFRA statute. [FN57] As for the federalism concerns associated with the statute, Neuman argues that "[t]he mere fact that the treaty may require the extension of religious exemptions within areas of traditional state regulation creates no obstacle to its validity" in light of the well‑settled proposition that the treaty power is "free from any 'invisible radiation from the general terms of the Tenth Amendment.' " [FN58] In other words, even though the Supreme Court has declared the RFRA statute to be "a considerable congressional intrusion into the States' traditional prerogatives and general authority to regulate for the health and welfare of their citizens," [FN59] this intrusion is entirely proper, says Neuman, as long as it is enacted pursuant to a treaty. This is what the nationalist view entails, and Neuman has only scratched the surface.

*402 C. Other Examples

The RFRA example is just one of many instances in which the treaty power might be used to overcome federalism restraints on domestic lawmaking. Below I consider a number of other examples, some actual and some speculative, where this issue might arise.

1. Human Rights Standards

There are numerous instances in which Congress might use human rights treaties to overcome federalism restraints on its lawmaking power. Consider, for example, the Convention on the Rights of the Child, [FN60] which became effective in 1990 and has now been ratified by almost every nation in the world. This treaty‑‑which the United States has signed but has not yet ratified‑‑contains a number of provisions that may be inconsistent with current U.S. family law. [FN61] This inconsistency has prompted federalism concerns [FN62] because family law is a subject that largely has been regulated in this country at the state rather than federal level. [FN63]

*403 Similar concerns have been raised with respect to the Convention on the Elimination of All Forms of Discrimination Against Women (CEDAW), [FN64] various sections of which "provide for the elimination of discrimination against women in areas of government that have historically been considered political functions of the several states, such as regulation of family relations and education." [FN65] The Convention includes provisions addressing, among other things, discrimination "in recreational activities, sports and all aspects of cultural life," as well as "in all matters relating to marriage and family relations." [FN66] The United States has signed but has not yet ratified this Convention, although the Clinton administration has been pressing the Senate for approval. [FN67]

Still another example is the relationship between international law and the controversial issue of affirmative action. Some states, such as California, have begun to cut back on race‑based and other affirmative action programs. [FN68] In response, two scholars recently made the claim that such affirmative action, even if no longer required as a matter of U.S. domestic law, may be mandated by human rights treaties to which the United States is a party. [FN69]

2. Criminal Law and Punishment

American criminal law, another area of law primarily regulated in this country at the state level, also has become the subject of treaty‑making. For example, the statute implementing the Genocide Convention makes it a federal crime to kill or cause serious *404 bodily harm in this country "with the specific intent to destroy, in whole or in substantial part, a national, ethnic, racial, or religious group as such." [FN70] A similar example is the recently‑enacted statute implementing the Hostage Convention, which federalizes garden‑variety kidnappings in this country whenever a foreign citizen is involved. [FN71] It is possible that statutes such as these raise federalism concerns, given that, "[u]nder our federal system, the 'States possess primary authority for defining and enforcing the criminal law.' " [FN72]

Criminal punishment, especially the death penalty, presents an additional area of potential conflict between treaty law and federalism. Despite international criticism, [FN73] the United States is imposing the death penalty with increasing regularity, primarily at the state level. In light of recent Supreme Court precedent, it seems unlikely that the federal government has the power under either Article I or the Fourteenth Amendment to abolish state use of the death penalty, assuming states are not imposing it in a manner that violates the Eighth Amendment. [FN74] According to the nationalist view, however, this limitation on the federal government's power would disappear if the federal government entered into a treaty outlawing capital punishment. [FN75]

*405 Although involving a more traditional treaty, a recent incident dramatically illustrates the potential conflict between the treaty power and state authority to impose the death penalty. In April 1998, the International Court of Justice in The Hague (the "ICJ") ordered the United States to "take all measures at its disposal" to stay the execution in Virginia of Angel Breard, a Paraguayan national, while the Court considered a suit brought against the United States by Paraguay. [FN76] The suit alleged that Virginia had violated a multilateral treaty, the Vienna Convention on Consular Relations, [FN77] by failing to advise Breard of his right to contact and request assistance from the Paraguayan consulate, and that, as a result, Breard's conviction and sentence should be vacated. [FN78]

In response to the ICJ's order, the State Department asked Virginia's governor voluntarily to stay the execution, but he refused to do so, stating that delaying the execution "would have the practical effect of transferring responsibility from the courts of the Commonwealth [of Virginia] and the United States to the International Court." [FN79] The U.S. Supreme Court also refused to stay the execution, concluding that "[i]f the Governor wishes to wait for the decision of the ICJ, that is his prerogative. But nothing in our existing case law allows us to make that choice for him." [FN80] The Justice Department, in a brief submitted to the Supreme Court, stated that even if the ICJ's order were binding, the federal government did not have the power to compel Virginia's compliance with it because "our federal system imposes limits on the federal government's ability to interfere with the criminal justice systems of the States." [FN81] *406 A number of commentators, by contrast, have argued that, in light of Holland, such federalism limits were not applicable. [FN82]

3. Commerce and Trade

Commercial and other private law treaties also have the potential to intrude on traditional state prerogatives. As one commentator recently explained, "[a] t issue in the ratification process . . . is nothing less than federal arrogation of traditional state competence in the law governing private, and in particular commercial, relations." [FN83] The United States already is a party to the Convention on Contracts for the International Sale of Goods. [FN84] This Convention governs the sale of goods in a variety of international contract situations, although contracting parties are allowed to opt out of its provisions. [FN85] In this country, the Convention is considered a self‑executing treaty, and thus, when it applies, it preempts inconsistent state law, including the Uniform Commercial Code. [FN86]

Another sector of private law that may become the subject of a treaty is the enforcement of judgments. A number of countries, including the United States, currently are negotiating a proposed multilateral treaty, in connection with The Hague Conference on Private International Law, that would establish uniform standards for the recognition and enforcement of foreign judgments. [FN87] This is a subject that has been regulated in this country primarily by the states, [FN88] and commentators expressed concern as late as the 1950s that a treaty on this subject might exceed federal power. [FN89] A similar *407 example is the Convention on the Law Applicable to Succession to the Estates of Deceased Persons, [FN90] which would establish choice of law rules concerning inheritance issues. This "Convention would seem to change hallowed rules of U.S. state law without the scrutiny that such a change would get in a state legislature." [FN91]

An additional example in commercial law is the protection of intellectual property, a subject increasingly regulated by treaty. [FN92] Intellectual property protection is governed extensively in this country by federal law, but there are recognized limits to the scope of permissible federal protection. [FN93] This has prompted one commentator recently to wonder whether, in light of Holland, the federal government could by treaty confer stronger intellectual property rights than it would otherwise have the power to do pursuant to its domestic lawmaking powers. [FN94]

Federalism concerns also exist with respect to the GATT and NAFTA trade agreements, which were concluded by the President by means of executive agreements rather than the Article II treaty process. [FN95] Both of these agreements contain provisions that affect *408 state regulatory authority. [FN96] Objections were made to the latest GATT agreement, for example, on the ground that it "impinged on matters that had been, and are generally, governed by state law, such as product‑safety regulation, banking and insurance, and local 'tax breaks' and other subsidy practices." [FN97] Similarly, one commentator recently noted that the agreement may override state voter initiatives concerning the labeling of products. [FN98]

4. Environmental Protection

The potential conflict between the treaty power and federalism is also evident in the area of environmental protection. The United States is a party to a number of treaties relating to the environment, some of which Congress has invoked or could invoke as a basis for its enactment of federal environmental legislation. In enacting the Endangered Species Act, [FN99] for example, Congress cited as a basis for its authority various treaties including the Convention on Nature Protection and Wild Life Preservation in the Western Hemisphere. [FN100] In light of the Supreme Court's decision in United States v. Lopez, it is arguable that some of these statutes, or at least certain applications of them, exceed Congress's powers under the commerce clause. [FN101] If so, the constitutionality of these statutes may depend on the validity of the nationalist view of the treaty power. [FN102]

*409 5. Commandeering of State Governments

Finally, consider the "commandeering" regulations invalidated by the Supreme Court in New York v. United States [FN103] and Printz v. United States. [FN104] It is not inconceivable that the federal government would conclude treaties addressing the subject matter of those cases‑‑disposal of radioactive waste and background checks for handgun ownership. Indeed, although not directly on point, the United States recently signed a treaty that provides that the parties to it "shall adopt the necessary legislative or other measures to establish as criminal offenses under their domestic law the illicit manufacturing of and trafficking in firearms, ammunition, explosives, and other related materials." [FN105] According to the nationalist view, the ratification of relevant treaties presumably would allow Congress to reenact the provisions invalidated in New York and Printz, notwithstanding the Supreme Court's determination that they intruded too deeply on state sovereignty. [FN106]

* * * I do not mean to suggest that any of these examples necessarily should be viewed as an improper exercise of the treaty power. I mention them rather because they highlight the increasing tension between international law and this country's federalist system. In light of this tension, the relationship between the treaty power and American federalism merits reexamination.

II. Historical Understanding of the Treaty Power

Proponents of the nationalist view of the treaty power sometimes rely on history to support their claims. In particular, they argue that the Founders did not intend either a states'‑rights or a subject matter limitation on the treaty power and that the Supreme Court repeatedly rejected such limitations during the nineteenth *410 century. [FN107] In fact, the historical record reveals a fairly consistent understanding that the treaty power was limited either by subject matter, states' rights, or both. This historical record may not by itself require contemporary rejection of the nationalist view, but it does undermine any strong historical claims for that view.

A. Founding Period

In the materials relating to the drafting and ratification of the Constitution, the only substantial discussions of the scope of the treaty power are contained in the records of the Virginia Ratifying Convention. Neither the records of the Federal Convention nor the Federalist Papers contain much discussion of this issue. Rather, the references to treaties in these materials primarily concern the process by which the federal government would conclude treaties and the proper governmental actors to be involved in this process. [FN108] Thus, for example, there was debate over whether to include the House of Representatives in the treaty process and over what proportion of the Senate should be required to approve a treaty. [FN109]

Several general themes do emerge even from these materials, however, that may be relevant to the scope of the treaty power. First, the Founders believed that treaties should be difficult to make. As Professor Henkin explains, "the prevailing mood at the Convention was that it should not be too easy to make treaties. Even the 'nationalists' among the Framers neither desired nor expected many treaties." [FN110] Thus, Gouverneur Morris observed that "[t] he more difficulty in making treaties, the more value will be set *411 on them," [FN111] and James Madison noted that it had been too easy to make treaties under the Articles of Confederation. [FN112] This theme was echoed in the state ratification debates. During the Pennsylvania Convention, James Wilson, in commenting on the likely workload of the Senate, stated that the treaty power "should be very seldom exercised‑‑. . . it will be but once in a number of years, that a single treaty will come before the senate." [FN113]

Second, the Founders contemplated that treaties would govern truly inter‑ national relations. The categories of treaties mentioned by the Founders concerned issues such as "war, peace, and commerce." [FN114] Even Alexander Hamilton, who was no great defender of states' rights, emphasized in The Federalist that treaties are "not rules prescribed by the sovereign to the subject, but agreements between sovereign and sovereign." [FN115] This understanding was a natural adjunct of the Founders' belief that there was a clear distinction between domestic and foreign affairs. Indeed, it is this distinction that helped ensure, in the Founders' minds, that the national government's power would be limited and, correspondingly, that states' rights would be protected. [FN116] As Professor Zechariah Chafee has explained: "The vital distinction between foreign affairs and domestic matters was taken for granted throughout [the drafting of the Constitution]. Indeed, this distinction was ingrained in their minds long before they met in Philadelphia." [FN117]

*412 Third, the Founders placed substantial emphasis on the role of the Senate in protecting states' rights. [FN118] Among other things, they noted that each state was to have equal representation in the Senate and that the senators would be elected by state legislatures. [FN119] Based on this, they reasoned that state interests would be safeguarded in the treaty process by assigning the treaty power to the Senate. [FN120] In carrying out its state‑protecting role, the Founders envisioned that the Senate would be actively involved with the President in the process of negotiating and concluding treaties. [FN121] Indeed, the Founders envisioned that the Senate would act as "a council‑like body in direct and continuous consultation with the Executive on matters of foreign policy." [FN122]

Fourth, supporters of the Constitution repeatedly expressed the view that the Constitution delegated only limited powers to the national government. Perhaps most famously, Madison stated in Federalist No. 45 that "[t]he powers delegated by the proposed Constitution to the federal government are few and defined. Those which are to remain in the State governments are numerous and indefinite." [FN123] The Federalist proponents of the Constitution thought this proposition so evident that it precluded the need for a Bill of Rights. [FN124] There is no evidence that the Founders believed the treaty power to be exempt from this general proposition.

*413 In addition to reiterating these general themes, the records of the Virginia Ratifying Convention contain specific discussions of the scope of the treaty power. [FN125] These discussions confirm that the Founders did in fact envision limitations on the treaty power. During the Virginia convention, Anti‑Federalists like Patrick Henry charged that under the proposed Constitution, the treaty‑makers could "make any treaty. . . . as they please." [FN126] In response, Federalist defenders of the Constitution strenuously denied that the treaty power was unlimited. Madison stated, for example: "The exercise of the power must be consistent with the object of the delegation. . . . The object of treaties is the regulation of intercourse with foreign nations, and is external." [FN127] He further explained that the Founders had not specified the proper subject matters of treaties in order to preserve flexibility, not because the power was unlimited. [FN128] Consistent with this view, Edmund Randolph remarked that "neither the life nor property of any citizen, nor the particular right of any state, can be affected by a treaty." [FN129] And, perhaps most broadly, George Nicholas stated that no treaty could be made "which shall be repugnant to the spirit of the Constitution, or inconsistent with the delegated powers." [FN130] Thus, when the question of the treaty power's scope was specifically discussed, the Founders did express the view that it was subject to constitutional limitations. As Charles Lofgren has explained, "[t]hose Virginia Federalists who *414 discussed the issue thus perceived constitutional limits to treaties beyond the minimal requirements that they be properly made." [FN131]

Statements made shortly after the ratification period, although not technically part of the original understanding, may nevertheless shed light on that understanding. It is significant, therefore, that the views expressed by the Federalists during the Virginia Convention were reiterated in the subsequent debates in 1796 over whether the House of Representatives had the power to inquire into the making of the Jay Treaty. This treaty‑‑the first adopted by the United States after ratification of the Constitution‑‑was between the United States and Great Britain and was negotiated by John Jay while he was serving concurrently as Minister to Great Britain and Chief Justice of the United States. [FN132] The treaty addressed, among other things, the withdrawal of military forces from American and British territories, indemnification of war‑time creditors, and certain boundary disputes. The treaty was extremely controversial in the United States because, among other things, it was perceived by many people as being too conciliatory to the British. In 1796, President Washington asked the House of Representatives to appropriate funds to implement the Treaty. In response, the Republicans in the House introduced a resolution asking the President to provide the House with the executive papers reflecting the negotiating history of the treaty. [FN133]

During the debates over this resolution, numerous representatives expressed the view that the treaty power was limited in scope. James Hillhouse, for example, stated that a treaty must relate "to objects within the province of the Treaty‑making power, a power *415 which is not unlimited." [FN134] He further stated that "[t]he objects upon which it can operate are understood and well defined, and if the Treaty‑making power were to embrace other objects, their doings would have no more binding force than if the Legislature were to assume and exercise judicial powers under the name of legislation." [FN135] Other representatives making similar statements included Daniel Buck, [FN136] James Madison, [FN137] Theodore Sedgwick, [FN138] and Samuel Smith. [FN139] Importantly, Madison observed that a "candid and collected view" of the state ratification debates made clear that the treaty‑making power had been understood as being limited in scope. [FN140]

Restrictions on the treaty power also were made a part of the Senate's Manual of Parliamentary Procedure, drafted by Thomas Jefferson between 1797 and 1801, when he was Vice‑President (and thus President of the Senate). [FN141] In the Manual, Jefferson noted that the Constitution did not define "[t]o what subject this power extends" and that the Founders were not "entirely agreed among ourselves" on this issue. [FN142] He explained that the treaty power was nevertheless impliedly limited in at least four ways. First, it had to "concern the foreign nation party to the contract." [FN143] Second, it covered "only those subjects which are usually regulated by treaty." [FN144] Third, it did not cover "the rights reserved to the states; for surely the President and Senate cannot do by treaty what the whole government is interdicted from doing in any way." [FN145] Fourth, it did not apply to "those subjects of legislation in which [the Constitution] *416 gave a participation to the House of Representatives." [FN146] Jefferson's fourth suggested limitation‑‑that the treaty power did not apply to matters that could otherwise be regulated by Congress‑‑"has been consistently rejected." [FN147] But both the subject matter and federalism limitations he suggested appear to have been consistent with the prevailing views of the time. [FN148]

Early opinions of the Attorney General also suggested limitations on the treaty power. An 1819 opinion from the Attorney General, for example, stated that the federal government could not alter by treaty state inheritance law concerning real property. [FN149] And an 1831 opinion stated that the federal government is "under a constitutional obligation to respect [the reserved powers of the states] in the formation of treaties." [FN150] Other government officials, by contrast, endorsed a subject matter rather than states'‑rights limitation on the treaty power. John Calhoun, for example, stated in 1816 (while he was a member of the House of Representatives) that treaties were proper as long as they "concern[ ] our foreign relations." [FN151]

The belief that there were subject matter or federalism limitations on the treaty power also was reflected in the views of early constitutional scholars. William Rawle stated that the treaty power was appropriate only for those subjects "which properly arise from intercourse with foreign nations," and he listed those subjects as "peace, alliance, commerce, neutrality, and others of a similar nature." [FN152] Joseph Story stated in his constitutional law treatise that, "though the [treaty] power is thus general and unrestricted, it is not to be so construed, as to destroy the fundamental laws of the state" *417 and "cannot supersede, or interfere with any other of [the Constitution's] fundamental provisions." [FN153] Subsequent nineteenth‑ century constitutional scholars echoed these comments. [FN154] As some of these scholars explained, the Founders did not enumerate subject matter limitations on the treaty power in order to preserve flexibility, not because they thought the power unlimited. [FN155]

In sum, although the primary focus of the Founders' attention was on the treaty process rather than on the scope of the treaty power, when they addressed that issue the Founders, as well as early scholars and government officials, made clear that the treaty power was not unlimited in scope. In particular, there appears to have been an understanding that the treaty power was limited either by subject matter, by the reserved powers of the states, or both. It could hardly have been otherwise, for, as noted above, the Constitution was viewed as delegating limited powers to the federal government. [FN156] Without subject matter or federalism restrictions on the treaty power, the federal government would have had essentially unrestricted authority vis‑a‑vis the states, thereby contradicting one of the core assumptions of the Founders. [FN157]

*418 B. Nineteenth‑Century Understanding

Proponents of a broad view of the treaty power sometimes also rely on nineteenth‑century Supreme Court decisions. They argue that, even if the Founding materials might support limitations on the treaty power, the Supreme Court repudiated such limitations during the nineteenth century. Professor Henkin, for example, asserts that any federalism limitation on the treaty power was "repeatedly rejected by the Supreme Court" in the nineteenth century. [FN158] In fact, while it is true that the Supreme Court in the nineteenth century upheld the validity of a number of treaties, the Court frequently expressed the view that the treaty power was limited either by subject matter, states' rights, or both. In addition, this assumption appears to have been shared by the treaty makers themselves.

The Supreme Court never held in the nineteenth century that the treaty power was immune from federalism limitations. It did emphasize, as Henkin and others have pointed out, that treaties are supreme over state law. [FN159] This proposition is, of course, evident from the constitutional text. [FN160] To say that treaties are supreme federal law, however, is not to say that the treaty power is unlimited in scope. As one early twentieth‑ century scholar explained: "[W]hile many of the decisions contain broad general statements to *419 the effect that treaties are the supreme law of the land, there is always the accompanying qualification that it must be a constitutional treaty, in order to be so considered." [FN161] Indeed, such a limitation was essential given the Court's repeated reaffirmation during this period, both before and after the Civil War, that the powers of the federal government are limited and enumerated. [FN162]

Consistent with this proposition, the Court in numerous cases stated that the treaty power was limited in scope. Some of the Court's decisions suggest a subject matter limitation on the treaty power. Thus, for example, the Court stated in several decisions that treaties must concern "proper subjects of negotiation." [FN163] In other decisions, the Court suggested a states'‑ rights limitation. Perhaps the best example is the Court's decision in New Orleans v. United States. [FN164] In that case, the United States had by treaty acquired the province of Louisiana from France, which had in turn acquired it from Spain. Prior to the treaty, the King of Spain had held certain public properties in the City of New Orleans in trust for the City. The issue was whether, pursuant to the treaty, the federal government had acquired similar trust rights over the properties, or whether the properties instead were now fully in the control of the City. In holding that the City had control, the Court explained that the federal government "is one of limited powers" and that its authority cannot be "enlarged under the treaty‑making power." [FN165] The Court concluded that the Commerce Clause did not encompass the trust power that had been exercised by the King of Spain and that "the treaty cannot give this power to the federal government." [FN166]

Similar statements, albeit often in dicta, are found in other Supreme Court decisions during this period. [FN167] Moreover, a *420 number of decisions during this period suggested both a subject matter and a states'‑ rights limitation on the treaty power. In a case involving the power of a state to extradite a person to a foreign country, for example, the Court stated that the treaty power covers "all those subjects, which in the ordinary intercourse of nations had usually been made subjects of negotiation and treaty; and which are consistent with the nature of our institutions, and the distribution of powers between the general and state governments." [FN168] Similarly, in a case involving a land treaty with the Cherokee Indians, the Court stated that treaties may deal with "all those objects which in the intercourse of nations had usually been regarded as the proper subjects of negotiation and treaty, if not inconsistent with the nature of our government and the relation between the States and the United States." [FN169]

Despite this language, proponents of a broad view of the treaty power point out that the Court upheld treaties during this period that concerned matters typically regulated by the states. Most notably, the Court upheld a number of treaties giving foreigners certain property rights equal to those enjoyed by citizens. [FN170] These treaties did intrude to some extent on state regulation of property ownership, and federalism concerns sometimes were invoked in opposition to these treaties. [FN171] But the treaties did not purport to regulate the relationship between states and their own citizens, or even citizens from other states in the nation. They regulated only the treatment of aliens, in return for similar treatment of U.S. citizens residing abroad. In that sense, the treaties were quite naturally viewed as regulating this country's inter‑national relations. As one court explained, after reviewing these cases: "If there is one object more than another which belongs to our political relations, and which ought to be the subject of treaty regulations, it is the extension of this comity which is so highly favored by the liberal spirit of the age, and so conducive in its tendency to the peace and amity of nations." [FN172]

*421 Even in the context of these treaties, the U.S. treaty makers expressed concern regarding states' rights, and in a number of instances acted to protect such rights. Ralston Hayden has documented how, between 1830 and 1860, "the Senate and the executive entertained grave and increasing doubts concerning their authority to make treaties" concerning rights to real property and how "in every particular instance in which conflict arose the treaty in question was amended to bring it more nearly into accord with the states' rights theory." [FN173] Thus, for example, he explains that, when President Fillmore submitted a proposed treaty between the United States and Switzerland to the Senate in 1850, he asked for and obtained amendments from the Senate to protect the reserved powers of the States. [FN174] William Mikell similarly has documented several instances in the nineteenth century when the treaty makers made treaties contingent on state agreement. [FN175] He describes, for example, an 1853 Consular Convention with France, in which the provision giving a property right to French citizens was made contingent on state laws permitting such a right. As for those states that did not permit such a right, "the President engage[d] to recommend to them the passage of such laws as may be necessary for the purpose of conferring this right." [FN176]

In the early part of the twentieth century, perhaps prompted by the increasingly active role of the United States in international trade and international politics, there was substantial debate over whether the treaty power was limited by the reserved powers of the states. Scholars lined up on both sides of this issue. The most prominent supporters of the states'‑rights view included Henry St. George Tucker, John Bassett Moore, and William Mikell. [FN177] The most prominent opponents of this view included Charles Butler, Edward Corwin, and Quincy Wright. [FN178] Few of the opponents of *422 the states'‑rights view, however, embraced the other component of the nationalist view‑‑that the treaty power also is unlimited by subject matter. [FN179] And other constitutional scholars in this period expressed the view that the treaty power was in fact limited by subject matter. [FN180] Moreover, U.S. officials continued to express the view that the United States could not regulate by treaty matters reserved to the states. [FN181]

In sum, although the Supreme Court upheld a number of treaties in the nineteenth century, it did not suggest that the treaty power was unlimited. Indeed, in many of these cases the Court expressly stated that the power was limited by subject matter, states' rights, or both. Such limitations also were reflected, as noted above, in the views of the treaty makers themselves. The states'‑rights issue became highly contested in the academic community in the early twentieth century, but even then many commentators appeared to assume at least a subject matter limitation.

III. Rejection of Tenth Amendment and Subject Matter Limitations

In this Part, I explain how each of the two components of the nationalist view of the treaty power became orthodoxy. I first discuss the rejection of Tenth Amendment limitations on the scope of the treaty power, and I then discuss the rejection of any meaningful *423 subject matter limitation on the treaty power. As I explain, these two components of the nationalist view have developed largely in isolation, and there has been little attempt to consider the implications for American federalism of combining them.

A. Rejection of Tenth Amendment Limitations

As discussed above, the existence of states'‑rights limitations on the treaty power was assumed at various times during the nineteenth century. [FN182] This assumption was reflected not only in Supreme Court decisions, but also in the statements and actions of the political branches. [FN183] The states'‑ rights issue then became the subject of substantial academic debate during the early part of the twentieth century, with leading commentators on both sides. [FN184]

The conventional wisdom is that the death knell to the states'‑rights view came in 1920, with the Supreme Court's decision in Missouri v. Holland. [FN185] The Holland case, "perhaps the most famous and most discussed case in the constitutional law of foreign affairs," [FN186] involved the constitutionality of the Migratory Bird Treaty Act of 1918. [FN187] This Act implemented a 1916 treaty between the United States and Great Britain that was designed to protect birds migrating in the United States and Canada. [FN188] Among other things, the Act made it unlawful to hunt or capture any migratory birds covered by the terms of the Convention. The State of Missouri argued that the Act unconstitutionally interfered with states' rights in violation of the Tenth Amendment. Missouri pointed out that two district courts already had held a similar statute, which had been enacted before the treaty, to be beyond Congress's commerce powers. [FN189]

*424 In rejecting Missouri's argument, the Court, in an opinion by Justice Holmes, acknowledged that the Act might be outside the scope of the Commerce Clause. [FN190] The Court stated, however, that any limits on the treaty power "must be ascertained in a different way" from limits on domestic powers. [FN191] The Court pointed out that the treaty in question did not "contravene any prohibitory words to be found in the Constitution" but rather was alleged to violate "some invisible radiation from the general terms of the Tenth Amendment." [FN192] To show a violation of states' rights here, the Court said, "it is not enough to refer to the Tenth Amendment . . . because by Article II, § 2, the power to make treaties is delegated expressly, and by Article VI treaties made under the authority of the United States . . . are declared the supreme law of the land." [FN193] In concluding that there was no violation of states' rights, the Court observed that the treaty concerned "a national interest of very nearly the first magnitude" that could be protected "only by national action in concert with that of another power." [FN194]

There was some concern after this decision that the treaty power might not be subject to any constitutional restraints, including the individual rights provisions of the Bill of Rights. [FN195] This concern was due in part to an observation by the Court, based on its reading of the Supremacy Clause, that "[a]cts of Congress are the supreme law of the land only when made in pursuance of the Constitution, while treaties are declared to be so when made under the authority of the United States." [FN196] As one commentator noted after the decision, the Court's "hint that there may be no other test to be applied than whether the treaty has been duly concluded indicates*425 that the court might hold that specific constitutional limitations in favor of individual liberty and property are not applicable to deprivations wrought by treaties." [FN197]

This concern was partially alleviated by the Supreme Court in a 1957 decision, Reid v. Covert. [FN198] In the consolidated cases at issue, [FN199] military courts had convicted two wives of U.S. servicemen of murdering their husbands on foreign bases. Executive agreements between the United States and the host countries permitted U.S. military courts to exercise jurisdiction over the offenses. Nevertheless, the wives argued that their trials were unconstitutional because they were conducted without a grand jury indictment and a trial by jury. The Court agreed, with a plurality of the Court stating that "no agreement with a foreign nation can confer power on the Congress, or on any other branch of Government, which is free from the restraints of the Constitution." [FN200] The plurality distinguished Holland, noting that the treaty in Holland "was not inconsistent with any specific provision of the Constitution." [FN201] There had been no Tenth Amendment problem in Holland, explained the plurality, because "[t]o the extent that the United States can validly make treaties, the people and the States have delegated their power to the National Government and the Tenth Amendment is no barrier." [FN202]

In distinguishing Holland, the plurality in Reid may have broadened it. The plurality read Holland as standing for the proposition that a valid treaty is not subject to any Tenth Amendment limitations. It was possible to read Holland, however, as holding simply that there are some actions that the federal government can take pursuant to the treaty power that it cannot take pursuant to the *426 commerce power. In other words, the treaty power, far from conferring on the federal government unlimited power vis‑a‑vis the states, confers a different (and in some cases broader) power than that conferred pursuant to the Commerce Clause. [FN203] The Court in Holland did, after all, note that "[w]e do not mean to imply that there are no qualifications to the treaty‑making power," [FN204] and it also emphasized the national interest at stake in the treaty in question. [FN205] Moreover, the Court stated that "[w]e must consider what this country has become in deciding what [the Tenth] Amendment has reserved," [FN206] perhaps suggesting that the Tenth Amendment's restrictions had changed over time but were not irrelevant. [FN207] Nevertheless, the construction of Holland by the Reid plurality is the one that has generally been accepted by lower courts [FN208] and commentators, [FN209] and the Supreme Court has yet to revisit the issue.

During the 1950s, there was a vigorous effort, led by Senator Bricker of Ohio and Frank Holman of the American Bar Association, to overturn Holland by means of a constitutional amendment. [FN210] There were numerous versions of the "Bricker *427 Amendment," one of which provided that "[a] treaty shall become effective as internal law in the United States only through legislation which would be valid in the absence of a treaty." [FN211] The proposed amendment never received the approval of the Senate, although one version came close. [FN212] During the time this amendment was being considered, there was, once again, substantial academic discussion of the proper scope of the treaty power. [FN213]

The Bricker Amendment controversy died out for several reasons. First, the Reid decision, discussed above, addressed the concern expressed by some proponents of the amendment that the treaty power might not be subject even to the individual rights protections of the Constitution. [FN214] Second, during this period, the Supreme Court was increasingly recognizing expansive domestic federal power, including the power to regulate extensively in the human rights area. [FN215] Third, the treaty makers indicated that they would exercise substantial self‑restraint with respect to overriding state prerogatives. Thus, for example, the Senate attached a reservation in 1951 to the U.S. ratification of the Charter of the Organization of American States stating that

none of its provisions shall be considered as enlarging the powers of the Federal Government of the United States or limiting the powers of the several states of the Federal Union with respect to any matters recognized under the Constitution as being within the reserved powers of the several states. [FN216]

A similar example is the Senate's statement in connection with its consent in 1961 to the Convention on the Organization for Economic Cooperation and Development that "nothing in the convention*428 . . . confers any power on the Congress to take action in fields previously beyond the authority of Congress." [FN217]

The approach of the federal government to human rights treaties has been even more restrictive. The Eisenhower administration, through Secretary of State John Foster Dulles, announced in 1953 that it had no intention of becoming a party to the then‑proposed human rights treaties. [FN218] And for many years thereafter, the Senate refused to consent to any of the major human rights treaties submitted to it. [FN219] The Senate recently has begun to ratify some of these treaties, but only subject to a now‑standard set of reservations, understandings, and declarations (RUDs) that limit the treaties' effect on domestic law. Among other things, these RUDs typically include a declaration that the treaty is non‑self‑executing, as well as a statement that the United States understands that the treaty shall be implemented by the federal government only to theextent that it possesses legislative and judicial power over thematters in question, and otherwise by the state and local governments. [FN220] These RUDs reflect "a desire not to effectuate*429 changes to domestic law by means of the treaty‑making power." [FN221]

A final reason that the federalism controversy died out was that many officials and commentators assumed that there was at least a subject matter limitation on the treaty power, which reduced concerns about its use to regulate internal matters. [FN222] As discussed below, this limitation too has now been rejected, at least in the academic community.

B. Rejection of Subject Matter Limitation

The second component of the nationalist view of the treaty power‑‑that there is no meaningful subject matter limitation on that power‑‑has become well accepted only recently. As discussed above, a subject matter limitation appears to have been assumed both during the Founding and at times during the nineteenth century. [FN223] The Court in Holland also may have assumed the existence of a subject matter limitation. [FN224] Other decisions from the early twentieth century further reflected a subject matter limitation (albeit a broad one), [FN225] as did some of the academic commentary. [FN226]

The existence of a subject matter limitation on the treaty power was subsequently suggested in 1929 by Charles Evans Hughes. [FN227] In a speech to the American Society of International Law, [FN228] Hughes stated that the treaty power might be limited to matters of "international concern" and thus not allow for the regulation of matters "which normally and appropriately were within the local *430 jurisdiction of the States." [FN229] These remarks did not provoke substantial controversy at the time. Rather, as Professor Henkin notes, Hughes's remarks were "accepted as authority." [FN230]

Indeed, opponents of the Bricker Amendment during the 1950s invoked the subject matter limitation in support of their argument that the Amendment was unnecessary. Thus, for example, an American Bar Association committee that opposed the Amendment argued that it was unnecessary in part because of "the very fundamental limitation upon the extent to which a treaty can affect internal law in that it must concern a matter which is 'properly the subject of negotiation with a foreign country."' [FN231] Similarly, Secretary of State Dulles stated during the Bricker hearings that a treaty could not regulate matters "which do not essentially affect the actions of nations in relation to international affairs, but are purely internal." [FN232]

The U.S. Court of Appeals for the D.C. Circuit came close to formally endorsing the Hughes view in 1957. In Power Authority of New York v. Federal Power Commission, [FN233] the court considered the effect of a reservation attached by the Senate to its ratification of a treaty with Canada concerning use of the waters of the Niagara River. The reservation stated, among other things, that "no project for redevelopment of the United States share of such waters shall be undertaken until it be specifically authorized by Act of Congress." [FN234] The court observed that enforcement of this reservation would raise a "constitutional concern" [FN235] because the reservation "relate[ d] to a matter of purely domestic concern." [FN236] The court avoided ruling on the constitutional question [FN237] by construing the reservation as merely "an expression of the Senate's desires and not *431 a part of the treaty." [FN238] The court distinguished Holland as involving a treaty "related to a 'national interest of very nearly the first magnitude' which 'can be protected only by national action in concert with that of another power.' " [FN239]

Although the position of the State Department on this issue varied to some extent during this period, it too appeared to agree that there was a subject matter limitation, at least as a matter of policy. Thus, for example, it stated in 1955 that "[t]reaties are not to be used as a device for the purpose of effecting internal social changes or to try to circumvent the constitutional procedures established in relation to what are essentially matters of domestic concern." [FN240] A number of commentators in this period also assumed that there was a subject matter limitation on the treaty power. [FN241]

The notion of a subject matter restriction on the treaty power perhaps reached its highest level of acceptance in 1965, when the Restatement (Second) of the Foreign Relations Law of the United States expressly endorsed it. In its black‑letter‑law provisions, the Restatement (Second) provided that the treaty power is limited to matters "of international concern." [FN242] The Restatement (Second) further stated, in its comments, that international agreements "must relate to the external concerns of the nation as distinguished from matters of a purely internal nature." [FN243] In support of this limitation, the Restatement (Second) noted, among other things, that "[n]o power granted to the United States by the Constitution is unlimited." [FN244] It also linked the purported subject matter limitation with the lack of a Tenth Amendment limitation on the treaty power, *432 noting that a treaty is constitutional, notwithstanding the reserved powers of the states, so long as "it deals with matters of international concern and is not in conflict with any express limitations on the powers of the government." [FN245] The existence of a subject matter limitation on the treaty power was further supported in 1967 by the American Bar Association. [FN246]

After the publication of the Restatement (Second), however, such a limitation also became the subject of academic criticism, most notably from Professor Henkin. Henkin criticized this idea in an article in the late 1960s, [FN247] as well as in his famous 1972 book on foreign affairs law. [FN248] Henkin argued that "every treaty, regardless of subject, serves the external purposes of the United States," and he questioned how one could in any event distinguish "external" from "internal" matters. [FN249]

The idea of a subject matter limitation was then expressly rejected in the Restatement (Third) of Foreign Relations Law, which was first published in draft form in 1980 and subsequently published in final form in 1987. The Restatement (Third), for which Professor Henkin was the Chief Reporter, declares that, "[c]ontrary to what was once suggested, the Constitution does not require that an international agreement deal only with 'matters of international concern.' " [FN250] The Restatement (Third) does not cite any intervening authority for this proposition, [FN251] and, indeed, the one decision on point between the Restatement (Second) and the Restatement (Third)‑‑the Power Authority decision‑‑had come out essentially the other way. [FN252]

In what has become a rather disturbing phenomenon in the development of American foreign affairs law, the new Restatement (Third) position, adopted without authority, is now being treated as *433 if it were black‑letter law. In his recent revision of his foreign affairs book, for example, Professor Henkin cites the Restatement (Third) for the proposition that the subject matter limitation has "now been authoritatively abandoned." [FN253] Unfortunately, this is not the only instance in which the Restatement (Third) has been used in this self‑fulfilling manner. [FN254] In any event, the Restatement (Third)'s rejection of a subject matter limitation on the treaty power now appears to be the accepted view, at least among academic commentators. [FN255] Importantly, no one seems to have considered the implications of combining the lack of Tenth Amendment limitations with the lack of a subject matter limitation.

IV. Critique of the Nationalist View

As we have seen, there are two components to the nationalist view of the treaty power‑‑the lack of Tenth Amendment limitations, and the lack of any meaningful subject matter limitation. When combined, these components give the treaty‑makers essentially unlimited power vis‑a‑vis the states. In our federalist system, which is premised on the principle of limited and enumerated powers, this result requires justification. In this Part, I examine the three principal justifications for the nationalist view and I find that, especially in light of recent developments, none of them is convincing.

*434 A. Delegation Argument

The purported immunity of the treaty power from federalism limitations is often premised on what I will call the delegation argument. This argument is that, because the treaty power has been delegated to the federal government, exercises of that power do not implicate the Tenth Amendment's reservation of powers to the states. As Professor Henkin states, "[s]ince the Treaty Power was delegated to the federal government, whatever is within its scope is not reserved to the states: the Tenth Amendment is not material." [FN256] This reasoning is reflected in both Holland and Reid. The Court in Holland stated that "it is not enough to refer to the Tenth Amendment, reserving the powers not delegated to the United States, because by Article II, § 2, the power to make treaties is delegated expressly." [FN257] Similarly, the plurality in Reid noted that, "[t]o the extent that the United States can validly make treaties, the people and the States have delegated their power to the National Government and the Tenth Amendment is no barrier." [FN258]

The delegation argument has superficial appeal. The treaty power, after all, is expressly delegated to the national government, and the federal government's ratification of treaties therefore would seem to fall within the principle of limited and enumerated powers. Moreover, the language of the Tenth Amendment only purports to reserve to the states "[t]he powers not delegated to the United States by the Constitution." [FN259] Relying on these and other points, Professor Henkin goes so far as to describe the delegation argument as "clear and indisputable." [FN260]

The fundamental flaw with the delegation argument, however, is that it fails to provide any reason for giving special Tenth Amendment immunity to the treaty power. Although it is true that the treaty power has been delegated to the federal government, this is true of all federal powers. The power to regulate interstate commerce, for example, also has been delegated to the federal government, but this, by itself, has not made it immune from Tenth Amendment scrutiny. And, indeed, notwithstanding this delegation, the Supreme Court has held in a number of cases, including *435 cases decided near the time of Holland, that the Tenth Amendment operates as a substantive restraint. [FN261] In sum, the fact of delegation says nothing about the scope of the delegation, and it certainly does not establish that the delegation is unlimited in scope. [FN262]

Of course, the willingness of the Supreme Court to enforce the Tenth Amendment as an independent limitation on any federal power has varied. Although the Court has at times treated the Tenth Amendment as an independent restraint on delegated powers, [FN263] at other times it has described the Tenth Amendment as stating "but a truism that all is retained which has not been surrendered." [FN264] Under this analysis, states' rights are protected by policing the scope of the delegated powers, not by giving immunity to reserved state powers per se. The Court's current position on this issue is not entirely clear, although the Court appears to be treating the Tenth Amendment as an independent restraint. [FN265] In any event, the key point here is that whatever limitations the Tenth Amendment does impose, it is not evident from the delegation argument why these limitations should not apply to the treaty power.

Some proponents of this aspect of the nationalist view have pointed out that, unlike its treatment of certain other delegated powers, the Constitution expressly denies states the power to enter *436 into treaties. [FN266] They argue that, even if some delegated powers are subject to Tenth Amendment limitations, this should not be the case for delegated powers that, like the treaty power, are held exclusively by the federal government. This argument, however, is a non sequitur. While it is true that the states have not reserved the power to enter into treaties, this does not mean that they have not reserved other regulatory powers that might be infringed by certain exercises of the federal treaty power. In other words, the Constitution's denial of state power to enter into treaties "proves that the federal power to make treaties is exclusive, but it does not prove that it is unlimited, or that [it] is not limited by the tenth amendment." [FN267] The exclusivity argument breaks down even further when it is remembered that the Constitution allows the states some ability to enter into international agreements, [FN268] and that many international agreements made by the federal government today take the form of executive agreements rather than Article II treaties. [FN269]

*437 The problematic nature of the delegation argument becomes further evident when its logic is applied in the context of separation of powers rather than federalism. [FN270] One could argue that, like federalism restrictions, separation of powers restrictions do not apply to powers delegated exclusively to a branch of the federal government. This argument was accepted to some extent by the Supreme Court in a much‑criticized decision, United States v. Curtiss‑Wright Export Corp. [FN271] In that case, Congress by Joint Resolution purported to confer on the President the authority to implement a criminal prohibition on the sale of arms in the United States to countries engaged in a conflict in Latin America. In response to the argument that this constituted an unlawful delegation of legislative authority to the President, a plausible argument at that time in light of the Court's precedent, [FN272] the Court sharply distinguished between federal power over external affairs and federal power over internal affairs.

In particular, the Court observed that "[t]he broad statement that the federal government can exercise no powers except those specifically enumerated in the Constitution, and such implied powers as are necessary and proper to carry into effect the enumerated powers, is categorically true only in respect of our internal affairs." [FN273] The Court explained that, whereas the states had delegated certain limited powers over internal affairs to the federal government, this was not the case with respect to powers relating to external affairs: "Since the states severally never possessed international *438 powers, such powers could not have been carved from the mass of state powers but obviously were transmitted to the United States from some other source." [FN274] One such power is the "very delicate, plenary and exclusive power of the President as the sole organ of the federal government in the field of international relations." [FN275] The existence of this power, reasoned the Court, meant that the usual separation of powers limitations on delegations of authority to the executive branch do not apply with respect to foreign affairs matters. [FN276]

The reasoning in Curtiss‑Wright is to some extent the separation of powers analogue to Holland. Just as the federalism restrictions operated differently with respect to external matters in Holland, so did the separation of powers restrictions in Curtiss‑Wright. Moreover, in both decisions, the Court distinguished between internal matters, over which the states retained some sovereign authority, and external matters, over which the states did not. In assessing the validity of the delegation argument, it is therefore important to note that the reasoning of Curtiss‑Wright has received "withering criticism" from commentators. [FN277] Among other things, commentators have criticized the reasoning for "carv[ing] a broad exception to the historic conception, often reiterated, never questioned, and explicitly reaffirmed in the Tenth Amendment, that the federal government is one of enumerated powers only." [FN278]

Moreover, the Supreme Court arguably has repudiated the logic of Curtiss‑Wright. In Youngstown Sheet & Tube Co. v. Sawyer, [FN279] the Court found a presidential order authorizing a seizure of the nation's steel mills to be invalid, notwithstanding the President's claim that the seizure was necessary for national defense given the ongoing conflict in Korea. The Court stated that "[t]he President's power, if any, to issue the order must stem either from an act of Congress or from the Constitution itself." [FN280] In his famous concurrence in that case, Justice Jackson observed that much of the *439 Court's opinion in Curtiss‑Wright had been "dictum," [FN281] and he reiterated the traditional view that "the executive branch, like the Federal Government as a whole, possesses only delegated powers." [FN282] In its subsequent decision in Dames & Moore v. Regan, [FN283] a case involving a challenge to an executive agreement, the Court largely endorsed the Youngstown approach rather than the Curtiss‑Wright approach. [FN284]

Given the criticism and erosion of the separation of powers reasoning in Curtiss‑Wright, it is perhaps not surprising that proponents of the nationalist view do not argue that the treaty power is immune from implied limitations arising from separation of powers. Professor Henkin, for example, states that "[t]he Treaty Power . . . is not limited by the powers of Congress, but it is assumed to be subject to other radiations from the separation of powers." [FN285] Indeed, notwithstanding Curtiss‑Wright, Henkin notes specifically that "[i]t has been assumed that constitutional limitations on delegation of legislative power apply as well to delegation by treaty." [FN286] These proponents fail to explain, however, why the implied limitations of separation of powers should govern when the implied limitations of federalism do not. [FN287] The delegation argument does not provide such an explanation.

*440 B. Political Process Argument

In addition to making the delegation argument, proponents of the nationalist view typically advance the political process argument. This argument is that the political process provides sufficient protection of federalism, such that no substantive federalism restriction is necessary. Proponents of this view observe, for example, that two‑thirds of the Senate must consent to treaties, and that the states have equal representation in the Senate. [FN288] Although some proponents of this view correctly point out that the Founders placed significant emphasis on the process for concluding treaties as a protection for the states, [FN289] they do not appear to claim that the Founders intended for this process protection to be exclusive, and any such claim likely would be unpersuasive given the Founding materials discussed above. Instead, they appear to make more of a stare decisis argument: there is no need to overturn the settled decision in Holland because states' rights are not in fact threatened by the treaty power.

This sort of argument is not, of course, unique to the treaty context. A similar political process argument was made by commentators, and eventually accepted by the Supreme Court, with respect to federal legislative power in general. The theory was first advanced in the 1950s by Herbert Wechsler, [FN290] and then later more thoroughly developed by Jesse Choper. [FN291] Citing these writings, a 5‑4 majority of the Supreme Court endorsed the political process theory in its 1985 Garcia decision. [FN292] The Court concluded that "the fundamental limitation that the constitutional scheme imposes on the Commerce Clause to protect the 'States as States' is one of process *441 rather than one of result," and the Court said that "[a]ny substantive restraint on the exercise of Commerce Clause powers must find its justification in the procedural nature of this basic limitation, and it must be tailored to compensate for possible failings in the national political process rather than to dictate a 'sacred province of state autonomy.' " [FN293]

Despite the Court's general endorsement of this theory in Garcia, there are several problems with relying on this argument today to support the nationalist view of the treaty power. First, the Supreme Court appears to have backed away from Garcia in recent years. Starting with its "clear statement" requirement in Gregory v. Ashcroft, [FN294] through its latest "commandeering" decision, Printz v. United States, [FN295] the Court has steadily chipped away at the proposition that federalism protections are to be left solely to the political process. As Professor John Yoo recently explained, "[i]n these cases, the Court has articulated its intention to establish areas of state control that are to remain immune from federal regulation, and it has suggested that these areas can be identified by policing Congress' use of its enumerated powers." [FN296] The political process theory also has been the subject of substantial academic criticism in recent years. [FN297] A number of these critics have examined the mechanics of our political process and have concluded that "[t]he structural protections identified by Wechsler, Choper, and company are marginal at best." [FN298] Professor Steven Calabresi has pointed out, for example, that the campaign finance system, with its national special interest groups and PACs, undercuts the representation *442 of state interests, and that the incentive structure for federal politicians favors the expansion of federal power rather than the protection of states' rights. [FN299]

Second, the political process argument can draw little, if any, support from the original intent of the Founders. Although the Founders did intend for the process of making treaties‑‑especially the two‑thirds Senate advice and consent requirement‑‑to protect states' rights, [FN300] there is no evidence that they intended this to be the only protection. [FN301] More importantly, several key assumptions of the Founders concerning the treaty process no longer hold true. The Founders envisioned direct state representation in the Senate, something that "basically evaporated with the adoption of the Seventeenth Amendment in 1913." [FN302] The Founders also believed that the Senate would play a major role in advising the President regarding treaties, but its role fairly quickly became one merely of consent. [FN303] An even more significant change is that the vast majority of international agreements concluded by the United States today are executive agreements rather than Article II treaties and thus do not depend on two‑thirds Senate consent at all. [FN304] Nevertheless, these agreements are widely considered to have equal status with Article II treaties, at least if they are approved in advance or after the fact by a majority of both houses of Congress. [FN305]

Third, it is arguable that the political process rationale is less persuasive in the treaty and executive agreement context than in the context of domestic legislation. As a number of commentators have pointed out, the treaty and executive agreement process is more opaque and less representative than the normal federal legislative process. [FN306] Negotiations are typically conducted by the Executive, *443 often without much public disclosure. And, as Professor Friedman has noted, "[t]here is not much in the political process of electing a President that suggests any particular sensitivity to state concerns." [FN307] Moreover, the Senate in the case of treaties, and the entire Congress in the case of congressional‑executive agreements, is often confronted with what is essentially an up‑or‑down vote. Unlike much domestic legislation, Congress has essentially no role in drafting treaties and it has relatively little ability to modify their terms. This has been particularly true in the context of so‑called "fast‑track" legislation. [FN308] Moreover, the costs of a down vote may involve not only a failure to commit to the agreement in question, but also an inability to participate in the international organization associated with the agreement. In these situations, the bargaining power of the Senate or Congress is likely to be fairly low, and it may feel substantial pressure to approve the commitments already made by the President. [FN309] In addition, treaties, especially multilateral treaties, may be more likely than domestic legislation to contain vague and aspirational language, making their effect on state prerogatives harder to anticipate during the ratification process. [FN310] Finally, the construction and implementation of many treaties are delegated to international bodies that may lack domestic, let alone state, accountability. [FN311]

Notwithstanding these objections to the political process argument, it must be acknowledged that the Senate often has acted to protect states' rights in the treaty context, especially with respect to *444 human rights treaties. [FN312] Human rights agreements have, to date, largely been concluded by means of treaties rather than executive agreements. The Senate has been slow to consent to these treaties, in part because of federalism concerns. And when it has consented, the Senate has sought to protect state interests by declaring the treaties to be non‑self‑executing and by attaching federalism clauses. [FN313] As Professor Peter Spiro has noted, there has been "a consistent refusal [of the U.S. treaty makers] to displace state law with international human rights treaty obligations." [FN314]

Of course, that the Senate has acted in a certain way is no guarantee that it will continue to do so as its membership and relationship with the President change. [FN315] Moreover, as discussed above, human rights treaties are not the only international agreements that raise federalism concerns, and it is far from clear that the Senate's practices with regard to human rights treaties will be carried over into other contexts. Indeed, if the GATT and NAFTA agreements are any indication, it cannot be assumed that the President will even use the Article II treaty process for these other agreements. [FN316] It should also be noted that proponents of the nationalist view have vigorously challenged the Senate's limitations on human rights treaties, so they may be in no position to rely on these limitations to support their political process argument. [FN317]

More importantly, the limitations imposed by the Senate to date on human rights treaties do not prevent a majority of Congress from relying on the treaties, in conjunction with the Necessary and Proper Clause, as a source of lawmaking power. Indeed, this is exactly *445 what Congress did in Holland. Thus, as noted above, a number of commentators have suggested recently that Congress rely on already‑ratified treaties as the basis for overcoming limitations on its domestic lawmaking power. [FN318] This is the process, for example, recommended by Professor Neuman for reenactment of the RFRA statute. [FN319] In this context, there is no super‑majority protection in the Senate, so we are left with whatever political process checks exist in the majority of Congress‑‑checks viewed by the current Supreme Court and many commentators as insufficient by themselves to protect the states.

C. One‑Voice Argument

A third argument against imposing federalism limitations on the treaty power is that it would hamper the ability of the United States to speak with one voice in international negotiations. According to this one‑voice argument, if the treaty power were limited by states' rights, the Executive's ability to negotiate and conclude vital international agreements would be severely compromised. As Professor Neuman has argued, "[r]equiring the unanimous agreement of . . . all the states for ratification of any treaty that includes a provision addressing 'local' concerns would greatly hamper American participation in international treaty regimes." [FN320]

This one‑voice argument has strong intuitive appeal. Foreign affairs, after all, concern the entire nation. Moreover, effective international bargaining may well require that we have a national representative with the power to make binding commitments. Consistent with this understanding, the Founders of the Constitution envisioned, in the words of James Madison, that "[i]f we are to be one nation in any respect, it clearly ought to be in respect to other nations." [FN321] In addition, the Supreme Court has emphasized in a number of cases the need for exclusive federal control over foreign relations. [FN322]

*446 It is likely, however, that the one‑voice metaphor has never been very accurate. At the federal level, it is common for all three branches of the government to be involved in foreign affairs matters, preventing any single federal voice. As Professor Goldsmith has observed, "[f]oreign relations law is replete with struggles between the statute‑makers, the treaty‑makers, the President, and sometimes the courts, for control of the federal foreign relations voice." [FN323] In addition, states are involved in a host of regulatory activities that sometimes affect foreign affairs, such as criminal prosecution of aliens, taxation of multinational corporations, and adjudication of cases involving foreign parties or transactions. [FN324]

In any event, the intuitive appeal of the one‑voice argument varies dramatically depending on the type of treaty‑making involved. It is strongest with respect to traditional treaty‑making. Most of us would agree, for example, that the Executive should not be hampered by federalism concerns when negotiating a peace treaty. But, as we have seen, the nature of treaty‑making has changed. It now purports to regulate many subjects formerly considered domestic in nature, especially in the human rights area. With respect to that sort of treaty‑making, the one‑voice argument loses much of its intuitive appeal. It is not at all obvious, for example, that it is necessary or desirable that the country speak through the Executive with respect to the regulation of religious freedom.

Whatever federalism limitations are imposed on the treaty power, and I discuss some of the options below, they are likely to be relevant almost entirely with respect to the new types of treaty‑making, where the one‑voice argument has less appeal. This might not always have been the case. When states' rights were viewed more expansively, and federal power viewed more narrowly, it is possible that federalism limitations, if indiscriminately applied, would have interfered with the nation's ability to speak with one voice in international relations. Interestingly, during that same period of time, there was nevertheless support for limitations on the treaty power. [FN325] Today, with federalism at best a weak restraint on federal power, any threat to flexibility posed by subjecting the treaty power to federalism limitations is substantially lower.

*447 Moreover, as a political matter, treaty‑making today is already subject to federalism limitations on an ad hoc basis. Thus, for example, the Senate routinely attaches federalism clauses to human rights treaties. [FN326] These clauses have not hampered this country's ability to conclude these treaties. Indeed, it may not have been politically possible to conclude these treaties without such clauses. [FN327] Rejecting the nationalist view would simply give the force of law to this practice without substantially interfering with the nation's ability to conclude treaties.

The one‑voice argument also has less doctrinal support than first meets the eye. As noted above, the Court has pulled back from the broad reasoning of decisions like Curtiss‑Wright. [FN328] Moreover, the Court's one‑voice statements have always been broader than the Court's actual decisions, which have not in fact allowed the federal government unfettered power in foreign affairs. [FN329] Perhaps most significantly, the Court recently sounded a formal retreat from a strong one‑voice position. In Barclays Bank PLC v. Franchise Tax Board, [FN330] the Court considered a challenge, under the dormant foreign commerce clause, to the constitutionality of California's "worldwide combined reporting" method for taxing multinational corporations. In prior decisions, the Court had held that state tax laws would be subject to dormant preemption if they interfered with the federal government's ability to "speak with one voice when regulating commercial relations with foreign governments.' " [FN331] Relying on that proposition, the petitioners in Barclays Bank pointed out that the California law had provoked enormous diplomatic controversy with the United States' closest trading partners and had been opposed in numerous executive pronouncements. [FN332] The Court nevertheless rejected their challenge to the law, in large part because Congress had shown a willingness to tolerate state taxation methods like the one at issue. [FN333] Importantly, the Court emphasized that the usual method in our constitutional *448 system for overcoming state disuniformity is for Congress to enact a statute within the scope of its Article I powers, and the Court indicated that it was unwilling to deviate from that approach just because foreign affairs were involved. [FN334] As other commentators have noted, the Court in Barclays Bank largely repudiated the strong "one voice" doctrine suggested in some of its earlier foreign commerce clause decisions. [FN335]

A variation of this one‑voice argument would be to claim that the scope of the treaty power is a nonjusticiable political question. Under the political question doctrine, courts will not review issues where there is a "textually demonstrable constitutional commitment of the issue to a coordinate political department; or a lack of judicially discoverable and manageable standards for resolving it." [FN336] While there are indeed broad statements by the Supreme Court suggesting that foreign affairs is a subject largely outside of the scope of judicial review, [FN337] the key problem here, once again, is that the Court's statements are premised on the existence of a bright line between foreign and domestic affairs. Once this line becomes blurred, as it has in the treaty‑making area, the justifications for judicial abstention diminish. Even if the courts lack competence and constitutional authority with respect to truly inter‑national relations, this may not be the case with respect to international agreements *449 that regulate the internal relationship between governments and their citizens.

In any event, the Court has made clear that "it is error to suppose that every case or controversy which touches foreign relations lies beyond judicial cognizance." [FN338] Rather, the political question doctrine shields from judicial review only "those controversies which revolve around policy choices and value determinations constitutionally committed for resolution to the halls of Congress or the confines of the Executive Branch." [FN339] States' rights, according to the current Supreme Court, are very much within the scope of proper judicial review. [FN340] The Court also has specifically observed that the construction of treaties and executive agreements is a proper subject for judicial review, even if such construction might have consequences for foreign relations. [FN341] While a plurality of the Court did conclude in Goldwater v. Carter [FN342] that a challenge by members of Congress to the President's unilateral termination of a treaty presented a political question, it did so because the challenge involved the respective powers of coordinate federal branches of government. [FN343] The scope of the treaty power, by contrast, involves the powers of the federal government as a whole vis‑a‑vis the constituent state governments. As the Court has noted, "it is the relationship between the judiciary and the coordinate branches of the Federal Government, and not the federal judiciary's relationship to the States, which gives rise to the 'political question.' " [FN344]

For all of these reasons, neither the one‑voice argument nor the related political question doctrine justifies giving the treaty power immunity from federalism restrictions. This is particularly true with respect to the new forms of treaty‑making that address issues that in the past largely have been regulated by states. As for those treaties, the one‑voice argument simply begs the question: the argument *450 is premised on the need to protect federal prerogatives in foreign affairs, yet whether the new forms of treaty‑making are in fact federal prerogatives is exactly what is at issue.

V. Protecting American Federalism

As discussed above, the nationalist view of the treaty power is unsupported by history, and its principal legal and policy justifications are questionable. Consequently, to the extent that American federalism continues to be worth protecting, there does not appear to be any good justification for exempting the treaty power from this protection. Indeed, at least some of the reasons for protecting federalism, such as the desirability of having a check against the centralization of power, would seem to be especially relevant to the treaty power, which entails the centralization of power even further away from the average U.S. citizen than does domestic legislation. [FN345]

The question remains, however, how federalism should be protected. As we have seen, the political process does not appear by itself to offer sufficient protection. Because there are two components of the nationalist view of the treaty power, there are, at least in theory, two additional avenues for protecting federalism. The treaty power could be subjected to some sort of a subject matter test, or it could be subjected to structural federalism limitations. In this Part, I consider both possibilities.

As I explain, the answer to this question may be different today than it would have been in the past. Historically, it often was assumed that the treaty power was appropriate only for certain externally oriented subjects. As the distinction between foreign and domestic affairs wanes, however, it becomes increasingly difficult to maintain any meaningful subject matter limitation. As a result, I argue that it makes more sense today to focus on structural federalism limitations on the treaty power, and, for a variety of reasons, I argue that the treaty power should be subject to the same federalism limitations that apply to Congress's legislative powers. In other words, my argument is that the federal government should not be able to use the treaty power (or executive agreement power) to create domestic law that could not be created by Congress. To the extent that this conclusion would require overruling Holland, I *451 argue that the justifications for stare decisis are weak in this context.

A. Reviving a Subject Matter Limitation

One option for protecting federalism in this context would be to revive a subject matter limitation on the treaty power. After all, such a limitation has substantial historical support and was rejected by commentators only recently. Moreover, this limitation has never been squarely rejected by the Supreme Court and thus presumably could be revived without revisiting settled Supreme Court precedent. Indeed, Holland itself arguably assumed that there was such a limitation on the treaty power. [FN346]

There are two possibilities here. The first would be to limit the treaty power to matters usually the subject of treaties at the time the Constitution was adopted. This was one of the limitations suggested by Jefferson. [FN347] This suggested limitation has never received much acceptance, however. [FN348] Moreover, as a policy matter, this limitation would seem to be highly undesirable. Not only would it exclude U.S. participation in human rights treaties, it also would presumably exclude U.S. participation in treaties relating to the environment, terrorism, and private international law, to name a few subjects. Moreover, it is doubtful that this limitation could be justified even under a strict originalist interpretation of the Constitution. There is no evidence that the Founders intended the treaty power to be frozen to 1780s issues, and the inflexibility such a limitation would impose on the national government makes it highly unlikely that the Founders would have had such an intent. And, of course, other constitutional powers‑‑such as the commerce power‑‑have not been limited in this fashion. [FN349]

The other possibility would be to limit the treaty power to matters that are truly "international" in nature. This is the approach suggested, for example, by Charles Evans Hughes and the Restatement (Second). [FN350] Although this approach may deserve further exploration, it suffers from a substantial problem: Today, almost any *452 issue can plausibly be labeled "international." Given the globalization of trade, technology, and travel, among other things, nations are indisputably connected to each other in a variety of ways. As a result, "domestic" actions by nations are often matters of concern to the international community. As Professor Tribe has observed, "[w]ith global interdependence reaching across an ever broadening spectrum of issues," any requirement that the treaty power be restricted to matters of international concern "seems unlikely to prove a serious limitation." [FN351]

This may be true even in the area of human rights. Human rights is, of course, a matter today of international negotiation and agreement. Unlike some proponents of the nationalist view, [FN352] I do not find that fact alone dispositive of the question whether this subject falls within the scope of the federal government's treaty power. If it were dispositive, it would mean that the federal government's power in this regard would be determined entirely by the international community rather than by domestic‑law standards, something at odds with this country's "dualist" approach to international law. [FN353] Nevertheless, it is difficult to dispute that, in this day and age, how one nation treats its citizens has international effects. [FN354] These effects may be direct and physical‑‑for example, an influx of refugees, or instability in the region. Or they may be more abstract and emotional‑‑for example, a sense of moral outrage, or an empathetic loss. Anyone who has observed recent events in Tiananmen Square, Somalia, or Rwanda will find it difficult to deny the existence of such effects.

*453 But perhaps a focus on effects is not the right test. Another test would be a focus on need. The question here might be whether the issue requires international cooperation in order to be addressed. This is, after all, how the Court described the migratory bird treaty at issue in Holland. Under this test, human rights treaties might be suspect, since their implementation involves action by individual governments within their territories rather than cooperative action. [FN355] But this line is fuzzy at best. It is arguable that there is in fact a demonstrable need for cooperative international action to address even "domestic" issues such as human rights. Without reciprocal agreements, along with international monitoring and other enforcement mechanisms, many nations might well continue to engage in human rights abuses. Perhaps treaties are required to obtain results even here. To be sure, the proliferation of human rights treaties has not eliminated human rights abuses in the countries that have ratified them, but it is certainly possible that it has helped improve conditions.

In any event, even if there were a workable distinction in theory between international and domestic matters, it seems unlikely that U.S. courts would feel competent to contradict the political branches on this issue. It is far from clear, for example, what standard the courts could use to draw such a line. Unlike some grants of federal power, the wording of the treaty power does not itself suggest any particular subject restriction. In construing the scope of the Commerce Clause, for example, one might look to the movement of goods or the exchange of money. [FN356] But other than the form of the agreement itself, what is an attribute of a permissible treaty? International law does not appear to offer any help. A treaty is defined in international law simply as an agreement concluded between nation‑states. [FN357] The only subject matter limitation under international law is that treaties cannot violate certain fundamental jus cogens norms, none of which is likely to have much relevance *454 to U.S. treaties. [FN358] Moreover, at least since the New Deal, the Supreme Court has imposed few subject matter restraints even on the federal powers that seem amenable to such a limitation. [FN359]

The difficulty of judicial line‑drawing here is compounded by the changing nature of the subject. What is considered "international" will undoubtedly vary over time, as world conditions and relationships between nations change. As Professor Henkin has explained: "What is of international concern, what affects American foreign relations and is relevant to American foreign policy, what matters the United States wishes to negotiate about, differ from generation to generation, perhaps from year to year, with the ever‑changing character of relations between nations." [FN360] In other areas of foreign affairs law, it is evident that the courts tend to defer to the political branches and, indeed, attempt to shift decisionmaking on such issues to those branches. [FN361] They are particularly likely to do so with respect to the scope of this country's international relations.

A recent decision that illustrates this point is the Second Circuit's decision in United States v. Lue. [FN362] The defendant in that case, a foreign national, was convicted under the Hostage Taking Act, a statute enacted as part of this country's implementation of *455 the International Convention Against the Taking of Hostages. [FN363] The basis for the conviction was the defendant's attempted abduction of a person in New York for the purpose of compelling the victim's relatives to pay ransom. Consistent with the terms of the Convention, the Hostage Taking Act applies even to domestic kidnappings if done "in order to compel a third person or a governmental organization to do or abstain from doing any act," except that the Act does not apply where "each alleged offender and each person seized or detained are nationals of the United States, and each alleged offender is found in the United States, unless the governmental organization sought to be compelled is the Government of the United States." [FN364] Since the defendant in this case was a foreign national, this exception did not apply.

In response to the defendant's argument that the Act's regulation of domestic crimes was beyond the scope of the treaty power, the court stated that "[t]he defendant relies far too heavily on a dichotomy between matters of purely domestic concern and those of international concern, a dichotomy appropriately criticized by commentators in the field." [FN365] While acknowledging that there "must be certain outer limits" to the scope of the treaty power, the court explained that the Act's regulation of hostage taking could easily be classified as a matter of international concern:

Whatever the potential outer limit on the treaty power of the Executive, the Hostage Taking Convention does not transgress it. At the most general level, the Convention addresses‑‑at least in part‑‑the treatment of foreign nationals while they are on local soil, a matter of central concern among nations. More specifically, the Convention addresses a matter of grave concern to the international community: hostage taking as a vehicle for terrorism. [FN366]

Given that the subject matter could plausibly be characterized as one of international concern, the court believed that it was inappropriate for the judiciary to second‑guess the treaty‑makers and thereby "impinge upon the Executive's prerogative in matters pertaining to foreign affairs." [FN367]

As this decision illustrates, the line between what is domestic and what is international is difficult to define, the scope of what can plausibly be labeled international has grown substantially in recent *456 years, and courts as a result are unlikely to restrict the treaty power much, if at all, based on this distinction. If federalism is to be protected in the treaty context, another approach must be found.

B. Parity with Federal Legislation

Another option for protecting federalism, and the option I favor, would be to subject the treaty power to the same federalism restrictions that apply to Congress's legislative powers. Under this approach, the treaty power would not confer any additional regulatory powers on the federal government, just the power to bind the United States on the international plane. Thus, for example, it could not be used to resurrect legislation determined by the Supreme Court to be beyond Congress's legislative powers, such as the legislation at issue in the recent New York, Lopez, Boerne, and Printz decisions. [FN368]

As mentioned above, this approach was endorsed by George Nicholas during the Virginia Ratifying Convention, [FN369] Thomas Jefferson in his Manual on Parliamentary Practice, [FN370] and the Supreme Court in its 1836 decision, New Orleans v. United States. [FN371] It also is essentially the law in Canada, where the treaty power has been construed not to give the national government legislative power over matters reserved to the provinces. [FN372]

In addition to this historical and comparative‑law support, there are several conceptual and doctrinal reasons why it may make sense today to subject treaties to the same federalism limitations as federal statutes. First, unlike traditional treaties that were generally bilateral and addressed the relations between nations, both the form and substance of modern treaty law resembles domestic legislation. As discussed above, treaty law today regulates the relations between nations and their citizens, it covers many of the same subjects as domestic law, and it is even made in a kind of legislative way, through mechanisms such as multilateral drafting conferences. [FN373]

*457 Second, as the Restatement (Second) of Foreign Relations Law recognized, [FN374] the immunity of the treaty power from states'‑rights limitations is premised on the existence of a meaningful distinction between the foreign and the domestic. Yet proponents of the nationalist view themselves, probably correctly, deny that we can continue to make this distinction. Once that is denied, however, there is a much stronger case, based upon the limited and enumerated powers doctrine, for subjecting the treaty power to the same limitations that apply to other federal law.

Third, there is strong doctrinal support for the equal treatment of federal statutes and treaties. Since at least the late 1800s, the Supreme Court has consistently held that treaties and federal statutes have an equal status in the U.S. system, such that, in the case of a conflict, the last in time is the controlling domestic law. [FN375] This well‑settled equality of treaties and federal statutes supports treating them as equal as well for federalism purposes. [FN376] As the Supreme Court has observed, its decisions "generally have regarded treaties as on much the same plane as acts of Congress, and as usually subject to the general limitations in the Constitution." [FN377] Indeed, not treating them the same with respect to federalism limitations presents a doctrinal puzzle: If treaties can be made that go beyond what Congress could do pursuant to its legislative powers, then what happens to Congress's ability to supersede the treaty with subsequent legislation? As William Mikell explained many years ago:

If, however, a treaty were made which affected the reserved rights of the states, it is, to say the least, doubtful if such a treaty could be abrogated at all without the consent of the President, for Congress having no power to pass a law, affecting the reserved rights of the states, could enact no law either in affirmance or derogation of the treaty. [FN378]

*458 Further doctrinal support for imposing on the treaty power the same federalism limitations as those imposed on Congress's Article I powers can be found in cases involving the Eleventh Amendment. This Amendment provides that "[t]he judicial power of the United States shall not be construed to extend to any suit in law or equity, commenced or prosecuted against one of the United States by Citizens of another State, or by Citizens or Subjects of any Foreign State." [FN379] This language refers only to suits by citizens and subjects, but it has been held to apply to suits by foreign nations themselves against U.S. states. [FN380] Moreover, although the language refers to jurisdiction, the Amendment has been read as embodying more general principles of state sovereignty. [FN381] Importantly, the Supreme Court has held that the federal government cannot overcome the Eleventh Amendment (as so defined) by means of Congress's commerce power. [FN382] The question remains, however, whether the federal government can overcome Eleventh Amendment immunity by means of its treaty power. The distinction made in Holland and Reid between federalism limitations on Article I powers and those on the treaty power at least raises the possibility that the treaty power should be treated differently from the commerce power with respect to the Eleventh Amendment. [FN383] Recently, however, the Supreme Court made clear that treaty claims, even when brought by foreign governments, are indeed subject to the Eleventh Amendment. [FN384]

Notwithstanding this doctrinal support, a principal disadvantage of this proposal is that it might require overruling at least some portion of the Holland decision. The Court in Holland was unclear about many things, but one thing it did make clear is that the treaty *459 power is not subject to the same federalism restrictions as Congress's lawmaking powers. [FN385] My proposal's deviation from Holland, however, is not as great as it first might appear. The specific holding of the Holland decision‑‑that the migratory bird treaty was a proper exercise of the treaty power‑‑presumably would be preserved, given the much narrower scope of the Tenth Amendment today. Moreover, as discussed above, although Holland has been construed as giving the treaty power complete immunity from federalism limitations, the decision itself can be read much more narrowly. [FN386] Recent confirmation of this can be found in the brief filed by the Clinton administration in the recent Breard case stating that there are in fact federalism limits on the national government's power to enforce treaties. [FN387]

In any event, the Supreme Court has said that stare decisis carries less weight with respect to constitutional decisions, since Congress cannot overturn them. [FN388] This is especially so when fundamental assumptions in the first decision no longer hold true. [FN389] That is exactly the case with respect to Holland.

The world in which Holland was decided looks very different from our own. The Court in Holland assumed that the power of Congress to regulate interstate commerce was relatively narrow. As a result, the Court thought it necessary that the treaty power be more expansive than the commerce power, in order to address "matters of the sharpest exigency for the national well being that an act of Congress could not deal with." [FN390] Today, of course, Congress's commerce power is extremely broad, and it is likely to remain so notwithstanding the Supreme Court's renewed commitment to federalism.

Holland also was decided at a time when customary international law, rather than treaties, was the dominant form of international law. [FN391] Since then, however, there has been a proliferation of *460 treaties, as well as the rise of numerous multilateral treaty regimes, such that treaties "have become the principal vehicle for making law for the international system." [FN392] This shift from custom to code has meant that international law is now more specific and regulates a substantially wider range of subjects. As Professor Mark Janis has observed, today "[v]irtually every human activity is to some degree the object of some treaty." [FN393] These changes have thus substantially heightened the significance of the treaty power in this country.

The Court in Holland also appeared to assume that treaties would deal only with matters concerning truly inter‑national relations. Thus, the Court emphasized that the treaty there concerned a problem that "can be protected only by national action in concert with that of another power." [FN394] Since then, however, we have seen the rise of international human rights law, which regulates the relations between nations and their citizens. Among other things, this means that there is today a significantly greater overlap and potential for conflict between treaty law and U.S. domestic law.

Finally, the Court in Holland placed great emphasis on the delegation language of Article II, [FN395] something problematic on its own terms but especially so given that most international agreements concluded by the President today take the form of executive agreements rather than Article II treaties. Even if it were true that the treaty clause in Article II was intended to delegate unlimited lawmaking power to the federal government, subject only to certain process protections, this clause has become much less relevant to American treaty‑making.

In sum, there is a strong case‑‑based on history, doctrine, and policy‑‑for subjecting the treaty power to the same federalism limitations that apply to Congress's legislative powers. This approach would involve overruling some of the reasoning in Holland, but that reasoning has become questionable in light of changes in both the nature of treaty‑making and the scope of federal legislative power. It is particularly questionable if, contrary to the Court's apparent assumption in Holland, there is no subject matter limitation on the treaty power. Importantly, this proposal would not interfere substantially with the treaty power. It might preclude some of the broadest intrusions on state power, such as Professor Neuman's recent *461 proposal concerning RFRA, but it would leave the political branches with substantial flexibility to conclude and implement international agreements in the national interest.

Conclusion

The treaty power in this country is a power to make supreme federal law. For much of our history, courts, commentators, and government officials have assumed that this power is limited by subject matter, states' rights, or both. In recent years, however, conventional wisdom has denied any such limitations. The result of this view is that the treaty makers have essentially unlimited power vis‑a‑vis the states. Such unlimited power, however, is inconsistent with a central principle of American federalism‑‑that the national government's powers are limited and enumerated. This inconsistency is particularly significant today, in light of the rapidly expanding nature of this country's treaty commitments.

Faced with this conflict between foreign affairs orthodoxy and federalism, we could, of course, abandon our commitment to protecting federalism. Perhaps the increased globalization and interdependence of nations renders federalism obsolete. On the other hand, these forces might actually increase the desire for local democracy and experimentation and thus make federalism even more attractive. [FN396] In any event, we must make a choice. As we continue with what is in essence an "international New Deal," we must decide whether federalism is worth preserving. If it is, the nationalist view of the treaty power should be reconsidered.

This is not the only foreign affairs issue that may require reconsideration. As I have explained, the nationalist view of the treaty power is but one example of an approach that could be called "foreign affairs exceptionalism." This approach, which is largely a product of academic and judicial thinking during the 1920s through the 1940s, distinguishes sharply between domestic and foreign affairs. As we enter the next century, that distinction, and the foreign affairs exceptionalism that it justifies, will become increasingly difficult to maintain.

[FNa1]. Associate Professor, University of Colorado School of Law. B.A. 1985, University of Colorado; J.D. 1988, Harvard.‑‑Ed. For their helpful comments and suggestions, I would like to thank Kathryn Bradley, Ann Estin, David Fidler, Martin Flaherty, Barry Friedman, Jack Goldsmith, Mark Loewenstein, Hiroshi Motomura, Bob Nagel, Dale Oesterle, Sai Prakash, Pierre Schlag, Steve Smith, Peter Spiro, Phil Weiser, Ted White, John Yoo, and participants in a faculty workshop held at the University of Colorado School of Law. I would also like to thank Sarah Good and Tyler Hand for their excellent research assistance.