The Intellectual Property Clause's External Limitations

Could Congress enact perpetual copyright or patent protection? Could it grant copyright or patent rights to an entity other than a work's author or inventor, respectively? Could it protect against the copying of works that are not fixed in a tangible medium? A look at the Constitution's Intellectual Property Clause (IP Clause) alone suggests a probable-to-definite no to each of these questions. The IP Clause empowers Congress "To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries." reach beyond the scope of the power conferred by the IP Clause to "secur[e] for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries." 6 The evidence marshaled by this Article shows that if Congress seeks, via legislation, to promote the progress of science and useful arts, the only way it may do so is by enacting laws that secure to authors and inventors exclusive rights in their writings and discoveries for limited times.

The IP Clause's text and placement within the constitutional structure suggest that Congress possesses power to pursue the goal of promoting the progress of science and useful arts, but only by using the means specified by the Clause itself: namely, securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries. Additionally, the IP Clause's history indicates that Congress cannot use other means to promote the specified end of promoting the progress of science and useful arts. For one thing, the Framers rejected other possible means of achieving this end, such as the ability to award grants and prizes. 7 This choice implies that the Framers intended for Congress to use only the means explicitly identified in the Clause to achieve the Clause's ends. Congress's refusal in its earliest years to enact certain laws that would have sought to promote the progress of science and useful arts through means other than those laid out in the IP Clause-even means allotted to Congress by other constitutional provisions, such as the Commerce Clause-further confirms this understanding.

Although the Supreme Court has never directly confronted the limitations that the IP Clause might impose on Congress's other powers, its jurisprudence on the IP Clause, as well as its general understanding of clashes between two constitutional provisions, supports this interpretation. 9 Finally, the policy embedded in the IP Clause of maintaining a careful balance between intellectual-property rights for creators on the one hand and the broader public benefit on the other demands that Congress not upset this balance by resorting to its other powers to further the IP Clause's ends. In all, the evidence to which courts and scholars normally turn to interpret and construe the Constitution suggests that the IP Clause externally limits Congress from using its other powers to promote the progress of science and useful arts in ways that go beyond the IP Clause's means.

That said, there is a competing side to the story of Congress's power in this area. Congress can sometimes have legitimate interests in enacting laws that promote the progress of science and useful arts that extend beyond the IP Clause's means. For one thing, the Commerce Clause's reach has grown greatly since the United States' early years. 11 The Commerce Clause can reasonably be understood to extend to intellectual goods traditionally governed by the IP Clause because of those goods' use in commerce. Alternatively, globalization has led to the escalating use of treaties and other foreign agreements to govern intellectual property. Congress might view commerce or trade interests or foreign-relations concerns as distinct from the concerns of the IP Clause and thus as requiring the employment of means other than those specified in the IP Clause. Any theory of the extent of Congress's power to regulate in the area of intellectual property ought to address the fact that Congress might not be concerned merely with promoting the progress of science and useful arts, but also, for example, with promoting commerce or foreign relations.

These competing considerations give rise to an analytical framework under which courts, legislators, and others can assess the constitutionality of federal legislation. As a general matter, when legislation has the structural purpose of promoting the progress of science and useful arts, it must restrict itself to the means specified in the IP Clause. But with respect to laws that have multiple constitutional purposes, there ought to be a presumption against the constitutionality of laws that promote the IP Clause's ends but subvert its means, a presumption that may be overcome only by clear and convincing evidence that Congress, pursuant to its other more permissive powers, intentionally chose to supersede the IP Clause's means because of paramount, legitimate interests. This presumption ought to be significantly harder to overcome when a law's chosen means interfere with the IP Clause's means instead of merely diverging from the means included in the IP Clause. This framework suggests that a number of modern federal laws, such as the trade-secrecy laws and antibootlegging laws, might be unconstitutional. The framework also demonstrates how to assess the constitutionality of laws that would protect databases, laws passed pursuant to international agreements, and laws that establish federal funding for scientific and artistic works. Because Congress, since the late twentieth century, has increasingly been legislating at or beyond this constitutional boundary, having a framework with which to assess the constitutionality of certain laws at the outer margins is increasingly important. This Article's proposed framework is different from analyses set out by previous works. Professor Thomas Nachbar, for instance, maintains that the IP Clause does not limit Congress's other powers. 12 I disagree.

Other scholars focus on the internal limitations of the IP Clause. For example, Professor Lawrence Solum maintains that the Clause is characterized by a unique means-end structure that constrains the granted power. 13 Professor Dotan Oliar similarly argues that restrictions inhere in the IP Clause-specifically, that the IP Clause allows Congress to enact intellectual-property laws only if they promote the progress of science and useful arts-but does not push further to understand how the IP Clause's boundaries might or might not limit Congress's other powers. 14 Given the expansiveness of Congress's other powers, such as the powers conveyed by the Commerce and Spending Clauses, this question is even more critical than the question of the IP Clause's internal meaning alone. Some works propose that the IP Clause restrains Congress's other powers. These works, however, tend to be undertheorized, do not extend as far as my analysis does, or suggest a complex analysis. The recurring suggestion that the IP Clause's specific power limits more general powers elsewhere in the Constitution is undertheorized. 15 In an in-depth analysis, Professor Yochai Benkler extrapolates from some Supreme Court decisions and builds on policy to read the IP Clause's grant of power as limiting Congress to granting exclusive rights in information only when that information makes an "original contribution[] to the wealth of human knowledge" not already in the public domain. 16 Similarly, former Professor William Patry reads the IP Clause as preventing Congress from undermining the public's right to copy unoriginal material. 17 My analysis suggests going further in some ways and less far in others. Professors Paul Heald and Suzanna Sherry perceive a complex constellation of principles in the IP Clause that tie Congress's hands in enacting other legislation. In their view, the IP Clause is the only relevant power with respect to "legislation that imposes monopolylike costs on the public through the granting of exclusive rights," and any law passed under it must, in exchange for granting a right to a contribution's creator, provide a new contribution to the public, in the Reading, 57 UCLA L. REV. 421, 423 (2009) (characterizing his own previous work on the IP Clause as "ma[king] one major argument": that"[t]he Framers intended the progress language in the Clause-'to promote the progress of science and useful art'-to limit Congress's power to grant IP rights").

15. See, e.g., Rochelle Cooper Dreyfuss, A Wiseguy's Approach to Information Products: Muscling Copyright and Patent into a Unitary Theory of Intellectual Property, 1992 SUP. CT. REV. 195, 230 ("Restrictions on constitutional grants of legislative power, such as the Copyright Clause, would be meaningless if Congress could evade them simply by announcing that it was acting under some broader authority."); Malla Pollack, The Right To Know?: Delimiting Database Protection at the Juncture of the Commerce Clause, the Intellectual Property Clause, and the First Amendment, 17 CARDOZO ARTS & ENT. L.J. 47, 60 (1999) ("Accepting that Congress may not do an end run around a limitation in one clause of the Constitution by invoking a more general clause, Congress may not grant (at least some types of) exclusive rights to something close to, but not quite, the writings of authors or the discoveries of inventors.").

16. Yochai Benkler, Constitutional Bounds of Database Protection: The Role of Judicial Review in the Creation and Definition of Private Rights in Information, 15 BERKELEY TECH. L.J. 535, 538-46 (2000). 17. William Patry, The Enumerated Powers Doctrine and Intellectual Property: An Imminent Constitutional Collision, 67 GEO. WASH. L. REV. 359, 364-66 (1999).

form of access to the work presently and free access eventually when it enters the public domain.

Part I uses the IP Clause's structure, text, history, judicial doctrine, and policy to build the case that the Clause ought to be understood as both a grant of limited power and a limitation on Congress's authority to use its other enumerated powers to promote the progress of science and useful arts by means other than those specified in the Clause. Part II uses this evidence to construct a framework for assessing the constitutionality of legislation and suggests how to evaluate the potential conflicts that might arise between the IP Clause power and other constitutional powers, most notably the commerce, spending, necessary-and-proper, and treaty powers. This Part incorporates the insight that a law regulating intellectual property might have multiple legitimate constitutional purposes. Part III builds on this framework to suggest that a number of existing and potential federal laws are of questionable constitutionality.

Most scholars and courts that have examined the IP Clause have dwelled on whether the first half of the Clause-"To promote the Progress of Science and useful Arts"-is a restraint on Congress's authority to enact intellectual-property laws. That is, can Congress enact intellectual-property laws if they do not, in actuality, promote this progress? For example, both Justice Breyer and Professor Oliar rely on history, constitutional structure, and policy considerations to conclude that the progress provision is an independent restriction. 19. See, e.g., Eldred v. Ashcroft, 537 U.S. 186, 242-67 (2003) (Breyer, J., dissenting) (arguing that the IP Clause limits congressional power based on the history and text of the IP Clause and policy considerations); JAMES BOYLE, THE PUBLIC DOMAIN: ENCLOSING THE COMMONS OF THE MIND 170, 208, 210 (2008) (arguing, after considering the constitutional text, "Jeffersonian ideals," and policy concerns, that the IP Clause limits congressional power); WILLIAM PATRY, MORAL PANICS AND THE COPYRIGHT WARS 123 (2009) ("The Constitution is quite clear that Congress can only grant copyright to promote the progress of science, and thus this 'burden' exists from the inception of the rights, and follows those rights for the duration of the copyright."); Oliar, supra note 1, at 1772-73, 1778-79 (arguing, after considering the Framers' intent, that the IP Clause limits congressional power); Malla Pollack, What Is Congress Supposed To Promote?: Defining "Progress" in Article I, Section 8, Clause 8 of the United States Constitution, or Introducing the Progress Clause, 80 NEB. L. REV. 754, 770-71 (2001) (reasoning that Congress cannot use its Commerce Clause powers to enact intellectual-This particular line of research focuses chiefly on the IP Clause's internal limits, not on its external limits-in other words, whether the Clause limits Congress's other powers. Such scholarship is compatible with this Article's analytical framework, but it ultimately falls outside of this Article's scope.

This Article understands the IP Clause's first part as a purposive channeling mechanism. That is, if Congress desires to promote the progress of science and useful arts, it may do so only by awarding to authors and inventors the exclusive right in their writings and discoveries for limited times. Moreover, Congress may not use its other enumerated powers to promote the progress of science and useful arts by using means other than those specified in the IP Clause. In this way, the IP Clause externally limits Congress's other powers. As I discuss in this Part, text, history, constitutional structure, Supreme Court doctrine, and policy strongly support this understanding.

Others have concluded otherwise. According to Professor Nachbar, the Constitution does not restrain Congress from enacting laws that do not conform to the IP Clause's specified means. 20 He states that no good evidence-historical or otherwise-supports the proposition that the IP Clause was intended to limit Congress's other powers, such as its powers under the far-reaching Commerce Clause.

Nachbar asserts that the IP Clause was not much debated at the Constitutional Convention and that it is "a relatively insignificant form of economic regulation allocating quasi-property rights between private entities." 22 Therefore, he sees the IP Clause as being far from a central structural concept in the constitutional framework. 23 Moreover, Nachbar detects no constitutional norms-in either the IP property protection that does not promote the progress of science and useful arts); Edward C. Walterscheid, The Preambular Argument: The Dubious Premise of Eldred v. Ashcroft, 44 IDEA 331, 333 (2004) ("[T]he preambular argument which treats the 'to promote' language as a preamble which places little or no constraint on the legislative power of Congress is the most dubious and least tenable of the various interpretations that have been given to the Science and Useful Arts Clause."); cf. Andrew M. Hetherington Clause or the First Amendment-that ought to restrain Congress from enacting intellectual-property laws beyond the IP Clause's limits, whereas he argues that Congress, as a matter of policy, ought to decide the IP Clause's boundaries vis-à-vis Congress's other powers. 24 Walterscheid takes a somewhat related position, although via another analytical route. He suggests that although the IP Clause ties Congress's hands broadly by requiring that certain laws not impose "monopoly-like costs on the public," among other restrictions, it also gives Congress power to promote the progress of science and useful arts by various means, including but not limited to the copyright and patent powers described in the Clause.

This Part disagrees by showing that the IP Clause externally limits Congress's other powers. The following sections in turn analyze the Clause's text, structure, and history, as well as Supreme Court doctrine and policy, to make the case.

The U.S. Constitution confers upon Congress power "To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries." 26 To understand the IP Clause's reach, one should first examine this text and its surrounding constitutional structure. As Professor Laurence Tribe explains, "[A]ny 'interpretation' of a constitutional term or provision must at least seriously address the entire text out of which a particular fragment has been selected for interpretation, and must at least take seriously the architecture of the institutions that the text defines." 27 This examination, although not conclusive, suggests a clause that allows Congress to promote the progress of science and useful arts 24. Id. at 318, 344. According to Professor Nachbar, the First Amendment limits congressional power under the IP Clause, but it does not thereby impose external limits on Congress's other powers. only by the means specified in the Clause, even if Congress might otherwise assert another textual source of authority for its action. Some view the first part of the IP Clause as nothing more than a nonbinding preamble. 28 This interpretation cannot be correct. For one thing, it requires reading the first half both unnaturally and as a nullity, contrary to the basic rule of construction that every word of the Constitution must be given meaning. 29 This construction also confers a different grammatical understanding on the IP Clause than on the other Article I, Section 8 powers. As James Monroe long ago observed, each of the enumerated powers listed in Article I, Section 8 begins with a grant of power. 30 Read correctly, the IP Clause fits this scheme because, like all of the other enumerated powers, the first part of the Clause begins with a "To," followed by the power that is being granted.

In another way, however, the IP Clause is structured differently from the other Article I, Section 8 powers. It is the only clause that specifies the means for carrying out the allotted power:

32 "by securing 'To promote the progress of science and useful arts . . .' must be read as largely in the nature of a preamble, indicating the purpose of the power but not in limitation of its exercise." (omission in original) (footnotes omitted)).

29. See, e.g., Oliar, supra note 1, at 1787 ("The last justification for the limiting operation of the Progress Clause has been to use the general interpretive presumption according to which each and every word and phrase in the Constitution should be given meaning. Some argued that interpreting the Progress Clause as a non-binding preambular statement of purpose would render it meaningless and thus should be rejected." (footnote omitted)); Solum, supra note 13, at 22 ("Given the plain language of the Intellectual Property Clause and the structure of Article I, Section 8, the notion that the Clause consists of a preamble followed by a power grant is simply unsustainable."); Hetherington, supra note 19, at 465 ("[I]t seems strange that the power to promote progress would be mentioned at all, unless it was intended as a purposive element in the clause.").

30. James Monroe, Views of the President of the United States on the Subject of Internal Improvements (May 4, 1822), reprinted in 2 A COMPILATION OF THE MESSAGES AND PAPERS OF THE PRESIDENTS 1789-1897, at 144, 163 (James D. Richardson ed., 1896).

31. Solum, supra note 13, at 13-20; see also Walterscheid, IP Clause, supra note 12, at 767 ("The clause is . . . treated as though the grant of power is: 'To secure for limited times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries.' . . . A major disadvantage is that this is not literally the language of the clause, and such interpretation renders its grammatical construction quite different than that afforded to every other enumerated congressional power.").

32. Solum, supra note 13, at 19-20; Walterscheid, IP Clause, supra note 12, at 774-75. The Militia Clause has a related structure: it confers upon Congress power "To provide for calling forth the Militia to execute the Laws of the Union, suppress Insurrections and repel Invasions." U.S. CONST. art. I, § 8, cl. 15. Professor Lawrence Solum understands this Clause to enable Congress to execute the Union's laws, suppress insurrections, or repel invasions by calling forth the militia. Solum, supra note 13, at 19. Unlike the IP Clause, this Clause's placement of means for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries." The natural reading of this clause, given its unique structure, is that Congress has the power to promote the progress of science and useful arts using solely the specified means. 33 The word "by," preceded by a comma, links the first part of the Clause to the second part, suggesting a means-end relationship between the two. 34 39 Similar limitations arguably exist elsewhere in Article I, Section 8, such as in the provision granting Congress power to raise and support armies, but with no appropriation of money before ends carries no suggestion that Congress may not use its other powers to execute the nation's laws, suppress insurrections, and repel invasions. Cf. id. at 20 ("Unlike the second Militia Clause, the Intellectual Property Clause grants the power to pursue a goal and then qualifies that power by specifying the permissible means."). Were the IP Clause to have the same structure as the Militia Clause, by providing Congress power "to secure for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries to promote the Progress of Science and useful Arts," it would not carry the same textual implication-that the Clause's goals may be promoted only through one specified avenue. See id. ("The alternative [to the IP Clause's structure] would be a power to employ any means that furthered the goal, and such a power would not fit the general design of Article I, Section 8, which lays out a scheme of limited and enumerated powers.").

33. lasting for more than two years; in the provision giving it power to establish a uniform rule of naturalization; and in the provision granting power to make uniform bankruptcy laws. 40 As discussed in Section B, the historical evidence more comprehensively supports this means-end textual reading of the IP Clause. Together, these sections suggest that the IP Clause authorizes only the specified means to promote the progress of science and useful arts.

Even so, might Congress use its other powers to employ different means to promote the progress of science and useful arts? James Madison-a central figure in the Constitution's making 41 -believed that a congressional power ought to be enumerated in a separate provision only if it did not fall within any of the other enumerated powers. 42 This line of thinking suggests that Congress cannot use its other powers to regulate intellectual property, at least with regard to the original understanding.

Relatedly, some scholars argue that the IP Clause limits some of Congress's other enumerated powers, most notably its Commerce Clause power. They reason that the IP Clause grants a more specific power than that provided by the Commerce Clause, and the more specific ought to take precedence over the more general. 43 This reasoning, however, does not provide sufficient evidence that the IP 40. See U.S. CONST. art. I, § 8, cl. 12 (authorizing Congress "To raise and support Armies," provided that "no Appropriation of Money to that Use shall be for a longer Term than two Years"); id. art. I, § 8, cl. 4 (granting Congress power "To establish an uniform Rule of Naturalization, and uniform Laws on the subject of Bankruptcies throughout the United States"); see also Patry, supra note 17, at 373 ("Claiming that [Clauses 4, 8, and 12] do not act as a limitation on Congress's power ignores their unambiguous text and repeated Supreme Court opinions holding that Clause 8 is both a grant of power to Congress and a limitation on that power.").

41. Clause externally limits Congress's other powers. With respect to its interstate restriction, the Commerce Clause is a more specific power than the IP Clause. Applying the general-versus-specific reasoning to that fact, however, would lead to the incorrect conclusion that Congress is prohibited from using the IP Clause to regulate purely intrastate commerce, on the ground that such regulation would evade the Commerce Clause's restriction on regulating commerce that is interstate, a restriction that is absent from the IP Clause.

By contrast, others considering the general issue of overlap among Article I, Section 8 powers tend to conclude that the enumerated powers can and do overlap. Professor Nachbar observes that these powers currently overlap, given the broad contemporary reach of the Commerce Clause, even if less overlap existed in earlier times:

If one compares Section 8 as it was applied in 1789 with Section 8 as it is applied today, any number of its provisions have become superfluous. Certainly Congress could establish post offices and post roads today under either the commerce or spending powers, enact bankruptcy and naturalization statutes, fix standards for weights and measures, and criminalize counterfeiting of U.S. currency under the interstate commerce power, or even punish felonies on the high seas under the foreign commerce power. There is no reason to think that any of these powers were considered superfluous at the time, but they have become so. The growth of the commerce power means that we have to get comfortable with redundancy in Section 8.

Professor Jack Balkin similarly reasons that the enumerated powers are listed together, perhaps redundantly, in support of a larger principle authorizing Congress to act when the states are incompetent to do so individually.

This possibility of overlap, however, does not mean that Congress can use its other powers to reach beyond what seem to be limitations in the IP Clause. In fact, the opposite would seem to be 44. See Nachbar, supra note 12, at 313 ("That one of the Section 8 limitations must be read externally does not, as a matter of logic, dictate that they all must be; to so infer would lead to impossible results. . . . [Applying this rule] would prevent federal regulation of purely intrastate commercial conduct (such as sales or distribution) involving intellectual property because to allow such regulation would be to 'eradicate from the Constitution a limitation on the power of Congress to' regulate commerce." (footnotes omitted) (quoting Ry. Labor Executives' Ass'n v. 47 Thus, the possibility of two different constitutional powers both conferring authority on Congress to take a particular action, as Professors Nachbar and Balkin discuss, is materially distinct from a situation in which one constitutional power seems to authorize action while the other seems to restrict or forbid it.

Professor Nachbar nonetheless emphasizes that each of the Article I, Section 8 powers has express limits. He elaborates:

Along with offenses against the law of nations, the tenth clause permits Congress to "define and punish Piracies and Felonies," but that power is expressly limited to felonies "committed on the high Seas." Does that mean that Congress cannot punish landlubbers' felonies pursuant to its commerce power? The Commerce Clause itself contains limitations, permitting the regulation of only foreign commerce, Indian commerce, or commerce "among the several States" . . . .

Nachbar goes on to suggest that the internal limitations on one power ought not necessarily be construed to limit other powers without evidence that the limitations were meant to reach beyond the power itself.

49 Nachbar reasons that limitations on a grant of one power are all too easily converted into a general lack of power to legislate in other areas. 50 Without any clear signal to make such a conversion, he 52 Moreover, as I elaborate in subsequent Sections, early American history, judicial doctrine, and policy together loudly reinforce this signal.

Some argue that notwithstanding the unique phrasal structure of the IP Clause, which suggests limitations on Congress's power, the broader structure of Article I indicates otherwise. They reason that Section 8 contains grants of congressional legislative power, whereas Section 9 contains limitations. 53 On this reasoning, Congress may invoke any power that is granted by at least one clause in Section 8 and is not thereafter limited in Section 9. 54 Thus, given that Section 9 does not limit Congress's authority to promote the progress of science and useful arts, Congress might use means authorized in powers outside of the IP Clause to do so. Former Professor William Patry rejects this argument on the basis that "the limitations in Article I, Section 9 are not closely related to the grants of power in Section 8; instead, they involve different issues, specialized issues that may or may not relate to Section 8, or issues that are general in nature." 55 Under this line of reasoning, to use Section 9 to reject the possibility that the IP Clause limits Congress's other powers seems unreasonable.

In sum, ample textual and inferential evidence suggests that the IP Clause on its own permits only its specified means to achieve its designated end. The evidence from Article I's broader structure, however, is less conclusive. All in all, the IP Clause's text and the constitutional structure volunteer a suggestive-but not ironcladargument that the Clause's unique construction operates externally to forbid Congress from using its other powers to promote the progress of science and useful arts beyond the means specified in the Clause. I turn now to historical evidence that this understanding is the correct one.

This Section discusses the IP Clause's adoption at the Constitutional Convention along with early congressional activity construing the Clause and its structural relationship to other enumerated powers. This evidence demonstrates that the IP Clause was intended to restrain Congress from using its other powers to employ means other than those in the IP Clause to promote the progress of science and useful arts.

1. Constitutional Convention. During the period in which the states lived under the Articles of Confederation, the infeasibility of states' varying intellectual-property laws became increasingly clear. Three primary reasons contributed to this reality: First, different states would assign rights in the same or similar creations to different creators, or they would assign rights to the same person for the same creation for differing terms. 56 Most famously, the steamboat patent was assigned in some states to John Fitch and in others to James Rumsey. 57 These inconsistencies were both inefficient and confusing. Second, the protection of a creation in a single state was ineffective at preventing unauthorized copies of the creation in other states.

Finally, for the same creator to apply in each state to secure effective nationwide protection was uneconomical. 59 61 James Madison made four proposals: First, in the realm of patents, Madison suggested a provision allowing Congress "To secure to the inventors of useful machines and implements the benefits thereof for a limited time." Second, as to copyrights, he advocated language that would empower Congress "To secure to literary authors their copyrights for a limited time." Third, in the area of education, he proposed a clause allowing Congress "To establish an University." And fourth, he imagined a clause authorizing the creation of a system of grants and prizes under which Congress would be permitted "To encourage, by proper praemiums and provisions, the advancement of useful knowledge and discoveries." 62 Charles Pinckney of South Carolina proposed four similar provisions that would have allowed Congress to act in the same areas envisioned by Madison. Pinckney's suggested language would have authorized Congress "To grant patents for useful inventions"; "To secure to Authors exclusive rights for a . . . certain time"; "To establish seminaries for the promotion of literature and the arts & sciences"; and "To establish public institutions, rewards and immunities for the promotion of agriculture, commerce, trades and manufactures."

63 Not only were these similar proposals by Madison and Pinckney made on the same day, 64 suggesting some coordination, but in each case, the four proposals were also made together, signifying a similarity of subject matter.

Madison's and Pinckney's proposals were referred to the Committee of Detail, and by the end of August, they had been passed on to the third Committee of Eleven. 65 On September 5, 1787, the third Committee of Eleven recommended-without any recorded debate-the inclusion of an IP Clause in the Constitution that would provide Congress with power "To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Despite the near-silent treatment given to the IP Clause in the Convention's records, 71 in reworking the eight proposals into their final form, the Framers made a number of choices that strongly suggest the IP Clause's intended meaning. For one thing, the IP Clause, as ultimately adopted, is more limited than plenary copyright and patent powers would have been.

72 Rather than granting a blanket power to Congress to enact copyright and patent laws, as Madison and Pinckney had each originally proposed, the Framers prefaced the Clause's operative provision with "To promote the Progress of Science and useful Arts, by." The inclusion of such prefatory language indicates that the Framers meant for the Clause to achieve something other than merely granting to Congress power to enact copyright and patent laws. As Section A's textual and structural reading of the text argues, the IP Clause's preface seems to be a limitation. 73 Similarly, had the Clause simply given the power "to promote the Progress of Science and useful Arts," Congress presumably would have had the power to enact legislation encompassing all eight of Madison's and Pinckney's proposals and then some. 74 The actual text and structure emphasize that the IP Clause is more limited than it would have been if it had consisted of either the prefatory language or the operative language by itself. Congress is granted authority to promote the progress of science and useful arts, but only in a limited sense: by securing for limited times to authors and inventors exclusive rights in their writings and discoveries. 75 Moreover, the first part of the IP Clause, which concerns promoting progress, is derived from Madison's and Pinckney's other proposals related to intellectual property-namely, those to allow Congress to establish universities and to provide grants or prizes for innovation. 76 The Constitution's final version did not expressly adopt these proposals. 77 Professor Oliar primarily uses this observation to conclude that the IP Clause ought to be understood as granting Congress power to enact copyright and patent laws only if those laws promote the progress of science and useful arts. 78 But the more straightforward understanding strongly suggests the converse, a possibility that Oliar secondarily recognizes: 79 if Congress wants to promote the progress of science and useful arts, it can do so only by securing to authors and inventors exclusive rights in their writings and discoveries for limited times. The Framers rejected multiple other ways to promote the progress of science and useful arts-via universities or governmental grants and prizes, for example-yet kept the language about promoting progress that had originally been associated with those proposals. The Framers therefore appear to have sought to limit the ways in which Congress can promote the progress of science and useful arts-and in particular, to limit Congress to the means specified in the IP Clause itself. 80 Exactly why the rejected proposals were cast aside was the subject of only minimal discussion. Roger Sherman of Connecticut explained that there was no need to give the university power to the federal government, as the states could sufficiently establish universities instead. 81 Gouverneur Morris of Pennsylvania observed that despite the elimination of this university power, the federal 75. government would nonetheless have a limited ability to create a university at the federal seat of government, an area over which the government would have exclusive control. 82 As for the rejected power to award grants and prizes, some contemporaneous evidence suggests that the Constitution did not include this power because it would have been more expensive for a cash-poor federal government to implement than a system of awarding copyrights and patents would have been.

83 This reason appears to have been grounded in a rejection on the merits rather than a judgment that the federal government would otherwise fully possess this power under alternative provisions.

Scholars offer conflicting accounts of the Framers' views on monopolies and of how those views ought to shade a modern audience's understanding of the IP Clause. To the extent that the Framers were concerned about monopolies, one might see even more reason to conclude that the IP Clause's carefully calibrated means of promoting the progress of science and useful arts ought to be the only means of doing so, for fear that pursuing alternative legislative strategies might lead too far toward monopolization. Some commentators emphasize that the Framers-prominently George Mason-and Thomas Jefferson feared monopolies and sought to approve of monopolies only narrowly by granting limited rights in 82. 2 id. at 616. 83. Walterscheid, To Promote, supra note 12, at 6 n.21. Scholars offer other theories for this decision, such as a desire to limit monopolistic tendencies, a fear of Congress's favoritism toward particular individuals or states, and a desire to minimize the government's role in the marketplace. E.g., Oliar, supra note 1, at 1800-02.

84. The means authorized by the IP Clause indicate how Article I, Section 8 more broadly focuses on solving problems of collective action among the states by placing power in those areas in the hands of Congress. See generally Balkin, supra note 46 (discussing this theory with regard to Article I, Section 8 generally); Robert D. Cooter & Neil S. Siegel, Collective Action Federalism: A General Theory of Article I, Section 8, 63 STAN. L. REV. 115 (2010) (same). Many collective-action problems were associated with the preconstitutional implementation of copyright and patent laws by states. See supra text accompanying notes 56-60. By giving Congress power to enact copyright and patent laws, the Constitution mitigated these problems. Relatedly, this theory of collective action explains the intellectual-property powers that the Framers rejected: education and grants and prizes. Neither of these powers would have implicated any significant collective-action concerns, as the states could have implemented both of the contemplated schemes without encountering any unusually worrisome negative externalities. This insight does not mean that Congress's powers extend only to problems of collective action or to every problem of collective action, but rather that the Framers were convinced that some such problems had to be fixed by Congress. intellectual property. 85 In support of their views, these scholars point to earlier English copyright and patent laws enacted to curb monopolies.

Others suggest that the Framers had inconsistent and complex views on monopolies. 87 For one thing, the Framers did not include a constitutional provision that would have forbidden monopolies altogether. 88 For another, George Mason and others with asserted antimonopoly views were, according to Professors Paul Schwartz and Dean William Treanor, "concerned with what they feared would be a broad implied congressional power under the Necessary and Proper Clause, rather than the express power to authorize patents and copyrights."

89 Some Framers may have even thought monopolies were good in various circumstances. 90 This group of scholars thus suggests that antimonopolistic concerns ought not to be relied upon in construing the IP Clause. 91 They deemphasize the importance of Thomas Jefferson's views in this context, not least because he did not participate in the Constitutional Convention.

As Professor Nachbar points out, even the antimonopolistic views of some of the Framers would not, on their own, establish that the IP Clause was meant to place affirmative limits on other federal powers. Nachbar comments that "those who would apply the Intellectual Property Clause's limits to all of Article I must identify a 91. See, e.g., Nachbar, supra note 12, at 349 ("[T]he granting of monopolies was not deemed to be a power different enough in kind or degree from other powers to be worthy of limits beyond those inherent in representative government.").

92. E.g., Schwartz & Treanor, supra note 42, at 2378. Jefferson did, however, see a draft of the Constitution that contained the IP Clause, sent to him by Madison. Walterscheid, IP Clause, supra note 12, at 769. Jefferson did not mention the IP Clause when he responded to Madison, but he urged the inclusion of a bill of rights, among whose provisions would be certain restrictions on monopolies. Id. link between that general distaste and the need to constrain government from having the power, a link never made by the Framers."

93 At most, then, the existence of antimonopolistic views among the Framers provides mere shading to supplement the other historical evidence on the IP Clause.

The Framers' views on monopolies may have ambiguous implications for various federal powers. Nonetheless, James Madison's writing on the topic is useful for shedding light on the reach of the IP Clause: Monopolies tho' in certain cases useful ought to be granted with caution, and guarded with strictness ag[ain]st abuse. The Constitution of the U.S. has limited them to two cases, the authors of Books, and of useful inventions, in both which they are considered as a compensation for a benefit actually gained to the community as a purchase of property which the owner might otherwise with[h]old from public use. There can be no just objection to a temporary monopoly in these cases: but it ought to be temporary, because under that limitation a sufficient recompence and encouragement may be given.

Madison's discussion provides evidence that whatever the Framers' overall views on monopolies, Madison himself thought that Congress would be unable to grant monopolistic rights in intellectual property other than via the means provided in the IP Clause. Although Professor Nachbar agrees that this interpretation is facially plausible, he writes that a viable alternative account would be that Madison thought that no other enumerated power gave Congress the power to grant exclusive rights. 95 But as discussed in the next Section, a more complete look at Madison's views demonstrates overwhelmingly that Madison thought the only way for Congress to grant exclusive rights would be through the means provided in the IP Clause. To cite one pertinent example, Madison likely believed that the exclusive rights that would be provided by a particular legislative proposal for patents of importation were within the reach of the Commerce Clause but were otherwise restricted by the IP Clause. 96 Therefore, the possibility All in all, the history that culminated in the IP Clause's adoption suggests that the Clause ought to be understood as granting Congress power to promote the progress of science and useful arts, but only by employing copyright and patent protection as limited by the Clause's specified means. The Framers rejected other possible means of promoting the progress of science and useful arts, and their rejection seems to have been a judgment on the merits that those means would be unwise. Moreover, the Clause ultimately retained purposive, endsoriented language similar to that contained in the rejected powers-"To promote the Progress of Science and useful Arts." The inclusion of such language underscores that the IP Clause's ends should not be effectuated by means other than the adopted ones.

This discussion would seem strongly to establish the IP Clause's internal limits, but what of the ability of Congress to invoke its other powers to use alternative means, including the rejected means? The same history seems strongly to support the inference that the Framers intended Congress to possess no means to promote the progress of science and useful arts other than those adopted in the IP Clause. And even stronger evidence, taken from the earliest congressional activity, encourages the drawing of this inference. I turn now to that evidence.

Evidence from the Constitutional Convention is not the only history helpful for understanding the IP Clause. As the Supreme Court emphasized in construing the Clause in a 2003 decision, constitutional construction can be aided by looking to "a contemporaneous legislative exposition of the Constitution when the founders of our Government and framers of our Constitution were actively participating in public affairs." 97 Early congressional activity can thus shed light on how the Framers understood the IP Clause itself and its connection with the other congressional powers. 98 This evidence demonstrates that the Framers understood the IP Clause to operate externally to forbid 98. See Currie, supra note 41, at 777 (noting that "the first Congress was a sort of continuing constitutional convention," in part "because so many of its members . . . had helped to compose or to ratify the Constitution itself").

Congress from using its other powers to employ means of promoting the progress of science and useful arts not specified in the IP Clause.

A critical piece of evidence supporting the IP Clause's external limitations concerns patents of importation, exclusive rights for limited times to the first person to import foreign technologies not previously known or in use in the United States. Although a draft of the Patent Act of 1790 provided for patents of importation, 99 the final version of the law that Congress passed during its first session did not authorize them. 100 The provision was removed on March 5, 1790, after debate in the House of Representatives. 101 Representative Thomas Fitzsimons explained that it had been removed because of "the Constitutional power being Questionable." 102 Correspondence reveals that James Madison-and possibly others-doubted the constitutionality of patents of importation. Tench Coxe, the assistant secretary of the treasury and a strong supporter of patents of importation as a means of encouraging American manufactures, wrote in a letter a few weeks later to Madison: I saw with regret the truth of [Madison's] apprehension, that the benefit of a patent could not be constitutionally extended to imported objects-nor indeed, if it were within the verge of the powers of Congress, do I think any clause to that effect could be safely modified. Private acts would be wise and safe, if they could be thought constitutional; but I think they cannot without an Amendment, by striking out all of the clause that follows the word "by" in the 8th parag. of the 8th Sec. of the first Article-or something to that purpose. 103 Coxe's letter makes evident that Madison did not believe that Congress could constitutionally provide for patents of importation because these patents seemed to lie outside of the means specified in the IP Clause, which allows patent rights to be conferred on inventors, not on importers of already-created inventions. For this reason, Coxe specified that the only constitutional way to legislate patents of importation would be to remove the means limitation specified in the IP Clause, thereby allowing the promotion of the progress of science and useful arts by any means, including through patents of importation.

104 A subsequent draft 105 of what became the 1791 Report on the Subject of Manufactures 106 by Alexander Hamilton also expressed doubt over the constitutionality of patents of importation.

107 This evidence has consequences for both the IP Clause's scope and its relationship to the other enumerated powers. As to the former, James Madison and others fairly clearly believed that Congress could not rely on the IP Clause to implement any means of promoting the progress of science and useful arts beyond the means specified in the Clause. Patents of importation represented one example of such impermissible means.

What has gone unobserved, however, is that this evidence also shows that Madison-and apparently others in the First Congressbelieved that Congress could not invoke any of its other enumerated powers to enact means of promoting the progress of science and useful arts. But for the IP Clause, Congress would likely have had the authority to provide for patents of importation pursuant to the Commerce Clause. Even under the more limited understanding of the Commerce Clause at that early juncture in American history, 108 the foreign-commerce powers would have authorized provisions for patents of importation, as those kinds of provisions would have regulated items of trade being imported from foreign countries into 108. See Gonzales v. Raich, 545 U.S. 1, 15-16 (2005) ("For the first century of our history, the primary use of the [Commerce] Clause was to preclude the kind of discriminatory state legislation that had once been permissible."); Richard A. Epstein, The Proper Scope of the Commerce Power, 73 VA. L. REV. 1387, 1389 (1987) (arguing that Alexander Hamilton's understanding of the word "commerce" was "restrictive by modern standards"). But see Balkin, supra note 46, at 5 (arguing that the commerce power was understood broadly during the eighteenth century).

109 That Madison and likely others thought that Congress could not establish patents of importation under the Commerce Clause strongly indicates that they thought that the IP Clause externally limited Congress's ability to promote the progress of science and useful arts by any means other than those laid out in the IP Clause itself.

The same conclusion can be drawn from Madison's views on granting land to encourage the importation of inventions. In his series of letters with Coxe following the excision of the provision concerning patents of importation from what became the first federal patent law, Coxe advocated that Congress "appropriat[e] a district of territory to the encouragement of imported inventions" if patents of importation could not directly be allowed. 110 Madison replied with skepticism as to the constitutionality of such land grants. He stated explicitly that Congress's powers of encouraging invention are limited to the means specified in the IP Clause, thus excluding the possibility of land grants: I can not but apprehend . . . that the clause in the constitution which forbids patents for [importation] will lie equally in the way of your expedient. Congress seem[s] to be tied down to the single mode of encouraging inventions by granting the exclusive benefit of them for a limited time, and therefore to have no more power to give a further encouragement out of a fund of land than a fund of money. This fetter on the National Legislature tho' an unfortunate one, was a deliberate one. The Latitude of authority now wished for was strongly urged and expressly rejected. 111 109. See e.g., United States v. Lopez, 514 U.S. 549, 585 (1995) (Thomas, J., concurring) ("At the time the original Constitution was ratified, 'commerce' consisted of selling, buying, and bartering, as well as transporting for these purposes."); Gibbons v. Ogden, 22 U.S. (9 Wheat.) 1, 189-90, 196-97 (1824) (finding that "commerce" includes "the commercial intercourse between nations"); Anthony J. Colangelo, The Foreign Commerce Clause, 96 VA. L. REV. 949, 962 (2010) (arguing that Congress's foreign-commerce power is robust because historically there has been an "overriding concern that the federal government speak with one voice when regulating foreign commerce"); Epstein, supra note 108, at 1394-95 (discussing the importance and purpose of the foreign-commerce power). The Supreme Court has noted that "[a]lthough the Constitution, [Article I, Section 8, Clause 3], grants Congress power to regulate commerce 'with foreign Nations' and 'among the several States' in parallel phrases, there is evidence that the Founders intended the scope of the foreign commerce power to be the greater." Japan Line, Ltd. v. County of Los Angeles, 441 U.S. 434, 448 (1979) ), pursuant to its power "To exercise exclusive Legislation in all Cases whatsoever, over such District (not exceeding ten Miles square) as may, by Cession of particular States, and the acceptance of Congress, become the Seat of the Government of the United States," U.S. CONST. art. I, § 8, cl. 17. With regard to Washington, D.C., the seat of the U.S. government, Congress's powers resemble the powers reserved to the states and include those otherwise forbidden to support the premise that the Constitutional Convention's rejection of a university power as a means of promoting the progress of science and useful arts precluded Congress from fulfilling President Washington's request.

Similar is the earliest Congress's view as to whether the federal government could fund artistic, scientific, and technological work, a means not specified in the IP Clause. In 1789, a man named John Churchman asked the House of Representatives to support his invention for determining longitude based on the magnetic variation at places of known latitude. Churchman asked Congress to help him in two ways: first, by awarding him exclusive rights in the invention; and second, by funding a voyage to Baffin Bay to help him further develop the invention. 116 A congressional committee debated this request, readily concluding that Churchman ought to receive an exclusive right for a limited term in exchange for publishing his invention.

117 Representative Thomas Tucker of South Carolina, however, questioned whether the IP Clause would allow the legislature "to go further in rewarding the inventors of useful machines, or discoveries in sciences, than merely to secure to them for a time the right of making, publishing and vending them." 118 Similarly, Roger Sherman expressed doubt that Congress could comply with Churchman's request for funding:

[Our grant of an exclusive right to him] appears gone as far as proper to go at this time, as far as warranted by the Constitution. . . . If have a right to go further and lay out money it must be upon-Gentleman has fruitful invention. Large sums might be expended which finally might be to no advantage. The committee thought fit to go as far as this to promote the progress; they did not think proper to give any further power to encourage this useful discovery. 119 Congress. Therefore, even if Congress is forbidden via the IP Clause from creating universities pursuant to its other enumerated powers, its power to do so pursuant to its plenary power over the District of Columbia seems to trump the IP Clause's external limitations. Cf Other representatives questioned the power to fund as well. 120 The committee tabled the funding request given "the present deranged state of [the federal government's] finances." 121 No resolution in hand, Churchman renewed his request for funding, and a congressional committee ultimately rejected it in 1791 because of the constitutional questions it raised, questions the committee did not want to explore. 122 In 1796, a congressional committee stated this concern more pointedly: "[A]pplication to Congress for pecuniary encouragement of important discoveries, or of useful arts, cannot be complied with, as the Constitution of the United States appears to have limited the powers of Congress to granting patents only."

123 Of all those who voiced an opinion on the matter, none appears to have thought that the means specified in the IP Clause might encompass the funding of artistic, scientific, and technological works.

This discussion is instructive when placed alongside another broader contemporaneous constitutional debate. At issue in that debate was whether the Spending Clause-authorizing Congress to "provide for the common Defence and general Welfare of the United States" 124 -enables Congress to spend on things beyond what the enumerated powers otherwise permit. James Madison argued that because "the United States is a government of limited and enumerated powers, the grant of power to tax and spend for the general national welfare must be confined to the enumerated legislative fields committed to the Congress." 125 Taking the opposite position, Alexander Hamilton contended that "the clause confers a power separate and distinct from [other enumerated powers] . . . [and is] limited only by the requirement that it shall be exercised to 120. For example, Representatives Joshua Seney, John Page, and Alexander White all questioned the government's authority to fund the invention's further development. Id provide for the general welfare of the United States." 126 Whether the Spending Clause allowed spending on items not included in the enumerated powers-or in the specific context here, on items not permitted by the IP Clause's means-was very much a live issue at the time. 127 In the context of Churchman's funding request, however, the committee members thought that Congress likely lacked the authority to fund Churchman's scientific research despite the Spending Clause.

In sum, the evidence from early congressional activity overwhelmingly suggests that the Framers believed that the IP Clause could be employed to promote the progress of science and useful arts only by using the means specified in that Clause and that Congress could not use its other enumerated powers to employ any other means.

The Supreme Court has not yet addressed the IP Clause's external limitations directly. Nonetheless, its cases construing the IP Clause and constitutional structure generally are consistent withand oftentimes are supportive of-an IP Clause that externally limits Congress's other powers.

The Supreme Court has stated numerous times that the power conferred on Congress by the IP Clause is specifically a power "to promote the Progress of Science and useful Arts" and that the means to achieve that goal are specified in the latter part of the Clause.

The Supreme Court has also examined the reach of the IP Clause. But the Court's discussions-frequently in dicta-have usually occurred in the context of determining the internal constraints of the IP Clause itself, not in the context of determining whether the IP Clause might impose affirmative limits on Congress's other enumerated powers. For example, in a 1989 preemption case, the Supreme Court noted,

[T]he Clause contains both a grant of power and certain limitations upon the exercise of that power. monopolies of unlimited duration, nor may it "authorize the issuance of patents whose effects are to remove existent knowledge from the public domain, or to restrict free access to materials already available." 129 Similarly, the Court stated in an earlier case interpreting patent law's nonobviousness requirement that "[t]he [IP] clause is both a grant of power and a limitation" and that "Congress in the exercise of the patent power may not overreach the restraints imposed by the stated constitutional purpose." 130 Whether one ought to infer from these statements that the IP Clause limits Congress's overall ability to regulate intellectual property outside of the means provided in the Clause is not entirely clear.

The IP Clause's external limitations can be more readily inferred from the Supreme Court's 1879 decision in the Trade-Mark Cases.

In that case, the Court struck down as unconstitutional the first federal trademark law, which Congress had enacted pursuant to the IP Clause.

132 The Court's reasoning was muddled, but it suggested that the decision was resolved this way because trademark protection has nothing-at least proximately-to do with promoting the progress of science and useful arts. The Court indicated that the IP Clause designates the means that Congress may use-"by securing for limited times, to authors and inventors, the exclusive right to their respective writings and discoveries"-to promote the progress of science and useful arts. 133 The Court observed that "[t]he ordinary trade-mark has no necessary relation to invention or discovery" and that a trademark should not be understood as a "writing[]" within the bounds of the IP Clause. 134 Because trademark law's purpose is not to This construction is consistent with-and supportive of-the notion of an IP Clause that externally limits Congress's other powers. It suggests that if a law is not aimed at promoting the progress of science and useful arts, then Congress may not enact the law pursuant to the IP Clause and may enact the law only if it finds another enumerated power as a hook. 137 This reasoning constructs a channeling mechanism of sorts: laws concerned with the promotion of the progress of science and useful arts ought to be channeled into the purview of the IP Clause and must comply with its limiting means, and those that are not must be channeled elsewhere, into the purview of another enumerated power. 138 In the Trade-Mark Cases, the Court was called upon to deal with the latter situation. It has not, however, been called upon to address directly the former situation.

The Supreme Court's discussion of the IP Clause in Eldred v. Ashcroft 139 in 2003 is also consistent with my understanding. In that case, the Court upheld the constitutionality of Congress's 1998 copyright-term extension for existing works protected by copyright, a move that had the effect of placing those existing works on equal durational footing with new works that were also covered by the term those goods." Id. at 37. The Court rejected the latter, broader possible interpretation because it would "creat[e] a species of perpetual patent and copyright, which Congress may not do." Id 140 Relying on text, history, and precedent, the Court held that the "limited Times" for which Congress may grant copyright protection includes the practice of extending copyright protection for already-protected works from one limited time to another. 141 More relevant for the purposes of this Article, however, is the tripartite framework the Supreme Court applied to evaluate the law's constitutionality. First, the Court looked to the structure of the termextension law to see if there was any "cause to suspect that a purpose to evade the 'limited Times' prescription [had] prompted Congress" to enact it. 142 This analytical move suggests that future analysis might do well to look to an intellectual-property law's structure and motivation to evince the law's purpose. Second, although the Supreme Court accorded no deference to Congress in analyzing what "limited Times" means and whether the copyright law at issue truly did provide protection for only "limited Times," the question of whether the law was a proper exercise of Congress's power under the IP Clause was one the Supreme Court evaluated by measuring-with substantial deference to the legislature-whether the law was "a rational exercise of the legislative authority conferred by the [IP Clause]."

143 Third, the Court recognized, as it had in previous decisions, 144 that promotion of the progress of science and useful arts is the purpose for which Congress may enact intellectual-property legislation and that Congress may do so by using any of the means outlined in the Clause.

The combination of these three analytical components supports my understanding of the IP Clause. By invoking these components, the Court suggested that when Congress acts via legislation to promote the progress of science and useful arts, it must choose to do so by the means specified in the IP Clause. The Court will uphold any legislation that has the end of promoting the progress of science and useful arts as long as it secures exclusive rights to authors and inventors in their writings and discoveries for limited times. If legislation with this purpose lacks any of these qualities-(1) securing exclusive rights (2) to authors and inventors (3) for their writings and discoveries (4) for limited times-as these qualities have been defined by the courts, then the Court will strike down the law as unconstitutional.

This understanding refutes Justice Stevens's claim in his Eldred dissent that the majority was reading the IP Clause "to provide essentially no limit on congressional action under the Clause." 146 To the contrary, the majority indicated that when Congress seeks to promote the progress of science and useful arts, it must act within the Clause's means-by providing rights only for "limited Times," as construed by the Court, for example. The Court's opinion suggests a substantial limit on congressional action, one the First Congress willingly imposed on itself with regard to patents of importation, land grants, research funding, and establishment of a national university.

147 It also suggests a significant restraint for modern times, one that I explore in questioning the constitutionality of multiple federal laws.

The Supreme Court has not said much more on limitations that the IP Clause imposes on other enumerated powers. Yet Court opinions in other constitutional areas shed light on the structural relationship between the IP Clause and other congressional powers, a relationship that supports the notion of an IP Clause that imposes external limitations on those powers.

In the landmark civil-rights case of Heart of Atlanta Motel, Inc. v. United States, 148 the Supreme Court upheld federal legislation banning racial discrimination in public accommodations. 149 The Court reasoned that the ban was a valid exercise of Congress's commerce power, as it was "carefully limited to enterprises having a direct and substantial relation to the interstate flow of goods and people." 150 The Court refused to accept the argument that the law should be struck down as outside the scope of Congress's powers merely because the Civil War amendments had not expressly granted the power to enact the law. 151 The Court reasoned that as long as the Commerce Clause 146. Id. at 230 (Stevens, J., dissenting); see also id. at 243 (Breyer, J., dissenting) ("The majority believes these conclusions rest upon practical judgments that at most suggest the statute is unwise, not that it is unconstitutional.").

147. could be held to grant the power, the specific authority conferred by the amendments was irrelevant. 152 Some scholars read the Trade-Mark Cases and Heart of Atlanta Motel as permitting Congress to legislate under one power when another power is closed off. For example, Professor Nachbar states that these cases suggest that Congress can use its other Article I, Section 8 powers to legislate when the IP Clause's limitations bar it from acting under that power.

153 Professor Nachbar's reading, however, is unwarranted. In Heart of Atlanta Motel, no reason existed to support the implication that the Civil War amendments' failure to ban racial discrimination in public accommodations prevented Congress from using one of its enumerated powers to accomplish that goal. The Court perceived no basis to infer a conflict between those amendments and Congress's other powers. The situation is different in the context of the IP Clause. The history, text, and structure discussed in Sections A and B provide strong reasons to see just such a conflict between the IP Clause and Congress's other enumerated powers. The analogy to Heart of Atlanta Motel, therefore, is inapposite.

More on point is the Supreme Court's decision about the relationship between the Bankruptcy 154 and Commerce 155 Clauses in Railway Labor Executives' Ass'n v. Gibbons. 156 In that case, Congress had provided economic benefits for the employees of a railroad company that had declared bankruptcy. 157 The Court rejected the argument that the action had been an exercise of Congress's commerce power on the basis that the clause's structural purpose is to allocate resources in a bankruptcy situation. 158 Next, the Court examined whether the law was a valid exercise of Congress's bankruptcy power. The Court ruled that it was not because the law had made bankruptcy laws nonuniform by causing the particular provisions at issue to conflict with the general bankruptcy provisions' priorities. 159 The Court reasoned that Congress could not invoke the Commerce Clause to enact nonuniform bankruptcy provisions because "the Bankruptcy Clause itself contains an affirmative limitation or restriction upon Congress' power: bankruptcy laws must be uniform throughout the United States." 160 Although Professor Nachbar suggests that Gibbons was incorrectly decided or should not be broadly applied to other Article I, Section 8 powers, 161 the Court's reasoning applies even more strongly to the IP Clause. The historical and other evidence that the IP Clause should limit Congress's other enumerated powers is more robust than any evidence of the negative implications one might draw from the uniformity requirement of the Bankruptcy Clause. 162 As such, the structural reasoning that the Court employed in the bankruptcy context ought to apply even more readily to the IP Clause to forbid Congress from evading the limited means allowed for promoting the progress of science and useful arts by invoking another enumerated power.

Having discussed the relevant judicial doctrine, I now turn to an analysis of policy considerations informing the IP Clause to provide another window into understanding the Clause's external limitations. (1935), which involved an alleged clash between Congress's power to establish a monetary system, U.S. CONST. art. I, § 8, cl. 5, and its power "To borrow Money on the credit of the United States," id. art. I, § 8, cl. 2. The Court in Perry held that "Congress can[not] use th[e first] power so as to invalidate the terms of the obligations which the government has theretofore issued in the exercise of the power to borrow money on the credit of the United States." Perry, 294 U.S. at 350. It reasoned that doing so would negate the promise inherent in the federal government's power to borrow money on U.S. credit. Id. at 353.

161. See Nachbar, supra note 12, at 308, 313 (reasoning that Gibbons would be "subject to failure if broadly applied" and finding that "blindly applying Gibbons would prevent federal regulation of purely intrastate commercial conduct . . . involving intellectual property").

162. Cf. Jane C. Ginsburg, No "Sweat"?: Copyright and Other Protection of Works of Information After Feist v. Rural Telephone, 92 COLUM. L. REV. 338, 370 (1992) ("Justice Rehnquist declared: 'If we were to hold that Congress had power to enact nonuniform bankruptcy laws pursuant to the Commerce Clause, we would eradicate from the Constitution a limitation on the power of Congress to enact bankruptcy laws.' Under this approach, one could contend that a law protecting compiled information under the Commerce Clause would similarly be invalidated as an attempt to elude a substantive limitation on Congress' power to grant copyrights." (footnote omitted) (quoting Gibbons, 455 U.S. at 468-69)).

If the provided rights were exceedingly extensive, society would be hurt and social welfare diminished. 170 Exclusive rights in intellectual property prevent competition in protected works, often allowing the rights holder to charge a premium for access and ultimately limiting these valuable works' diffusion into society. 171 Moreover, given that knowledge is frequently cumulative, society benefits when subsequent creators are not prevented from building on previous artistic, scientific, and technological creations to generate new works. 172 Therefore, copyright and patent laws ensure both that the works that they protect will fall into the public domain in due course and that third parties will be free to use the protected works for certain socially valuable purposes.

173 At bottom, utilitarian theories of intellectual property rest on the premise that the benefit to society of creators' crafting valuable works offsets the costs to society of the incentives the law offers to those creators. 174 The power conferred on Congress by the IP Clause reflects a desire to reach a balance between granting creators the exclusive rights to create and disseminate publicly valuable works and restricting those rights so that the public is not hurt. 175 The Supreme Court has stated as much in numerous decisions. 176 This principle is longstanding.

177 Congress may thus act to secure intellectual-property protection as long as it does not upset the carefully calibrated balance established by the IP Clause. ("It is self evident that on the expiration of a patent the monopoly created by it ceases to exist, and the right to make the thing formerly covered by the patent becomes public property. It is upon this condition that the patent is granted.").

In this vein, the Supreme Court has strongly indicated that the IP Clause contemplates the existence of a public domain that cannot be undone. For example, in the context of discussing the relationship between federal powers under the IP Clause and state authority, the Court has noted that the ultimate goal of the patent system is to bring new designs and technologies into the public domain through disclosure. State law protection for techniques and designs whose disclosure has already been induced by market rewards may conflict with the very purpose of the patent laws by decreasing the range of ideas available as the building blocks of further innovation. . . . To a limited extent, the federal patent laws must determine not only what is protected, but also what is free for all to use. 179

The policy of maintaining a balance between public benefit and intellectual-property rights for creators is deeply utilitarian in promoting the progress of science and useful arts. Laws that upset this calibrated balance by transgressing the limits set by the IP Clausesay, by offering rights of unlimited duration or by offering rights to someone other than the creator-are therefore particularly troublesome. In this sense, policy interests support the idea that Congress is disallowed from using any of its powers to promote the progress of science and useful arts other than the power conferred by the IP Clause itself.

That said, one might readily imagine Congress's using means other than those specified in the IP Clause that, as a policy matter, do not upset the balance envisioned by the IP Clause. For example, federal funding of scientific research or cultural works would seem, as a policy matter, to promote the progress of science and useful arts without undermining public utility. At the right price, the cost of such funding would be exceeded by the public benefit, all without the monopoly-like effects that can result from patent and copyright laws.

On the whole, then, intellectual property's utilitarian basisreflected in the IP Clause-supports the Clause's external limitations, at least as far as Congress might use its other powers to subvert the means set out in the IP Clause. To the extent that Congress would use its other powers to employ means that do not affect the calibrated 179. Bonito Boats, 489 U.S. at 151; see also Graham v. John Deere Co., 383 U.S. 1, 6 (1966) ("Congress may not authorize the issuance of patents whose effects are to remove existent knowledge from the public domain, or to restrict free access to materials already available."). balance set out in the IP Clause's text, these same policy concerns are not necessarily implicated.

This Part demonstrates that the combination of textual and structural analysis, historical evidence, judicial doctrine, and policy favors the existence of external limitations in the IP Clause that operate to ban Congress from using its other enumerated powers to promote the progress of science and useful arts via means not authorized by the IP Clause.

Some other scholars working on this topic conclude that the IP Clause limits Congress's other powers, but they take different approaches from the one advanced in this Article-approaches that either require more complex analytical frameworks or are less grounded in the foregoing evidence. For one, former Professor Patry does think that the IP Clause restricts Congress's other powers. He reads the Supreme Court's doctrine as holding that the IP Clause "contains both positive and negative rights: a positive right to grant authors a limited monopoly in their original material, and a negative right in the public to copy unoriginal material." 180 Patry elaborates that "these two rights are mutually exclusive: either an author has an exclusive right over given material, or the public has a right to copy that material. Both rights cannot be exercised at the same time with respect to the same material." 181 As such, Patry claims that Congress may not provide for perpetual intellectual-property protection, as doing so would contravene the "limited Times" restriction of the IP Clause. 182 Similarly, according to Patry, databases lacking sufficient constitutional originality to qualify for copyright protection should not be protectable under legislation enacted pursuant to one of Congress's other powers.

183 Former Professor Patry's analysis does not go far enough in light of the evidence presented in this Part. By insisting that Congress's powers are restricted only with regard to unoriginal material, Patry does not address the evidence supporting an understanding that all means that are not specified in the IP Clause are off-limits. Like former Professor Patry, Professor Yochai Benkler proposes that the IP Clause is "a specifically limited grant available only to protect original contributions to the wealth of human knowledge" 184 and that Congress cannot legislate beyond those bounds, even using its other powers. 185 More specifically, Benkler suggests that Congress may create [IP] rights if and when they are likely to encourage information production more than they inhibit it, if it makes them available only to those who make original contributions to the wealth of our collective knowledge, and if the exclusive right enacted does not remove, or burden free access to, information already in the public domain.

Benkler suggests that Congress can use its other powers to "regulate information markets" only if the laws that it creates are different in kind from those authorized under the IP Clause. 187 To derive these limitations, Benkler relies on doctrinal and policy analysis, 188 along with what seems to be an implicit argument that the IP Clause's restrictions would be too easy to evade if they did not limit Congress's other powers. Under his framework, Benkler concludes that some possible legislation protecting databases is unconstitutional. 189 Professors Paul Schwartz and William Treanor call Professor Benkler's analysis "a highly unconventional originalist linguistic argument, which turns on the Framers' alleged understanding of a single word: 'Progress. '" 190 In any event, like former Professor Patry's position, Benkler's analysis does not go as far as it might in light of the myriad pieces of evidence suggesting that all means not specified in the IP Clause are off-limits to Congress.

Professors Paul Heald and Suzanna Sherry also think that the IP Clause limits Congress's other powers. They rely on early American history, which they read as reflecting a widespread suspicion toward governmental grants of exclusive rights, to suggest that "the Intellectual Property Clause was included as both a grant of power to Congress and also an absolute limitation of its power." 191 Based on this reading, they identify five principles that delineate the reach of Congress's IP powers: the "Suspect Grant Principle," which provides that "the limiting language of the Clause . . . [applies] only to legislation that imposes monopoly-like costs on the public through the granting of exclusive rights"; 192 the "Quid Pro Principle," under which laws enacted pursuant to the Clause must be premised on "the author or the inventor . . . giv[ing] the public something it did not have before to earn a grant of exclusive rights from Congress"; 193 the "Authorship Principle," which dictates that the creator must receive the grant of exclusive rights;

194 the "Public Domain Principle," which specifies that the rights should last for a limited time, after which point the protected work must fall into the public domain;

195 and the "Flexibility Principle," which provides Congress with broad discretion to decide how to wield its powers under the IP Clause. 196 Heald and Sherry conclude that Congress may not use its other powers, such as the commerce or treaty powers, to enact laws that flout these principles. 197 They suggest that these principles can holistically guide decisions on various laws' constitutionality. 198 For example, they use their framework to suggest that the copyright-term extension that the Supreme Court upheld in Eldred 199 is unconstitutional because it violates the Quid Pro Principle and the Public Domain Principle. 200 Although the holistic framework of Professors Heald and Sherry shares points of commonality with the more streamlined approach I take, it is too complex both to conceptualize and to apply.

To sum up, this Part surveys considerable evidence from multiple vantage points-text, structure, history, judicial doctrine, and policy-that supports the IP Clause's external limitations. This Part thus refutes the claims of Professor Thomas Nachbar that the IP Clause does not reach out to limit Congress's other powers. Moreover, it yields a framework for evaluating laws' constitutionality, a subject to which I now turn.

This Part builds on Part I's evidence of the IP Clause's external limitations to suggest a framework for evaluating a law's constitutionality. Generally, if a law has only the structural purpose of promoting the progress of science and useful arts, it must comply with the means described in the IP Clause. If it does not comply with those means, it is unconstitutional. If the law lacks the structural purpose of promoting the progress of science and useful arts, it is outside the IP Clause's purview and must be enacted pursuant to another of Congress's powers, as is the case with trademark law. If, in addition to the structural purpose of promoting the progress of science and useful arts, the law has another legitimate purpose pursuant to Congress's other powers, such as promoting trade or foreign relations, Congress may act beyond the IP Clause's means only if it overcomes a presumption against the law's constitutionality. That is, the law ought to be deemed constitutional only if clear and convincing evidence demonstrates that Congress intentionally chose to supersede the IP Clause's means because of paramount interests pursuant to its other more permissive powers. This presumption ought to be extremely hard to overcome when a law's means subvert those in the IP Clause-such as by granting perpetual copyright protection-as compared to a law whose means merely are not contained in the IP Clause-such as funding for scientific research.

Section A of this Part discusses how to assess whether a law's function is to promote the progress of science and useful arts through the lens of structural purpose. It also addresses how to analyze laws with multiple legitimate constitutional purposes, creating a presumption framework for assessing constitutionality in those cases. Section B then turns to some of the most likely sources of conflict with the IP Clause: the commerce, spending, necessary-and-proper, and treaty powers.

Honoring the IP Clause's external limitations requires an understanding of when Congress is seeking to promote the progress of science and useful arts to help determine whether a given law should be understood to be within the IP Clause's purview. 202 This Section explores how to understand the first part of the IP Clause. Broken down, "To promote the Progress of Science and useful Arts" generally refers to the goal of encouraging the advancement of systematic knowledge, cultural knowledge, and technology. I then work out how to assess whether a law has this end and thus falls within the scope of the IP Clause, meaning that it may use only the IP Clause's specified means.

According to the Supreme Court, "to promote" as used in the IP Clause means "'to stimulate,' 'to encourage,' or 'to induce.'" 203 Most commentators understand "progress" to mean advancement. 204 As Professor Solum explains further, "progress" can be understood either as "advancement of learning [with a] focus on the results of scientific activity" or as "encouraging the activity itself [with a] focus on the process itself." 205 The first understanding can further mean improvement in a knowledge base's quality or quantity.

Rejecting an approach like Professor Solum's, Professor Malla Pollack concludes on the basis of some historical linguistic evidence that "progress" means spread, diffusion, or distribution. 206 She rejects Solum's possible meanings in large part because under those meanings, she argues that "to promote the progress of science and useful arts" would have the same meaning as "to promote science and useful arts." 207 Pollack's objection, however, is unpersuasive. The former phrase might straightforwardly be understood to mean that the Framers wanted to provide Congress the power to promoteboth directly and indirectly-the movement of science and useful arts in forward directions. This capacious ambition would not have been communicated by the latter language. Moreover, as Solum observes, the linguistic evidence Pollack presents is consistent with his favored understanding and in fact represents the more usual understanding of the term "progress."

208 I thus presume that Solum's understanding is correct, although my framework would also work with Pollack's definition.

"Science" as it appears in the IP Clause did not originally have the meaning contemporary Americans associate with it-biology, chemistry, and the like. Instead, at the time of the Framing, science meant knowledge or learning, particularly of the kind that is systematic and of enduring value. 209 That meaning makes particular sense in light of the fact that one of James Madison's four constitutional proposals that yielded the IP Clause sought "the advancement of useful knowledge." 210 For the most part, this goal of advancement corresponds to what the Constitution encourages in copyright law: the production of cultural goods and knowledge.

The term "useful arts," confusingly enough, has nothing to do with cultural or fine arts. Rather, at the time of the Framing, it meant "helpful or valuable trades," such as mechanical and civil engineering.

211 Useful arts were understood to be skills that were practical rather than theoretical.

They were principally what modern Americans have come to understand as technology, 213 and they are what patent law is intended to promote.

In sum, then, the evidence seems to show that a law is about promoting the progress of science and useful arts if it seeks to encourage advancement in areas of systematic knowledge, including cultural knowledge or technology.

This understanding provides a definition to help assess whether a law is focused on promoting the progress of science and useful arts so 208. See Solum, supra note 13, at 46-47 ("[G]iven that the common-sense interpretation of the term 'progress' involves a figurative use, the prevalence of the underlying literal use only reinforces the possibility that the term was used in the common figurative sense.").

; see also Solum, supra note 13, at 3 ("[Th]e meaning of science that best coheres with the constitutional text and the original understanding can be glossed as systematic knowledge or learning of enduring value . . . ."); id. at 51-52 (identifying the earliest federal copyright law's reference to "encouragement of learning" as a synonym for "promot[ing] the Progress of science" (internal quotation marks omitted)).

210. as to tell whether Congress must respect the IP Clause's external limitations. But observers have many ways with which to measure whether a law has this purpose, such as by gauging different legislators' views or by looking to legislation's stated purposes. Canvassing different legislators' views is very slippery, as assigning a unified purpose to a multimember body is difficult. 214 At the other extreme, looking conclusively to a law's stated purpose-or, similarly, to its stated constitutional authority-gives Congress the easy escape valve of stating one purpose-or power-when in fact it has another in mind. Relying too much on this measurement would permit Congress to evade the IP Clause's external limitations. Another possible test would be to investigate whether a law actually has the effect of promoting the progress of science and useful arts. Such a test is not ideal, however, both because judging legislation's actual effect can be difficult at the outset and because courts might not be institutionally competent to make such a determination.

Additionally, the Supreme Court has emphasized in numerous cases that Congress should not be second-guessed on whether its laws enacted under the IP Clause actually promote the progress of science and useful arts. 216 Instead, I suggest that the optimal way to measure whether a law aims to promote the progress of science and useful arts is to assess its structural purpose. This approach is consistent with the way the Supreme Court typically evaluates a law's purpose to determine its constitutionality, including in the context of intellectual property. 217 In Eldred, the Court reasoned that a helpful mode of analysis was to 214. See assess a law's purpose structurally. 218 In Gibbons, the Court assessed the relevant law's constitutionality by reasoning that its structural purpose was to allocate resources in a bankruptcy situation and that it must therefore comply with the Bankruptcy Clause's limitations.

220 numerous Court decisions also suggest that an analysis of structural purpose is essential to ensure that Congress is acting within its powers.

As Professor John Manning notes, "Few would deny the possibility of gleaning a statute's overall purpose from its structure or from the aims suggested by the text itself." 222 Although reasonable minds might differ as to a law's structural purpose, the law ultimately passed by Congress often gives significant clues as to the purpose or power used. For example, relevant pieces of information include the intended effects or causes of the legislation, the stated effects or causes, and the statute's desired placement alongside other laws with a similar purpose or power.

This reasoning suggests that if Congress is acting with the structural purpose of promoting the progress of science and useful arts, it must comply with the IP Clause's means. If it employs other means for that purpose, it is acting beyond its powers.

How, though, should courts analyze laws that have multiple material purposes? For example, a law providing copyright or patent protection to foreign creators might be aimed at promoting the progress of science and useful arts, but it might also be aimed at promoting interstate and foreign commerce. Or a law regulating intellectual property pursuant to a treaty obligation of the United States might be directed both at promoting the progress of science 221. See id. at 423 ("Should Congress, in the execution of its powers, adopt measures which are prohibited by the constitution; or should Congress, under the pretext of executing its powers, pass laws for the accomplishment of objects not entrusted to the government; it would become the painful duty of this tribunal . . . to say that such an act was not the law of the land."); see also Soon Hing v. Crowley, 113 U.S. 703, 710 (1885) ("[T]he rule is general with reference to the enactments of all legislative bodies that the courts cannot inquire into the motives of the legislators in passing them, except as they may be disclosed on the face of the acts, or inferrible from their operation, considered with reference to the condition of the country and existing legislation."); The Head Money Cases, 112 U.S. 580, 595 (1884) (looking to the structural purpose of a federal law levying a per-passenger fee on vessels from foreign ports to determine whether the law was an exercise of Congress's commerce power).

222. John F. Manning, The Absurdity Doctrine, 116 HARV. L. REV. 2387, 2408 n.75 (2003).

and useful arts and American foreign-relations interests. Congress, in such cases, has multiple legitimate constitutional interests, and those interests may point in competing directions. For example, foreignrelations or trade interests might suggest that a law ought to extend beyond the IP Clause's means, whereas intellectual-property interests would counsel otherwise. In these cases, there are competing considerations. On the one hand, the IP Clause, as demonstrated herein, externally limits Congress's other powers. On the other, Congress can have purposes other than intellectual-property regulation in mind in enacting a law.

To resolve these competing considerations, a presumption framework is helpful to ensure that if Congress is evading the IP Clause's means, it is doing so because it feels compelled to do so due to competing constitutional interests. 223 That is, laws that have a purpose of promoting the progress of science and useful arts ought to be presumed unconstitutional unless clear and convincing evidence shows that Congress considered superseding the IP Clause's means important due to other legitimate, material constitutional interests, such as foreign relations or commerce. Of course, these other interests must be legitimate and not merely asserted as a sham to bypass the IP Clause's external limitations. 224 Otherwise, laws with multiple purposes would easily evade the IP Clause's limited means, which together operate as a sensible restriction on Congress's authority.

This presumption framework would, as Professors Samuel Issacharoff and Richard Pildes note in a different context, place "the judicial emphasis on second-order issues of appropriate institutions and processes, through which courts seek mainly to ensure that the 167, 189 (1992) (refusing to distinguish between laws with a particular single purpose and those with multiple purposes, "lest a legislature be able to piggy-back legislative purposes to enable it to do that which the Constitution forbids").

226 This process-oriented approach would take cognizance of Congress's better placement, as compared to courts', to evaluate how to trade off between multiple policy interests when more than one constitutional authority is at stake in the context of intellectual property. 227 This presumption framework would transform the constitutional inquiry into one of statutory interpretation, one with which courts ought to feel more comfortable.

This presumption of unconstitutionality ought to be harder to overcome in certain circumstances than in others. In particular, Congress might enact laws that subvert the IP Clause's means, such as by enacting perpetual copyright protection, granting patent rights to someone other than the inventor, or granting protection to works that are neither writings nor discoveries.

229 Such laws would contravene the IP Clause's external limitations most directly, by upsetting the calibrated balance of the IP Clause, a balance between a grant of incentives to authors and inventors and a broader public benefit. 230 In such situations, the presumption of unconstitutionality ought to be extremely hard to overcome. By comparison, the presumption ought to be somewhat easier to overcome for a law whose means merely are not contained in the IP Clause, such as funding for scientific research.

In sum, then, legislation with the material structural purpose of promoting the progress of science and useful arts must use only the The means specified in the IP Clause require "securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries." 232 In relevant part, then, as elaborated in the next Section and in Part III, to comply with those means, a law must include a grant of an exclusive right that is limited in time to someone who qualifies as an author of something that is his or her writing, or to someone who qualifies as an inventor of something that is his or her invention.

Before delving in the next Part into some federal laws of dubious constitutionality in light of the IP Clause's external limitations, I first analyze the four constitutional powers most likely to conflict with, and to provide cover for evading, the Clause's external limitations: the commerce, spending, necessary-and-proper, and treaty powers. After setting out the nature of these possible collisions-in the simple case of a law's having the single purpose of promoting the progress of science and useful arts-I return to the presumption framework. I return to that framework to examine how extensively it might efface the IP Clause's external limitations in light of a nuanced understanding of the varied constitutional interests that might be invoked in enacting a law.

1. Commerce Clause. Although most think the Commerce Clause, as originally understood, described only a narrow power, 233 its reach has grown to be extensive. The Commerce Clause has been described by the Supreme Court as authorizing three broad classes of regulation: (1) laws that "regulate the channels of interstate commerce"; (2) laws that "regulate and protect the instrumentalities of interstate commerce"; and (3) laws that "regulate activities that substantially affect interstate commerce."

234 Included in this third category are "purely local activities that are part of an economic 'class of activities' that have a substantial effect on interstate commerce." 235 Furthermore, the Court has said that "Congress can regulate purely intrastate activity that is not itself 'commercial,' in that it is not produced for sale, if it concludes that failure to regulate that class of activity would undercut the regulation of the interstate market in that commodity." 236 Although these categories might betray some limitations on Congress's commerce power, the power is, nevertheless, quite extensive.

But for the IP Clause's external limitations, most intellectualproperty legislation could likely be enacted under the cover of this expansive understanding of the Commerce Clause. Artistic, scientific, and technological works that are encouraged by intellectual-property legislation can typically be linked to activities that substantially affect interstate commerce.

If a law were aimed at promoting the progress of science and useful arts but failed to comply with the IP Clause's means, then the federal government would be likely to defend that law by reference to the Commerce Clause. 237 As the discussion in Part I reveals, however, Congress is acting outside of its constitutional authority when it attempts to use the Commerce Clause to promote the progress of science and useful arts while not restricting itself to the IP Clause's means, unless Congress can show clearly and convincingly that it had legitimate material interests under the Commerce Clause that necessitated a transgression of the IP Clause's limited means.

The collision between the Commerce and IP Clauses is direct. The other three powers that might collide with the IP Clause-the spending, necessary-and-proper, and treaty powers-to which I will now turn, clash less obviously or directly, if at all. Spending Clause expansively, ensuring that even these limitations are narrowly applied.

Although Madison and Jefferson may in fact have had the more appropriate understanding of the Spending Clause, particularly in view of early congressional practice, the Supreme Court's holdings seem to foreclose the return of that view. This reality, however, does not necessarily mean that Congress may spend to promote the progress of science and useful arts in violation of the IP Clause's means. Much in the same fashion that the Commerce Clause is constrained by the IP Clause's limited means, a strong case can be made that Congress cannot rely on the Spending Clause to initiate spending to promote the progress of science and useful arts. Even accepting that the Spending Clause is an independent power, one might still see the IP Clause as providing an external limit on that Article I power. 257 In fact, the Supreme Court's understanding that other constitutional limitations can restrict the Spending Clause's reach could be understood to prohibit congressional spending to promote the progress of science and useful arts using means other than those laid out in the IP Clause. Nonetheless, given the Supreme Court's expansive understanding of the Spending Clause over time, the Court might not as readily find that the IP Clause's limitations do in fact inhibit Congress's authority under the Spending Clause. 'convenient, or useful' or 'conducive' to the authority's 'beneficial exercise.'" 259 As the Court explained in McCulloch v. Maryland, "necessary" does not require that the action be "absolutely necessary." 260 Rather, as McCulloch explained, "Let the end be legitimate, let it be within the scope of the constitution, and all means which are appropriate, which are plainly adapted to that end, which are not prohibited, but consist with the letter and spirit of the constitution, are constitutional."

261 A law is proper under this Clause, then, if it "constitutes a means that is rationally related to the implementation of a constitutionally enumerated power." 262 Does this power enable Congress to legislate to promote the progress of science and useful arts beyond the means specified in the IP Clause? Not wholly. Rather, the Supreme Court's jurisprudence suggests that, at most, the Necessary and Proper Clause gives Congress the power to legislate to help implement the IP Clause. Because the IP Clause specifies the limited means by which Congress may legislate to achieve a specified end, Congress can use the Necessary and Proper Clause to promote the progress of science and useful arts, but only by legislating in ways that support the means specified in the IP Clause. The Clause does not allow Congress to use other means more broadly to promote the progress of science and useful arts. That is, Congress can establish a patent office to examine patents before granting them and can require or encourage deposits of copyrighted materials in a central repository, even though these means are not literally within the IP Clause's scope. What is less clear is whether Congress can use other means further afield from the implementation of copyright and patent laws. I return to this issue in the next Part as I consider federal funding of artistic and scientific works.

This analysis is consonant with the Supreme Court's interpretation of the Necessary and Proper Clause as a source of congressional power. That is, congressional exercises of power under the Clause are not constricted by the IP Clause insofar as they seek to help execute the means and ends of the IP Clause. Consider that in McCulloch, Chief Justice Marshall indicated that the Necessary and Proper Clause cannot be invoked by Congress to enact a law that is forbidden by the Constitution, even if the Clause would otherwise permit it. 263 Because, as this Article argues, Congress is forbidden from using means other than those laid out in the IP Clause to promote the progress of science and useful arts, in certain instances Congress may likewise be forbidden from pursuing the same goal generally via the Necessary and Proper Clause. Congress may, however, use the Necessary and Proper Clause to support the means it employs in service of the IP Clause's ends.

4. Treaty Powers. The federal government's treaty powers also might come into conflict with the IP Clause's external limitations. As I discuss here, the federal government's traditional treaty power is not a congressional power in the way that the commerce, spending, and necessary-and-proper powers are. When the federal government is exercising this traditional power, a reasonable argument can be made for the proposition that the IP Clause does not externally limit this power. This argument is somewhat less powerful, however, with respect to non-self-executing treaties, which Congress implements through legislation. Furthermore, the IP Clause even more likely imposes external limits on Congress when it enacts congressionalexecutive agreements.

According to Article II of the Constitution, 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."

The Supreme Court, in Missouri v. Holland, 265 approved a broad understanding of this treaty power. In upholding a treaty between the United States and Great Britain regulating the killing of migratory birds, the Court rejected the argument that "what an act of Congress could not do unaided, in derogation of the powers reserved to the States, a treaty cannot do."

266 That is, even if Congress lacks authority to regulate the killing of migratory birds with its Article I, Section 8 powers-particularly the Commerce Clause-that fact alone does not prohibit the federal government from exercising its Article II treatymaking powers to accomplish the same effect. 267 Nonetheless, as the Supreme Court recognized in Reid v. Covert, 268 the Article II treaty power does have limitations. In that case, the Court struck down an executive agreement between the United States and Great Britain that had "permitted United States' military courts to exercise exclusive jurisdiction over offenses committed in Great Britain by American servicemen or their dependents" because the agreement forced the wives of soldiers to be tried in military court, depriving them of their Fifth and Sixth Amendment rights. 269 The Court reasoned that "[t]he prohibitions of the Constitution were designed to apply to all branches of the National Government and they cannot be nullified by the Executive or by the Executive and the Senate combined." 270 In other words, the Constitution's prohibitions on governmental power trump the treaty power.

When a treaty passed pursuant to the Article II treatymaking power seeks to promote the progress of science and useful arts, one could make a plausible case that the IP Clause does not externally limit the treaty to using only the means specified in the IP Clause itself. 272 As Professor Graeme Dinwoodie observes, "[T]he Treaty Clause presents a weaker case for the imposition of structural constraints not only because the limits in the [IP] Clause appear unrelated to the structure of government, but also because the independent exercise of the Treaty Clause appears affirmatively supportive of a governmental vision contained in the Constitution." 273 Specifically, the treaty power is situated among the president's 267. Id 2007) (arguing that copyright laws passed pursuant to the treaty powers are likely to be valid if they "seek[] to ensure domestic compliance with real international obligations" and are "adopted through a process involving real political checks on legislative lawmaking," even if they do not comply with the IP Clause's means); Timothy R. Holbrook, The Treaty Power and the Patent Clause: Are There Limits on the United States' Ability To Harmonize?, 22 CARDOZO ARTS & ENT. L.J. 1, 4 (2004) (arguing that the IP Clause should not limit the Treaty Clause when the subject matter of a given treaty is international).

273. Dinwoodie, supra note 272, at 372. powers in Article II, not Congress's powers in Article I, Section 8. It specifies a treatymaking process that involves the president and the Senate, not the whole Congress. As such, one might question whether the external limitations that the IP Clause imposes on Congress's other powers also restrict the treaty power, given that treatymaking is not a congressional power, even if the Senate is constitutionally integral to the process. The question can be restated more generally: Does the IP Clause establish external limitations only on congressional legislation-as with the legislation passed under the Commerce Clause in Missouri v. Holland-or more broadly on federal governmental action, including the exercise of the treaty power-just as the Sixth Amendment created such a restriction in Reid v. Covert? The evidence set forth in Part I does not seem to shed light directly on the intersection between treatymaking and the IP Clause. Structurally speaking, the IP Clause-situated in Article I-might limit Congress's powers when the legislature acts bicamerally, but not the Article II treatymaking powers employed jointly by the president and the Senate. 274 Although this structural analysis might seem formalistic, an exercise of the Article II treatymaking powers is designed to ensure that the president and Senate represent the country's interests broadly with regard to other countries, as compared with the more regionally grounded House of Representatives, which by design has a greater hand in Article I legislation. 275 As Professor Dinwoodie notes, "[T]he values that lawmakers might need to pursue at the international level, or the means through which they might pursue them, might depart from those ends or means that are desirable domestically." 276 Still, the argument that the IP Clause does not limit treaties is not airtight. For one thing, if the argument is correct, then the federal government could easily circumvent the IP Clause's limitations through treatymaking instead of Article I legislation. 277 199, 210 (2003) ("[T]he federal government could enter into a treaty for the purpose of circumventing limitations on Congressional lawmaking authority. There is good reason to believe that the scope of the treaty enacted for this purpose would not be restricted to external matters."). historical evidence discussed in Part I shows that the IP Clause was believed in one circumstance to limit externally a power exercised under Article IV, beyond Article I. 278 It might similarly limit a power exercised under Article II; but then, the Article II treaty power is granted to the president as well as to the Senate, whereas the Article IV power in that historical example was provided solely to Congress.

Furthermore, even if Article II self-executing treaties are not limited externally by the IP Clause, non-self-executing treaties might be so limited. Although both kinds of treaties are made pursuant to Article II, 279 the former become law without any need for Congress to activate them via Article I, 280 whereas the latter come into effect only if Congress uses Article I's Necessary and Proper Clause to enact appropriate legislation. 281 Therefore, one might argue that whenever Congress uses its Article I powers to implement non-self-executing treaties-the only powers it has at its disposal to do so-the IP Clause externally limits the legislature to the Clause's specified means. 282 The reasoning goes that whereas the non-self-executing treaty in Holland could be implemented via the Necessary and Proper Clause even though the Commerce Clause was unavailable, the same could not be said of the IP Clause, because unlike the Commerce Clause, the IP Clause externally limits the other Article I powers. This argument for different treatment, however, is weak, primarily because of the strange state of affairs that would result if Article II were to allow the United States to take on a treaty obligation that it could not then implement via Article I.

Regardless, my analysis suggests that good reasons exist to think that the IP Clause externally limits the government's power to enact treaty-like congressional-executive agreements. A congressionalexecutive agreement is an agreement negotiated by the president with one or more other countries, which is then adopted bicamerally by that the Senate ratified and Congress implemented in 1988. 290 As just discussed, the arguments about whether congressional implementation of this non-self-executing treaty should be limited by the IP Clause's external limitations are weakly colorable.

In sum, the federal government's treaty powers and the IP Clause's external limits may collide. One would most expect a collision when Congress enacts a congressional-executive agreement; one would expect it less when Congress implements a non-selfexecuting treaty; and one would least expect it when the federal government enacts a self-executing treaty.

Although other powers might come into conflict with the IP Clause's external limits, the commerce, spending, necessary-andproper, and treaty powers are those most likely to clash with them because those powers are broad and are responsible for much federal lawmaking. Each is capable of being invoked in ways that the IP Clause would seem to restrain. This Part's analysis has mostly assumed that enacted laws have a singular purpose and that collisions between the IP Clause's external limitations and other constitutional powers take place when Congress uses those powers as a cover for promoting the progress of science and useful arts using means not authorized in the IP Clause. A nuanced analysis, however, allows for the possibility that Congress might have legitimate interests in enacting legislation in addition to promoting the progress of science and useful arts. 291 Many of these legitimate interests are likely to be pursuant to those powers just discussed that are likely to collide with the IP Clause's external limitations. In those instances, the presumption framework outlined in Section A should apply to permit Congress to enact legislation in spite of the IP Clause's external limitations, as long as clear and convincing evidence exists that Congress considered those other interests legitimate enough to override the IP Clause's external limitations. As a practical matter, this framework might enable Congress to bypass the IP Clause's external limitations whenever it wants merely by giving lip service to other more permissive constitutional powers, such as the Commerce Clause or the treaty powers, to override the IP Clause's limited means. This situation might sometimes happen, as unfortunate as it may be. There is perhaps no better solution, however, given the recognition that Congress might have truly legitimate trade, foreign-relations, or other interests in enacting laws with the IP Clause's ends that extend beyond its means. At worst, the presumption framework would force Congress to consider clearly-and one would hope carefullywhether it is transgressing the IP Clause's important external limitations. Moreover, because the presumption would be extremely difficult to overcome for laws whose means subvert the IP Clause's means, those laws would be less likely to be deemed to be constitutional.

Bearing this general analysis in mind, I now turn to an exploration of some existing federal laws that might be unconstitutional in light of the foregoing analysis.

For most of the nation's history, few federal laws were passed that might even questionably transgress the limited means specified in the IP Clause. Whatever the cause-be it globalization, an expansive Commerce Clause, or recklessness-starting in the late twentieth century, Congress proposed or enacted a number of federal laws of dubious constitutionality in light of the IP Clause's external limitations. 292 This Part explores some of those laws-trade-secrecy, antibootlegging, copyright-restoration, and database-protectionanalyzing their constitutionality in light of the analytical framework proposed in Part II. This Part also investigates the constitutionality of federal laws establishing the federal funding of scientific and artistic works. As the analysis in this Part demonstrates, the framework proposed by this Article is straightforward in its application. Even when the conclusion to be drawn is not obvious, use of the framework focuses and sharpens the analysis.

This Section describes Congress's legislation protecting trade secrets from misappropriation. I show that this law has a strong 292. Cf. Heald & Sherry, supra note 18, at 1168 n.359 (observing that the earlier absence of questionable laws "might itself be evidence of the unconstitutionality of such statutes"). structural purpose of promoting the progress of science and useful arts, but that it does not comply with the IP Clause's limited means. Moreover, although the law also has a weighty legitimate purpose of promoting trade and interstate commerce, Congress did not leave clear evidence that it thought commerce interests ought to take precedence over the IP Clause's external limitations. As such, the legislation is constitutionally problematic. Given that President Obama's Justice Department has called for greater use of this law in prosecuting trade-secret misappropriation, 293 its constitutional status is ever more important.

In 1996, Congress passed the Economic Espionage Act, 294 providing the first general federal protection for trade secrecy. 295 Among other things, it provides that

[w]hoever, with intent to convert a trade secret, that is related to or included in a product that is produced for or placed in interstate or foreign commerce, to the economic benefit of anyone other than the owner thereof, and intending or knowing that the offense will, injure any owner of that trade secret, knowingly . . . [misappropriates] shall . . . be fined under this title or imprisoned not more than 10 years, or both.

The law defines trade secrets as all forms and types of financial, business, scientific, technical, economic, or engineering information . . . if-(A) the owner thereof has taken reasonable measures to keep such information secret; and (B) the information derives independent economic value . . . from not being generally known to, and not being readily ascertainable through proper means by, the public.

The Senate report on the bill explains that this law has two purposes: "to promote the development and lawful utilization of United States proprietary economic information produced for, or placed in, interstate and foreign commerce by protecting it from . . . misappropriation" and "to secure to authors and inventors the exclusive right to their respective writings and discoveries." 298 The first purpose corresponds with Congress's power under the Commerce Clause, and the second corresponds with Congress's power under the IP Clause. 299 The report explains further that Congress has heretofore confined its protection of intellectual property to patented and copyrighted material. With this legislation, Congress extends the protection of Federal law to the equally important area of proprietary economic information. During the course of the Committee's hearings, we documented that proprietary economic information is vital to the prosperity of the American economy, that it is increasingly the target of thieves, and that our current laws are inadequate to punish people who steal the information.

As the report confirms, then, with the Economic Espionage Act, the Senate expressly set out to protect a broader swath of material in a manner akin to Congress's protection of patentable and copyrightable material. Although the law is primarily occupied with promoting the country's economic status, it is also focused on safeguarding information that might otherwise be protected by existing copyright and patent law.

The Economic Espionage Act comports with a general understanding of trade-secret laws' purposes. As the Supreme Court has reasoned, along with "maintenance of standards of commercial ethics," a primary purpose of trade-secrecy protection is "the encouragement of invention."

302 Scholars emphasize trade secrecy's important role in fostering scientific and technological innovation. 303 Like patent law, trade-secrecy protection encourages investment in scientific and technological research by "giv[ing] the developer of new and valuable information the right to restrict others from using it, and therefore the prospect of deriving supracompetitive profits from the information." 304 This discussion indicates that the law's structural effects are to protect economic development and to promote the progress of science and useful arts. As to the former effect, the law protects many things that are primarily helpful in facilitating commerce, such as customer lists and customer research. 305 As to the latter, it promotes the progress of science and useful arts by encouraging investment in scientific and technological research because it authorizes punishment for those who misappropriate such information. 306 In fact, nearly every reported judicial opinion resulting from a prosecution under this law seems to involve this latter variety of trade secret, the variety whose protection is intended to promote the progress of science and useful arts. 307 Although courts and commentators have debated whether trade-secrecy protection interferes with patent law's operation, 308 few dispute that the two regimes have, in large part, the same aim: the constitutional end laid out in the IP Clause of promoting the progress of science and useful arts. As such, to the extent that the law is protecting scientific and technological information from misappropriation, it must comply with the IP Clause's specified means unless clear and convincing evidence exists that commerce interests superseded the IP Clause's external limitations.

Federal trade-secrecy law does not confine itself to the Clause's means. Putting aside the issue of whether the law's criminal and injunctive relief confer exclusive rights on inventors 309 as the IP Clause requires, the law provides protection of potentially unlimited duration. As long as the holder of a trade secret complies with the law's secrecy requirements and as long as a third party does not uncover the secret by proper means, such as independent discovery or reverse engineering, the law's protection endures. 310 Because one can conceive of scenarios involving protected trade secrets in which protection would not endure for "limited Times," the law contravenes the IP Clause's prescribed means. Moreover, there is no evidence that Congress deliberately chose to contravene the IP Clause's means to protect weightier commercial interests, suggesting that the presumption of unconstitutionality is not overcome here. 311 By contrast, former Professor Patry argues that federal tradesecrecy laws are not constitutionally problematic under the IP Clause.

312 Patry asserts that the purpose of trade-secrecy laws is distinct from that of patent and copyright laws: the goal of the former is to protect against industrial espionage, whereas the goal of the latter is to advance knowledge and technology. 313 In Patry's view, trade-secret laws do not protect "Writings" per se but rather protect 309. Professors Paul Heald and Suzanna Sherry think that federal trade-secrecy laws are constitutional. Heald & Sherry, supra note 18, at 1194-95. They also reason that the Economic Espionage Act does not establish exclusive rights, but rather merely supplements existing common-law remedies, and therefore does not fall within the ambit of the IP Clause's limitations. Id. Yet if Professors Heald and Sherry are correct that trade-secrecy protection is not an exclusive right, then that premise supplies all the more reason for concluding that the law transgresses the IP Clause's limited means of promoting innovation. 311. Congress is less likely to provide this clear evidence absent an adopted framework encouraging it to do so. Cf. United States v. Lopez, 514 U.S. 549, 561-63 (1995) (encouraging Congress to make express findings supporting a law's link to interstate commerce by underscoring the findings' relevance to a determination whether Congress acted permissibly within its commerce power).

the right to be free from theft, especially because material that is reverse engineered-and thus that is available to the public without misappropriation-is not protectable as a trade secret. 314 Because, according to Patry, these laws do not protect the secrets themselves so much as they protect this more abstract right to be free from theft, they are not within the reach of the IP Clause's limitations and are thus constitutionally unproblematic.

This analysis is insufficiently nuanced. Trade-secrecy protection surely seeks to protect against industrial espionage, but the statute's structural purpose also includes the goal of advancing knowledge and technology. As such, my framework turns the analysis of former Professor Patry on its head: if Patry is correct that the trade-secret statute does not protect "Writings," then as a law aimed at promoting the progress of science and useful arts, it actually complies with one fewer of the IP Clause's means. Thus, on former Professor Patry's theory, trade-secret laws are constitutionally problematic for an additional reason.

In sum, Congress's incursion into trade-secret protection with the Economic Espionage Act is constitutionally problematic. The statute has a strong structural purpose of promoting the progress of science and useful arts, but it exceeds the IP Clause's means. To the extent that Congress attempts to promote the IP Clause's goals, it cannot use its other more expansive powers, such as its power under the Commerce Clause, to protect information that promotes the progress of science and useful arts without making a clear decision to supersede the IP Clause's means because of prevalent commerce interests. By contrast, Congress's protection of other information, such as customer lists, is not constitutionally problematic because that protection is about promoting commerce, not promoting the progress of science and useful arts.

Another provision of doubtful constitutionality is the federal antibootlegging provision. In 1994, Congress enacted this provision in the Uruguay Round Agreement Act (URAA), 316 which implemented aspects of a number of multilateral trade agreements, including TRIPS, a treaty concerned with intellectual property. 317 The TRIPS agreement requires member countries to establish minimum standards for copyright, patent, trademark, and other intellectualproperty laws.

318 Among these standards is the ability of performers to prevent unauthorized "fixation of their unfixed performance and the reproduction of such fixation." 319 Pursuant to TRIPS, Congress implemented the first federal antibootlegging protection by enacting the URAA. According to this provision,

[a]nyone who, without the consent of the performer or performers involved-

(1) fixes the sounds or sounds and images of a live musical performance in a copy or phonorecord, or reproduces copies or phonorecords of such a performance from an unauthorized fixation,

(2) transmits or otherwise communicates to the public the sounds or sounds and images of a live musical performance, or

(3) distributes or offers to distribute, sells or offers to sell, rents or offers to rent, or traffics in any copy or phonorecord fixed as described in paragraph (1), . . . shall be subject to the remedies provided in sections 502 through 505, to the same extent as an infringer of copyright.

The remedies mentioned in this provision include injunctions, impoundment, and damages. 321 The URAA also includes a parallel criminal provision authorizing possible fines, a prison term of up to ten years, forfeiture, and destruction of the recordings.

The criminal provision's constitutionality has been challenged in prosecutions under the provision. beyond the limited ones specified in the IP Clause. They have claimed both that the law provides protection for performances that are not "Writings" and that the protection is perpetual, no durational provision having been included in the law. 323 Each final court that has considered a constitutional challenge has rejected it, reasoning that Congress had the authority to enact the law under the Commerce Clause, even if the IP Clause does not authorize the means employed. 324 This Section argues that in light of the IP Clause's external limitations, these courts were too quick to reject the constitutional challenges.

The structural purpose of the URAA's antibootlegging provision is indubitably the promotion of the progress of science and useful arts. For one thing, when Congress enacted this law, it labeled both the civil and criminal laws as "[c]opyright [p]rovisions." 325 The Senate Committee on the Judiciary that helped craft the implementation of TRIPS stated that it was doing so "to safeguard intellectual property rights." 326 Congress chose to place the civil provision in Title 17 of the U.S. Code with extant copyright law. 327 The civil remedies the law authorizes are incorporated from existing copyright laws. 328 The antibootlegging provision thus seems to have been enacted pursuant to the IP Clause. The civil provision of the Act makes no mention of commerce, 329 and it was not enacted pursuant to the Article II treaty power. Because TRIPS is a congressional-executive agreement, the antibootlegging provision therefore must have been enacted pursuant to some Article I power, presumably the IP Most scholars are persuaded that unfixed performances lie outside the scope of "Writings." 340 Professor David Nimmer reasons that "[i]f the word 'writings' is to be given any meaning whatsoever, it must, at the very least, denote some material form, capable of identification and having a more or less permanent endurance." 341 Moreover, federal copyright laws, from the earliest to the more modern versions, have not protected unfixed works. 342 In fact, proposals to confer copyright protection on unfixed works have been met with staunch criticism that such works are impermissibly intangible. 343 The consistent decision not to protect unfixed works, combined with the term "Writings" in the IP Clause, strongly suggests that only fixed works are "Writings." 344 Fixation can be regarded as an integral prerequisite to copyright protection for a number of reasons. First, the benefits society receives from artistic works' being created and eventually moved into the public domain are unlikely to happen without fixation. Fixation ensures that cultural creations remain available for public use long after their creation. 345 Second, fixation can help as an evidentiary matter by delineating precisely a work's expression, so that it cannot later be disputed in an infringement claim.

That said, a minority of scholars think that "Writings" can encompass unfixed works. Some reason that once "Writings" is read nonliterally, as the Supreme Court has read it, no reason remains to insist on a distinction between fixed and unfixed works. 347 The reasoning advanced by Professor Nimmer, however, suggests otherwise. 348 Others respond that an unfixed performative work "clearly is a physical rendering of the fruits of aesthetic labor" because it "is capable of being perceived, reproduced, or otherwise communicated" through a performance. 349 That observation is true enough, but unfixed performances are unlike the various works the Court has found to constitute "Writings," and the Court has hinted that they do not belong to the same class.

If the majority's analysis is correct, then live performancesprotected by the antibootlegging provision of the URAA-are not "Writings." 350 As such, the antibootlegging provision is not in compliance with the IP Clause's external limitations because it seeks to promote the progress of science and useful arts without confining itself to the IP Clause's prescribed means.

Another reason exists to doubt the law's constitutionality: it lacks a provision limiting the duration of protection. 351 Absent a court's reading such a provision into the law, live musical performances could be protected eternally from bootlegging. Although some commentators suggest that copyright's durational limitation ought to be incorporated into the antibootlegging law, 352 courts have been hesitant to do so. 353 Because the antibootlegging provision regulates works that are likely not "Writings," because it lacks any durational limitation, and because the structural purpose of the law is to promote the progress of science and useful arts, the provision is of dubious constitutionality. Moreover, it cannot overcome any presumption against constitutionality because there is no clear and convincing evidence that Congress intended another legitimate constitutional interest to take precedence and thereby supersede the IP Clause's external limitations. The federal government's attempt to classify this law ex post as an exercise of Congress's power under the Commerce Clause 354 does nothing to change the fact that the actual and only structural purpose reflected in the law is to promote the progress of science and useful arts.

Another law that raises constitutional doubts is the copyrightrestoration provision Congress passed in 1994, also in the URAA.

The law restored copyright protection for certain foreign works that had fallen into the public domain in the United States-due, among other things, to "lack of national eligibility" or "noncompliance with formalities imposed at any time by United States copyright law"-but whose copyrights had not yet expired in the creators' home countries. 356 Included among the works whose copyrights were so restored were "films by Alfred Hitchcock and Federico Fellini, books by C.S. Lewis first interest was compliance with international obligations, whereas the next two were promoting the progress of science and useful arts: the former by giving greater incentive to American authors to create and the latter by giving an incentive to foreign authors to develop and market their works.

Because Congress passed the law to comply with TRIPS, one might suggest that it was thus using an Article I power-presumably the IP Clause-as its source of authority for implementing a congressional-executive agreement. Yet the story is more complicated than that because TRIPS required the United States to comply with the Berne Convention on copyright restoration. 368 The United States acceded to Berne pursuant to its Article II treatymaking power. 369 Therefore, as an alternative to legislating under the IP Clause, by implementing copyright restoration pursuant to TRIPS, Congress may have been using the Necessary and Proper Clause. 370 As discussed in Part II.B.4, although a case might be made that Article II treaties, particularly non-self-executing ones, need to comply with the IP Clause's external limitations, one could mount a respectable argument that they do not. 371 To the extent that the IP Clause limits the means that Congress may employ in this context, Congress's passage of the copyrightrestoration provision within the URAA might seem to exceed the means authorized by the Clause. In fact, the law's challengers argue that extending copyright restoration to works already in the public domain violates the IP Clause's "limited Times" restriction. 372 These challengers assert that once a work has moved into the public domain, its limited time of protection is at an end and cannot be restored.

The Supreme Court, however, rejected this argument, effectively reasoning that a one-time restoration of copyrights to some works for to avoid a trade-enforcement proceeding in the World Trade Organization on that basis. a finite time is well within Congress's power to confer rights in intellectual property for limited times.

374 By interpreting the "limited Times" provision as it did, 375 the Supreme Court avoided-and perhaps consciously ducked-having to address whether and how the IP Clause's external limitations tie Congress's hands and prevent it from restoring copyright protection to works that have already fallen into the public domain.

Legislation involving database protection is another area of statutory law that raises constitutional uncertainty. As of 2012, Congress had not yet passed into law any such protection, but many bills had been introduced on the topic in the preceding years. In 1991, the Supreme Court cast doubt on the possibility of using copyright law broadly to protect information databases. The Court ruled that databases constitutionally fall within the purview of the copyright laws only when the database compiler for whom the protection is intended "chooses which facts to include, in what order to place them, and how to arrange the collected data so that they may be used effectively by readers . . . . so long as [the choices] are made independently by the compiler and entail a minimal degree of creativity." 376 Absent such conditions, protection for databases 374. Id. at 884. 375. Another possible interpretation of the provision was colorable. The historical evidence on the precise definition of "limited Times" in this context is scant, Golan I, 501 F.3d at 1190-91, and before Golan III, the Supreme Court had not ruled on it. It had stated in other contexts, however, that the precise point of copyright and patent laws enacted pursuant to the IP Clause is to ensure that the protected works will fall into the public domain for others to use freely. transgresses the "Authors" and "Writings" requirements of the IP Clause.

After this decision, Congress proposed various bills to protect databases more comprehensively, although none of them passed into law. 378 Those bills relied on the Commerce Clause for their authority 379 because the IP Clause had been ruled off-limits for noncreative databases. The remaining question is whether even that reliance on the Commerce Clause was permissible in light of the IP Clause's external limitations. Scholarship on the constitutional question is divided. For example, Professors Heald and Sherry think that legislation protecting databases is constitutionally problematic because "facts have historically been considered to form a critical part of the public domain, available for anyone to use."

380 Professor Pollack, however, thinks that database protection is constitutional under the Commerce Clause as long as it is limited to cases of market failure.

Much of this debate implicitly centers on whether the structural purpose of database-protection laws is promoting the progress of science and useful arts or fostering interstate commerce. If their purpose is only the former, the laws must comply with the IP Clause's 377. See id. at 346 (explaining that the Supreme Court has "made it unmistakably clear" that "the crucial terms 'authors' and 'writings'" both "presuppose a degree of originality"). But cf. Jane C. Ginsburg, Creation and Commercial Value: Copyright Protection of Works of Information, 90 COLUM. L. REV. 1865, 1867 (1990) (using history to show that early copyright law protected "works of little personal authorship yet considerable expenditure of labor and capital").

378. means, meaning no protection can be extended for databases that are insufficiently original. If their purpose is just the latter, however, then Congress has a wide berth to legislate to protect databases. Moreover, if the laws have both purposes, Congress may constitutionally enact them only if it sets out clear and convincing evidence that commercial interests dictate superseding the IP Clause's external limitations, a requirement that will be very difficult to satisfy if the laws subvert the IP Clause's means rather than merely supplementing them. Overall, which structural purpose is at issue in any given piece of legislation depends heavily on the legislation's particulars. For example, if a database-protection law were to prohibit only business competitors from extracting information from a database without authorization, that fact would suggest that the law was occupied with promoting commerce by encouraging investment in database development. 382 By contrast, the same inference would be absent if the law were to prevent anyone from extracting information, even for noncompetitive purposes; in that case, the law would be less about promoting commerce and more about promoting the progress of science and useful arts. 383 Similarly, one could make the case that a law protecting people who put a substantial investment into creating any sort of database-not just people who make a creative or technical contribution-is less interested in promoting the progress of science and useful arts than in encouraging commerce. 384 Whatever the outcome, the framework for evaluating the IP Clause's external limitations crisply queues up the issues that must be answered.

Possibly the most prominent category of federal action that raises constitutional concerns in light of the IP Clause's external limitations is federal funding of artistic and scientific works. As this Section shows, this practice seems to be constitutionally proper, if not 382. Cf. Benkler, supra note 16, at 597 (arguing that "[i]f in fact the important governmental interest is to prevent undercutting of database producers by competitors who provide the same product without paying the cost of its development," then a provision that would extend beyond "competition among producers of the same product" to affect "producers of different products to the extent that one product is an improvement of, or uses as an input some of, the other product" would be "unnecessarily broad").

383. See id. at 596-97 (observing that if a particular bill "aims to prohibit use by users who are not competitors, then the law is not an unfair competition law, but a property right, which cannot be passed outside the Intellectual Property Clause").

384. E.g., J.H. Reichman & Pamela Samuelson, Intellectual Property Rights in Data?, 50 VAND. L. REV. 51, 54-55 (1997).

Project. 391 Even though this research would often trickle down to public uses, the funding was initiated to support the federal government's military function. 392 Vannevar Bush, who was involved in the Manhattan Project, proposed that the government begin funding, through a federal agency, scientific research done by universities and other research centers during peacetime. 393 As a result, in 1950, Congress created the National Science Foundation (NSF). 394 Among the goals established for the NSF by Congress were "to initiate and support basic scientific [and engineering] research and programs to strengthen scientific [and engineering] research potential and . . . education programs" and "to foster the interchange of scientific and engineering information among scientists and engineers in the United States and foreign countries."

395 Moreover, Congress articulated that it was founding the NSF because "the Nation's capacity to conduct high quality research and education programs and to maintain its competitive position at the forefront of modern science, engineering, and technology [was being] threatened by [a] research capital deficit." 396 As of 2010, the NSF was funding about half of colleges' and universities' basic nonmedical-research activities. 397 The federal government has a longer history of funding medical research. In 1878, Congress appropriated funds for "investigating the origin and causes of epidemic diseases, especially yellow fever and cholera." 398 The next year, Congress created the National Board of Health, the federal government's first general and organized funder [Vol. 61:1329 of medical research. 399 Having since changed its name to the National Institutes of Health, the agency provides funding for a wide variety of medical research, 400 approximating $31.2 billion annually.

Funding for scientific and technological research also comes through the military-for example, through the Defense Advanced Research Projects Agency.

This funding is concerned with developing technologies to promote national security.

When otherwise-copyrightable works are produced pursuant to these grants, the copyright law does not speak expressly about whether such works are copyrightable. The legislative history to the 1976 revision of the Copyright Act, however, states that either Congress or the agency responsible for the funding may decide whether to allow the funded party to secure copyright in the funded works. 404 Nonetheless, one commentator points to a federal policy "favoring broad copyrightability in federally supported works." 405 Following federal grants in both arts and sciences, the government typically allows the grantee to copyright resulting work-including articles discussing research-with the government reserving a royaltyfree, nonexclusive license. Since 1980, patent law has similarly allowed nonprofit, university, and small-business grantees to patent inventions resulting from federal grants after the inventors have complied with various formalities.

407 Before this scheme was implemented, the federal government would typically own the rights to these inventions and would either commit them to the public domain or freely license resulting patents nonexclusively to interested parties.

Could this extensive funding of artistic and scientific works be unconstitutional in light of the IP Clause's external limitations? Recall that the Founders rejected just this sort of funding as a means of promoting the progress of science and useful arts, 409 and the First Congress was worried about its constitutional ability to fund scientific research. 410 Nonetheless, even if the Spending Clause does not permit this funding, the Necessary and Proper Clause likely does.

Indeed, the Federal Circuit has rejected the notion that such funding is unconstitutional. 411 In an analysis directly contrary to this Article's reasoning, that court has reasoned that Art. I, § 8, cl. 8 does not state that the Government may promote the progress of the useful arts only through the patent and copyright system. Ample constitutional power for Government funding of research and development can be found in art. I, § 8, cl. 1 (provide for the common Defense and general Welfare), cl. 3 (Commerce), cl. 12 (Army), cl. 13 (Navy) and cl. 18 (necessary and proper clause). The court then concluded that federal funding for research and development also falls within the Spending Clause's purview.

Yet as this Article maintains, the text, structure, history, judicial doctrine, and policy surrounding the IP Clause yield a plausible argument that the Spending Clause is in fact limited externally by the IP Clause. 414 If that argument is correct, then federal funding of artistic and scientific works-which is done to promote the progress of science and useful arts-would seem to be impermissible because it falls outside of the IP Clause's limited means. At the same time, however, given how expansively the Supreme Court has interpreted the Spending Clause, 415 the Court very well might not allow the IP Clause to limit Congress's spending power.

In any event, federal funding of artistic and scientific works, as typically structured, is almost certainly constitutionally authorized by the Necessary and Proper Clause. As I analyze in Part II.B.3, the Necessary and Proper Clause authorizes Congress to promote the progress of science and useful arts through means other than those specified in the IP Clause, but only to support the Clause's means for the specified end. 416 Accordingly, as conventionally structured, federal funding of artistic and scientific works seeks to promote the progress of science and useful arts so as to secure exclusive rights to authors and inventors for limited times in their works. It does so by providing funding to authors and inventors to create their works in the first place and by enabling them generally-although not requiring them-to seek copyright and patent protection in those works as further incentive to create and disseminate them. 417 2006) (disallowing an inventor a patent when "another inventor . . . establishes . . . that before [the filing person's] invention thereof the invention was made by such other inventor and not abandoned, suppressed, or concealed" (emphasis added)); Apotex USA, Inc. v. Merck & Co., 254 F.3d 1031, 1038-39 (Fed. Cir. 2001) ("Even though there is no explicit disclosure requirement in § 102(g), the spirit and policy of the patent laws encourage an inventor to take steps to ensure that 'the public has gained knowledge of the invention which will insure its preservation in the public domain' or else run the risk of being dominated by the patent of another." (quoting Palmer v. Dudzik, 481 F.2d 1377, 1387 (C.C.P.A. 1973))). Because a trade secret cannot lead to a patent except in a tenuous way, between the IP Clause's means and ends is as much as the Necessary and Proper Clause requires; by maintaining that congruency, the Necessary and Proper Clause enables a law that is "'convenient, or useful' or 'conducive' to the [underlying] authority's 'beneficial exercise.'" 418 In fact, federal funding for scientific works was perhaps more constitutionally problematic before 1980, when rights to research vested in the federal government, not in inventors. When that vesting occurred, the funding was not promoting exclusive rights to inventors, as the Necessary and Proper Clause seems to require when constrained by the IP Clause's external limitations.

In sum, the vast practice of federal funding of artistic and scientific works, as commonly structured, seems to lie well within Congress's authority.

As this Part shows, the constitutionality of a number of federal laws is left in doubt by the IP Clause's external limitations, particularly the federal trade-secrecy and antibootlegging laws. By contrast, federal funding of artistic and scientific works is likely constitutional. Moreover, the Supreme Court's interpretation of the IP Clause in upholding the copyright-restoration laws obviated any clash between the IP Clause and other constitutional sources of congressional authority. Even when the analytical framework proposed herein does not yield an obvious, clear-cut answer as to a law's constitutionality, as it does with the proposed databaseprotection legislation, the framework still sharpens and focuses the issues to be resolved.

This Article relies on the text, structure, and history of the IP Clause, as well as subsequent governmental activity, Supreme Court doctrine, and policy, to show that the IP Clause operates to forbid Congress from using its other powers "To promote the Progress of Science and useful Arts" through laws that reach beyond the scope of the IP Clause's prescription to "secur[e] for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries." This evidence shows that if Congress seeks, via legislation, to promote the progress of science and useful arts, the only way it can do so is by enacting laws that secure to authors and inventors exclusive rights in their writings and discoveries for limited times. This Article provides an analytical framework under which courts, legislators, and others could assess the constitutionality of federal legislation. If the structural purpose of a law is to promote the progress of science and useful arts, the law may use only the means specified in the IP Clause to pursue that purpose. This understanding of external limitations imposed by the IP Clause yields varying collisions with other constitutional provisions, primarily the Commerce, Spending, and Necessary and Proper Clauses and the treaty powers. To evaluate laws that have multiple constitutional purposes, a presumption ought to exist against the constitutionality of laws that promote the IP Clause's ends but subvert its means, a presumption that may be overcome only by clear and convincing evidence that Congress intentionally chose to supersede the IP Clause's means because of paramount, legitimate interests pursuant to its other more permissive powers. This presumption ought to be extremely difficult to overcome when a law's means interfere with the IP Clause's means instead of merely diverging from the means included in the IP Clause. This Article's framework suggests that a number of existing federal laws, such as federal trade-secrecy provisions and antibootlegging laws, might be unconstitutional. The framework also suggests how to assess the constitutionality of laws that would protect databases, laws passed pursuant to international agreements, and federal funding for scientific and artistic works.