CONSTITUTIONAL CONCERNS RAISED BY THE COLLECTIONS OF
INFORMATION ANTIPIRACY ACT


     The proposed Collections of Information Antipiracy Act raises difficult and novel constitutional questions concerning Congress's power to restrict the dissemination of information. Congress may not, pursuant to the Intellectual Property Clause of the Constitution, create "sweat of the brow" protection for compiled facts, at least insofar as such protection would extend to what the Supreme Court has termed the nonoriginal portion of such a compilation. Either or both the Intellectual Property Clause and the First Amendment may impose limitations on the exercise of congressional power under the Commerce Clause that would raise serious constitutional concerns regarding the constitutionality of the bill.


July 28, 1998


MEMORANDUM FOR THE ASSOCIATE WHITE HOUSE COUNSEL


      You have asked for our views on the constitutionality of the Collections of Information Antipiracy Act, H.R. 2652, 105th Cong. (1998), which passed the House on May 19, 1998. H.R. 2652 raises very difficult, and quite novel, constitutional questions, which are the subject of this memorandum. The following analysis is preliminary and general. We would, of course, be pleased to provide views directed to more specific questions that you might have.

      The object of H.R. 2652 is, in effect, to provide a quasi-property right in certain collections of information that required great effort to compile. H.R. 2652 would impose liability upon anyone who "extracts, or uses in commerce, all or a substantial part, measured either quantitatively or qualitatively, of a collection of information gathered, organized, or maintained by another person through the investment of substantial monetary or other resources, so as to cause harm to the actual or potential market of that other person, or a successor in interest of that other person, for a product or service that incorporates that collection of information and is offered or intended to be offered for sale or otherwise in commerce by that other person in commerce, or a successor in interest of that person." Id. 2 (proposed 17 U.S.C. 1202).

      In assessing the constitutional concerns raised by the bill, which would provide what is known as "sweat of the brow" protection for certain compilations of factual material, we address three related questions:

(i) whether the bill constitutes a valid exercise of Congress's power under the Intellectual Property Clause of the Constitution, Art. I, 8, cl. 8, which provides that Congress shall have the 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";

(ii) whether, if the bill does not constitute a valid exercise of Congress's power under the Intellectual Property Clause, it constitutes a valid exercise of Congress's power under the Commerce Clause, or whether the Intellectual Property Clause precludes such Commerce Clause legislation; and

(iii) whether, if the Intellectual Property Clause does not preclude Congress from exercising its commerce power to enact such legislation, the First Amendment restricts such an exercise of the commerce power.

      As to the first question, the Supreme Court's decision in Feist Publications, Inc. v. Rural Tel. Serv. Co., 499 U.S. 340 (1991), indicates that Congress may not, pursuant to the Intellectual Property Clause of the Constitution, create such "sweat of the brow" protection for compiled facts, at least insofar as such protection would extend to what the Court termed the nonoriginal portion of such a compilation. As to the second and third questions, Supreme Court precedents do not provide clear guidance; it is fair to say, however, that either or both the Intellectual Property Clause and the First Amendment may impose limitations on the exercise of congressional power under the Commerce Clause that would raise serious constitutional concerns regarding the constitutionality of H.R. 2652. (1)

I. Description of H.R. 2652

      The stated purpose of H.R. 2652 is to "complement" the protection that copyright law provides to collections of information. See Collections of Information Antipiracy Act, H.R. Rep. No. 105-525, at 5 (1998) ("House Report"). According to the House Report on H.R. 2652, the Supreme Court's decision in Feist (described in more detail below) has substantially reduced the incentives for the creation of compilations of information at the same time that "[c]opying large quantities of materials from another's collection, and using it in a competing information product -- behavior that copyright protection may not effectively prevent -- is cheaper and easier than ever, through digital technology now in widespread use." House Report at 7. The House Report recognizes that "[v]arious legal and technological options exist today for producers of collections of information to protect their investments" -- namely, copyright and state contract law. Id. (2) The House Report concludes, however, that these other existing tools are not "adequate to address the crux of the problem," and that there are "meaningful gaps in protection that can best be filled by a new federal statute." Id. at 7-8. In particular, "the coverage of copyright law is limited after Feist, and the protection of a contract binds only the parties to that contract." Id. at 7.

      The asserted "goal" of H.R. 2652 "is to stimulate the creation of more collections [of information], as well as increased dissemination to the public, and to encourage more competition among producers." House Report at 8. In particular, the object of H.R. 2652 is to "restore a modified form of the `sweat of the brow' protection available in the past as a separate doctrine and then under copyright law, but under appropriate Constitutional power and with appropriate limitations." Id. at 9. The House Report asserts that the Act would not "create a property right like copyright," but would instead establish "a tort-based cause of action against misappropriation." Id.

      H.R. 2652 would establish a new chapter in Title 17, to be entitled "Misappropriation of Collections of Information." The principal provision would establish a "misappropriation" tort, to be codified as 17 U.S.C. 1202:

Any person who extracts, or uses in commerce, all or a substantial part, measured either quantitatively or qualitatively, of a collection of information gathered, organized, or maintained by another person through the investment of substantial monetary or other resources, so as to cause harm to the actual or potential market of that other person, or a successor in interest of that other person, for a product or service that incorporates that collection of information and is offered or intended to be offered for sale or otherwise in commerce by that other person, or a successor in interest of that person, shall be liable to that person or successor in interest for the remedies set forth in section 1206.

      Any person injured by a use or extraction of information in violation of 1202 could file a civil action in federal district court. Id. (proposed 1206(a)). Such courts would have the power to issue injunctions enjoining any uses or extractions of information that would contravene 1202. Id. (proposed 1206(b)). Those courts also would be able to "impound[]" any "copies of contents of a collection of information extracted or used in violation of 1202." Id. (proposed 1206(c)). A prevailing plaintiff in a civil action would be entitled to treble damages, as well any profits realized by the defendant, costs and attorneys' fees. Id. (proposed 1206(d)). Willful violations would, under certain circumstances, be subject to criminal felony sanctions, including five years imprisonment. Id. (proposed 1207(b)). No criminal or civil action could be maintained by virtue of a use or extraction "that occurs more than 15 years after the investment of resources that qualified the . . . collection of information for protection under [H.R. 2652]." Id. (proposed 1208(c)). But this limitation might, for all intents and purposes, create perpetual liability, since every time the collection of information is "maintained, "id. (proposed 1202), that would be an "investment of . . . resources" that qualifies the "collection of information" for protection under proposed 1202. Thus, if the collector "expand[s]" or "refresh[es]" the collection, arguably the 15-year period would start anew. See House Report at 21.

      The proposed legislation sets forth six categories of what it terms "permitted acts." See proposed 1203(a)-(f). The first subsection provides that the legislation shall not prevent "the extraction or use of an individual item, or other insubstantial part of a collection of information, in itself," but notes that repeated or systematic uses or extractions of individual items or insubstantial portions may not be used in a manner that would circumvent the general prohibition against uses or extractions. Id. (proposed 1203(a)). The second subsection makes clear that the legislation shall not "restrict any person from independently gathering information or using information obtained by means other than extracting it from a collection of information gathered, organized, or maintained by another person through the investment of substantial monetary or other resources." Id. (proposed 1203(b)). The third subsection provides that the legislation shall not restrict a person from using or extracting information contained in a compilation "for the sole purpose of verifying the accuracy of information independently gathered, organized, or maintained by that person." Id. (proposed 1203(c)). The fourth subsection provides that extractions or uses "for nonprofit educational, scientific, or research purposes" shall not be prohibited unless such extractions or uses would "harm the actual or potential market for the product or service." Id. (proposed 1203(d)). The fifth subsection provides an exception for uses or extractions "for the sole purpose of news reporting" in certain specific circumstances. Id. (proposed 1203(e)). The sixth subsection permits "the owner of a particular lawfully made copy of all or part of a collection of information from selling or otherwise disposing of the possession of that copy." Id. (proposed 1203(f)).

      The proposed bill also contains a separate exclusion (with limited exceptions) for "collections of information gathered, organized, or maintained by or for a government entity, whether Federal, State, or local, including any employee of agent of such entity." Id. (proposed 1204(a)). This "exclusion," would be confined to collections of information gathered, organized, or maintained "in the course of performing governmental functions," and thus would not appear to exempt factual databases -- even databases made available to the public -- that were compiled by private parties using government funding, or pursuant to government contract. Finally, another section of the bill provides, in pertinent part, that an exclusion for "collections of information gathered, organized, or maintained in the course of performing governmental functions other than education or scholarship, by or for a government entity, whether Federal, State, or local, including any employee or agent of such entity, or any person exclusively licensed by such entity, within the scope of the employment, agency, or license." However, the exception for "education or scholarship" would mean that 1202's "use or extraction" tort would still make the prohibition applicable to information compiled entirely by public colleges and universities. See also House Report at 17 (confirming that the statute would apply to information collected by "Federal or State educational institutions in the course of engaging in education or scholarship").

      Particularly in light of the constitutional limitations that might apply to the type of protection afforded by H.R. 2652, the precise nature of the prohibitions, permissions and exemptions that are contained in the proposed bill are of critical importance. However, many of the critical, proposed statutory terms are not well-defined. Because of the ambiguity of many of these terms, it is impossible to know for certain how wide-ranging H.R. 2652's application would be. Nevertheless, in the remainder of this section, we identify some of the broadest and most ambiguous provisions of H.R. 2652 in order to clarify its possible scope.

      To begin with, "information" would be defined to mean "facts, data, works of authorship, or any other intangible material capable of being collected and organized in a systematic way." Proposed 1201((2). As a result, unlike the Copyright Act, the proposed legislation would provide protection that would not be limited to compilations of what have been termed expressive or original materials, concepts that we discuss in more detail below. The legislation would instead also provide protection to ordinary facts, which are not now subject to copyright protection and may be unsuited to such protection as a matter of constitutional law. In addition, the definition of "information" would not, from its face, appear to be limited to those compilations of information that are accessible only for a fee.

      The proposed legislation also does not define either the term "extracts" or the phrase "uses in commerce." Given their seemingly expansive, ordinary meanings, these words would, standing alone, appear to give H.R. 2652 quite a broad scope. See House Report at 12 (explaining that the provision would cover any "dissemination to others"). Moreover, the bill does not expressly provide that the prohibition on uses or extractions would apply only to uses or extractions for commercial purposes. In addition, the bill would not expressly limit liability to uses of information that is conveyed for a fee, or that is conveyed subject to contractual conditions on its further dissemination. (3) Finally, the provision would prohibit certain "uses" or "extractions" of even quantitatively insubstantial parts of a compilation, if the part in question is "qualitatively" substantial. The House Report provides the following elaboration on this point:

Only portions of the collection that are substantial in amount or importance to the value of the collection as a whole would be covered. Qualitative harm may occur through the extraction of a qualitatively small but valuable portion of a collection of information. For example, the Physician's Desk Reference, a work that compiles generally available information about every prescription drug approved by the FDA, contains some several thousand drugs and is available to both consumers and medical professionals. If a second comer extracted information about the thousand most commonly prescribed medications and offered it for sale to the general public -- for example under the title "Drugs Every Consumer Should Know" -- that extraction and use, although a fraction of the total collection of information, would cause the kind of market harm that the Committee intends H.R. 2652 to prevent. Similarly, the extraction or use of real-time quotes for all technology stocks from a securities database, while constituting a relatively small portion of actively traded or volatile securities, may be of such "qualitative" importance to the value of the database that it creates the type of commercial harm that the Committee intends section 1202 to prevent.

House Report at 12. (4)

      At the same time, the bill only prohibits extractions or uses in commerce that would "harm the actual or potential market" of the person who gathered, organized or maintained the collection of information. Proposed 1202. The scope of this important limitation is unclear. The legislation would define "potential market" to mean "any market that a person claiming protection under section 1202 has current and demonstrable plans to exploit or that is commonly exploited by persons offering similar products or services incorporating collections of information." Proposed 1201(3). This definition is arguably an expansive one that would justify a very broad construction of what would constitute harm to the potential market. Under such a broad construction, even an individual's decision to download information that had been offered for sale, purchased, but then posted on the internet for free use by the purchaser could give rise to liability on the theory that such an "extraction" would decrease the "potential market" by depriving the initial seller of a potential buyer. So construed, even the provisions in H.R. 2652 that would exempt certain uses and extractions for scientific or educational purposes would do little to confine the reach of the bill. As noted above, these exemptions are themselves limited by the requirement that such uses or extractions not harm the potential market of the original compiler, and it would appear that any educational or scientific sharing of information could deprive a potential seller of a potential buyer.

      In addition, H.R. 2652 does not include anything resembling the express exemptions found in the Copyright Act for uses that Congress previously has considered to be of particular public benefit. See, e.g., 17 U.S.C. 108 (concerning reproduction by libraries and archives), 110(1) (concerning face-to-face teaching activities), 110(2) (concerning performances and transmissions for educational purposes), 110(3) (concerning performances in the course of religious services and assemblies), 118 (concerning uses by noncommercial broadcasters). The absence of these express exemptions in what would be a statutory scheme closely related to the Copyright Act could be read to suggest that Congress intended to prohibit such uses.

      There are, however, factors that counsel against a broad construction of "potential market," and thus that point toward a more limited construction of the scope of the protection that would be provided by H.R. 2652. As an initial matter, the broadest possible construction would raise very serious constitutional concerns that we discuss in the following sections, and thus courts may be likely to avoid such a construction for that reason alone.

      In addition, the Copyright Act itself identifies harm to the "potential market" as one of the four statutory factors to be weighed in determining whether the "fair use" standard has been met, see 17 U.S.C. 107(4), and thus the appearance of this same phrase in the proposed legislation may signal Congress's intention to incorporate the definition that has been developed in the copyright context. Moreover, H.R. 2652 would contain, in addition to the "harm to the potential market" requirement, the requirement that a use or extraction be of a substantial portion of the compilation. This limitation also appears to be analogous to one of the four statutory factors for determining "fair use" under the Copyright Act. See 17 U.S.C. 107(3) (describing the factor as "the amount and substantiality of the portion used in relation to the copyrighted work as a whole). (5) Thus, there would appear to be some textual basis for concluding that H.R. 2652 is intended to incorporate, albeit implicitly, something like the "fair use" provision of the Copyright Act, and thus to limit to a significant degree the scope of the protection that the statute would provide.

      If so, the Court's recent decision in Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 590-94 (1994), would be relevant to the construction of H.R. 2652. The Court there suggested that the potential market factor is satisfied for purposes of the Copyright Act when a copyrighted work is used in a way that would create, in effect, a substitute product in direct competition with the original. See id. at 590-94. The Court added, however, that when "the second use is transformative, market substitution is less certain, and market harm may not be so readily inferred." Id. at 591. Applying that same approach here, H.R. 2652 would arguably reach, with some exceptions, only non-transformative uses for commercial purposes, as it would be only such uses that, in light of the "fair use standard" developed in copyright law, would result in harm to the potential market within the meaning of H.R. 2652. It is important in this regard to emphasize that the fair use standard in copyright is an equitable one that requires a sensitive weighing of the statutory factors in light of the specific factual context at issue, see id. at 577 ("The task is not to be simplified with bright-line rules, for the statute, like the doctrine it recognizes, calls for case-by-case analysis."), and that a determination as to fair use may also depend upon a an evaluation of the "good faith" of the use, see id. at 585 n.18.

      In sum, while it is clear that H.R. 2652 is intended to cover nonoriginal, factual material, which the Copyright Act does not (and, as we explain below, for constitutional reasons, probably could not be extended to reach), the scope of the protection that H.R. 2652 is intended to afford to such factual materials is far less clear. The ambiguity concerning the scope of the intended protection for factual material arises in large part because the legislation does not make clear whether it is intended to incorporate a version of the fair use provision that is contained in the Copyright Act or whether it is instead intended to reach broadly to encompass individual uses by non-competitors for non-commercial purposes.

     Suffice it to say that, notwithstanding the ambiguities in the text, to the extent the provision would prohibit extractions or uses of substantial portions of factual compilations by direct competitors, it is much more likely to be held constitutional than if it would prohibit extractions or uses by potential consumers for noncommercial purposes. By contrast, if the provision were construed to provide protection against uses by potential consumers, and not simply direct competitors, it would appear to be of almost limitless scope and therefore to raise constitutional concerns that would appear insurmountable. (6) We explain, however, that even if the protection provided by H.R. 2652 were construed as limited direct competitors and to somehow distinguish between "fair" and "unfair" uses of collections of information, there would remain substantial constitutional questions concerning the degree to which any reuse of factual information that would not infringe on the originality of a work may be deemed by Congress to be in some sense "unfair" and therefore subject to regulation. In other words, it is unclear what "unfair uses" of factual material could be constitutionally prohibited. There is also little precedent to guide interpretation as to where the line between fair and unfair uses of factual information is intended to be drawn precisely because the Copyright Act, which codifies the "fair use" standard, does not provide protection for facts.

      With this background concerning the proper construction of H.R. 2652 in place, we now turn to the constitutional analysis of the bill.

II. The Intellectual Property Clause

      We understand that the proposed legislation is not necessarily intended to constitute an exercise of Congress's power under the Intellectual Property Clause, and that it is instead apparently premised on Congress's power to regulate interstate commerce. Nevertheless, it is instructive for purposes of analysis to examine, as an initial matter, whether the legislation could be premised on Congress's power under the Intellectual Property Clause. It is only to the extent that the legislation would fall outside the permissible scope of the power conferred by that clause that it would give rise to concerns that, as an exercise of the Commerce Power, it would impermissibly infringe on an implicit limitation contained in the Intellectual Property Clause.

      The key precedent for assessing whether this proposed legislation would constitute a valid exercise of Congress's power under the Intellectual Property Clause is Feist. In Feist, the Supreme Court considered the extent to which the Copyright Act, 17 U.S.C. 101-1101 (West 1996 & Supp. 1998), protected the listings in telephone directory white pages from copying by a competitor. In answering that statutory question, the Court did not confine itself to a conventional consideration of congressional intent. Instead, the Court first examined the constitutional limitations inherent in the power conferred by the Intellectual Property Clause, on which the Copyright Act was premised. (7) Only after having considered these background constitutional limitations on the exercise of the copyright power did it reach the conclusion that Congress did not intend the Copyright Act to extend protection to such listings. There is language in the opinion, however, that indicates that the Court also predicated its decision on a judgment that the Intellectual Property Clause would not empower Congress to provide copyright protection to either the listings themselves, or the facts contained in the listings, even if Congress intended to extend such protection.

      In addressing the background constitutional limitations on the scope of the power conferred by the Intellectual Property Clause, the Court acknowledged that copyright protection may extend to factual compilations and to other "fact-based works," but concluded that the prerequisite for such protection is that the selection or arrangement of the facts is in some degree "original." 499 U.S. at 344-51. The Court explained that "[o]riginality is a constitutional requirement." Id. at 346. In order to satisfy this constitutional prerequisite of originality, the Court opined, the work in question must "possess[] at least some minimal degree of creativity." Id. at 345. In a factual compilation, this creativity can be present in the manner in which the compiler selects or arranges the facts. Id. at 348. Indeed, "[t]he vast majority of works make the grade quite easily, as they possess some creative spark, no matter how crude, humble, or obvious it might be." Id. at 345 (internal quotation marks omitted). The Court noted that "[o]riginality does not signify novelty; a work may be original even though it closely resembles other works so long as the similarity is fortuitous, not the result of copying." Id.

      Under Feist, however, even if a compilation is in some sense original, and* thereby entitled to some copyright protection, "the copyright in a factual compilation is thin." Id. at 349. That is because, in such circumstances, the bulk of the material that comprises the work will, by definition, be facts that in and of themselves lack the originality that justifies protection pursuant to the Intellectual Property Clause. As the Court explained:

The mere fact that a work is copyrighted does not mean that every element of the work may be protected. Originality remains the sine qua non of copyright; accordingly, copyright protection may extend only to those components of a work that are original to the author. Thus, if the compilation author clothes facts with an original collocation of words, he or she may be able to claim a copyright in this written expression. Others may copy the underlying facts from the publication, but not the precise words used to present them. . . .

. . . Notwithstanding a valid copyright, a subsequent compiler remains free to use the facts contained in another's publication to aid in preparing a competing work, so long as the competing work does not feature the same selection and arrangement. As one commentator explains it: "[N]o matter how much original authorship the work displays, the facts and ideas it exposes are free for the taking . . . . [T]he very same facts and ideas may be divorced from the context imposed by the author, and restated or reshuffled by second comers, even if the author was the first to discover the facts or to propose the ideas."

Id. at 348-49 (citation omitted) (quoting Jane C. Ginsburg, Creation and Commercial Value: Copyright Protection of Works of Information, 90 Colum. L. Rev. 1865, 1868 (1990)). Accordingly, as applied to a factual compilation that has nonoriginal written expression, the Court concluded that "only the compiler's selection and arrangement may be protected; the raw facts may be copied at will. This result is neither unfair nor unfortunate. It is the means by which copyright advances the progress of science and art." Id. at 350.

      Against this backdrop, the Court rejected the argument that the Copyright Act incorporated the "sweat of the brow" doctrine -- namely, that, whether or not a factual compilation contained any degree of creativity, copyright still attached in order to compensate compilers for the hard work and resources that they expended in the course of compiling the facts. Id. at 352-54. Such a doctrine was not tethered to the originality requirement that the Court concluded was the sine qua non for copyright protection.

      On the basis of its constitutional and statutory analysis, the Court concluded that the white pages at issue in Feist contained none of the creativity that would suffice to render a work "original." It therefore concluded that the listings were entitled to no protection under the Act, despite the fact that the defendant had copied significant portions of the plaintiff's compilation for use in its own competing white pages. The Court noted that the listings at issue fell into the "narrow category of works in which the creative spark is utterly lacking or so trivial as to be virtually nonexistent." Id. at 359. It explained that the white pages at issue are "entirely typical. Persons desiring phone service area fill out an application and Rural issues them a telephone number. In preparing its white pages, Rural simply takes the data provided by its subscribers and lists it alphabetically by surname. The end product is a garden-variety white pages directory, devoid of even the slightest trace of creativity." Id. at 362. The Court further explained that Rural could not claim "originality in its coordination and arrangement of facts. . . . [T]here is nothing remotely creative about arranging names alphabetically in a white pages directory. It is an age-old practice, firmly-rooted in tradition and so commonplace that it has come to be expected as a matter of course." Id. at 363.

      The Court therefore concluded that both the compilation itself, and the particular pieces of information contained therein, lacked sufficient originality to warrant protection. The Court summarized its judgment as follows:

We conclude that the names, towns, and telephone numbers copied by Feist were not original to Rural and therefore were not protected by the copyright in Rural's combined copyright in white and yellow pages directory. As a constitutional matter, copyright protects only those constituent elements of a work that possess more than a de minimis quantum of creativity. Rural's white pages, limited to basic subscriber information and arranged alphabetically, fall short of the mark. As a statutory matter, 17 U.S.C. 101 does not afford protection from copying to a collection of facts that are selected, coordinated, and arranged in a way that utterly lacks originality. Given that some works must fail, we cannot imagine a more likely candidate. Indeed, were we to hold that Rural's white pages pass muster, it is hard to believe that any collection of facts could fail.

Id. at 363-64.

      Despite the strong language contained in the opinion, an argument can be made that the Court's constitutional pronouncements in Feist were dictum because they were unnecessary to the disposition of the case. The Court in Feist was asked only to resolve a statutory issue concerning the scope of statutory protection for compilations under the Copyright Act. On the other hand, the Court in Feist plainly stated at numerous points that originality and creativity are constitutional prerequisites for copyright protection under Article I, Section 8, Clause 8 of the Constitution. (8) Those statements strongly indicate that the Court's decision rested on a constitutional, rather than merely a statutory, judgment.

      Because the proposed bill would clearly provide protection for "collections of information" without regard to whether they are original, and because it would define "information" quite expansively, it would appear to protect even the type of noncreative white pages listing at issue in Feist, as well as similarly unoriginal factual compilations or facts within otherwise original compilations. In this respect, the prohibition in proposed section 1202 would go well beyond the "thin protection" for factual compilations recognized in Feist. (9) Accordingly, to the extent that the proposed bill would attempt to provide protection, pursuant to the Intellectual Property Clause rather than some other power, to the very type of unoriginal factual materials that were at issue in Feist, it would run afoul of recent Supreme Court precedent that is, if not binding, at a minimum a clear indication of how the Court would likely rule.

III. Possible Intellectual Property Clause Limitations on the Commerce Power

      The House Report asserts that H.R. 2652 may be enacted "within Congress' authority to regulate interstate commerce under Article I, Section 8, Clause 3 of the Constitution." House Report at 9-10. Absent some external constitutional limitation, the bill would appear to constitute a valid exercise of the commerce power, as we understand that extractions, or uses in commerce, of substantial portions of collections of information would, in the aggregate, substantially affect interstate commerce. See United States v. Lopez, 514 U.S. 549 (1995). (10) This section examines the question whether the Intellectual Property Clause places an external limitation on such an exercise of the commerce power.

      Feist does not provide clear guidance on the question. Nothing in Feist holds that the Intellectual Property Clause limits the scope of Congress's power under other Clauses, such as the Commerce Clause, and the opinion may be read to state limits that pertain to the exercise of the Intellectual Property Clause itself. At the same time, some language in Feist might also fairly be read to suggest, not only that the Intellectual Property Clause does not authorize sweat-of-the-brow protection for either unoriginal factual compilation or facts in otherwise original compilations, but also that the Intellectual Property Clause prohibits Congress from relying on any other constitutional power to afford copyright-like protection to facts and to the nonoriginal parts of factual compilations.

      For example, the Court noted that "all facts . . . `may not be copyrighted and are part of the public domain available to every person." 499 U.S. at 348 (emphasis added; citation omitted). See also id. at 349 ("`[N]o matter how much original authorship the work displays, the facts and ideas it exposes are free for the taking. . . . [T]he very same facts and ideas may be divorced from the context imposed by the author, and restated or reshuffled by second comers, even if the author was the first to discover the facts or to propose the ideas.'") (emphasis added) (quoting Ginsburg, Creation and Commercial Value, 90 Colum. L. Rev. at 1868). The Court also opined that it is a "constitutional requirement" that persons be permitted to use "the fruit of the [factual] compiler's labor" without compensation:

It may seem unfair that much of the fruit of the compiler's labor may be used by others without compensation. As Justice Brennan has correctly observed, however, this is not "some unforeseen byproduct of a statutory scheme." Harper & Row, 471 U.S., at 589 (dissenting opinion). It is, rather, "the essence of copyright,"id., and a constitutional requirement.

Id.

      The Court further explained that the constitutional objective is realized not only by providing intellectual property rights in expression, but also by permitting ideas and information to be disseminated freely:

This principle, known as the idea/expression or fact/expression dichotomy, applies to all works of authorship. As applied to a factual compilation, assuming the absence of original written expression, only the compiler's selection and arrangement may be protected; the raw facts may be copied at will. This result is neither unfair nor unfortunate. It is the means by which copyright advances the progress of science and art.

Id. at 350.

      Accordingly, one possible reading of the Feist decision is that a system in which the "raw facts" in a compilation may not be "copied at will" is a system that necessarily undermines the object of the Intellectual Property Clause -- the progress of science and art -- and is therefore unconstitutional. On this view, the clause would constitute not only a grant of power to Congress but also a limitation on Congress. Cf. Bonito Boats, Inc. v. Thunder Craft Boats, Inc., 489 U.S. 141, 146 (1989) (discussing scope of federal preemption of state intellectual property law and stating that "[a]s we have noted in the past, the [Intellectual Property] Clause contains both a grant of power and certain limitations upon the exercise of that power"); Graham v. John Deere Co., 383 U.S. 1, 5-6 (1966) (explaining, again with reference to federal preemption of state law, that "[t]he clause is both a grant of power and a limitation. . . . 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."); Compco Corp. v. Day-Brite Lighting, Inc., 376 U.S. 234, 237 (1964) (discussing scope of federal preemption of state law and explaining that "[t]o forbid copying [under state law] would interfere with the federal policy, found in [Article] I, [section] 8, [clause] 8, of the Constitution and in the implementing federal statutes, of allowing free access to copy whatever the federal patent and copyright laws leave in the public domain.").

      If the Intellectual Property Clause precluded Congress from providing protection against the copying of nonoriginal portions of factual compilations, even pursuant to a power other than that conferred by that Clause, then Congress would not be able to use the Commerce Clause to avoid the implicit strictures of the Intellectual Property Clause that the Court in Feist could be said to have recognized, just as Congress may not use the Commerce Clause to avoid the Bankruptcy Clause's express requirement that bankruptcy laws be uniform, see Railway Labor Executives' Ass'n v. Gibbons, 455 U.S. 457, 468-69 (1982). Under this reading, Congress's reliance on the commerce power would not obviate any of the constitutional problems concerning the exercise of congressional power under the Intellectual Property Clause that we have already identified. (11)

      On the other hand, prior to Feist, the Court had recognized intellectual property interests not grounded in the Intellectual Property Clause. There are at least four notable circumstances outside the copyright context in which the Court has recognized such interests. Although these examples, together, indicate that there is no categorical prohibition on Congress's power to restrict the dissemination of data and other forms of "intellectual property" that are not copyrighted, neither do they make clear that Congress would have the power to enact legislation like H.R. 2652 under the Commerce Clause against a claim that the Intellectual Property Clause imposes a limitation. With one exception, the cases are distinguishable, and even that case does not, by itself, support legislation of this scope.

      First, the Court has sanctioned federal limitations on the dissemination of information where the person who wishes to disseminate it received such information only on the condition that it remain secret or confidential, whether such condition was expressly set forth by contract or impliedly recognized as a matter of law. For example, the government is able to afford protection to factual information pursuant to its commerce power in order to protect trade secrets. See Bonito Boats, Inc. v. Thunder Craft Boats, Inc., 489 U.S. 141, 155-57 (1989) (discussing compatibility of state trade secret protection with the federal intellectual property regime). See also Seattle Times Co. v. Rhinehart, 467 U.S. 20 (1984) (federal court may impose protective order restricting party from revealing trade secrets that it obtained pursuant to compulsory discovery process). (12) These types of protection would appear to be distinguishable, however, from the type of protection that H.R. 2652 would provide.

      H.R. 2652 would provide protection to compilers of information so that they would be able to offer the information to the public for a fee. By contrast, provisions that protect trade secrets do not restrict the manner in which information that is offered to the public in the market may be used. Such provisions instead simply provide protection to those persons who wish to keep information confidential and therefore to persons who have no interest in offering to the wider public for sale. As a result, trade secret protections do not interfere, at least directly, with the manner in which information that is made available for sale to the public might be used. By contrast, H.R. 2652 would impose direct limitations on the manner in which members of the public might use information that is, in some sense that may be constitutionally relevant, already in the public domain. Cf. Bonito Boats, 489 U.S. at 155-157 (explaining that state trade secret protection is not preempted by federal patent laws because trade secrets protection does not interfere with policy that "matter once in the public domain must remain in the public domain."). (13)

      Second, protection may be afforded pursuant to the commerce power to deter false representation, and/or to protect consumers from confusion, as the trademark laws demonstrate. See The Lanham Act, 15 U.S.C. 1501-1540 (1994). An analogy between H.R. 2652 and trademark protection would appear questionable, as the bill plainly provides protection that is not directed at avoiding confusion as to the identity of the source of the information. See Bonito Boats, 489 U.S. at 154-55 (distinguishing between traditional trade dress regulation and laws aimed at protecting factual information that would not sow confusion). Indeed, the provision would provide protection even if it were made perfectly clear, and no consumer could reasonably conclude otherwise, that the copier of the collection of information had not exerted personal effort in compiling the facts provided but had instead merely copied them from someone who had exerted such personal effort.

      In one notable case, protection analogous to that afforded trademarks has been extended to a word, the use of which would not cause consumer confusion. See San Francisco Arts & Athletics, Inc. v. United States Olympic Comm., 483 U.S. 522, 539 (1987). That case, however, is distinguishable. There, Congress had provided statutory protection for the use of the word "Olympic" in order to protect the commercial interests of the United States Olympic Committee and "the value [that] the USOC's efforts have given to [that word]." Id. at 541. That case did not involve protection of facts, as such, but rather of the special commercial value associated with the use of a particular word in a particular context. H.R. 2652, however, would appear to provide such broad protection that it would protect facts not for any special value apart from their ordinary meaning that has been given to them by the compiler's efforts but rather merely because the compiler expended effort in collecting them.

      Third, state law has been used to provide protection against dissemination of certain "copied" materials, in order to protect what has been termed "the right of publicity." See Zacchini v. Scripps-Howard Broadcasting Co., 433 U.S. 562, 569 (1977). The right of publicity protection, however, only guards the use of an individual's "personality" and personal talents against unauthorized commercial exploitation. See id. For example, Zacchini concerned the legality of a news service's airing of film of an individual's paid human cannonball performance, against the wishes of the performer. The case therefore involved protection of a depiction of the performer's original performance, a "fact" -- the performance -- that existed only because of the performer's own efforts. The Court expressly noted that the right of publicity would not serve to prevent reporting of facts about the cannonball act, as opposed to display of the act itself in its entirety,id. at 574, and that the right was analogous to copyright's protection of original expression, id. at 577 n.13. See also id. at 569 (a case involving description of the act would be "a very different case"). By contrast, the protection provided by H.R. 2652 would extend to factual data that exists independently of the compiler's efforts.

      Finally, competitive misappropriation of so-called "hot news" information has also been afforded protection by the Supreme Court as a matter of federal common law. See International News Serv. v. Associated Press, 248 U.S. 215 (1918). International News Serv. might provide some authority for the argument that Congress may use its Commerce Clause power to create certain torts relating to "misappropriation" of facts, even where the facts themselves may not be copyrighted pursuant to the Intellectual Property Clause.

      In International News Serv., the Court, without relying on the Intellectual Property Clause, recognized the permissibility of a certain limited form of liability for copying publicly disclosed information. The case arose prior to the Court's decision in Erie R.R. Co. v. Tompkins, 304 U.S. 64 (1938), and it represented an exercise of the Supreme Court's power to make federal common law pursuant to the grant of diversity jurisdiction. The case concerned a dispute that arose from a practice of the International News Service. The news agency systematically reviewed east coast editions of newspapers published by subscribers to the Associated Press, copied or rewrote the stories contained therein, and published the stories in its own west coast newspapers, some of which were delivered and sold before rival Associated Press newspapers in the same cities. Id., 248 U.S. at 231. The Associated Press had not copyrighted its stories, id. at 233, and there was no established cause of action that the Associated Press could invoke to stop the International News Service practices.

      The Court held that, even if the Associated Press did not have any property interest in its reported facts "as against the public," it had a "quasi property" right vis-a-vis the International News Service, which was "seeking to make profits at the same time and in the same field." Id. at 236. The Court used this quasi-property right to justify an injunction against the International News Service's "misappropriation" of Associated Press's reportage, because the International News Service was "endeavoring to reap where it has not sown." Id. at 239. The Court's holding "only postpone[d] participation by [the Associated Press's] competitor in the processes of distribution and reproduction of news that it has not gathered, and only to the extent necessary to prevent that competitor from reaping the fruits of [the Associated Press's] efforts and expenditure, to the partial exclusion of [the Associated Press]." Id. at 241. (14)

      Although the legal status of the quasi-property right recognized in International News Serv. -- and, more particularly, the scope of that right -- is not entirely clear, Feist suggested that the so-called "hot news" misappropriation tort, at least as it was recognized in International News Serv. itself, could survive. The Feist Court explained that the International News Serv. Court had acknowledged that the news articles themselves were "copyrightable," but had then "flatly rejected" the view "that the copyright in an article extended to the factual information it contained." 499 U.S. at 353-54. Nevertheless, the Court noted that "[t]he Court ultimately rendered judgment for Associated Press on noncopyright grounds that are not relevant here." Id. at 354 n.*.

      More generally, the Feist Court suggested that an "unfair competition" theory could be the basis for some anti-copying protection of nonoriginal factual compilations:

Protection for the fruits of such research . . . may in certain circumstances be available under a theory of unfair competition. But to accord copyright protection on this basis alone distorts basic copyright principles in that it creates a monopoly in public domain materials without the necessary justification of protecting and encouraging the creation of "writings" by "authors."

499 U.S. at 354 (quoting Melville B. Nimmer & David Nimmer, Nimmer on Copyright 3.04, at 3-23 (1990) (footnote omitted)). The passage is obscure, and thus it is not exactly clear what protection might be available "under a theory of unfair competition," or even what the Court intended by the phrase "a theory of unfair competition." It is possible, however, that the passage provides further support for an argument that the misappropriation tort recognized in International News Serv. survives the Feist analysis as an example of permissible regulation of unfair competition.

      Some insight into the possible meaning of the phrase "unfair competition" as it appears in Feist may be gleaned from Justice O'Connor's decision for the Court, two years prior to her opinion in Feist, in Bonito Boats. Justice O'Connor identified the "usual sense [in which] the term `unfair competition' is understood" by tying it to trade dress protection:

The law of unfair competition has its roots in the common-law tort of deceit: its general concern is with protecting consumers from confusion as to source. While that concern may result in the creation of "quasi-property rights" in communicative symbols, the focus is on the protection of consumers, not the protection of producers as an incentive to product innovation. Judge Hand captured the distinction well in Crescent Tool Co. v. Kilborn & Bishop Co., 247 F. 299, 301 (CA2 1917), where he wrote:

"[T]he plaintiff has the right not to lose his customers through false representations that those are his wares which in fact are not, but he may not monopolize any design or pattern, however trifling. The defendant, on the other hand, may copy plaintiff's goods slavishly down to the minutest detail: but he may not represent himself as the plaintiff in their sale."

489 U.S. at 157 (emphasis added). As the Court in Bonito Boats concluded, "unfair competition" thus does not describe the object of a statute "aimed directly at preventing the exploitation of [publicly disclosed factual information]." Id. at 158. H.R. 2652 would be such a statute.

      If this limited meaning of "unfair competition" were all that the Court intended to cover in the passage quoted above from Feist, then it would be difficult to rely on that passage as authority for the type of "unfair competition" protection contemplated here. On the other hand, the tort recognized in International News Serv. does appear to have been premised on the notion that the International News Service had engaged in "unfair competition," and thus that a legal remedy could be provided for such conduct even though the copyright power would not provide the basis for such protection. As a result, the general reference in Feist to protection against unfair competition emanating from the exercise of a valid power other than the Intellectual Property Clause provides some basis for the congressional creation of a misappropriation tort, at least along the lines recognized in International News Serv.

      If, as seems fair to be the case, Feist does not foreclose Congress from enacting something approximating the misappropriation tort recognized in International News Serv. itself, there remains the question concerning the permissible scope of an extension of such a tort. There is little precedent to provide direct guidance on this point in part because there has been little legal development in the misappropriation tort itself since Feist. Indeed, even prior to Feist, due to the Court's decision in Erie R.R. limiting the authority of federal courts to engage in common lawmaking, federal courts had not occasion to expand upon the tort recognized in International News Serv. State courts have also had little occasion to expand upon the tort recognized in International News Serv., in part because of the preemptive effect of the Copyright Act. As explained in the recent case of See NBA v. Motorola, Inc., 105 F.3d 841, 852 (2d Cir. 1997), a state-law tort that would not be preempted by the Copyright Act must have the following essential elements: (i) the plaintiff generates or collects information at some cost or expense; (ii) the value of the information is highly time-sensitive; (iii) the defendant's use of the information constitutes free-riding on the plaintiff's costly efforts to generate or collect it; (iv) the defendant's use of the information is in direct competition with a product or service offered by the plaintiff; and (v) the ability of other parties to free-ride on the efforts of the plaintiff would so reduce the incentive to produce the product or service that its existence or quality would be substantially threatened. (15)

      Unburdened as it is by limitations on judicial common lawmaking or federal preemption doctrine, Congress might have greater freedom than federal courts or states to expand upon the tort recognized in International News Serv. It is plain, however, that H.R. 2652 would constitute, not a modest extension of the "hot news" misappropriation tort, but a dramatic extension of the tort recognized in the case. See House Report at 17 (explaining that H.R. 2652 would "preserve the holding" of International News Serv., but would reach far beyond that case to make impermissible much conduct that does not fall within the "narrow scope," id., of that holding.) H.R. 2652 would not require, a civil plaintiff or a federal prosecutor, to prove that the value of the information be highly time-sensitive, or that the ability of other parties to free-ride on the efforts of the plaintiff would so reduce the incentive to produce the product or service that its existence or quality would be substantially threatened. Moreover, due to the ambiguity as to the scope of the limitation that there must be a demonstration of harm to the potential market, it is not at all clear that H.R. 2652 would even require proof that the offending use or extraction be committed by a person in direct competition with a product or service offered by the plaintiff, or even that a use was nontransformative and for a commercial purpose. In contrast, the Court in International News Serv. repeatedly emphasized that the tort it was identifying would not extend to the copying and dissemination of news stories by members of the public, as opposed to by competitors of the Associated Press. 248 U.S. at 239-41. (16)

      Accordingly, to the extent that Feist may be read to have construed the Intellectual Property Clause to have established a kind of constitutionally prescribed public domain for factual material on which Congress may not infringe (absent, perhaps, private contractual agreements), a broad expansion of the "hot news" tort would appear to raise serious constitutional concerns. H.R. 2652 -- which would apply well beyond the context of direct competitors, let alone the context of time-sensitive direct competition -- would therefore raise substantial questions under Feist (and under the First Amendment, see infra) that would not be raised by a less ambitious statute that codified a limited International News Serv.-like tort. See Reichman & Samuelson, Intellectual Property Rights in Data?, 50 Vand. L. Rev. at 139-45.

IV. Possible First Amendment Limitations on the Commerce Power

      Even if the Intellectual Property Clause does not itself impose constraints on Congress's Commerce Clause power, the First Amendment might nevertheless limit the type of protection that Congress can provide against the "use" and "extraction" of factual compilations.

      One of the principal aims of the First Amendment is to "secure `the widest possible dissemination of information from diverse and antagonistic sources.'" New York Times Co. v. Sullivan, 376 U.S. 254, 266 (1964) (quoting Associated Press v. United States, 326 U.S. 1, 20 (1945)). In accordance with this objective, the First Amendment imposes significant constraints on the ability of the government to restrict the dissemination of information that has been publicly disclosed and that the disseminator has lawfully obtained. For example, although the Supreme Court has been careful never to hold categorically that publication of lawfully obtained truthful information "is automatically constitutionally protected," see The Florida Star v. B.J.F., 491 U.S. 524, 541 (1989), the Court has, on several occasions, held that "the government may not generally restrict individuals from disclosing information that lawfully comes into their hands in the absence of a `state interest of the highest order.'" United States v. Aguilar, 515 U.S. 593, 605 (1995) (quoting Smith v. Daily Mail Pub. Co., 443 U.S. 97, 103 (1979)). See also Butterworth v. Smith, 494 U.S. 624, 632 (1990). (17) And even if the state has such an interest, "punishment may lawfully be imposed, if at all, only when narrowly tailored to a state interest of the highest order." Florida Star, 491 U.S. at 541. (18) What is more, even in situations in which the government hypothetically could impose subsequent sanctions for the publication or copying of certain information, there is a particular concern about imposing a prior restraint on a secondary recipient from disseminating non-commercial speech. See, e.g., New York Times Co. v. United States, 403 U.S. 713 (1971). That is true even where the information was unlawfully obtained as an initial matter. Id.

      To be sure, cases such as New York Time v. Sullivan and Florida Star are not directly on point. Those cases involved governmental attempts to suppress certain types of information from being disseminated on the basis of content. By contrast, H.R. 2652 would not target any particular types of messages for suppression. It would instead prescribe the means under which collections of information that had been complied may be used by others. Namely, it would require users, in certain circumstances, to expend great effort independently before using information contained in a collection that itself had been compiled only after great effort.

This ground of distinction hardly dispenses with the concern that H.R. 2652 trenches on First Amendment rights. Copyright protection similarly does not seek to suppress certain types of messages. It, too, merely prescribes the means by which information may be used by others. Nevertheless, the Court has concluded that the First Amendment may impose limitations on the types of material that may be copyrighted. Most significantly, in Harper & Row Publishers, Inc. v. Nation Enterprises, 471 U.S. 539 (1985), the Court explained that the First Amendment and the Copyright Act can be reconciled by virtue of the fact that copyright law already embodies a distinction between original forms of expression -- which are copyrightable -- and facts (and ideas) -- which are not:

[C]opyright's idea/expression dichotomy "strike[s] a definitional balance between the First Amendment and the Copyright Act by permitting free communication of facts while still protecting an author's expression." No author may copyright his ideas or the facts he narrates. 17 U.S.C. 102(b).

Id. at 556 (citation omitted). See also New York Times Co., 403 U.S. at 726, n.* (Brennan, J., concurring); Feist, 499 U.S. at 344-45 ("The most fundamental axiom of copyright law is that `[n]o author may copyright his ideas or the facts he narrates.'") (quoting Harper & Row, 471 U.S. at 556). Thus, for example, the Court held that although direct quotations from President Ford's biography were subject to copyright, the historical facts contained in that biography were not subject to copyright and could be freely copied. See Harper & Row, 471 U.S. at 565-66 & n.8 (applying copyright analysis only to "verbatim quotes" from the biography, and excluding from infringement consideration historical quotations attributed to third parties and to government documents). See also Zacchini, 433 U.S. at 574 (right of publicity would not serve to prevent reporting of facts about the cannonball act, as opposed to display of the act itself in its entirety); id. at 577 n.13 (noting analogy to copyright's expression/idea distinction).

      The distinction referenced in Harper & Row may be understood to reflect the Court's understanding that, in order to reconcile and accommodate copyright and the First Amendment, no intellectual property rights can extend to facts that have been released in the public domain. (19) Moreover, even in the context of creative forms of expression that can be copyrighted (as opposed to factual information, which cannot), First Amendment values are further protected in the copyright law by virtue of the "latitude for scholarship and comment traditionally afforded by fair use." Harper & Row, 471 U.S. at 560. Furthermore, the Intellectual Property Clause ensures that expression itself must enter the public domain after the "limited times" for which copyright protection is available. Indeed, where important factual information could not satisfactorily be conveyed except by certain unique expression, the First Amendment might even ensure that copyright protection for such expression be denied or limited. See Nimmer on Copyright 1.10[C][2], at 1-85--1-92.

      Accordingly, H.R. 2652, by providing protection for facts, raises serious First Amendment concerns. It would restrict the ability of persons to use and disseminate factual materials that are not protected by copyright, and it arguably would do so even in circumstances where the copyright law would not protect creative expression.

      We can imagine two arguments that might be made in support of H.R. 2652 against a First Amendment challenge. First, it remains the case that in International News Serv., the Court permitted a tort for the dissemination of information, as such. It is unclear to what extent the International News Serv. tort can be reconciled with modern First Amendment doctrine. Nevertheless, that case was approvingly cited in San Francisco Arts & Athletics, where the Court recognized the possibility that the unauthorized use of an Olympic logo could impermissibly undermine the "owner's" legitimate commercial interests, even in the absence of a demonstration that such a use would be confusing to consumers. 483 U.S. at 541. As the Court there explained, "[t]here is no question that this unauthorized use could undercut the [United States Olympic Committee's] efforts to use, and sell the right to use, the word in the future, since much of the word's value comes from its limited use." Id. at 539. Thus, "[e]ven though this protection may exceed the traditional rights of a trademark owner in certain circumstances, the application of the Act to . . . commercial speech is not broader than necessary to protect the legitimate congressional interest and therefore does not violate the First Amendment." Id. at 540. The Court went on to reject the claim that the restriction violated the First Amendment because it reached noncommercial, promotional uses of the word. "The mere fact that [petitioner] claims an expressive, as opposed to purely commercial, purpose, does not give it a First Amendment right to `appropriat[e] to itself the harvest of those who have sown.'" Id. at 541 (quoting International News Serv., 248 U.S. at 239-40). San Francisco Arts & Athletics did not consider a broad prohibition against the dissemination of factual information of the type that is at issue here; therefore it did not implicate the First Amendment doctrine discussed above. Nonetheless, that case's favorable reference to International News Serv. in response to a different First Amendment argument indicates that the former case provides some authority for a possible intellectual property exception to certain First Amendment constraints that would apply outside the intellectual property context.

      Even if International News Serv. does indicate that the First Amendment permits some anti-copying protection for nonoriginal factual information, however, it must be emphasized that H.R. 2652 raises serious constitutional concerns because it provides protection that is much broader than that at issue in International News Serv. Unlike the "hot news" misappropriation tort that the Court recognized in International News Serv., the bill would not create liability only for "competitive and systematic interference with dissemination of unpublished, partially published or access-controlled information," where "the timeliness of the information makes its commercial value of short duration." Ginsburg, No "Sweat"?, 92 Colum. L. Rev. at 357;see also NBA v. Motorola, 105 F.3d at 852.

      Second, there is an important consideration that might distinguish H.R. 2652 from the prototypical First Amendment case where the government acts to limit the use of publicly available information. As we explained above, in the usual First Amendment cases, such as the ones cited at beginning of this section, the government's restriction on the dissemination of information has the intent and effect of constricting the total quantum of information that the public could put to lawful and valuable use by singling out certain disfavored messages for suppression. Because H.R. 2652 would simply regulate the means by which information generally may be re-used, it arguably could be defended as a legitimate attempt to recognize individual rights in intellectual property in order to ensure in an overall increase in the amount of available, valuable factual information (because of the heightened incentives to compile facts). See Ginsburg, No "Sweat"?, 92 Colum. L. Rev. at 386. It could be argued that such a statute -- like copyright's protection of creative expression -- would secure a wider "dissemination of information from diverse and antagonistic sources," New York Times, 376 U.S. at 266, than would result from a regime in which factual compilations are protected against reproduction only by "thin" copyright and (perhaps) by state contract law.

      We should note, however, that the above-stated rationale -- that protection against reuse and copying of factual compilations could increase, rather than decrease, the existence of useful knowledge -- would be in some tension with the premises of the Court's holdings in Feist and with the Court's and Congress's exclusion of copyright protection for facts, reflected in the Copyright Act and in cases such as Harper & Row. In those contexts, the Congress and the Court have concluded that, whereas protection against reuse of expression has the effect of increasing the output of unique and original writings, analogous protection of facts would, on the whole, impede the progress of knowledge. In addition, the strength of the argument would not doubt turn in large part on the scope of the protection afforded by H.R. 2652. To the extent it would apply even to non-commercial, transformative uses, it would appear to be far more vulnerable to constitutional attack on First Amendment grounds.

V. Possible Ways in Which H.R. 2652 Might be Narrowed

      It may be the case that any "misappropriation" statute such as H.R. 2652 enacted pursuant to the Commerce Clause "may prove difficult to reconcile with Feist's constitutionally derived endorsement of free-riding on previously gathered information." Ginsburg, No "Sweat", 92 Colum. L. Rev. at 341. Such a statute might also raise serious First Amendment problems, no matter how it is crafted. We do, however, believe that H.R. 2652 could be narrowed in several ways that would lessen the risk of constitutional invalidity. Each of these suggestions would have the effect of preserving more of what is now understood to constitute the public domain, in which facts could freely be copied in furtherance of important scientific, educational, and analogous objectives.

     The easiest and most direct way to cabin the constitutional issues would be to limit the statutory liability to the sort of "hot news" misappropriation tort that the Court recognized in International News Serv. The law could, for example, create liability for "competitive and systematic interference with dissemination of unpublished, partially published or access-controlled information," where "the timeliness of the information makes its commercial value of short duration." Ginsburg, No "Sweat"?, 92 Colum. L. Rev. at 357. The elements of a claim under such a statute could be: (i) that the plaintiff generates or collects information at some cost or expense; (ii) that the value of the information is highly time-sensitive; (iii) that the defendant's use of the information constitutes free-riding on the plaintiff's costly efforts to generate or collect it; (iv) that the defendant's use of the information is in direct competition with a product or service offered by the plaintiff; and (v) that the ability of other parties to free-ride on the efforts of the plaintiff would so reduce the incentive to produce the product or service that its existence or quality would be substantially threatened. See NBA v. Motorola, 105 F.3d at 852.

      Absent such a fundamental change in H.R. 2652, the following changes would tend to alleviate some of the constitutional concerns (20):

1. The provision could dispense with the time-sensitivity element of the International News Serv. tort of misappropriation, but still require proof that the defendant's use of the information constitutes free-riding on the plaintiff's costly efforts to generate or collect it; that the defendant's use of the information is in direct competition with a product or service offered by the plaintiff; and that the ability of other parties to free-ride on the efforts of the plaintiff would so reduce the incentive to produce the product or service that its existence or quality would be substantially threatened. Again, we emphasize that a statute of even this more limited scope would still raise substantial constitutional concerns for the reasons provided in the previous sections.

2. The prohibition in section 1202 could be expressly limited to nontransformative uses and
extractions by direct competitors in the particular market for the database in question. (21) This could be accomplished in part by expressly including a "fair use" exception akin to that contained in the Copyright Act, and other like statutory exceptions, at least as expansive as those found in the Copyright Act. (22) As noted at the outset, it may well be that H.R. 2652 is intended to incorporate something approximating the fair use standard for copyright by virtue of its reference to two of the four statutory fair use factors contained in the Copyright Act's fair use provision. Nonetheless, in light of the difficulties in determining how a fair use exception would apply to facts, given that it has thus far developed in the context of copyright, which does not protect facts, it would be advisable to be far more clear on this point than the present statute is. Moreover, for constitutional reasons, it would probably be advisable to provide even greater protection for the public's interest in freely exchanging information here than would be necessary outside the context of a statute that would provide intellectual property interests in factual information. As a result, a broad definition of fair use would be appropriate.

3. The duration of the protection could be substantially shortened, to the briefest period that would provide sufficient incentives for the data collection. Perpetual protection probably is unnecessary to provide sufficient incentive to the creation of databases.

4. Instead of effectively prohibiting certain use or extraction by subjecting it to potential treble-damage judgments, Congress could consider permitting widespread copying on reasonable terms and conditions, under a system of compulsory, nondiscriminatory licensing. That would allow the compiler to receive fair value for the cost of compiling, but might not unreasonably deter valuable reuses of the information. (23)

5. The provisions for injunctive relief and impoundment could be eliminated, in light of the disfavored status under the First Amendment of prior restraints. Cf. Campbell, 510 U.S. at 578 n.10.

6. The prohibition in section 1202 should be narrowed so that it extends, at most, only to those portions of a compilation that were gathered, organized or maintained through investment of substantial resources. There is little apparent justification for constraining third parties' use of portions of a collection that were not the result of such an investment.

VI. Conclusion

      H.R. 2652, the Collection of Information Antipiracy Act, raises difficult and novel questions of constitutional law. It is clear, however, that, under current Supreme Court case law, the bill, in its current form, raises serious constitutional concerns.


WILLIAM M. TREANOR
Deputy Assistant Attorney General



1. It is a matter of some contention whether, and to what extent, the incentives that would be created by H.R. 2652 are necessary to stimulate a significant quantum of valuable compilations of facts that otherwise would remain uncompiled, or whether currently available incentives and legal protections are sufficient to ensure the continued wide dissemination of factual compilations in the public domain. See, e.g., J.H. Reichman & Pamela Samuelson, Intellectual Property Rights in Data?, 50 Vand. L. Rev. 51, 113-36 (1997); Jessica Litman, After Feist, 17 U. Dayton L. Rev. 605, 611-13 (1992); Jane C. Ginsburg, No "Sweat"? Copyright and Other Protection of Works of Information after Feist v. Rural Telephone, 92 Colum. L. Rev. 338 (1992). This memorandum does not address the merits of this dispute, but, as we explain below, courts would be more likely to uphold the legislation against constitutional challenge if they were persuaded that it would increase, rather than decrease, the collection and use of information.

2. Increasingly, compilers of information -- particularly those who incorporate such compilations in electronic form -- package such compilations with a so-called "shrinkwrap" license (or "click-on" license, for documents posted on-line). This sort of "contract" purports to condition consumers' use of the product on the consumers' implicit agreement not to copy the information or disseminate it to others. See generally ProCD, Inc. v. Zeidenberg, 86 F.3d 1447 (7th Cir. 1996). Such contract-based restrictions might have a significant impact on the ability of users of factual compilations to copy or distribute the facts contained therein. However, because of several unresolved questions concerning the enforceability of these contracts, the efficacy of this approach is unclear. See, e.g., id. at 1453-55 (discussing whether contract claim is preempted by the Copyright Act and holding that it is not); see also Cohen v. Cowles Media Co., 501 U.S. 663, 669-71 (1991) (holding that the First Amendment does not prohibit a state from applying a "generally applicable" law of promissory estoppel to impose damages on a newspaper that revealed the identity of a source to whom it had promised confidentiality).

3. Indeed, the proposed statute is intended to supplement, rather than to supplant, any contractual remedies that the compiler might have. See 1205(a)-(b) (expressly providing that state contract law is not preempted). Accordingly, it would prevent the "use" or "extraction" of data from a collection even if the creator of the collection had disseminated it freely, without any contractual limitations.

4. The prohibition against extracting or using such information would not (at least not expressly) be limited to the use or extraction of those parts of a collection that were compiled "through the investment of substantial monetary or other resources"; instead, the prohibition apparently would apply to uses or extractions of a substantial part of a compilation, so long as the compilation itself (rather than the extracted components thereof) was "gathered, organized, or maintained . . . through the investment of substantial monetary or other resources."

5. The two other statutory factors that are identified in the fair use provision of the Copyright Act are also arguably incorporated by H.R. 2652. The first factor is "the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes." 17 U.S.C. 107(1). The second factor is "the nature of the copyrighted work." Id. 107(2). These factors are also arguably implicitly encompassed by H.R. 2652, which applies to collections of information, broadly defined, with specific provisions permitting certain acts such as educational and scientific uses (to the extent that they would not harm the potential market).

6. Read literally, for example, 1202 would appear to prevent the library from disseminating "substantial" (including "qualitatively" substantial) portions of the compilation to its patrons, and might prevent the patrons from using such compilations, since such patrons are part of a market or "potential market" for purchase of the book. Or, imagine a book that contains a great deal of unearthed factual material -- such as valuable, accurate information on the dangers of prescription drugs (see House Report at 13), a thorough historical chronology of important events, or a comprehensive amalgamation of geographical or topographical data. If a subsequent researcher, scientist or historian concludes that a "qualitatively substantial" portion of such facts are important, and therefore posts them to the World Wide Web or includes them in a later work -- or, possibly, if that later historian so much as "extracts" the facts by taking notes -- he or she might possibly violate 1202, whether or not that later work uses, incorporates, or transforms the facts in a manner that the compilation did not. See J.H. Reichman & Pamela Samuelson, Intellectual Property Rights in Data?, 50 Vand. L. Rev. 51, 135, 143 n.424 (1997). Of course, the extreme nature of these examples may counsel in favor of a construction of H.R. 2652 that would exclude them.

7. There was no contention in Feist that the Copyright Act was premised on any source of power other than the Intellectual Property Clause.

8. See, e.g., 499 U.S. at 346 ("Originality is a constitutional requirement."); id. at 363 ("[a]s a constitutional matter," copyright protection requires "more than a de minimis quantum of creativity"). See also Paul Goldstein, Copyright, 55 Law & Contemp. Probs. 79, 88 (1992) (noting that Feist Court indicated 13 times that originality was a constitutional requirement, and indicated 16 times that creativity was a requirement of originality).

9. It is important to note, however, that, due to the breadth of the definition of "information," which expressly includes works of authorship, the bill also would appear to provide protection to many factual compilations that do possess the requisite creativity necessary for copyright protection under Feist. In addition, it would appear that at least some, and perhaps many, extractions or uses barred by the bill might infringe on sufficiently original characteristics of such work -- such as unique arrangements or selections of the facts copied. We caution, however, that these valid applications of the bill might not provide much greater protection than would already be provided under the Copyright Act, although H.R. 2652 would also provide for criminal sanctions. Moreover, even these seemingly valid applications of the bill would be authorized under the Intellectual Property Clause only insofar as the legislation satisfied the requirement that the "exclusive Right[s]" being conferred were for "limited Times." Article I, Section 8, Clause 8 of the Constitution.

10. Of course, as we discuss below, too broad a construction of "harm to the potential market" would give rise to serious First Amendment concerns, and might, if particularly extreme, raise concerns under Lopez as well.

11. See Ginsburg, No "Sweat"?, 92 Colum. L. Rev. at 368 ("Feist's claim that its standard of originality is `constitutionally mandated' may impede enactment of a federal law protecting unoriginal compiled information under the Commerce Clause."); id. at 349 ("Justice O'Connor's opinion appears to enshrine a policy of free-riding in the Constitution").

12. See also Carpenter v. United States, 484 U.S. 19 (1987) (conspiracy to trade on newspaper's confidential information is within reach of federal mail and wire fraud statutes); Snepp v. United States, 444 U.S. 507 (1980) (government can, as a condition of employment, extract enforceable promise that employees will not reveal classified information they learn during their employ).

13. We note, however, that in the specific context of libel law, a plurality of the Court in one notable case drew significance from the fact that information was provided only to a limited number of subscribers for a fee. "[S]ince the credit report was made available to only five subscribers, who, under the terms of the subscription agreement, could not disseminate it further, it cannot be said that the report involves any strong interest in the free flow of commercial information." Dun & Bradstreet, Inc. v. Greenmoss Builders, Inc., 472 U.S. 749, 762 (1985) (plurality opinion) (internal quotation marks and citations omitted) (permitting recovery of damages for defamatory statement not involving matters of public concern absent a showing of actual malice).

14. International News Service did not discuss the Intellectual Property Clause, except to note that:

[i]t is not to be supposed that the framers of the Constitution, when they empowered 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' (Const. Art. 1, 8, par. 8), intended to confer upon one who might happen to be the first to report a historic event the exclusive right for any period to spread the knowledge of it.

Id. at 234.

15. Indeed, the "unusual circumstances" in International News Serv. itself may not have been limited to misappropriation simpliciter. The Associated Press alleged that the International News Service had done far more than simply republish the facts conveyed in the Associated Press's stories. The International News Service allegedly had bribed employees of Associated Press subscribers for an early look at breaking news, 248 U.S. at 231, occasionally had sold Associated Press's stories "bodily," i.e., without rewriting them, id., and had falsely represented to its readers that the news transmitted was the result of International News Services's own investigation,id. at 242. Such factors, the Court acknowledged, "accentuat[ed] the wrong," even if they were not "the essence of it." Id.

16. For example:

Defendant insists that when, with the sanction and approval of complainant, and as the result of the use of its news for the very purpose for which it is distributed, a portion of complainant's members communicate it to the general public by posting it upon bulletin boards so that all may read, or by issuing it to newspapers and distributing it indiscriminately, complainant no longer has the right to control the use to be made of it; that when it thus reaches the light of day it becomes the common possession of all to whom it is accessible; and that any purchaser of a newspaper has the right to communicate the intelligence which it contains to anybody and for any purpose, even for the purpose of selling it for profit to newspapers published for profit in competition with complainant's members.

The fault in the reasoning lies in applying as a test the right of the complainant as against the public, instead of considering the rights of complainant and defendant, competitors in business, as between themselves. The right of the purchaser of a single newspaper to spread knowledge of its contents gratuitously, for any legitimate purpose not unreasonably interfering with complainant's right to make merchandise of it, may be admitted; but to transmit that news for commercial use, in competition with complainant -- which is what defendant has done and seeks to justify -- is a very different matter.

Id. at 239.

17. This same restriction does not necessarily apply if the information is secret, confidential, or classified, and is provided to another on the express condition that it not be further disclosed. For example, the Court has upheld the constitutionality of governmental restrictions on its own employees' activities to ensure that those employees do not disclose classified information belonging to the government itself. The Court explained in Snepp v. United States, 444 U.S. 507, 509 n.3 (1980), that such restrictions on employee conduct generally will not violate the First Amendment so long as they are a "reasonable means" of protecting the government's "compelling interest in protecting . . . the secrecy of information important to our national security." Similarly, a court may provide trade secrets to a plaintiff as part of discovery in a civil lawsuit, subject to the condition that the plaintiff not further disseminate such secrets. Seattle Times Co. v. Rhinehart, 467 U.S. 20 (1984). And even a private party can create enforceable limits on the right to publish confidential information that it shares with another, pursuant to state laws of contract or promissory estoppel that are "generally applicable" (i.e., that do not single out speech for disfavored treatment). See Cohen v. Cowles Media Co., 501 U.S. 663, 669-71 (1991) (whereas First Amendment is not implicated by application of "generally applicable laws" to violations involving speech or the press, there is a greater constitutional problem where, as in Florida Star, the "State itself define[s] the content of publications that would trigger liability"). These cases would not be directly applicable to the proposed bill, however, in that they involved restrictions on the person to whom the information had been distributed under the confidentiality agreement, and not to restrictions on third parties who would be subsequent users or disseminators of such information.

18. On occasion, the Court has indicated that this demanding standard applies only to information concerning "`a matter of public significance.'" See, e.g., Florida Star, 491 U.S. at 533 (quoting Smith, 443 U.S. at 103). See also Dun & Bradstreet, 472 U.S. at 759 (plurality opinion) (speech on matters of "purely private concern" entitled to less First Amendment protection in defamation cases); id. at 764 (Burger, C.J., concurring in pertinent part); id. at 773-74 (White, J., concurring in pertinent part). But see Florida Star, 491 U.S. at 541 (omitting the "matter of public significance" standard in the Court's ultimate holding, quoted in the text above).

19. Nimmer explains that this would be so even where a great quantity of labor and expense were necessary to research and compile the facts: "Would anyone seriously suggest that the Washington Post was entitled to a copyright on the facts of the Watergate incident because its reporters, Woodward and Bernstein, through considerable labor, expense and ingenuity, discovered such facts?" I Nimmer on Copyright 2.11[E], at 2-172.30-172.31.

20. We should not be understood as suggesting that any or all of these changes would, or would not, be preferable as a matter of policy.

21. See, e.g., Jessica Litman, After Feist, 17 U. Dayton L. Rev. at 615; Ginsburg, 92 Colum. L. Rev. at 386.

22. See Reichman & Samuelson, Intellectual Property Rights in Data?, 50 Vand. L. Rev. at 146; id. at 155-57.

23. See Ginsburg, 92 Colum. L. Rev. at 386-87; Reichman & Samuelson, Intellectual Property Rights in Data?, 50 Vand. L. Rev. at 146.