SMITHKLINE BEECHAM CONSUMER HEALTHCARE, L.P., Plaintiff-Appellant, v. WATSON PHARMACEUTICALS, INC., Watson Laboratories, Inc., and Circa Pharmaceuticals, Inc., Defendants-Appellees, 211 F.3d 21

WINTER, Chief Judge: This appeal arises out of a copyright action alleging infringement of appellant’s copyright in a user’s guide and audiotape developed for its Nicorette-brand gum. Appellees, in obtaining approval to sell a competing generic nicotine gum product, were directed by the Food and Drug Administration (“FDA”) to use labeling almost identical to appellant’s copyrighted guide and tape.

The FDA acted pursuant to the Hatch-Waxman Amendments to the Federal Food, Drug, and Cosmetic Act (“FFDCA”), see Drug Price Competition and Patent Term Restoration Act of 1984 § 101, 21 U.S.C. § 355© (“Hatch-Wax-man Amendments”). Judge Chin initially-enjoined appellees from selling or shipping its nicotine gum with the infringing label. See SmithKline Beecham Consumer Healthcare, L.P. v. Watson Pharms., Inc., 63 F.Supp.2d 467, 473 (S.D.N.Y.1999) (“SmithKline /”).

Subsequently, the district court determined that the balance of the hardships tilted in favor of appellees and dissolved its earlier injunction, see SmithKline Beecham Consumer Healthcare, L.P. v. Watson Pharms., Inc., No. 99 Civ. 9214, 1999 WL 1243894, at *5-*7 (S.D.N.Y. Dec.22, 1999) (“SmithKline II”), and SmithKline appealed. Appellees cannot be liable for copyright infringement because the Hatch-Waxman Amendments require generic drug producers to use the same labeling as was approved by the FDA for, and is used by, the producer of the pioneer drug. We therefore affirm.

BACKGROUND Appellant SmithKline Beecham Consumer Healthcare, L.P. (“SmithKline”) manufactures and sells Nicorette nicotine polacrilex gum (“Nicorette”), an over-the-counter (“OTC”) product designed to help smokers overcome the cigarette habit. The FDA administrative history of Nicor-ette, which was the subject of a patent, is as follows. On January 13, 1984, Smith-Kline obtained FDA approval to sell 2 mg strength Nicorette for prescription-only use. Later, on June 8, 1992, the FDA approved prescription-only use of 4 mg Nicorette.

Finally, on February 9, 1996, the FDA approved both 2 mg and 4 mg Nicorette for OTC sale. Pursuant to 21 U.S.C. § 355(c)(3)(D)(iv), SmithKline obtained a three-year period of exclusivity— essentially an extension of the effective term of SmithKline’s Nicorette patent based on additional clinical testing — for OTC sale of Nicorette. SmithKline’s user’s guide and audiotape were developed in the course of its research into producing a method of, and product for, quitting smoking. To obtain approval for the OTC sale of Nicorette, SmithKline submitted various versions of the guide and tape to the FDA for review.

See 21 U.S.C. § 355(b)(1)(F) (requiring application for approval of drug to include “specimens of the labeling proposed to be used for such drug”). Between July 1993 and February 1996, SmithKline made approximately 70 changes to the guide and the tape at the FDA’s request. Most of the changes related to factual matters, safety, and efficiency. The tape and guide were ultimately included as part of Nicor-ette’s FDA-approved OTC labeling. On April 21, 1998, SmithKline registered a federal copyright for the guide and audiotape script.

On February 9, 1999, the day when its exclusivity period for Nicorette expired, SmithKline registered a copyright for the words and music on the tape. Shortly thereafter, appellees Watson Pharmaceuticals, Inc., Watson Laboratories, Inc., and Circa Pharmaceuticals, Inc. (collectively “Watson”) obtained FDA approval for the OTC marketing of a generic version of nicotine gum intended to compete directly with Nicorette.

To obtain that approval from the FDA, Watson had to comply with the requirement imposed by the Hatch-Waxman Amendments that “the labeling proposed for [its] new drug [be] the same as the labeling approved for” Nicorette. 21 U.S.C. § 355(j)(2)(A)(v); see also 21 C.F.R. § 314.127(a)(7) (“FDA will refuse to approve an abbreviated application for a new drug under

section 505© of the act [if] .... information submitted in the abbreviated new drug application is insufficient to show that the labeling proposed for the drug is the same as the labeling approved for , the listed drug....”). Thus, Watson’s generic nicotine gum was “accompanied by a user guide and audio tape that [we]re virtually identical to SmithKline’s.” SmithKline I, 63 F.Supp.2d at 469. Before Watson could sell its product to the public, SmithKline initiated the present copyright action, alleging willful infringement of its guide and tape.

The district court granted a preliminary injunction that effectively stopped Watson from shipping or selling its product. See id. at 473. The district court relied on a March 1999 FDA letter recounting that the agency had explained to Watson that “the ‘same labeling’ requirement d[oes] not require that the generic’s behavioral support materials be identical to the innovator’s materials” and indeed that “generic sponsors, like all other sponsors of nicotine-based smoking cessation aids, have discretion to design their own audio support materials.” Id. at 470.

Based on this representation, the district court “concluded that the FDA would have permitted Watson to use a user’s guide and audio tape that deviated to some extent from SmithKline’s materials ... [and] that were sufficiently different in wording and otherwise to avoid copyright concerns.” Smith-Kline II, 1999 WL 1243894, at *3. Subsequently, the FDA altered its position. In the face of the preliminary injunction, Watson revised its guide and tape to render them “comparable, but not identical, to SmithKline’s.” Id. However, on November 23, 1999, the FDA rejected the revised user guide.

The FDA “advised Watson that it would approve a revised version of Watson’s ‘previously approved labeling,’ ie., the virtually identical user’s guide previously approved by the FDA.” Id. To assist Watson, the FDA “marked up a copy of the previously approved user guide” and bracketed certain portions of text which could be in appropriate cases deleted or “substituted with new text .... similar to the original in tone, content and length.” Id. Nevertheless, the bracketed guide gave Watson “very little leeway to deviate from the previously approved user guide.” Id.

In essence, therefore, the FDA “determined that Watson had to copy verbatim substantially all of the text used in the SmithKline” user’s guide. Id. at *4. In December, representatives of the FDA attended a conference with the district court, at which time the court asked the FDA to “‘revisit’ the question of whether portions of Watson’s proposed user’s guide could be rewritten to change the text ‘a little bit’ to address the copyright concerns.” Id.

On December 15, the FDA wrote the court and advised that it had “declinefd] to change its approach to Watson’s labeling” and that it could not address copyright concerns because it “ha[d] never been directed by Congress to consider potential copyright rights in approving generic drug labeling.” Id. In a supplemental decision issued the same day, the FDA “adhered to its decision to require Watson to copy verbatim most of the SmithKline” user guide. Id.

In light of the FDA’s position, the district court dissolved the preliminary injunction, citing also Watson’s efforts to revise its materials, the prejudice caused Watson by delay, and the public interest in a generic nicotine gum product. See id. at *5-*7. We granted SmithKline’s motion for a stay and expedited the appeal.

DISCUSSION A preliminary injunction may issue only if the plaintiff demonstrates irreparable harm, and either a likelihood of success on the merits, or sufficiently serious questions going to the merits to make them a fair ground for litigation and a balance of hardships tipping' decidedly in its favor. See Otokoyama Co. v. Wine of Japan Import, Inc., 175 F.3d 266, 270 (2d Cir.1999).

We may overturn a district court’s decision to dissolve a preliminary injunction only if it constitutes an abuse of discretion, “which usually involves either the application of an incorrect legal standard or reliance on clearly erroneous findings of fact.” ABKCO Music, Inc. v. Stellar Records, Inc., 96 F.3d 60, 64 (2d Cir.1996); accord Niagara Hooker Employees Union v. Occidental Chem. Corp., 935 F.2d 1370, 1374 (2d Cir.1991). In ruling on an interlocutory appeal, we generally “will go no further into the merits than is necessary to decide the interlocutory appeal.” American Cyanamid Corp. v.

Connaught Lab., Inc., 800 F.2d 306, 310 (2d Cir.1986) (quoting Hurwitz v. Dirs. Guild of Am., Inc., 364 F.2d 67, 70 (2d Cir.1966)). However, “this rule is subject to a general exception — [we] may dismiss the complaint on the merits if [our] examination of the record ... reveals that the case is entirely void of merit.” Id. (dismissing trademark infringement action on appeal of preliminary injunction (quoting Hurwitz, 364 F.2d at 70)). This appeal falls within that exception.

We do not doubt that SmithKline has demonstrated the existence of substantial issues under the copyright laws, at least when they are considered in isolation. See 17 U.S.C. §§ 101 et seq. SmithKline’s guide and tape are creative works in which it has a substantial investment, and they are integral to both the marketing and use of Nicorette. Watson’s guide and tape are concededly in large

part copies of Smith-Kline’s copyrighted materials. Moreover, Watson intends to use the guide and tape in marketing a product in direct competition with SmithKline’s gum. Absent more, the propriety of a preliminary injunction would seem clear. Watson asserts that this copying, having been dictated by the FDA, is a “fair use” protected under 17 U.S.C. § 107.

The United States, in its amicus curiae brief, argues instead that in submitting its copyrighted materials for FDA approval, SmithKline gave the FDA an implied, nonexclusive license to permit or require generic drug applicants to copy the user’s guide and audiotape in their own nicotine gum packaging. Neither fair use nor implied license is clearly a defense in the present circumstances.

Watson’s use of SmithKline’s copyrighted works in its labeling is rather different from the sorts of copying traditionally deemed to constitute a fair use, e.g., copying for “purposes such as criticism, comment, news reporting, teaching ..., scholarship, or research.” See 17 U.S.C. § 107. Moreover, courts have found implied licenses only in “narrow” circumstances where one party “created a work at [the other’s] request and handed it over, intending that [the other] copy and distribute it.” Effects Assocs., Inc. v. Cohen, 908 F.2d 555, 558 (9th Cir.1990); see also Korman v.

HBC Fla., Inc., 182 F.3d 1291, 1293 (11th Cir.1999) (determining that plaintiff gave defendant radio station implied, nonexclusive license to use jingles she had written for station). However, we see little need for further examination of these possible defenses. If either were to prevail, some new law, essentially judge-made, would have to be fashioned.

In our view, the case can more easily be disposed of on the straightforward ground that the Hatch-Waxman Amendments to the FFDCA not only permit but require producers of generic drugs to use the same labeling as was approved for, and is used in, the sale of the pioneer drug, even if that label has been copyrighted. Because those Amendments were designed to facilitate rather than impede the approval and OTC sale of generic drugs, the FDA’s requirement that Watson use much of SmithKline’s label precludes a copyright infringement action by Smith-Kline.

SmithKline’s copyright claim is therefore meritless, and we need not address either the fair use or implied license defenses. The FFDCA prohibits the sale of drugs that are not first approved by the FDA. See 21 U.S.C. § 355(a); Weinberger v. Hynson, Westcott & Dunning, Inc., 412 U.S. 609, 612-13, 93 S.Ct. 2469, 37 L.Ed.2d 207 (1973). To gain FDA approval, manufacturers of new drugs must generally submit a “new drug application” (“NDA”) that demonstrates the drug’s safety and efficacy for its intended uses. See 21 U.S.C. § 355(b).

The NDA must include, inter alia, drug samples, proposed labeling, and the results of clinical safety and efficacy tests. See id.; see also 21 C.F.R. § 314.50. Generic drugs are identical to pioneer drugs that have previously obtained FDA approval; they can be marketed once the pioneer drug’s patent protection and FFDCA exclusivity periods expire. See H. Rep. No. 98-857,

Part I, at 16 (1984), reprinted in 1984 U.S.C.C.A.N. 2647, 2649; see also 21 U.S.C. § 355(j)(2)(A)(vii). In 1984, the Hatch-Waxman Amendments were enacted “to make available more low cost generic drugs by establishing a generic drug approval procedure for pioneer drugs....” 1984 U.S.C.C.A.N. at 2647. The Amendments provided that, rather than filing a complete NDA, manufacturers of generic drugs need complete only an “abbreviated new drug application” (“ANDA”), which could essentially piggyback on the pioneer drug’s human clinical trials and labeling. See 1984 U.S.C.C.A.N. at 2649.

The Hatch-Waxman Amendments reflect the FDA’s view that clinical retesting of generic drugs was “unnecessary and wasteful because the drug ha[d] already been determined to be safe and effective,” as well as “unethical because it [would] require!] that some sick patients take placebos and be denied treatment known to be effective.” Id. Bypassing redundant human testing would also speed up FDA approval for generic entrants and thus introduce price competition more rapidly once the pioneer producer’s patent and exclusivity periods expired. See Mead Johnson Pharm. Group v.

Bowen, 838 F.2d 1332, 1333 (D.C.Cir.1988) (interpreting congressional purpose of Hatch-Wax-man Amendments to be “increas[ing] competition in the drug industry by facilitating the approval of generic copies of drugs”).

Except for human clinical tests, the ANDA requires a manufacturer to submit to the FDA the same items as required in the NDA, see 21 U.S.C. § 365(j)(2)(A)(vi); 1984 U.S.C.C.A.N. at 2649 (“The only difference between a NDA and an ANDA is that the generic manufacturer is not required to conduct human clinical trials.”), as well as information to show that the generic drug has the same active ingredients, means of administration, dosage form, strength, pharmacological or therapeutic class, and labeling as the already-approved pioneer drug, and that the generic drug does not infringe any outstanding patents.

See generally 21 U.S.C. § 355(j)(2)(A). With specific regard to labeling, the Hatch-Waxman Amendments require that an ANDA “show that the labeling proposed for the [generic] drug is the same as the labeling proposed for the [pioneer] drug ... except for changes required because of [approved] differences [between the pioneer and generic drug] or because the [generic] drug and [pioneer] drug are produced or distributed by different manufacturers.” 21 U.S.C. § 355(j)(2)(A)(v).

The F^DCA defines “label” and “labeling” for these purposes as “a display of written, printed, or graphic matter upon the immediate container of any article” or “accompanying such article,” 21 U.S.C. § 321(k) & (m), which the FDA has broadly interpreted to include “[b]rochures, booklets, ... sound recordings, ... and similar pieces of printed, audio, or visual matter descriptive of a drug.” 21 C.F.R. § 202.1(Z)(2). Applying the Hatch-Waxman Amendments to the present appeal, SmithKline’s copyright claim fails.

First, its copyrighted user’s guide and audiotape constitute “labeling” for purposes of the Hatch-Wax-man Amendments. SmithKline has not contended otherwise and understandably so. The guide and tape clearly fall within the statutory and regulatory

definitions quoted immediately above. Moreover, they were submitted to the FDA as part of SmithKline’s quest for administrative approval of OTC sales of Nicorette. The guide and tape were approved only after more than two years of administrative consideration and after that consideration had led to some 70 changes in the guide and tape at the FDA’s request. Second, the FDA’s requirement that Watson use copious amounts of Smith-Kline’s copyrighted material is not a misapplication of the Hatch-Waxman Amendments.

As noted, the Amendments require that the labeling for the generic drug be the “same” as the labeling for the pioneer drug. See 21 U.S.C. § 355(j)(2)(A)(v). To be sure, as Smith-Kline noted at oral argument, “same” may be something less than “identical.” However, where the language of a document is the “same” as that of a copyrighted work, the former usually infringes the latter. Certainly, a legislative drafter would believe that a sameness requirement would lead to the creation of works that would easily fall within the copyright law’s infringement test of “substantial similarity.” See, e.g., Hamil Am., Inc. v.

GFI, 193 F.3d 92, 99 (2d Cir.1999) (“To prove infringement, a plaintiff with a valid copyright must demonstrate that: (1) the defendant has actually copied the plaintiff’s work; and (2) the copying is illegal because a substantial similarity exists between the defendant’s work and the protectible elements of plaintiffs.” (internal quotation marks omitted)), cert. denied., — U.S. -, 120 S.Ct. 1171, 145 L.Ed.2d 1080 (2000).

Indeed, the legislative history of the Hatch-Waxman Amendments suggests that whatever difference may exist between “same” and “identical” is narrow and intended to prevent misstatements rather than infringement: [A]n ANDA must contain adequate information to show that the proposed labeling for the generic drug is the same as that of the listed drug. The Committee recognizes that the proposed labeling for the generic drug may not be exactly the same. For example, the name and address of the manufacturers would vary as might the expiration dates for the two products.

Another example is that one color is used in the coating of the listed drug and another color is used in that of the generic drug. The FDA might require the listed drug maker to specify the color in its label. The generic manufacturer, which has used a different color, would have to specify a different color in its label. H. Rep. No. 98-857,

Part I, at 22 (1984), reprinted in 1984 U.S.C.C.A.N. 2647, 2655. Third, if SmithKline’s copyright claim has merit, then Watson cannot realistically use the ANDA process to sell its generic nicotine gum because it will either have to change the label and lose FDA approval or be enjoined from using a label that infringes SmithKline’s copyright. We are thus faced with a conflict between two statutes. The Hatch-Waxman Amendments require generic drug producers to use labeling that will infringe upon copyrights in labels of pioneer drugs. The Copyright Act seems to prohibit such copying. However, applying the familiar canon that, where two laws are in conflict, courts should adopt the

interpretation that preserves the principal purposes of each, see, e.g., Zenith Elecs. Corp. v. Exzec, Inc., 182 F.3d 1340, 1347 (Fed.Cir.1999) (“Unless Congress clearly indicates which of two statutes is to prevail in event of conflict, our responsibility is to interpret and apply them ‘in a way that preserves the purposes of both and fosters harmony between them.’ ” (quoting Vornado Air Circulation Sys., Inc. v. Duracraft Corp., 58 F.3d 1498, 1507 (10th Cir.1995))), the conflict is less stark and more easily resolved than it might seem.

The purposes of the Hatch-Waxman Amendments would be severely undermined if copyright concerns were to shape the FDA’s application of the “same” labeling requirement. The Amendments were intended to facilitate the introduction of generic competitors once a pioneer drug’s patent term and exclusivity periods had ended by allowing the generic producer to piggy-back upon the pioneer producer’s successful FDA application. For example, human testing by the generic producer is not required because it would be time-and-resource-consuming even though redundant.

For the very same reason, the creation and approval of new labels is avoided by the “same” labeling requirement. If labels that were “substantially similar” to copyrighted labels on pioneer drugs had to be avoided, the administrative process of approving a new label would, in cases like the present one, drain the resources of the FDA and generic producer — not to mention the problem of successive generic producers avoiding infringement of multiple copyrighted labels.

Avoiding such infringement would also delay the introduction of the generic product without advancing public, health and safety to any perceptible degree. For that reason, Congress left no room for such redundant proceedings and adopted the “same” labeling requirement. The FDA cannot be faithful to that requirement, however, without requiring labels that will often violate copyrights. If copyright law were to prevail, producers of generic drugs will always be delayed in — and quite often prohibited from' — marketing the generic product, results at great odds with the purposes of the Hatch-Waxman Amendments.

No such severe undermining of the purpose of the copyright laws would follow from the rejection of SmithKline’s claim, however. The creation of labels to be approved by the FDA, such as Smith-Kline’s user’s guide and audio tape, is ancillary to the FDA’s administrative process. The creativity of the author is focused not only on pleasing and medicating ultimate consumers but also on obtaining the administrative approval of labeling necessary to FDA approval of a drug that will be protected from competition both for the period of the patent term and FFDCA exclusivity periods.

Our point here is not only that Congress would have provided explicitly that the Hatch-Waxman Amendments trump the copyright laws had it foreseen the statutory conflict exposed by the present action, although we firmly believe that to be obvious. Our point is also that the profit sought by the creator of the pioneer drug label flows primarily from the administrative approval of the drug and the patent and exclusivity periods free from competition that follow.

The pertinent purpose of the copyright laws — to encourage the production of creative works by according authors a property right in their works so that authors will not have to share profits from their labors with free riders, see, e.g., Mazer v. Stein, 347 U.S. 201, 219, 74 S.Ct. 460, 98 L.Ed. 630 (1954) (“The economic philosophy behind the [copyright] clause ... is the conviction that encouragement of individual effort by personal gain is the best way to advance public welfare .... ”); see also Twentieth Century Music Corp. v.

Aiken, 422 U.S. 151, 156, 95 S.Ct. 2040, 45 L.Ed.2d 84 (1975) (“[T]he ultimate aim is, by this incentive, to stimulate artistic creativity for the general public good.”); cf. Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 586, 114 S.Ct. 1164, 127 L.Ed.2d 500 (1994) (emphasizing that “some works are closer to the core of intended copyright protection than others” in assessing fair use defense to claim that parody infringed on valid copyright) — is not seriously implicated by allowing the “same” labeling requirement to trump a copyright under the Hatch-Waxman Amendments.

It is simply not conceivable that, if we reject SmithKline’s claim, pioneer drug producers will so fear the copying of labels by future generic drug producers that some pioneer producers — or even one of them— will lack the incentive to create labeling needed for FDA approval. We emphasize that we do not read the Hatch-Waxman Amendments to repeal other rights under the Copyright Act of copyright owners in SmithKline’s circumstances.

Even though such an owner cannot enforce its copyright against generic drug manufacturers who are required by the Hatch-Waxman Amendments to copy labeling and who do no more than that, it still retains a copyright, if otherwise valid, in the label and might well pursue copyright claims against potential infringers in other circumstances, e.g., use of the copyrighted material in non-labeling advertisements.

Because recognition of SmithKline’s claim here would severely undermine the Hatch-Waxman Amendments while its dismissal would not impair the copyright laws, we affirm and direct dismissal of SmithKline’s complaint for failure to state a claim. The stay is lifted forthwith. . The implied license issue is somewhat superfluous in that the license would have to be inferred from the Hatch-Waxman requirement that copyright labels be infringed.

If that view of Hatch-Waxman is correct, as the present opinion holds, then the outcome is dispositively determined without further discussion of the doctrine of implied license. . Some courts have held that infringement of a copyright in commercial labeling must involve verbatim or near-verbatim copying. See, e.g., Sassafras Enters., Inc. v. Roshco, Inc., 889 F.Supp. 343, 345 (N.D.Ill.1995) (in copyright labeling claim, "[ajbsent verbatim or near-verbatim copying, no infringement exists”).

However, this heightened standard typically involves a copyrighted label of a "fact-based work[],” id. at 347, and Smith-Kline’s guide and tape are substantially more creative than the typical commercial labels afforded copyright protection. We need not decide whether the traditional "substantial similarity” infringement test or a heightened “verbatim or near-verbatim” test applies to the instant case. As discussed infra, the plain language of the Hatch-Waxman Amendments, their legislative history, and their

interpretation by the FDA all require manufacturers of generic drugs to copy the labeling of pioneer drugs "near-verbatim” to obtain ANDA approval, and this statutory mandate necessarily trumps any copyright interest in the label at issue. . In addition to looking to conflicting statutes’ principal purposes, courts have traditionally given weight to statutes’ priority of enactment and specificity in reconciling conflicts. See, e.g., FDA v.

Brown & Williamson Tobacco Corp.,-U.S.-, 120 S.Ct. 1291, 1306,-L.Ed.2d-(2000) ("The classic judicial task of reconciling many laws enacted over time, and getting them to 'make sense’ in combination, necessarily assumes that the implications of a statute may be altered by the implications of a later statute.

This is particularly so where the scope of the earlier statute is broad but the subsequent statutes more specifically address the topic at hand.... [A] specific policy embodied in a later federal statute should control our construction of the earlier statute, even though it has not been expressly amended.” (internal quotation marks, brackets, and citations omitted)). Here, of course, both time of enactment and specificity favor our

interpretation: the Hatch-Waxman Amendments were enacted subsequent to the Copyright Act of 1976, and the Copyright Act’s broad generality contrasts with the Hatch-Waxman Amendments’ specific generic drug approval scheme. . Indeed, Congress enacted the Hatch-Wax-man Amendments as a partial amendment to the patent laws.

It extended patent length for drugs to accommodate the delay caused by the FDA’s testing process and slightly weakened patentholders’ rights by permitting prospective manufacturers of generic drugs to engage in preliminary testing pursuant to AN-DAs before pioneer drugs' patents or terms of exclusivity had fully expired. See 1984 U.S.C.C.A.N. at 2714 & n. 20. .

Although commercial labeling is clearly copyrightable, see 1 Nimmer on Copyright § 2.08[G], at 2-135 (“It is clear that [17 U.S.C. § ] 102(a)(5) includes prints and labels used for articles of merchandise under the general protection accorded to pictorial, graphic and sculptural works.” (citing H.R.Rep. No. 94-1476, at 54 (1976) [1976 U.S.C.C.A.N. 5659, at 5667] (“There is no intention whatever to narrow the scope of the subject matter now characterized in

section 5(

k) as ‘prints and labels used for articles of merchandise.' ”))), it has been recognized that the "danger lurking in copyright protection for labels is that the tail threatens to wag the dog — proprietors at times seize on copyright protection for the label in order to leverage their thin copyright protection over the text ... on the label into a monopoly on the typically uncopyrightable product to which it is attached.” Id. § 2.08[G][2], at 2-138. "Used in that fashion, the copyright serves 'primarily as a means of harassing competitors,’ and thus fails 'nine times out of ten.’ " Id. at 2-139.

Here, although the labeling at issue is more creative than that in the “familiar” commercial labeling cases, see Sassafras, 889 F.Supp. at 345, SmithKline’s copyright claim is arguably weaker than even the typical commercial labeling case, because the copyrighted text was submitted to obtain FDA approval and consequent márket exclusivity.