graphical material mistakenly will be equated with text-only pages for character counting. (Trial Tr. 1005-6) Plaintiffs again scored poorly, with their least expensive journal on a per page basis nearly three times more expensive than defendants' most expensive journal on a per page basis. (Def. Ex. C) In sum, regardless of the measure used, G&B's journals consistently scored at the bottom. Ribbe also demonstrated that plaintiffs' poor showing by such measures extended to journals in other disciplines. (Def. Exs. K, O) Finally, Ribbe averaged the data for each publisher, as was done in Table 2 in the 1988 Physics Today article. In every category, G&B received the worst score.
No evidence was received contesting Ribbe's analysis. Rather than challenging Ribbe's report, plaintiffs' expert King admitted that he had no factual basis for believing Barschall's results would have been different had he used total citations rather than impact (Trial Tr. 472), and Kingma forthrightly acknowledged that correcting for the alleged errors he identified would not affect Barschall's results. (Id. 138) Plaintiffs' own expert thus concurred with Ribbe's assessment that Barschall's results were reliable. (See supra note 7).
Weighing these factors, the Court easily concludes that Barschall's methodology reliably establishes the costs/kilocharacter and costs/impact of the journals in his survey, as well as the rankings of journals and publishers based on those measures. Plaintiff made no serious effort to attack the data relied upon by Barschall or his method of calculation, other than to assert that his position as an APS officer rendered him fatally biased. In the absence of affirmative evidence suggesting that Barschall's conclusions were erroneous, and particularly in light of Ribbe's verification of Barschall's results, G&B has failed to carry its burden of demonstrating that defendants' methodology was not sufficiently reliable to establish the results reached.
B. Whether Barschall's Study Establishes the Proposition for Which it was Cited
Most of G&B's evidence at trial was directed towards establishing that Barschall's methodology does not establish journal cost-effectiveness and fails to account for various differences among journals, making his analysis an inappropriate basis upon which physics journal subscribers should rely when making purchase decisions. Plaintiffs thus argue that defendants' promotional claim that Barschall's study proves their journals to be cost-effective is literally false, and that defendants' promotional use of Barschall's study carries a false implied message that Barschall's study, to the exclusion of all other factors, should be used as the basis for journal purchase decisions.
1. The Literal Message of Defendants' Promotions
In order to determine whether defendants' promotions were literally false, the Court first must determine whether plaintiffs accurately characterize the literal message conveyed by defendants' promotions. In making this determination, the Court considers the context of the challenged portion of the promotional material, scrutinizing it "within a meaningful and reasonable context, including not only the entire sentence of each portion of challenged text but also the entire text and overall message of the [material.]" JR Tobacco of America, Inc. v. Davidoff of Geneva, 957 F. Supp. 426, 432 (S.D.N.Y. 1997); see Avis, 782 F.2d at 385; Vidal Sassoon, Inc. v. Bristol-Myers Co., 661 F.2d 272, 276 (2d Cir. 1981).
Plaintiffs' contend that defendants' promotions claim that Barschall's method establishes overall cost-effectiveness. The Court disagrees. The 1988 Articles themselves convey a message only as to cost-effectiveness as measured by cost/kilocharacter or cost/impact. The Physics Today article states in its subtitle that the survey shows journal "cost-effectiveness, as measured by the ratio of the printed character to the frequency with which the articles are cited." (Pl. Ex. 3 at 56) (emphasis added) The article thus immediately qualified and defined its use of the term "cost-effective," stating the factors which entered its assessment of cost-effectiveness without implying that the study was the only conceivable measure of cost-effectiveness. Similarly, in the opening paragraph of the article, Barschall states that subscription decisions
are usually based on the research interests of the users of the library, but the decision-making process can be improved if a quantitative measure of the cost-effectiveness of journals is available. An oft-used measure is the cost per printed character; another is the frequency with which articles in the journal are cited, often referred to as the "impact." The ratio of these two measures is perhaps the best indicator of a journal's cost-effectiveness.
(Id.) (emphasis added) As the underscored language indicates, Barschall explicitly acknowledged that subscribers must consider their particular usage needs, stated that there are multiple measures of cost-effectiveness, and only opined that his ratio might be the best.
Defendants were also careful to qualify their claim in the cover letters which accompanied their promotional redistributions of the study. APS's May 1988 letter to society members reported that the Barschall study "shows that the APS publications are the greatest bargains around. Measured by the ratio of the number of citations to the cost of the journal, the APS publications score among the highest of all, with some of our competitors orders of magnitude lower." (Pl. Ex. 53) By juxtaposing the qualitative claim "greatest bargain" immediately next to a statement of the measures used in the study, the letter conveys only that APS publications are a great "bargain" as measured by the stated yardsticks, not that those yardsticks are the only conceivable measures one might use. A September 1988 APS letter to non-member subscribers similarly qualified its claim, stating that APS journals "remain extremely inexpensive when compared to other physics publications. When their utility is also included, are (sic) measured by the Citation Index, they are even less expensive." (Pl Ex. 65A) (emphasis added)
As for defendants' presentations to librarians, the evidence reflects that the librarians were shown the data from Barschall's study, allowing them to see the factors which entered into defendants' claim of cost-effectiveness. (E.g., Pl. Ex. 122) Moreover, plaintiffs adduce no evidence that defendants failed to qualify their claim precisely as they did in all of their printed material, dating back to their use of the 1986 Article. (See Pl. Exs. 33 and 34)
Plaintiffs make much of the cancelled September 1988 AIP letter to subscribers, which stated, without qualification, that Barschall's study placed AIP's publications "in the highest ranks of quality and value." (Pl. Ex. 65B; Pl. Post-Trial Mem. at 12-13) A quantitative test of certain cost factors does not necessarily prove the existence of subjective features such as "value" or "quality." Were defendants now to be asserting that Barschall's methodology proves that their product is superior to that of plaintiffs, serious concerns would be present. But the fact is that the sole material which spoke without qualification of "value" or "quality" was never disseminated, and thus, having not been put to promotional use, cannot substantiate a finding of a Lanham Act violation.
Nor would injunctive relief be necessary to ensure that defendants do not make such a claim. Defendants apparently now acknowledge that Barschall's analysis does not demonstrate abstract product superiority or quality. (Def. Post-Trial Mem. at 68) There is no need to enjoin the making of a claim which defendants have never in fact disseminated and do not threaten to disseminate in the future. For example, in Knickerbocker Toy Co., Inc. v. Azrak-Hamway Int'l, Inc., 668 F.2d 699 (2d Cir. 1982), the defendant's catalog allegedly used artwork illustrating the plaintiff's product in order to sell the defendant's competing product. The defendant, however, had withdrawn the catalog after only four days of distribution, had received no orders for its competing product, and contemplated no further use of plaintiff's artwork. Id. at 702-03. The district judge, in a conclusion affirmed by the Second Circuit, found it unnecessary to determine whether there had been a Lanham Act violation because, "since the irregularities had been eliminated and were unlikely to recur there was no basis for any injunctive relief." Id. at 702. The only compelling distinction between Knickerbocker and the instant case is that, unlike in Knickerbocker, defendants have made no distribution whatsoever of the potentially offending material, so that injunctive relief would be even more inappropriate in this case. See also Burndy Corp. v. Teledyne Indus., Inc., 748 F.2d 767, 774 (2d Cir. 1984) (injunctive relief "wholly unnecessary" given defendant's rectification of any infringement and the absence of any likelihood of repetition); Sanborn Mfg. Co. v. Campbell Hausfeld/Scott Fetzer Co., 997 F.2d 484, 489 (8th Cir. 1993) (district court was not erroneous in declining to grant an injunction where defendant had voluntarily taken remedial action); Reader's Digest Ass'n, Inc. v. Conservative Digest, Inc., 261 U.S. App. D.C. 312, 821 F.2d 800, 807 (D.C. Cir. 1987) ("when a defendant has ceased its infringing conduct and shows no inclination to repeat the offense," no injunction need issue); Schutt Mfg. v. Riddell, Inc., 673 F.2d 202, 207 (7th Cir. 1982) (no abuse of discretion for district court to find that "because [defendant] clearly did not threaten to persist in or resume the allegedly infringing or unfair conduct, equitable relief would be inappropriate"); In re Circuit Breaker Litigation, 860 F. Supp. 1453, 1456 (C.D. Cal. 1994) ("Courts grant injunctions if they suspect that defendant might infringe again, even if the infringement has ceased for the moment. If, on the other hand, the defendant infringed innocently, ceased before judgment and assured the court that it has no intention of infringing in the future . . ., courts usually deny requests for permanent injunctions.") (internal citations omitted), aff'd sub nom. Westinghouse Elec. Corp. v. General Circuit Breaker & Elec. Supply Inc., 106 F.3d 894 (9th Cir. 1997), petition for cert. filed, 65 U.S.L.W. 3085 (U.S. June 25, 1997) (No. 97-2).
The Court therefore finds that defendants' promotions claimed only that their journals were cost-effective as measured by cost/kilocharacter or cost/impact.
b. Factors Purchasers Should Consider
Most of plaintiffs' evidence was designed to show that there are legitimate reasons why their journals cost more as measured by Barschall's methodology, and that subscribers would be poorly-served if they relied only on the results of Barschall's methodology in making purchasing decisions.
For example, plaintiffs argue that they publish more specialized journals (so-called "niche" publications) catering to smaller audiences than do defendants. Additionally, unlike defendants, plaintiffs do not offset their publication costs by asking authors to pay "page charges" for published articles. Consequently, they must recover higher fixed costs of publication from a smaller pool of subscribers, resulting in higher subscription costs. Plaintiffs also emphasize that the physics community needs commercial publishers (Trial Tr. 535, 544), but that were purchase decisions based solely on Barschall's study, niche publications would be driven out of business. As a result, articles could only be published in more generalized journals, and thus subscribers needing only articles in a specialized subdiscipline would effectively be forced to pay for unneeded articles. (Id. 444-45)
As the Court suggested during trial (Trial Tr. 551), however, these arguments are simply irrelevant to the claim that defendants' material is literally false -- the Barschall articles contain no analysis of the reasons for the cost differentials. In an attempt to work around this considerable obstacle, plaintiffs argue that defendants' materials contain an implied message that journal purchase decisions should be based only on Barschall's study. This, they contend, is fallacious because their evidence demonstrates that decisions should be based on a more nuanced and complex analysis.
The Court finds no such implied message in defendants' use of the Barschall study. The 1988 articles, and the cover letters which accompanied defendants' promotional distribution of it, identify a feature of their journals which the study irrefutably demonstrates to be true -- cost-effectiveness as measured by cost/kilocharacter or cost/impact. Nothing in the articles or cover letters suggests to the reader that they use this information to the exclusion of all other factors in making purchasing decisions.
Furthermore, to the extent that the promotion necessarily contains an implied message urging consumers to buy the promoted goods based on their advertised qualities, every advertisement contains such an implied pitch. Urging consumers to make a particular purchase, however, without more, is not actionable under the Lanham Act. See 15 U.S.C. § 1125(a) (Lanham Act reaches only false or misleading descriptions or representations of fact). For example, in Castrol, the plaintiff prevailed because although the defendant's test showed its motor oil to have faster lubrication times than the motor oils of its competitors, the defendant made the further unsubstantiated claim that its oil better protected against engine wear. Castrol, 957 F.2d at 64-65. G&B, however, apparently would have the defendants in Castrol held liable if the plaintiffs could show that consumers ideally should not purchase motor oil based on "pumpability" considerations alone.
Plaintiffs misapprehend the purpose and scope of the Lanham Act's limits on commercial advertising. The Act contemplates a free market into which advertisers are not to inject false or misleading information, but in which, as in any free market, it is up to the consumer to see to it that only the product that best serves the consumer's needs is bought. If G&B believes librarians will make more optimal decisions if they consider information other than that provided by defendants, its solution is to augment rather than censor the available truthful information. G&B's arguments that librarians will not adequately discharge their responsibilities if they rely solely on the output of Barschall's methodology are thus best addressed to librarians, not this Court.
2. Whether the Test Establishes the Message
Because plaintiffs have failed to demonstrate that Barschall's methodology yielded unreliable cost/kilocharacter and cost/impact results, their claim that his study fails to support the proposition for which it was cited -- that, as measured by cost/kilocharacter and cost/impact, defendants' journals are cost-effective -- necessarily fails. Plaintiffs' evidence that there are reasons why their journals cost more in no way makes false the statement that they do cost more. Likewise, that librarians should consider more than cost-effectiveness as measured by cost/kilocharacter or cost/impact in making purchasing decisions does not make the cost/kilocharacter or cost/impact figures false. As defendants rightly submit, "the methodology is not false simply because the results must be used with care." (Def. Post-Trial Mem. at 49) The Court therefore finds for defendants on plaintiffs' cause of action, and does not reach defendants' affirmative defense.
Barschall's methodology has been demonstrated to establish reliably precisely the proposition for which defendants cited it -- that defendants' physics journals, as measured by cost per character and by cost per character divided by impact factor, are substantially more cost-effective than those published by plaintiffs. Plaintiffs have proved only the unremarkable proposition that a librarian would be ill-advised to rely on Barschall's study to the exclusion of all other considerations in making purchasing decisions. This consideration in no way makes Barschall's study or defendants' descriptions thereof false, and accordingly judgment is granted to defendants.
It is SO ORDERED.
Dated: August 26 1997
New York, New York
Leonard B. Sand