Playboy Enterprises, Inc. v. Chuckleberry Publishing, Inc., Publishers Distributing Corporation, Arcata Publications Group, Inc., Tattilo Editrice Spa

MANSFIELD, Circuit Judge

(concurring in part and dissenting in part):

Although I concur in upholding the injunction against use of the title PLAY-MEN, I do not subscribe to some of the reasoning of the majority and of the district court, particularly the reliance upon the concept of “subliminal association,” which in my view only seems to confuse rather than to clarify. In addition, I dissent from the majority’s affirmance of the injunction against the use of “Playmen” in the much smaller subtitle of the appellant’s male sophisticate magazine ADELINA (see Appendices A and B hereto, showing typical front pages of both magazines). In my view the subtitle poses no substantial likelihood of confusing an “appreciable number” of consumers. McGregor-Doniger Inc. v. Drizzle Inc., 599 F.2d 1126, 1138 (2d Cir. 1979). Accordingly, I would reverse the broader injunction and the award of attorneys’ fees.

The ADELINA and PLAYBOY titles, printed in equivalent large-sized block type letters across the tops of their respective magazine covers (see Apps. A & B), look nothing alike. Even when only the top left corners of the covers are visible, as is typical at a newsstand overlapped display, the first three letters of the two titles are completely different. There is thus no reasonable possibility that upon a quick glance, which is the first step in the four-step “impulse buying” process, a consumer would confuse the two publications. This is fully supported by the testimony of Playboy’s own expert Scott, who acknowledged unequivocally that “[tjhere is zero likelihood of confusion as to product(emphasis added). Scott also testified that any potential confusion as to source (i.e., that PLAYBOY sponsors ADELINA) caused by the subtitle “America’s Edition of Italy’s Playmen” is “tremendously less than” would be caused by the much larger title PLAYMEN. He quantified the extent of the possible source confusion at “[l]ess than ten per cent.” Scott’s testimony, considered along with the absence of any evidence of actual consumer confusion between ADELINA (several issues of which were marketed in the U.S.) *572and PLAYBOY, should have precluded a finding of likelihood of appreciable confusion arising from the subtitle.

The majority and the district court base their contrary conclusion on several considerations. First, and foremost, the majority endorses the district court’s finding that by reason of the subtitle consumers will be confused by “subliminal trademark association,” 486 F.Supp. at 428, which defendants could exploit to gain a toehold in Playboy’s market.1 Even assuming the validity and viability of this amorphous concept of “subliminal confusion,” its importance when applied to the small subtitle is highly dubious since Scott testified that subtitles play a minimal role in the consumer’s purchasing decision.2 More significantly, the prior decision relied on for the concept of subliminal confusion, Londontown Manufacturing Co. v. Cable Raincoat Co., 371 F.Supp. 1114 (S.D.N.Y.1974), referred to subliminal association merely as a psychological explanation of product or source confusion, and never viewed it as a separate type of confusion that could independently warrant a finding of infringement.3 It is error to assign such independent significance to psychological speculations of this sort in light of Scott’s unqualified denial of any product confusion and any appreciable source confusion.

Second, the majority and the district court give great weight to the defendant’s “exploitive purpose in adopting the subtitle.” Maj. Op. 571. I have trouble with various aspects of the district court’s finding that Tattilo adopted the PLAYMEN title and later the subtitle with the intention of causing and trading on any possible consumer confusion. For instance, in its first opinion the district court relied on a Time magazine article about the genesis of appellant’s Italian PLAYMEN in 1967, which is hardly reliable, much less admissible, evidence of Tattilo’s intent. 486 F.Supp. at 425. In its second opinion, the district court stated erroneously that “Tattilo claimed not to know why she and her employees had chosen the title Playmen." 511 F.Supp. at 491. In fact, she explained that in 1967, when PLAYBOY was still banned in Italy, she chose the title PLAY-MEN to provide reader identification with the pre-existing magazine MEN published by her.

In the end, the “proof” of bad intent relied upon by the majority and the district court consists of Tattilo’s failure to disprove an intent to confuse. While I agree that the use of a similar mark may sometimes give rise to an inference of bad intent absent a reasonable explanation, mere similarity cannot constitute conclusive proof. “In no case [has this Court] determined a senior user’s right to injunctive relief solely on the basis of the similarity of the marks and the proximity of the products.” Vitarroz v. *573Borden, 644 F.2d 960, 966 (2d Cir. 1981). See, e.g., id. (“BRAVOS” chips does not infringe “BRAVO’s” crackers); Affiliated Hospital Products, Inc. v. Merdel Game Manufacturer’s Co., 513 F.2d 1183 (2d Cir. 1975) (“KICK’ER” and “KIK-IT” allowed for similar table-top soccer games); Beech-Nut Inc. v. Warner-Lambert, Inc., 480 F.2d 801 (2d Cir. 1973) (“BREATH SAVERS” and “BREATH PLEASERS” allowed for similar breath mints).

Even assuming, however, that there was adequate proof of Tattilo’s bad intent, this only gives rise to a presumption of the ultimate fact to be proved — the actual likelihood of confusion. See Perfect Fit Industries, Inc. v. Acme Quilting Co., 618 F.2d 950, 954 (2d Cir. 1980). Such a presumption must yield in the face of evidence bearing more directly on the issue of likelihood of confusion. Here, Scott’s clear testimony and the absence of any proof of actual confusion constitute evidence requiring the rejection of such presumed confusion.

Third, the majority and district court rely on those factors justifying the injunction against the PLAYMEN title as further support for the injunction against the use of “Playmen” anywhere on the magazine’s cover. Such evidence, less than overwhelming even when considered with respect to the PLAYMEN title, provides little support for the general injunction against use of the word in a subtitle which is less than one-eighth the size of the title ADELINA and even there refers only to “America’s Edition of Italy’s Playmen” (emphasis added). Appellant is permitted by law to publish PLAYMEN as the magazine’s main title in Italy.4

Although PLAYBOY and PLAYMEN have competed head-to-head in Italy since 1972, with monthly circulations of roughly 135,000 and 98,000, respectively, Playboy, which had the burden of proof, introduced no evidence of actual confusion aside from a few readers’ letters. Nor did it conduct a market survey, which would be the most probative evidence, to test for any confusion between PLAYBOY and PLAYMEN or ADELINA and, if so, whether it was substantial. The burden is not on appellant to prove a negative, i.e., the absence of confusion. To the contrary, a plaintiff’s failure to prove actual confusion justifies an inference that there is no likelihood of confusion. McGregor-Doniger Inc. v. Drizzle Inc., supra, 599 F.2d at 1136. Moreover, there is currently a proliferation of magazines using the prefix “PLAY,” many of which are sex-oriented magazines that have been registered (e.g., “PLAYGIRL,” “PLAYGUY,” “PLAYERS,” “PLAYTHINGS”). While the fact that some of these appeal to different consumer markets may minimize the possibility of product confusion, Scott testified, and readers’ letters confirmed, that these magazines are just as likely as the PLAYMEN title to engender confusion as to source.5 Yet the Ninth Circuit, in HMH Publishing Co., Inc. v. Lamber, 482 F.2d 595, 597 (9th Cir. 1973), affirmed a finding of the district court that “the name ‘Playgirl’ is not confusingly similar to, nor would it be likely to cause confusion to members of the public with either ‘Playmate’ or ‘Playboy.’ ” Moreover, a later suit by PLAYBOY in the Northern District of Illinois against Lambert, the publisher of PLAYGIRL, was dismissed with prejudice, but without payment, pursuant to a settlement agreement in which both agreed that the mark “Playgirl in itself is not confusingly similar to Playboy or Playmate.”

*574Thus the prevalance of competing or related magazines using titles with the “PLAY” prefix reduces the distinctiveness of Playboy’s mark and the degree of protection to which it is entitled. See Callman, 3 Unfair Competition Trademarks and Monopolies § 82.1(1) at 759 n.42 (3d ed.). See, e.g., Beech-Nut Inc. v. Warner-Lambert, Inc., supra, 480 F.2d 801; Durox Co. v. Duron Paint Manufacturing Co., 320 F.2d 882 (4th Cir. 1963); Triumph Hosiery Mills, Inc. v. Triumph International Corp., 308 F.2d 196 (2d Cir. 1962). Finally, even with respect to the PLAYMEN title the likelihood of product confusion is limited by what Judge Sofaer found to be “[a] discernible difference in explicitness and tone [that] unquestionably exists,” with the result that “PLAYMEN seems likely to repel and shock many more.” 486 F.Supp. at 432-32 & n.23.6 Such equivocal evidence of the likelihood of confusion between PLAYBOY and PLAYMEN is insufficient to support an injunction against appellant’s reference in its magazine cover entitled ADELI-NA to “America’s Edition of Italy’s Play-men” in a subtitle less than one-eighth the size of the ADELINA title. (See Apps. A & B).

Finally, the majority and the district court find further support for the general injunction in Scott’s testimony that source confusion might well increase if the subtitle’s use of “Playmen” became more prominent with increased size or repositioning. Nevertheless, the undisputed testimony is that, as currently employed, the subtitle engenders no product confusion and at most minimal source confusion. The proper course is to enjoin the subtitle use if and when the likelihood of confusion becomes appreciable, and not to anticipate what may never occur as long as ADELINA remains the dominant title. Berlitz Schools of Language of America v. Everest House, 619 F.2d 211, 215-16 (2d Cir. 1980).

Accordingly, I would reverse as clearly erroneous the general injunction against the use of “Playmen” in the subtitle or anywhere else on the cover, and also reverse the award of attorneys’ fees.

*575APPENDIX “A”

*576APPENDIX “B”

. The fact that appellants may be able to enter the market with substantially less in advertising expenditures than Playboy has undertaken throughout the years to acquire its reputation is not a dispositive factor establishing infringement. HMH Publishing Co., Inc. v. Lambert, 482 F.2d 595, 598 (9th Cir. 1973) (quoting Carter-Wallace, Inc. v. Procter & Gamble Co., 434 F.2d 794, 800 (9th Cir. 1970)).

. It is significant that the district court first created the category of subliminal confusion with reference only to the PLAYMEN title, where the likelihood of source confusion was far greater than for the subtitle and was by itself sufficient to warrant a finding of infringement. 486 F.Supp. at 428-29.

. The full passage from Londontown Manufacturing Co. v. Cable Raincoat Co., 371 F.Supp. 1114, 1118 (S.D.N.Y.1974), is as follows, with the sentence omitted in the district court’s opinion italicized (see 486 F.Supp. at 428):

“If consumers come to think of a wire fence as a reminder of a cyclone, then a competitor may not remind them of his wire fence as a tornado. [Citation omitted]. The reason is that advertising and trademarks rely on impressions. The consumer does not memorize the mark. He has a feeling about it from past exposure. That feeling may be vague, subliminal it is said, but it comes to consciousness when the article is seen with the trademark affixed. The ultimate test is, of course, whether the public is likely to be confused by the similarity of the marks as to the source of the goods. (Latter emphasis added).

. The Italian courts have denied PLAYBOY any protection against appellant’s use of the title PLAYMEN in Italy.

. In this regard both the majority and the district court take somewhat inconsistent positions. While they discount the proliferation of “PLAY-” magazines on the ground that they are directed at different audiences than PLAYBOY and PLAYMEN, they nevertheless rely on evidence of confusion as the source concerning PLAYGIRL in the form of readers’ letters to support their conclusion about confusion as to source caused by the use of “Play-men.”

. The district court appears to fluctuate on this issue: to support its finding of irreparable harm warranting a preliminary injunction it found “discernible differences;” yet to support its later conclusion of product confusion justifying a permanent injunction it concluded that the difference “in terms of content, format and appearance ... were superficial.” Compare 486 F.Supp. at 431-32 & n.23 with 511 F.Supp. at 488.