Simon & Schuster, Inc. v. Dove Audio, Inc.

970 F.Supp. 279 (1997)

SIMON & SCHUSTER, INC. and William J. Bennett, Plaintiffs,
v.
DOVE AUDIO, INC., Defendant.

No. 95 Civ. 6012 (LBS).

United States District Court, S.D. New York.

July 11, 1997.

*280 *281 *282 Kay Collyer & Boose, LLP, New York City (Marcia B. Paul, Gregory J. Ikonen, of counsel), for Plaintiffs.

Kenneth David Burrows, New York City, (Hilary B. Miller, Stacy Grossman, of counsel), for Defendant.

OPINION AND ORDER

SAND, District Judge.

Plaintiffs Simon & Schuster, Inc. ("S & S") and William J. Bennett are the publisher and editor, respectively, of The Book of Virtues and related print and audiobook publications. Defendant Dove Audio, Inc. ("Dove") is the publisher of an audiobook, The Children's Audiobook of Virtues. When this action was filed, Dove also intended to publish a print book, The Children's Book of Virtues. Plaintiffs contend that Dove is liable for violation of Section 43(a) of the Lanham Act, 15 U.S.C. § 1125(a), common law trademark infringement and unfair competition, and violation of the New York anti-dilution statute, N.Y.Gen. Bus.Law § 368-d. The case is before the Court for findings of fact and conclusions of law after a bench trial.

PROCEDURAL BACKGROUND

Plaintiffs commenced this action on August 9, 1995. Pending final adjudication, Dove consented to be enjoined "from the manufacture, publication, advertisement, distribution and sale of any audio and/or print books under the title(s) The Children's Audiobook of Virtues and/or The Children's Book of Virtues, or under any other title confusingly similar to plaintiffs' title and alleged trademark `The Book of Virtues.'" Preliminary Injunction Upon Consent dated August 14, 1995. A condition of Dove's consent was that plaintiffs post a bond of $500,000 as surety for the preliminary injunction.

Dove subsequently moved for judgment on the pleadings, pursuant to Fed.R.Civ.P. 12(c). Dove contended, among other things, that plaintiffs' trademark "The Book of Virtues" is generic and therefore not entitled to trademark protection. Plaintiffs argued in opposition that the "The Book of Virtues" is a suggestive mark. Judge Allen G. Schwartz, to whom this case was then assigned, denied Dove's motion. See Simon & Schuster, Inc. v. Dove Audio, Inc., 936 F.Supp. 156 (S.D.N.Y.1996) (hereinafter, the "Opinion"). Judge Schwartz held that plaintiffs' mark "does not fall into the generic category." Id. at 161. Judge Schwartz also found that plaintiffs' trademark "does not rise to [the level of a suggestive mark], but rather falls into the category of descriptive marks." Id. at 162.

Dove subsequently filed a motion for reargument and clarification of the Opinion. Dove inquired whether it would be precluded from introducing evidence of the generic nature of plaintiffs' mark at trial. Judge Schwartz denied Dove's motion for reargument, holding that

[i]n its Opinion, the Court rejected defendant's argument that plaintiffs' mark is generic and found that, as a matter of law, "The Book of Virtues" mark is not generic. Accordingly, defendant is precluded from introducing evidence at trial to support its theory that plaintiffs' mark is generic; however, defendant is not precluded from introducing evidence showing that plaintiffs' mark lacks secondary meaning.

Order dated December 31, 1996, at 1-2. Judge Schwartz noted that, in reaching the earlier decision, "the Court considered certain factual materials submitted to the Court by defendant." Id. at 1 n. 1.

The case was subsequently transferred to the docket of the undersigned for trial. A bench trial commenced on January 27, 1997, and concluded on February 5, 1997. The following constitutes the Court's findings of fact and conclusions of law, pursuant to Fed. R.Civ.P. 52(a).

*283 FINDINGS OF FACT[1]

I. The Parties

S & S is a New York corporation engaged in the business of publishing with its principal place of business in New York City. SF 1. Bennett, a citizen and resident of the state of Maryland, is a well-known public figure who has served as Secretary of Education and Director of the Office of National Drug Control Policy. SF 2.

Dove is a California corporation engaged in the business of publishing with its principal place of business in Hollywood, California, and an office in Williamsville, New York. SF 3.

II. Plaintiffs' Publications and Adoption of the Mark

In November 1993, S & S published a hardcover print book entitled "The Book of Virtues" and subtitled "A Treasury of Great Moral Stories." SF 4. The book consists of an introduction by Bennett and ten chapters with headings such as "Self-Discipline," "Courage," "Honesty," and "Faith." Each chapter begins with a two-page commentary written by Bennett and includes a selection of stories, essays, poems, and other literary material, commented upon and edited by Bennett. The book "is intended to aid in the time-honored task of the moral education of the young." William J. Bennett, The Book of Virtues 11 (1993); Bennett Aff. ¶ 10.

Bennett chose the title "The Book of Virtues" by scribbling long lists of potential titles "until this one evolved. Immediately, I knew it was the right title: it conveyed my message, it stood out, it was distinctive, and it was effective and yet interesting." Bennett Aff. ¶ 14. He also testified that he considers himself "pretty well read" in philosophical literature, and he acknowledged that, before selecting the title, he had either read or was familiar with a number of philosophical tracts that have the word "virtue" in their titles. Tr. 15-18.

The Book of Virtues proved to be a tremendous commercial success. It reached the number one position on The New York Times best-seller list on January 16, 1994, and remained on that list for a total of 88 weeks. SF 8. The Book of Virtues also appeared on best-seller lists in USA Today and Publisher's Weekly. Martin Aff. ¶ 6. To promote The Book of Virtues, Bennett embarked on a national speaking tour, including numerous national television and radio appearances. SF 12, 27. By February 1994, four months after publication, the book was in its twelfth printing with 547,000 books in print. Overall, S & S has sold nearly 2.2 million hardcover copies since the book's publication. SF 7; JX 42. The book was an alternate selection for "The Book of the Month Club" and was advertised extensively in The New York Times, The Washington Post, The Los Angeles Times, and other national publications. SF 8; Martin Aff. ¶ 24.

Prior to the publication of The Book of Virtues, Bennett intended to publish a children's book entitled The Children's Book of Virtues, and he discussed that plan with his editor at S & S. SF 17. To date, plaintiffs have sold over 15 book products under the mark "The Book of Virtues" or a closely similar title, including the following hardcover books: The Children's Book of Virtues (published October 1995), an illustrated version of The Book of Virtues designed for very young children; The Moral Compass (published October 1995), a companion volume to The Book of Virtues; and The Book of Virtues for Young People (published March 1995), an educational reader published by an S & S affiliate for the school and library market. SF 4, 16, 19-22. S & S has also published five audiobooks: The Book of Virtues, Vol. I and The Book of Virtues, Vol. II (published May 1994 and November 1994, respectively); The Children's Book of Virtues *284 (published April 1996); The Moral Compass (published September 1995), and a compact disc entitled The Children's Book of Virtues Audio Treasury (published October 1995).[2] SF 14, 18-19.

The Book of Virtues and its related products have received substantial attention in the press, which has published numerous articles discussing the commercial success of the products as well as their identification with Bennett. Upon its release in November 1993 and in the following year, The Book of Virtues was reviewed in a wide variety of publications. JX 35. On December 20, 1993, Publisher's Weekly reported a retailer as saying "the book we can't get that everyone is screaming for is The Book of Virtues." JX 32. The book was included in the book section of New York magazine's "180 Best Christmas Gifts" in December 1993. SF 9. On March 7, 1994, a full-page essay in Time recounted the book's commercial success and popular appeal and concluded that "[t]he Bennett book ought to be distributed, like an owner's manual, to new parents leaving the hospital." JX 35. On June 13, 1994, Newsweek featured a cover story entitled "The Politics of Virtue: The Crusade Against America's Moral Decline," with caricatures of Bennett, Hillary Clinton, and Peggy Noonan on the cover. The Newsweek article stated that The Book of Virtues "remains hot beyond expectations," referred to Bennett as "a cottage industry of character education," and discussed his plans for a sequel, textbook versions, and a television series. SF 10; Bennett Aff. ¶ 24; JX 103. The Book of Virtues has also played a role in California and national politics. The book was featured in a 30-second campaign commercial which aired in California in 1994 for Representative Michael Huffington, a California Republican then running for the Senate.

Media attention continued in 1995 and 1996. In July 1995, The New Yorker reported that some of Bennett's friends refer to his "multimillion-dollar industry" as "Bennett, Inc." and that "his writings, and a heavy schedule of speaking engagements ... have transformed Bennett into a leading voice of the force that is driving American politics right now — the national hunger for a moral society." JX 37. An article headlined "The Chairman of Virtue, Inc." appeared in the September 16, 1996, issue of Time magazine, which reported the "88-week ride on The New York Times best-seller list" of The Book of Virtues, which "spawned two profitable sequels and a cartoon show ... on PBS." JX 103. The television series itself has received substantial media coverage. JX 108.

Plaintiffs' mark "The Book of Virtues" is currently an unregistered trademark, although plaintiffs are attempting to register it with the Patent and Trademark Office ("PTO"). On January 23, 1997, the PTO approved the mark "The Book of Virtues" for publication in the Official Gazette in the categories for print and audio books without requiring proof of secondary meaning, PX 13, Tr. 127-131, a step which indicates that a PTO trademark examiner has deemed the mark to be a suggestive mark. Defendant intends to file an opposition with the PTO's Trademark Trial and Appeal Board. Tr. 131.

Bennett has received numerous requests to license the mark "The Book of Virtues" for such diverse products as porcelain dolls, television series, motion pictures, public radio, animated videos, multimedia children's series, calendars, and apparel. SF 28. Bennett has granted only a few such requests. See, e.g., SF 13; Bennett Aff. ¶ 36; JX 24. He has licensed the mark to Porchlight Entertainment, Inc. ("Porchlight") for "Adventures From The Book of Virtues," an animated public television series for children which began airing in September 1996. SF 62-64. Videos of the first six episodes are available for sale to the public under that name and mark. SF 64. With Bennett's approval, Porchlight has sublicensed the right to sell t-shirts and certain novelty items associated with the television series, and Porchlight is negotiating with an additional 15 to 20 prospective *285 licensees. See Bennett Aff. ¶ 34, SF 68-70.

Bennett retains full control and approval rights for the "Adventures From The Book of Virtues" television series and the products licensed by Porchlight. Bennett Aff. ¶ 35. Bennett has exercised similar control over the content of The Children's Book of Virtues and the related calendar. Rosen Aff. ¶ 5. Plaintiffs have actively policed unauthorized uses of the mark "The Book of Virtues." Rayman Aff. ¶ 22; JX 50-52, 97.

III. Defendant's Publications and Adoption of the Mark

In the late summer of 1994[3], Michael Viner, the president of Dove, announced to his staff and the staff of Dove's distributor his plan to publish, in early 1995, an audiobook entitled The Children's Audiobook of Virtues under the "Dove Kids" imprint. SF 34; Gilbert Aff. ¶ 7. Viner was solely responsible for selecting the title for The Children's Audiobook of Virtues. SF 38. Viner also planned to release a companion print book, The Children's Book of Virtues, at or about the same time. Viner Aff. ¶ 10; Gilbert Aff. ¶ 20. Only the audiobook was released as scheduled; distribution and advertising commenced in February 1995. SF 41, 46.

The audiobook, which is subtitled "A Library of Moral Learning," is a compilation of ten public domain stories, each read by a well-known actor and musically scored. SF 36; JX 63. Each of the stories was previously recorded and included on at least one other audiobook previously published by Dove. SF 37. As of December 31, 1996, Dove had net sales of 7,422 units of The Children's Audiobook of Virtues, for a gross profit of $23,105.80. SF 75.

As to the conceptual origin of the Dove books and their titles, Viner testified that "virtues is a subject with which I had grown up and in which I have always been interested." Viner Aff. ¶ 8. According to Viner, one of his grade school teachers required that he and his classmates keep a notebook called either a "Book of Virtues" or "Virtues" in which they collected stories about good deeds. Id.; Tr. 202-05. On cross-examination, Viner was asked, in essence, to explain why he decided to publish a "Book of Virtues" in the summer of 1994 — some 45 years after grade school and six months after Bennett's success under the title "The Book of Virtues." Viner claimed to have been concerned with the idea of virtues throughout his lifetime, and that 1994 "was the first appropriate time" for publishing a "Book of Virtues" because it coincided with the beginning of Dove's children line. Tr. 204-05. He claimed that he was not "specifically aware" of Bennett's book The Book of Virtues at the time he decided to publish Dove's audiobook, and, although he "may have seen it in passing," he claimed to have "no recollection of it." Tr. 199. Viner acknowledged that he regularly reviewed best-seller lists, at least as to Dove's titles and sometimes more generally. Tr. 200. Plaintiffs' counsel asked Viner if it was his testimony that he "did not know that William Bennett's book, The Book of Virtues, was number one on virtually every best-seller list in the United States from the first week of January of 1994 for 88 consecutive weeks." Viner responded: "I assure you I absolutely did not know that until this moment.... I didn't know the length or the details about being on the best-seller list." Tr. 200-01.

We find it difficult to accept, as a matter of fact, Viner's testimony as to his extremely limited familiarity with plaintiffs' The Book of Virtues in the summer of 1994 and his explanation for his sudden interest in publishing a "Book of Virtues." We believe that Viner significantly understated his awareness of The Book of Virtues and its extraordinary success based on circumstantial evidence which permits us to draw certain adverse inferences and our observations of his demeanor during trial.

Both Sandi Gilbert, Dove's sales and marketing director, and Pat Collins, the sales administration director at Penguin USA, Dove's distributor, acknowledged that they *286 were aware of Bennett's book at the time they first learned that Viner intended to publish The Children's Audiobook of Virtues.[4] Gilbert Aff. ¶ 27; Tr. 140-41. In light of their testimony and the substantial publicity surrounding the publication of The Book of Virtues, including its prominent position on best-seller lists beginning in January 1994, it is hard to believe that a sophisticated publishing executive like Viner would not have been fully aware of the title, general contents, and commercial success of plaintiffs' book when he was deciding to publish books containing a similar mix of stories and fables with a morality message. In short, Viner's claim that he independently recalled the title and concept of his grade school project and that the timing of Dove's publications in relation to plaintiffs' success under a similar title was simply coincidental is not convincing.[5]

Viner testified that he decided to use "Book of Virtues" in Dove's title because "I thought it was a strong title which was likely to sell well, and because `audiobook of virtues' clearly describes the content of the work." Viner Aff. ¶ 12. He also testified that because he has been in the publishing business for more than 25 years, he was "aware of the fact that titles are — in general — not infringable or protectable," but nonetheless discussed his proposed title with legal counsel and relied on counsel's advice in publishing The Children's Audiobook of Virtues and in planning to publish the companion print book. Id. ¶ 13. Viner claimed that Dove "run[s] every title we publish past one of our attorneys." Tr. 209-10. Viner supplied no details as to the advice concerning The Children's Audiobook of Virtues, and no attorney testified concerning any legal advice given to Viner.

Plaintiffs first learned of Dove's audiobook and planned print book through an advertisement which appeared in the February 6, 1995, issue of Publisher's Weekly. Rayman Aff. ¶ 10; JX 114. By letter dated February 27, 1995, plaintiffs demanded that Dove cease and desist, asserting that Dove's titles were confusingly similar to plaintiffs' title and constituted unfair competition. Rayman Aff. ¶ 12; JX 56. The parties' attorneys exchanged letters through the spring and summer of 1995.

Viner spoke with Seth Gershel, the head of audiobooks at S & S, about Gershel's concern that Dove's planned print book would be confusingly similar to S & S's print book. Viner claimed to have reached an agreement with Gershel on steps Dove would take to avoid potential confusion between Dove's book and S & S's book: (1) the cover of Dove's book would have distinctive coloration, and (2) the "Dove Kids" logo would be moved from its usual place on the book's spine to the title, thereby changing the title so that the work would now be called "The Dove Kids Children's Book of Virtues." Viner testified that he sent a revised cover design to Gershel, and that Gershel responded that S & S's legal department "had reviewed [Dove's] proposed book cover and that they had no problem with it." Id. ¶ 16. Gershel disputed this version of events, testifying that he never reached any agreement with Viner on the proposed cover and title of Dove's book. Tr. 283. Although there is no corroborating documentary evidence supporting the existence of the agreement described by Viner, we find it possible, though unlikely, that Viner actually believed, from his conversations with Gershel, that S & S was satisfied with Dove's proposed changes. In any event, we find that there was in fact no such agreement.

IV. The Strength of Plaintiffs' Mark

Plaintiffs argued that Bennett played a role in resurrecting the word "virtue" from obscurity through the success of his book and his ideas about the importance of teaching *287 virtues. Bennett testified that he wanted to "reintroduce to the contemporary public a term ... which had fallen out of vogue." Bennett Aff. ¶¶ 2, 15. He further testified that "the public thinks of me, my ideas, and the values I publicly espouse, when they hear the words `The Book of Virtues.'" Id. ¶ 18. Indeed, Bennett testified that people have referred to him as "`The Book of Virtues guy'; it is almost as if the title of my book has become my own nickname." Id. ¶ 20.

Dove, on the other hand, introduced evidence — primarily the expert testimony of Lynda Boose, professor of English at Dartmouth College — in an effort to demonstrate that the "The Book of Virtues" lacks strength as a source-identifying mark. Boose testified that the word "book" or the phrase "the book of" in book titles is an extremely common literary convention, and that the word "of" in such titles almost always means "about" or "concerning." See Boose Aff. ¶¶ 6-13. According to Boose, hundreds, perhaps thousands, of book titles contain the phrase "the book of" where "of" means "about" — citing examples such as The Book of Mormon, The Book of Dogs, and The Book of Time — and that "very rarely is a book which includes `the book of' in its title the singular book about its subject matter, i.e., the book, so that the word `the' thus constitutes exactly the opposite of its usual meaning," connoting definitiveness. Id. ¶ 12.

Boose also testified that the phrase "book of virtues" or similar phrases have been used repeatedly as the title of works about virtues "since the earliest days of the recorded word." Id. ¶ 14. According to Boose, the earliest classical use of the phrase was Aristotle's book Of Virtue in the Nicomachean Ethics, the precursor of most philosophical works in the field of "virtue-ethics," and that Xenocrates (the third century head of Plato's school) and Theophrastus (successor to Aristotle) both wrote works with "the already standard, generic title Of Virtues." Id. ¶ 15. Cicero's The Book of Virtues ("De Virtutibus Libri") survives from the first century B.C., and the title Of Virtues was used by Epicurus (in Morals), Epictetus (in Discourses), Marcus Aurelis (in Meditations), and St. Thomas Aquinas (in his thirteenth-century Summa Theologica). Montaigne's Essays includes one entitled "Of Virtue," and similarly titled works appear from Spinoza (in the Ethics), John Locke (in Human Understanding) and Immanuel Kant (in The Metaphysics of Morals). Boose placed many of these works, both in title and content, into what she termed the "advice book" genre. She identified the features of this genre as setting forth the virtues defined by Aristotle, Plato, and other philosophers, and either discussing the virtues individually in laymen's terms or illustrating them with parables. Boose claimed that "plaintiffs' book is merely one more example of this type of book; it compiles non-original parables illustrative of the classical virtues and applies a non-original and time-tested `book of virtues' title."[6]Id. ¶ 20.

Notably lacking from Boose's testimony and report, however, was the mention of a single book sold to the American consuming public, other than plaintiffs', that has the title "The Book of Virtues" in the modern English language.[7] Tr. 172-76. At trial, Boose identified *288 one title in the modern English language, not mentioned in her report, which includes both the words "book" and "virtues" — The Book of Bad Virtues. Tr. 172. However, that book is a parody published by S & S of Bennett's book and is thus largely irrelevant to the issues presented by this case.

The Court finds that Boose's expert opinion as to the source-identifying strength of the mark "The Book of Virtues," is entitled to reduced weight for two principal reasons. First, Boose's testimony is premised, at least in part, on her dissection of the mark in order to analyze it. Boose separately analyzed the phrase "The Book of" and the word "Virtues," Tr. 175, 189-90, thus failing to recognize that even if separate components of a mark are generic, the composite may nonetheless be protectable. Also, Boose's conclusion that "the same or closely similar marks have been used ... over and over again since the earliest days of recorded work" appears to rely on an overly broad definition of "the same or closely similar marks" that includes any title which uses the word "virtue" or any title that uses the word "book" and the word "virtue." Tr. 174-75. As previously noted, however, Boose failed to note one book in the modern English language with the exact title of plaintiffs' book, "The Book of Virtues." Second, Boose's report opines as to how the mark would be viewed by "scholars" and "literary scholars," Boose Aff. ¶¶ 4, 22, ignoring the fact that the academic book market is not the intended market for plaintiffs' or defendant's products. SF 44, 57. Boose admitted that she has no knowledge of the book-purchasing habits of parents with small children. Tr. 161-62.

Substantial evidence was presented at trial of news articles and other media references to plaintiffs' The Book of Virtues. E.g., JX 18, 32-41, 46, 103, 105. However, not a single such article or reference refers to a class of books as "books of virtue"; all use the phrase "The Book of Virtues" to refer specifically to Bennett's book and its spin-off products.

As noted above, Bennett has received numerous requests to license his mark on diverse products, offering advances, fees, and royalties. Although a few small publishers planned to publish their own collections of fables and poems with a morality message under the mark "The Book of Virtues" or a closely similar title, S & S and Bennett's attorney have objected to all such uses of the mark, and no publisher — except for Dove — has published under the challenged title. SF 58-60. Accordingly, the Court finds that there are no unchallenged uses of plaintiffs' mark "The Book of Virtues."

Plaintiffs introduced evidence, largely uncontested by Dove, demonstrating that plaintiffs' various products under the mark "The Book of Virtues" achieved tremendous, rapid success in the marketplace, and that S & S and Bennett have engaged in extensive promotional efforts to promote the products sold under this mark. Bennett also testified that he has received hundreds of letters recounting what his book means to people and their families, how they have used it, and what other kinds of materials they would like to see produced by plaintiffs.[8] Bennett Aff. ¶ 23; JX 104.

Plaintiffs have engaged in substantial advertising and promotional efforts for these products, including advertising in national media, Bennett's appearances on national television and radio programs and at paid lectures throughout the United States, elaborate point-of-purchase displays for retail outlets, and specialized promotional efforts for "niche" markets such as schools and libraries. E.g., Martin Aff. ¶ 25; Dooling Aff. ¶ 4; Dana Aff. ¶ 10; SF 26; JX 10. Plaintiffs have expended over $1.1 million promoting and advertising various The Book of Virtues products. SF 29. S & S also promoted the books in its various catalogues. JX 7. Retailers of these products have engaged in their *289 own promotional efforts, as has Porchlight, the producer of the PBS television series.

V. The Relevant Market and Consumers

Both parties' print books and audiobooks relevant to this litigation were intended to be marketed in retail outlets throughout the country, and the books are sold through the same marketing channels, to the same wholesale and retail customers. SF 57. Indeed, "it is almost inevitable that these titles will appear on the same shelf or in the same location" in certain stories. Martin Aff. ¶ 29. S & S's books and Dove's books also have comparable prices. Id.

There was conflicting evidence on whether purchases of the books at issue are "impulse-driven" purchases or whether they are the result of more extensive consideration by consumers. Viner testified that parents purchasing books for their children "are apt to exercise an extremely high degree of care in making their decisions." Viner Aff. ¶ 20. However, there was also evidence that while a consumer may initially go to a store intending to purchase a specific book, once in the store, the consumer's purchases are often impulse-driven, resulting in the purchase of another book or additional books because the title or cover captures the eye. See Jacoby Aff. ¶ 52. This more complex explanation of purchasing habits seems credible.

VI. Consumer Survey Evidence

The parties presented the results of two consumer surveys designed to assess the likelihood of actual confusion between S & S's books and Dove's books. Jacob Jacoby, a professor of consumer behavior and the president of a consumer research firm, designed plaintiffs' survey (the "Jacoby Survey"). Norman Passman, the chairman of a market research firm and an instructor on marketing, prepared Dove's survey (the "Passman Survey"). Dove also presented the expert testimony of Michael Rappeport, a partner at a consumer marketing company, who was retained to critique the Jacoby Survey.

Not surprisingly, the results of the two surveys diverged markedly. The Jacoby Survey concluded that approximately 36 percent of potential consumers are likely to be confused into believing that plaintiffs' and defendant's products are the same or emanate from the same source for reasons with trademark significance. Jacoby Aff. ¶ 51. The Passman Survey concluded that a maximum of 6.6 percent of survey respondents were confused for trademark-related reasons about the source or sponsor of Dove's books. Passman Aff. ¶ 18.

Jacoby and Passman used quite different approaches in designing their respective surveys. The Jacoby Survey used a product line-up or "Squirt" format.[9] Respondents are generally shown a line-up of products, including both the junior and senior marks, and then asked whether the junior and senior marks are put out by the same or different companies. See Itamar Simonson, "The Effect of Survey Method on Likelihood of Confusion Estimates: Conceptual Analysis and Empirical Test," 83 Trademark Rptr. 364, 370-71 (1993) (hereinafter, "Simonson"). The Passman Survey employed a variation of the "Eveready" format.[10] In such a survey, respondents are generally shown only the junior product (or brand name) and asked questions such as: (1) Who do you think puts out this product? (2) What makes you think so? and (3) Name any other products put out by the same concern which puts out this product. Respondents naming the senior manufacturer or its products provide evidence of likelihood of confusion. Simonson, supra, at 368-69.

In the Jacoby Survey, respondents were first shown a line-up of three books, one of which was always plaintiffs' audiobook The Children Book of Virtues, plaintiffs' print book The Children Book of Virtues, or a reproduction of the print book's cover; the other two items were controls unrelated to this litigation. These items were then replaced by three other items, one of which *290 was always Dove's audiobook, The Children's Audiobook of Virtues, or a reproduction of the cover art for Dove's unpublished print book, The Children's Book of Virtues; the other two items were controls. Respondents were then asked with regard to each of the items in the second array of products: "Is it or isn't this another version of any of the books[11] that I showed you earlier?" If the answer was in the affirmative, the respondent was asked to explain his or her reasons for saying that it was such a version, and the item was put aside. Only if the answer was in the negative, i.e., that the item in the second array was not a version of any item previously shown, were additional questions asked, such as whether the product was "another in the same series as any of the books[12] that I showed you earlier?" or whether the item was prepared by the same author or editor as an item previously displayed, whether the item came from the same publisher that put out one of the previous items, and whether the products are connected in any way. if the respondent answered any one of these questions affirmatively, the respondent was asked to explain his or her answer, the item was put aside, and the remaining questions were not asked.

Dove attacked the Jacoby Survey on several fronts, but we find that the survey suffered from only one significant flaw: the ambiguous wording of the main question designed to test for likelihood of confusion.[13] The question asked was: "Is it or isn't this another version of any of the books that I showed you earlier?" The problem with this question is the ambiguity of the word "version" in the context of a likelihood of confusion survey. A respondent's identification of a Dove product as another "version" of plaintiffs' product does not necessarily mean that the respondent cannot distinguish the products or believes plaintiffs are the source of Dove's product. The statement that one publication is another "version" of another publication may carry any number of possible meanings: that it is an earlier (but identical) edition; that it is the same work in a different medium; that it is part of a series of two or more similar and related works; that it is a translation of a work into another language; or that it is a variant form of works in the same genre. Rappeport provided a useful example of this ambiguity in his testimony. A respondent might agree that Coke and Pepsi are different "versions" of soft drinks, meaning that they are both different examples of the same class of beverage. The same respondent might also agree that Coke and Diet Coke are different "versions" of soft drinks, meaning that they are different examples of the same beverage. Rappeport Aff. ¶ 14. In these two cases, the word "version" obviously means different things. Neither statement necessarily has trademark significance because in both cases it is impossible to say whether the respondent was confused into believing anything about the source or origin of Coke, Pepsi, or Diet Coke. Another example of the ambiguity in the word "version" was provided at trial, when Jacoby was asked if "Batman" was another version of "Superman." Jacoby agreed that "it could conceivably be, yes.... It depends on how you define that version." Tr. 93.

But the Jacoby Survey never defined the word "version" or elicited a definition of "version" from the survey respondents. Thus, there is no way of knowing which meaning of the word "version" the respondents intended. Although the Jacoby Survey asked other questions which could be considered indicators of trademark confusion — such as whether the books were in the same series, or whether the books were put out by the same publisher and editor — these questions were only asked if the respondent answered the ambiguous "version" question negatively. In other words, survey respondents who stated *291 that they believed the Dove publication was another "version" of plaintiffs' publication were not asked the remaining questions. Thus, the Court cannot simply exclude from the survey the respondents who answered the ambiguous "version" question affirmatively. Because of the ambiguity of the "version" question, the Court assigns significantly reduced weight to the Jacoby Survey's results.

In the Passman Survey, respondents were shown either defendant's audiobook, The Children's Audiobook of Virtues, or the cover art for defendant's proposed print book, The Children's Book of Virtues. Respondents were not shown any of plaintiffs' products. The Dove item was removed from sight, and respondents were questioned regarding possible confusion with respect to the origin or sponsorship of the product which had been shown to them. Specifically, all respondents were asked "Have you seen or heard of this book before?" If the respondent answered affirmatively, certain follow-up questions were asked, including (1) "What made you say so?" (2) "What else, if anything, do you know about the book?" (3) "Can you tell me the name of the author or editor of the book you have seen or heard of?" and (4) "Can you tell me the name of the publisher of the book you have seen or heard of?" Regardless of their answer to the question "Have you seen or heard of this book before?" All respondents were asked whether they connected defendant's books with any other book, and, if so, they were asked the name, author, editor, and publisher of the book they associated with defendant's book.

Because respondents were not shown any of plaintiffs' products, Passman conceded that his survey confronted only conscious confusion, or "top-of-mind" awareness. Tr. 247-48. The Passman Survey did not attempt to determine the level of confusion that would might result on an "aided" awareness basis. Id. Nor did it attempt to ascertain whether respondents believed that defendant's and plaintiffs' products emanated from a common, albeit anonymous, source. Tr. 249. These factors may have tended to underestimate significantly the likelihood of confusion found. Perhaps most importantly, the "Eveready" survey design selected by Passman may not have been the most appropriate format given the type of product: at issue in this litigation. As one author in this area has recognized, surveys of this type are "less likely to reveal confusion if respondents think that the junior mark is the senior mark ... and do not know the name (or products) of the company that puts out the senior mark."[14] Simonson, supra, at 369. Because the marks at issue in this litigation are the titles of books rather than the names of common household products, the Passman Survey's underestimation of the likelihood of confusion may be significant. Only those persons who have heard of plaintiffs' books could be expected to show any confusion with defendant's books.

DISCUSSION

I. The Lanham Act Claim

Section 43(a) of the Lanham Act prohibits any person from using

in connection with any goods ... any word, term, name, symbol, or device, or any combination thereof ... which ... is likely to cause confusion, or to cause mistake, or to deceive ... as to the origin, sponsorship, or approval of his or her goods ... by another person.

15 U.S.C. § 1125(a). Its purpose is "to prevent consumer confusion regarding a product's source and to enable those that fashion a product to differentiate it from others on the market." Centaur Communications, Ltd. v. A/S/M Communications, Inc., 830 F.2d 1217, 1220 (2d Cir.1987). This section is the only provision in the Lanham Act that protects an unregistered mark like plaintiffs'. Id. "[T]he general principles qualifying a mark for registration under § 2 of the Lanham Act are for the most part applicable in determining whether an unregistered mark *292 is entitled to protection under § 43(a)." Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763, 768, 112 S.Ct. 2753, 2757, 120 L.Ed.2d 615 (1992). In order to prevail on a claim of trademark infringement in violation of the Lanham Act, a plaintiff must show (1) that it has a valid mark that is entitled to protection, and (2) that use of the defendant's mark infringes, or is likely to infringe, the plaintiff's mark. See, e.g., Estee Lauder Inc. v. The Gap, Inc., 108 F.3d 1503, 1508 (2d Cir. 1997).

A. The Protectability of Plaintiffs' Mark

The first issue confronting us is whether plaintiffs' mark "The Book of Virtues" is eligible for protection under the Lanham Act. Since plaintiffs' mark is unregistered, they bear the burden of proving that "The Book of Virtues" is a valid trademark.[15]See Reese Publishing Co. v. Hampton Int'l Communications, Inc., 620 F.2d 7, 11 (2d Cir.1980). In Abercrombie & Fitch Co. v. Hunting World, Inc., 537 F.2d 4 (2d Cir. 1976), Judge Friendly set forth four categories of terms, each one conferring a differing degree of eligibility for trademark protection. "Arrayed in an ascending order which roughly reflects their eligibility to trademark status and the degree of protection accorded, these classes are (1) generic, (2) descriptive, (3) suggestive, and (4) arbitrary or fanciful." Abercrombie & Fitch, 537 F.2d at 9. Suggestive, arbitrary and fanciful marks, because their intrinsic nature serves to identify a particular source of a product, are deemed inherently distinctive and are entitled to protection. Bristol-Myers Squibb Co. v. McNeil-P.P.C., Inc., 973 F.2d 1033, 1040 (2d Cir.1992). Generic marks are never entitled to protection, while a descriptive mark is eligible for protection if it has become distinctive of the producer's goods in commerce. This distinctiveness is generally called secondary meaning.[16]Id.

Judge Schwartz found in his prior Opinion that plaintiffs' mark was not generic. See 936 F.Supp. at 161. After analyzing the case law on the categorization of marks — and taking into consideration factual material submitted in connection with Dove's Rule 12(c) motion — Judge Schwartz concluded, in effect, that there were no material issues of fact on this issue that precluded him from finding that the mark "The Book of Virtues" was not generic.

At trial, plaintiffs urged the Court to categorize their mark as suggestive, as a PTO trademark examiner has done, while defendant continued to press its disagreement with Judge Schwartz's ruling that the mark was not generic. At the conclusion of trial, the Court directed the parties to submit proffers of additional evidence they would introduce if the Court were to reopen the inquiry into whether the mark is generic. As noted above, see supra n. 6, Dove's proffer consisted only of the unredacted, original report of its expert, Lynda Boose, and testimony consistent therewith.

Consideration of Dove's proffer satisfies the Court that no further proceedings are necessary with reference to the issue of whether the mark "The Book of Virtues" is generic. Dove's proffer did not consist of any additional factual material but rather Boose's opinion, based on facts already in evidence, that the mark "The Book of Virtues" is generic. The Court has reviewed Boose's unredacted expert report (Court Exhibit *293 I). Independently of Judge Schwartz's prior determination, we conclude that plaintiffs have adequately demonstrated that their mark is not generic. The evidence before the Court does not establish that the mark "The Book of Virtues" "refers, or has come to be understood as referring, to the genus of which the particular product is a species." Abercrombie & Fitch, 537 F.2d at 9. Plaintiffs' The Book of Virtues is not literally a "book of virtues," nor is it referred to as such in common speech or in news articles and book reviews. Compare Reese Publishing, 620 F.2d at 11 n. 1 ("Video Buyer's Guide" was generic because "it names a whole class of magazines, i.e. video buyer's guides"); CES Publishing Corp. v. St. Regis Publications, Inc., 531 F.2d 11 (2d Cir.1975) ("Consumer Electronics Monthly" was generic as applied to trade publication; "one might intelligibly speak of both plaintiff's and defendant's magazines as `consumer electronics monthlies'"). Rather, the evidence shows that The Book of Virtues and Dove's publications are referred to by consumers and journalists as books containing short stories, fables, and poems with a morality message. Unlike the situation in CES Publishing, it would not be difficult for Dove and other publishers to identify their collections of short stories and fables without using the phrase "book of virtues" in their titles. Compare CES Publishing, 531 F.2d at 15. Furthermore, while there have been numerous treatises on the subject of virtue with the word "virtue" featured in the title, there is no evidence of another book in the modern English language with the title "The Book of Virtues."

Plaintiffs' contention that the mark is suggestive is also not persuasive. A suggestive mark suggests or conveys an impression of the product, though it, may take imagination to grasp the nature of the product. E.g., Gruner + Jahr USA Printing & Publishing v. Meredith Corp., 991 F.2d 1072, 1076 (2d Cir.1993); Estee Lauder, 108 F.3d at 1509. In contrast, a mark is classified as descriptive if it "tells something about a product, its qualities, ingredients, or characteristics." Id. Plaintiffs' mark cannot be categorized as suggestive because it does not require thought and imagination to recognize that a product sold under the mark "The Book of Virtues" is a book touching upon the subject of virtues. In sum, we conclude that "The Book of Virtues" is a descriptive mark because it conveys an immediate idea of the ingredients, qualities, or characteristics of the plaintiff's product — a book dealing in some fashion with the subject of virtues.

B. Secondary Meaning

A descriptive mark is entitled to protection under the Lanham Act only if it has acquired secondary meaning, that is to say, "an identity that consumers associate with a single source, even though the source itself may be unknown." Gruner + Jahr, 991 F.2d at 1076. A mark acquires secondary meaning when "it [is] shown that the primary significance of the term in the minds of the consuming public is not the product but the producer." Centaur Communications, 830 F.2d at 1221. Secondary meaning in a literary title occurs "where the title is sufficiently well known that consumers associate it with a particular author's work." Rogers v. Grimaldi, 875 F.2d 994, 998 (2d Cir.1989); Tri-Star Pictures, Inc. v. Leisure Time Productions, B.V., 749 F.Supp. 1243, 1252 (S.D.N.Y.), aff'd, 17 F.3d 38 (2d Cir.1994).

Plaintiffs must establish that their mark acquired secondary meaning "before its competitor commenced use of the mark." Paper-Cutter, Inc. v. Fay's Drug Co., 900 F.2d 558, 564 (2d Cir.1990); Black & Decker Corp. v. Dunsford, 944 F.Supp. 220, 227 (S.D.N.Y. 1996). Here, defendant began to advertise and sell its allegedly infringing audiobook in February 1995, so the question before the Court is whether plaintiffs have demonstrated that the mark "The Book of Virtues" acquired secondary meaning prior to February 1995.[17]

*294 The Second Circuit has identified six elements that a court should evaluate in determining the existence of secondary meaning: (1) advertising expenditures, (2) consumer studies linking the mark to a source, (3) unsolicited media coverage of the product, (4) sales success, (5) attempts to plagiarize the mark, and (6) length and exclusivity of the mark's use. Thompson Medical Co. v. Pfizer, Inc., 753 F.2d 208, 217 (2d Cir.1985); Centaur Communications, 830 F.2d at 1222. We consider each below.

1. Advertising Expenditures

As noted above, The Book of Virtues was advertised extensively in major publications. By the end of 1996, plaintiffs had spent over $1.1 million promoting and advertising their various products sold under the mark "The Book of Virtues." SF 29. Although plaintiffs have not broken down the $1.1 million figure to show how much was spent before February 1995 and how much was spent after that date, the record nonetheless reflects that plaintiffs promoted The Book of Virtues extensively in 1993 and 1994 through various means. Upon publication of The Book of Virtues, Bennett embarked on a national tour promoting the book on television and radio programs, and he publicized The Book of Virtues during dozens of paid lectures throughout the United States in 1993 and 1994. S & S also included The Book of Virtues in various of its catalogues, including its Fall 1993 catalogue. This factor favors plaintiffs.

2. Consumer Studies

Plaintiffs did not introduce any study attempting to demonstrate consumers' association of the mark "The Book of Virtues" with plaintiffs. Thus, this element is inconclusive.

3. Unsolicited Media Coverage

There was significant media coverage of The Book of Virtues from its publication in 1993 and continuing through early 1995 that appears to have been the result of independent editorial judgments that The Book of Virtues was newsworthy. Reviews of the book were printed in numerous newspapers and magazines, and New York magazine cited the book as one of 1993's best Christmas gifts. Time devoted a full-page essay to The Book of Virtues and the popular appeal of Bennett's message in March 1994. Newsweek featured Bennett on its cover in June 1994, stated that The Book of Virtues "remains hot beyond expectations," and referred to Bennett as "a cottage industry of character education" when discussing his plans for releasing future products in "The Book of Virtues" series. The Book of Virtues was also featured in a campaign commercial aired in California during the 1994 Senate race. The evidence establishes a significant amount of national and local media coverage of The Book of Virtues prior to defendant's entrance into the market.

4. Sales Success

Defendant concedes that this element favors plaintiffs. See Def. Mem. at 14. The Book of Virtues was a hit book upon its release in November 1993, reaching the number one position on The New York Times best-seller list in January 1994. By February 1995, The Book of Virtues had been on The New York Times best-seller list for 60 consecutive weeks S & S's two audiobooks based on The Book of Virtues published in May and November 1994 also sold well.

5. Attempts to Plagiarize

This element also weighs in plaintiffs' favor. There is compelling circumstantial evidence that Dove intentionally copied plaintiffs' mark "The Book of Virtues" in order to capitalize on plaintiffs' success. First, the titles and subtitles of the books at issue evidence deliberate copying. Plaintiffs' book, The Book of Virtues, is subtitled "A Treasury of Great Moral Stories." Dove's audiobook, The Children's Audiobook of Virtues, is subtitled *295 "A Library of Moral Learning." Second, the timing of Dove's publications in relation to plaintiffs' release of The Book of Virtues in hardcover and in audiobook format strongly supports an inference of deliberate copying. The Book of Virtues was published in hardcover in November 1993, and the first audiobook followed in May 1994. During the spring and early summer of 1994, the news media devoted significant attention to The Book of Virtues and the popular chord Bennett had apparently struck with the book. Sometime later that summer, Viner announced to Dove's staff his plan to publish The Children's Audiobook of Virtues and The Children's Book of Virtues. Viner testified that it was essentially a coincidence that Dove's publications were planned when they were, but this testimony is simply not credible. Viner equivocated when asked whether he was aware of plaintiffs' The Book of Virtues prior to deciding to publish Dove's books. As we have stated, supra at 285-286, the Court does not believe that a publishing executive as savvy as Viner, planning to release an audiobook containing children's stories with a morality message, would not have been familiar with plaintiffs' The Book of Virtues, which contained similar content and which had met with extraordinary commercial success. Viner's claim that he drew inspiration for the concept and title of Dove's books from a grade-school project 45 years before is also not persuasive. The nature of Dove's audiobook — which was merely a repackaging, under a new title and cover, of ten stories that had been previously released on earlier Dove audiobooks — suggests that Dove's audiobook was produced quickly from stock materials in order to capitalize on plaintiffs' success under The Book of Virtues mark, rather than being the product of Viner's independent creativity, inspired by grade-school memories.

While coincidences may sometimes occur, there is simply too much circumstantial evidence of deliberate copying to credit that explanation here.

6. Length and Exclusivity

"[N]o absolute time span can be posited as a yardstick in cases involving secondary meaning." Centaur Communications, 830 F.2d at 1225. Rather, the length and exclusivity of a mark's use must be evaluated "in light of the product and its consumers." Id. Although Dove's expert, Professor Boose, testified to uses of the word "virtue" and the word "book" in the titles of various classical and modern texts, she could point to no other book in the modern English language with plaintiffs' title, "The Book of Virtues." Thus, Boose's testimony does not significantly detract from plaintiffs' contention that they are the exclusive modern day user of the mark "The Book of Virtues." Plaintiffs used this mark on three separate products (the original hardcover book and two audiobooks) for 15 months before Dove published The Children's Audiobook of Virtues in February 1995. In the context of book sales, 15 months of exclusive use is a sufficiently lengthy period to conclude that this element weighs in favor of a finding of secondary meaning.

* * *

In sum, five of the six elements identified in Thompson Medical, supra, weigh in favor of a finding of secondary meaning in the title "The Book of Virtues." The absence of any consumer study bears only peripherally on the analysis of plaintiffs' mark; no single factor among the six is determinative, and "every element need not be proved." Thompson Medical, 753 F.2d at 217; Centaur Communications, 830 F.2d at 1222. Plaintiffs have presented sufficient evidence demonstrating that plaintiffs' print book and audiobooks sold under the title "The Book of Virtues" were well known among book consumers and that the title was closely associated in the public's mind with Bennett well before Dove released its audiobook in February 1995. Plaintiffs engaged in substantial promotional efforts, the press devoted significant unsolicited coverage to The Book of Virtues and its author, and The Book of Virtues was firmly entrenched on best-seller lists for much of 1994. Based on this evidence, as well as the circumstantial evidence of Dove's intentional copying of plaintiffs' title, the Court concludes that plaintiffs have demonstrated secondary meaning in their title "The Book of Virtues."

*296 C. Likelihood of Confusion

The test for trademark infringement is whether the defendant's use of a designation as a trademark creates a likelihood of confusion. Estee Lauder, 108 F.3d at 1508-09. "The issue of likelihood of confusion turns on whether `numerous ordinar[il]y prudent purchasers are likely to be misled or confused as to the source of the product in question because of the entrance in the marketplace of defendant's mark.'" Id. at 1510 (quoting Gruner + Jahr, 991 F.2d at 1077). Likelihood of confusion means a probability of confusion; "it is not sufficient if confusion is merely `possible.'" Id. (citation omitted).

1. First Amendment Concerns

Relying on Rogers v. Grimaldi, 875 F.2d 994 (2d Cir.1989), Dove contends that its titles cannot be found to violate the Lanham Act because they have artistic relevance to the underlying works and its titles do not "explicitly mislead" as to the source or content of Dove's works. See Def. Mem. at 2-4. While the Court agrees that concerns of free expression must inform our analysis, we disagree with Dove's contention that its titles must be found to be "explicitly" misleading in order to violate the Lanham Act.

In Rogers v. Grimaldi, the Second Circuit stated that "[t]he purchaser of a book, like the purchaser of a can of peas, has a right not to be misled as to the source of the product." Rogers, 875 F.2d at 997-98. Nonetheless, Rogers recognized that First Amendment concerns must "inform our consideration of the scope of the [Lanham] Act as applied to claims" involving the titles of artistic works. Id. at 998. Rogers involved a claim by Ginger Rogers that a movie entitled "Ginger and Fred" created the false impression that the film was about her and Fred Astaire or had her endorsement, when in fact it was an account of two fictional Italian cabaret performers who imitated Rogers and Astaire in their performances. The Second Circuit held that

in general the [Lanham] Act should be construed to apply to artistic works only where the public interest in avoiding consumer confusion outweighs the public interest in free expression. In the context of allegedly misleading titles using a celebrity's name, that balance will normally not support application of the Act unless the title has no artistic relevance to the underlying work whatsoever, or, if it has some artistic relevance, unless the title explicitly misleads as to the source or the content of the work.

Id. at 999. In a footnote following this text, the Circuit stated that "[t]his limiting construction would not apply to misleading titles that are confusingly similar to other titles. The public interest in sparing consumers this type of confusion outweighs the slight public interest in permitting authors to use such titles." Id. at 999 n. 5.

In light of the Circuit's limitation on the reach of the Rogers standard, Dove's argument that its titles do not "explicitly mislead" (i.e. Dove's books are not titled "William Bennett's Book of Virtues") misses the mark. Rogers did not require that such a test be applied when a title is misleadingly similar to another title. Rather, Rogers set forth a balancing approach "generally applicable to Lanham Act claims against works of literary expression." Cliffs Notes, Inc. v. Bantam Doubleday Dell Publishing Group, Inc., 886 F.2d 490, 494-95 (2d Cir.1989) (applying Rogers to allegedly infringing parody of study guide); see also Twin Peaks Prods., Inc. v. Publications Int'l, Ltd., 996 F.2d 1366, 1379 (2d Cir.1993) (applying Rogers to allegedly infringing guide to television program). A literary title can be misleading "in the sense that it induces members of the public to believe the [b]ook was prepared or otherwise authorized by [plaintiffs]. This determination must be made, in the first instance, by application of the venerable Polaroid factors." Twin Peaks, 996 F.2d at 1379 (referring to factors for likelihood of confusion set forth in Polaroid Corp. v. Polarad Elecs. Corp., 287 F.2d 492, 495 (2d Cir. 1961)). However, the finding of likelihood of confusion must be "particularly compelling to outweigh the First Amendment interest recognized in Rogers." Id. Accordingly, we consider the relevant Polaroid factors and whether plaintiffs have made a "particularly compelling" showing of likelihood of confusion that outweighs the public interest in free expression.

*297 2. The Polaroid Factors

a. Strength of the Mark

A mark's strength "ultimately depends on the degree to which the designation is associated by prospective purchasers with a particular source." Estee Lauder, 108 F.3d at 1510 (quoting Restatement (Third) of Unfair Competition § 21 comment I (1995)). The strength of a mark depends upon both its "conceptual" strength (i.e., whether the mark is categorized as generic, descriptive, suggestive, or arbitrary) and strength "in its commercial context." Centaur Communications, 830 F.2d at 1225-26; Hasbro, Inc. v. Lanard Toys, Ltd., 858 F.2d 70, 76 (2d Cir. 1988). In considering the strength of a trademark in a publication title, "the time and size of circulation and consumer identification are examined." Centaur Communications, 830 F.2d at 1226.

As a descriptive mark, the mark "The Book of Virtues" is, "by definition, somewhat weak. Yet, a mark's category is not alone controlling." Id. Here, as in Centaur Communications, there is persuasive evidence of the mark's strength in its commercial context: the tremendous and rapid sales success of The Book of Virtues, plaintiffs' extensive promotional efforts during 1993 and 1994, the substantial quantity of unsolicited media coverage during that period, and the number of third-party requests to license the mark. Considered together, we think this evidence demonstrates that an appreciable number of consumers associate the book title "The Book of Virtues" with Bennett, or, at least, a single anonymous source. The evidence of sales success and promotional activities is far more compelling here than it was in Centaur Communications, where the court nonetheless concluded that the descriptive magazine title "Marketing Week" had "achieved relative — if not great — strength in its market context." Id.

Despite the fact that "The Book of Virtues" is only a weak descriptive mark, plaintiffs' title had substantial source-identifying strength by the time Dove entered the market in February 1995.[18] Thus, this factor favors plaintiff.

b. Degree of Similarity

The second Polaroid factor is the degree of similarity between the senior and junior users' marks. "The comparison of marks is an inquiry designed to determine the `general impression conveyed to the purchasing public by the respective marks.'" Hasbro, 858 F.2d at 77 (quoting C.L.A.S.S. Promotions, Inc. v. D.S. Magazines, Inc., 753 F.2d 14, 17 (2d Cir.1985)).

Dove's titles "The Children's Audiobook of Virtues" and "The Children's Book of Virtues" — although not identical to plaintiffs' title "The Book of Virtues" — are nonetheless quite similar. The addition of the word "Children's" and the word "Audio" on Dove's audiobook do not fundamentally change the character of plaintiffs' mark. The two added words could foster a belief among consumers that the Dove books are special versions of The Book of Virtues emanating from the same source as plaintiffs' best-selling book. Moreover, the subtitles of the parties' respective books also share some similarities.

Defendant contends that the "Dove Kids" logo on the cover of its books gives its titles a degree of phonetic and visual distinction. However, the logo is relatively insignificant *298 in the overall scheme of the products' titles and covers. An employee of Dove's distributor testified, and the Court agrees, that the inclusion of the logo on the cover of Dove's proposed print book between the words "The" and "Children's Book of Virtues" did not make it part of the title. Tr. 57. Nevertheless, the presence of the Dove logo, and the absence of any reference to Bennett, are still relevant factors in assessing the likelihood of confusion.

On balance, this factor tips in plaintiffs' favor, especially when considered in conjunction with the products' close competitive proximity, discussed below.

c. Competitive Proximity

The third Polaroid factor is the "competitive proximity" of the products, which is measured, in part, with reference to the first two Polaroid factors. See, e.g., Centaur Communications, 830 F.2d at 1226. As Dove concedes, this factor favors plaintiffs. See Def. Mem. at 19. The parties stipulated that their respective publications were intended to be marketed in retail outlets throughout the country and that the books are sold through the same marketing channels, to the same wholesale and retail customers. Indeed, there was testimony that the parties' books will often appear on the same shelf or in the same location in certain stories.

d. Bridging the Gap

This factor looks to whether the senior user of the mark is likely to enter the market in which the junior user is operating, that is, "bridge the gap." Centaur Communications, 830 F.2d at 1227. "If the senior user can show such an intention, it `helps to establish a future likelihood of confusion as to source.'" Id. (quoting Lois Sportswear, U.S.A., Inc. v. Levi Strauss & Co., 799 F.2d 867, 874 (2d Cir.1986)).

Dove concedes the likelihood that plaintiffs would bridge the gap since the parties are direct competitors, but Dove argues that this factor is entitled to reduced weight because the parties are in the same business, book publishing. See Def. Mem. at 20. There is support for this view. See, e.g., Bristol-Myers Squibb, 973 F.2d at 1044. However, other cases indicate that although the parties operate in the same market, "this very fact indicates that a greater likelihood of confusion exists than if they operated in different markets." Hasbro, 858 F.2d at 78. Thus, we consider this factor to slightly favor plaintiffs.

e. Actual Confusion

The fifth Polaroid factor examines whether any consumers had actually been confused by the products bearing the allegedly confusing marks. Evidence is generally anecdotal in nature or takes the form of a market research survey. Centaur Communications, 830 F.2d at 1227. "Evidence of actual confusion is not required to prove the likelihood of confusion." Id.; see also Lois Sportswear, 799 F.2d at 875 ("actual confusion need not be shown to prevail under the Lanham Act, since actual confusion is very difficult to prove and the Act requires only a likelihood of confusion as to source").

Plaintiffs did not introduce any significant anecdotal evidence of actual consumer confusion. We decline to draw any negative inference from plaintiffs' lack of such evidence because of the relatively short time — approximately six months — that Dove's audiobook was on the market before sales were enjoined. See Hasbro, 858 F.2d at 78 (finding lack of actual confusion inconclusive where defendant's product was marketed for eight months); Lois Sportswear, 799 F.2d at 875 (where defendant's sales had been "minimal" thus far, "[i]t would be unfair to penalize [plaintiff] for acting to protect its trademark rights before serious damage has occurred").

The parties did, however, conduct consumer surveys designed to test the likelihood of confusion. The Jacoby Survey, commissioned by plaintiffs, demonstrated a fairly high 36 percent level of consumer confusion, but, as described above, supra at 290-291, there was an ambiguity in the wording of the Jacoby Survey's key question and we assign it significantly reduced weight. We cannot be assured that the 36 percent figure is a valid indicator of the percentage of survey respondents who were confused for trademark-related reasons. The Passman Survey, commissioned by Dove, found a relatively low 6.6 percent level of consumer confusion. However, as noted above, supra at 291, the *299 Passman Survey's methodology — a variant of the Eveready survey format, which measures "top-of-mind" awareness — may significantly underestimate the likelihood of confusion in cases where the survey respondents are not familiar with the senior users' product. Accordingly, we assigned reduced evidentiary significance to the results of the Passman Survey.

Evaluating the results of the two consumer surveys together, and giving due weight to their respective limitations, we conclude that the surveys are probative of a likelihood of confusion as to source, but only to a very limited degree. On balance, we consider this Polaroid factor to be neutral or, at most, tip only slightly in plaintiffs' favor.

f. Junior User's Good Faith

This factor examines the good faith of the junior user in selecting the mark. Evidence of intentional copying raises a presumption that a second comer intended to create a confusing similarity of appearance and succeeded. See, e.g., Warner Bros., Inc. v. American Broadcasting Cos., 720 F.2d 231, 246-47 (2d Cir.1983); Centaur Communications, 830 F.2d at 1228 ("awareness [of defendant's mark] can give rise to an inference of bad faith").

As discussed above in connection with secondary meaning, there is compelling circumstantial evidence that Dove copied plaintiffs' mark "The Book of Virtues" in order to capitalize on plaintiffs' success and the goodwill associated with plaintiffs' mark. See, e.g., Spring Mills, Inc. v. Ultracashmere House, Ltd., 689 F.2d 1127, 1134-35 (2d Cir. 1982) (discussing evidence that defendants adopted their mark "for no other purpose than to obtain a free ride on the good reputation of their successful competitor"). Viner's purported innocent explanation for the similarity of the marks and the timing of Dove's release — essentially, that it was a coincidence — is not persuasive. Viner also testified that he believed that book titles are, in general, not infringable or protectable and that he consulted with an attorney before proceeding with Dove's titles. It is hard to believe that an experienced publisher could seriously believe that book titles are "in general" not infringable or protectable. Had Viner received any legal advice concerning titles during his publishing career, he would undoubtedly have been informed, as the leading trademark treatise's section on literary and artistic rights begins, that "[i]n general, [literary] titles are protected according to the fundamental tenets of trademark and unfair competition law." 1 J. Thomas McCarthy, McCarthy on Trademarks and Unfair Competition § 10:1, at 10-4 (4th ed.1996). His claim that he consulted with an attorney is not entitled to any weight because of the paucity of details concerning the substance of this alleged legal advice.[19]

Accordingly, this factor weighs in plaintiffs' favor. There is substantial circumstantial evidence of Dove's bad faith in intentionally copying plaintiffs' mark, and Dove has not offered a "credible innocent explanation." Centaur Communications, 830 F.2d at 1228; International Star Class Yacht Racing Ass'n v. Tommy Hilfiger, U.S.A., Inc., 80 F.3d 749, 753-54 (2d Cir.1996).

g. Quality of Junior User's Product

Neither party presented evidence on the quality of Dove's goods. Accordingly, this Polaroid factor favors neither party.

h. Sophistication of Consumers

The final Polaroid factor, the sophistication of the relevant consumer group, *300 is grounded on the belief that unsophisticated consumers aggravate the likelihood of confusion. Hasbro, 858 F.2d at 79 (citing Centaur Communications, 830 F.2d at 1228). This is especially true when the competing products' marks are similar and the inexpensive products are in competitive proximity. Id.

Dove contends that this factor clearly favors its position that there is no likelihood of confusion, citing Girl Scouts of the United States v. Bantam Doubleday Dell Publishing Group, Inc., 808 F.Supp. 1112, 1127 (S.D.N.Y.1992), aff'd, 996 F.2d 1477 (2d Cir. 1993), which stated that "purchasers of children's books are undoubtedly literate; moreover, undoubtedly they have in mind supplying quality reading material to a five or six year old. In short, they are not necessarily `unsophisticated' because the books are low-priced." Girl Scouts, 808 F.Supp. at 1130. Because the books and audiobooks at issue are priced several times higher than the inexpensive books in Girl Scouts, Dove argues that this factor weighs heavily in its favor.

Evidence presented at trial, however, did not uniformly support Dove's assumption that children's book purchasers are discriminating in making their selections. There was evidence that books are often impulse purchases based upon a publication's cover or title. Moreover, there is no particular consumer sophistication attached to likely purchasers of the books at issue, which are geared to a wide segment of the American public. In the Court's view, this factor does not favor either party.

i. Other Factors

The eight Polaroid factors are non-exclusive. The senior user's delay, if any, in asserting its trademark claim is a factor which may also be considered. See, e.g., Hasbro, 858 F.2d at 79. Based on the facts in the record, the Court finds no undue delay by plaintiffs in discovering Dove's use of the mark "The Book of Virtues" or in commencing the instant action.

3. Conclusion

Most of the relevant Polaroid factors support the conclusion that there exists a significant likelihood of confusion as to source as a result of Dove's use of the book titles "The Children's Audiobook of Virtues" and "The Children's Book of Virtues." Although there have been no demonstrated instances of actual consumer confusion between Dove's books and plaintiffs' The Book of Virtues, and although the consumer surveys introduced at trial are only marginally probative of a likelihood of confusion, a number of other factors — including the strength of plaintiffs' title in its commercial context, the degree of similarity of the parties' titles, the extremely close competitive proximity of the products, and the evidence of Dove's intentional copying of plaintiffs' mark in bad faith — demonstrate that numerous ordinarily prudent purchasers are likely to be misled or confused as to the source of Dove's books.

Our finding that there is a likelihood of confusion is sufficiently compelling, in light of all the circumstances, to outweigh any potential First Amendment interests. In our view, Dove has reflexively invoked the First Amendment without offering a persuasive explanation of why free speech interests are seriously threatened by Lanham Act liability in this case. Although this case involves literary titles, it is not a case like Rogers, Cliffs Notes, or Girl Scouts, supra, where trademark liability runs the risk of unduly restricting authors of parody or fiction from writing about certain subjects and titling their works appropriately.[20] Unlike *301 the film title in Rogers, Dove's title is not an "integral element" of its books and their creator's "artistic expressions." Rogers, 875 F.2d at 1001. Rather, the evidence shows that Dove deliberately gave its children's story books confusingly similar titles in a blatant and ill-conceived effort to piggy-back on the goodwill associated with Bennett's best-selling title. The public interest in sparing consumers this type of confusion — confusion resulting from titles that are confusingly similar to other titles — "outweighs the slight public interest in permitting authors to use such titles." Rogers, 875 F.2d at 999 n. 5.

D. Relief

1. Injunctive Relief

Plaintiffs have carried their burdens of proof under Section 43(a) of the Lanham Act and demonstrated a risk of irreparable injury to their unregistered mark. We have considered the equities and balanced the conflicting interests of the parties, see Jim Beam Brands Co. v. Beamish & Crawford Ltd., 937 F.2d 729, 737 (2d Cir.1991), and concluded that an injunction is necessary to protect plaintiffs' mark.[21] Accordingly, Dove and its agents will be permanently enjoined from the manufacture, publication, distribution, and sale of any audio and/or print books under the titles "The Children's Audiobook of Virtues" and "The Children's Book of Virtues," or under any other title confusingly similar to plaintiffs' title "The Book of Virtues." Plaintiffs also seek an order requiring Dove to recall from circulation and destroy, or certify the destruction of, all copies of the covers of Dove's publications The Children's Audiobook of Virtues and The Children's Book of Virtues. See Amended and Supplemented Consolidated Pre-trial Order at 3. This additional relief is appropriate under the circumstances and will be granted.

2. Profits, Costs, and Attorneys' Fees

Plaintiffs also seek an accounting of Dove's profits of $23,105.80 from the sale of The Children's Book of Virtues and an award of their costs, including survey costs, and attorneys' fees. Under Section 35(a) of the Lanham Act, successful plaintiffs are entitled, "subject to the principles of equity, to recover (1) defendant's profits, (2) any damages sustained by the plaintiff, and (3) the costs of the action." 15 U.S.C. § 1117(a). The Act also provides that "[t]he court in exceptional cases may award reasonable attorney fees to the prevailing party." Id.

In this Circuit, an accounting of profits is normally available "only if the defendant is unjustly enriched, if the plaintiff sustained damages from the infringement, or if the accounting is necessary to deter a willful infringer from doing so again." George Basch Co. v. Blue Coral, Inc., 968 F.2d 1532, 1537 (2d Cir.1992) (citations and internal quotations omitted). Under any of these three theories for awarding profits, "a finding of defendant's willful deceptiveness is a prerequisite for awarding profits." Id.; see also International Star Class, 80 F.3d at 753. To recover under the unjust enrichment theory, a plaintiff must show that, were it not for defendant's infringement, the defendant's sales would otherwise have gone to the plaintiff. George Basch, 968 F.2d at 1538. To recover under the damage theory of profits, a plaintiff must demonstrate consumer confusion resulting from the infringement, id. at 1539, which may be shown through circumstantial evidence such as consumer surveys. See PPX Enterprises, Inc. v. Audiofidelity Enterprises, Inc., 818 F.2d 266, 271 (2d Cir. 1987).

We have examined the considerations for an accounting of profits set forth in George Basch and conclude that the equities weigh in favor of an accounting of profits under the deterrence theory.[22] We have already set *302 forth our finding that Dove infringed plaintiffs' title "The Book of Virtues" in bad faith in order to capitalize on the goodwill associated with Bennett's best-selling book, and that Dove failed to offer a credible innocent explanation. We believe that Dove's president did not testify forthrightly about his limited familiarity with plaintiffs' title and his intent in trading on its goodwill. We believe an accounting of profits in this case protects the public at large and "promote[s] the secondary effect of deterring public fraud regarding the source ... of consumer goods." George Basch, 968 F.2d at 1539. Accordingly, Dove will be awarded $23,105.80, the stipulated amount of Dove's infringing profits from the sale of The Children's Audiobook of Virtues.

Plaintiffs have also demonstrated an entitlement to the usual costs permitted under the local rules, which shall be taxed by the Clerk of the Court. However, plaintiffs have not demonstrated that they are entitled to recover the costs of conducting their consumer survey, and such costs shall not be recoverable.[23]

Deliberate and willful infringement can render a case "exceptional" and support an award of attorneys' fees. Centaur Communications, 830 F.2d at 1229; Quaker State Oil Ref. Corp. v. Kooltone, Inc., 649 F.2d 94, 95 (2d Cir.1981) (per curiam); Springs Mills, Inc. v. Ultracashmere House, Ltd., 724 F.2d 352, 357 (2d Cir.1983). Although we find that Dove willfully infringed plaintiffs' title "The Book of Virtues," we do not believe this case qualifies as "exceptional" in order to justify an award of attorneys' fees. This litigation presented a number of close and contested issues; the outcome was by no means a foregone conclusion. Under these circumstances, we decline to award plaintiffs' their attorneys' fees.

II. State Law Claims

Plaintiffs assert two claims under New York law: (1) common law trademark infringement and unfair competition, and (2) violation of the New York anti-dilution statute, N.Y. Gen. Bus. Law § 368-d.

Under New York law, the standards for a finding of unfair competition based on trademark infringement are substantially similar to those applied under the Lanham Act, except insofar as the state claim may require an additional element of bad faith or intent. See, e.g., Girl Scouts, 808 F.Supp. at 1131; Saratoga Vichy Spring Co. v. Lehman, 625 F.2d 1037, 1044 (2d Cir.1980). Just as in a Lanham Act claim, plaintiffs must demonstrate the use of their mark in a way likely to confuse consumers as to the source of the products. See Girl Scouts, 808 F.Supp. at 1131. For the reasons set forth above, the Court finds that Dove has engaged in unfair competition in violation of New York law.

To prevail on a Section 368-d dilution claim, "a plaintiff must prove, first, that its trademark either is of truly distinctive quality or has acquired secondary meaning, and, second, that there is a `likelihood of dilution.'"[24]Deere & Co. v. MTD Products, Inc., 41 F.3d 39, 42 (2d Cir.1994) (quoting Sally Gee, Inc. v. Myra Hogan, Inc., 699 F.2d 621, 624 (2d Cir.1983)). Predatory intent, while not precisely an element of the claim, is a relevant consideration. Id. at 42, 45-46. Dilution is defined as either the blurring of a trademark's product identification *303 or the tarnishment of the affirmative associations a mark has come to convey, Id. at 42-43. Blurring may occur "where the defendant uses or modifies the plaintiff's trademark to identify the defendant's goods and services, raising the possibility that the mark will lose:its ability to serve as a unique identifier of the plaintiff's product." Id. at 43. For the reasons set forth above, we find that plaintiffs' title "The Book of Virtues" has acquired secondary meaning and that there is likelihood of blurring. Dove's titles are so similar to plaintiffs' as to create the likelihood that plaintiffs' mark will lose its status as a unique identifier of plaintiffs' products. Further, we find evidence of predatory intent.

Accordingly, plaintiffs' two state law claims provide further support for their requested relief.

CONCLUSION

For the reasons set forth above, we find that Dove has infringed plaintiffs' trademark in the title "The Book of Virtues" through its use of the confusingly similar titles "The Children's Audiobook of Virtues" and "The Children's Book of Virtues." Dove's actions violate Section 43(a) of the Lanham Act, constitute common law trademark infringement and unfair competition, and violate the New York anti-dilution statute. Dove will be permanently enjoined from using the above-mentioned titles or any other confusingly similar titles. Plaintiffs shall be awarded $23,105.80 (the amount of defendant's infringing profits) and the usual costs permitted under the local rules.

Settle judgment on five days notice.

SO ORDERED.

NOTES

[1] The parties stipulated to certain facts in the "Amended and Supplemented Consolidated Pretrial Order." The Court will use the designation "SF" to refer to these stipulations. Documentary exhibits are identified as they are in the pretrial order, as either joint exhibits ("JX"), plaintiffs' exhibits ("PX"), or defendant's exhibits ("DX"). Adhering to a format initially established by Judge Schwartz and relied upon by the parties in their pre-trial preparation, the direct testimony of most trial witnesses came in by affidavit; the Court will refer to these trial affidavits as "Aff." "Tr." refers to the trial transcript.

[2] In 1996, S & S also published paperback versions of The Book of Virtues and The Moral Compass, four paperbacks in a series of 13 books entitled "Adventures From The Book of Virtues" based upon a PBS television series licensed by Bennett, and The Children's Book of Virtues calendar. Plaintiffs intend to publish additional products in "The Book of Virtues" series. Martin Aff. ¶ 18; SF 23, 71, 73.

[3] At trial, Viner claimed that he decided to publish the audiobook sometime before the summer of 1994. Tr. 201-02. However, because Viner supplied no details on the timing of this decision, the Court will utilize the summer 1994 date.

[4] Collins also testified that at a sales conference in December 1994 concern was raised about whether there would be a problem with Dove's proposed title because of its similarity to plaintiffs' title, and "we [Penguin] were assured that there was no conflict." Tr. 51-52. Collins recalled that Viner attended the conference on behalf of Dove. Tr. 53.

[5] Viner's claim at trial that plaintiffs stole the idea for The Children's Book of Virtues from him, see Tr. 217-18, is even more incredible and, in fact, contradicted by stipulated evidence that Bennett and his editor discussed a children's version of his book as early as 1993. SF 17.

[6] Boose's direct testimony summarized the findings set forth in her expert report, DX F. Boose's expert report was redacted at trial in light of Judge Schwartz's ruling that Dove was precluded from introducing evidence as to the generic nature of the mark "The Book of Virtues." The parties and the Court, however, understood that Judge Schwartz's ruling did not preclude the parties from introducing evidence as to the conceptual strength of the mark, which is relevant to other issues in the case. Accordingly, material relating solely to Boose's conclusion that "The Book of Virtues" is a generic mark was deleted from her original expert report.

During trial, however, the Court asked the parties to submit proffers of evidence they would introduce if the Court were to reopen the inquiry into whether or not plaintiffs' mark is generic. Defendant represented to the Court that Boose's original, unredacted report would be the only such evidence they would introduce, and that Boose's direct testimony would have conformed to her original report. Tr. 273-74. The original report, while containing essentially the same factual background and analysis as the report introduced at trial, concludes with Boose's opinion that the title "The Book of Virtues" is generic. The unredacted report is marked Court Exhibit I and is made part of the record.

[7] Boose did, however, identify a book written in 1405 by Christine de Pizan, Le Livre des trois Vertus, which has been translated under the title "The Book of Three Virtues." DX F at 7-8; Tr. 191-92.

[8] Consumer awareness of the mark "The Book of Virtues" is also indirectly evidenced by S & S's marketing plan for subsequent publications in "The Book of Virtues" series, which demonstrates plaintiffs' efforts to trade on public awareness of the mark by expanding into additional product lines. See Martin Aff. ¶ 9; Rosen Aff. ¶ 3; Dana Aff. ¶ 6.

[9] The "Squirt" format takes its name from the case Squirt Co. v. Seven-Up Co., 207 U.S.P.Q. 12, 20-21 (E.D.Mo.1979), aff'd, 628 F.2d 1086 (8th Cir.1980).

[10] The "Eveready" format takes its name from the case Union Carbide Corp. v. Ever-Ready, Inc., 531 F.2d 366, 381-82 (7th Cir.1976).

[11] The question was modified to refer to "book on tape" or "book cover," as appropriate depending on the item displayed.

[12] Again, the question was modified as appropriate to the item displayed.

[13] The Court does not find Dove's other criticisms of the Jacoby Survey compelling, including its view that the Jacoby Survey asked improper leading and suggestive questions. Rappeport Aff. ¶¶ 21-23. Aside from the ambiguity of the question addressed above, the Jacoby Survey asked appropriate questions that did not improperly suggest the answer plaintiffs were seeking. The Court also finds that the Jacoby Survey defined an appropriate universe of respondents.

[14] In contrast, a limitation of the Squirt format, used in the Jacoby Survey, is that "it explicitly leads respondents to consideration of the association between the two marks, a question that might not have occurred to them in a normal purchase situation." Simonson, supra, at 371. "This limitation can have a significant effect on confusion estimates when the awareness level of the senior mark is low." Id. at 386.

[15] While the PTO examiner's action in approving plaintiffs' mark for publication in the Official Gazette without requiring proof of secondary meaning may be marginally relevant to our analysis, we do not believe that it entitles plaintiffs to a rebuttable presumption that their mark is suggestive or has secondary meaning. Cf. McGregor-Doniger Inc. v. Drizzle Inc., 599 F.2d 1126, 1132 (2d Cir.1979) (decision of PTO to register a mark without requiring proof of secondary meaning "affords a rebuttable presumption that mark is more than merely descriptive"); Arrow Fastener Co. v. Stanley Works, 59 F.3d 384, 393 & n. 6 (2d Cir.1995) (rebuttable presumption of secondary meaning arises when PTO issues registration after requiring proof of secondary meaning). Here, no registration has yet been issued by the PTO for plaintiffs' mark, and an opposition proceeding is anticipated.

[16] In the case of literary titles, the cases suggest that a showing of secondary meaning is required even if a mark is found to be suggestive, but this issue remains unsettled in this Circuit. See Twin Peaks Prods., Inc. v. Publications Int'l, Ltd., 996 F.2d 1366, 1378 & n. 4 (2d Cir.1993). We need not decide the issue here because of our conclusion that plaintiffs' mark is descriptive.

[17] Dove contends that it commenced use of the mark in August 1994 "when it announced to the trade and to its distributor its intention of releasing an audiobook and book bearing the mark (Pre-Trial Order Stipulated Fact No. 34)." Defendant's Post-Trial Memorandum of Law ("Def. Mem.") at 13. However, SF 34 actually states: "At a production meeting in the late summer of 1994, Michael Viner, president of defendant, announced his plan to publish an audiobook entitled The Children's Audiobook of Virtues in early 1995." Thus, SF 34 does not establish that Viner announced anything to the "trade" in August 1994; at most, SF 34 establishes that Viner announced his plans to Dove's staff and, possibly, certain employees of Dove's distributor.

At trial, Viner took a different position than his counsel now takes. He claimed that Dove publicly announced its books in its "spring announcement list of 1994." Tr. 213. However, there are no documents in the record corroborating this claim, and Dove's counsel has apparently retreated from this testimony.

[18] Plaintiffs urge the Court to take into consideration the evidence that there now exists a "family" of related marks with "The Book of Virtues" as a "recognizable common characteristic" that is associated with Bennett, see J & J Snack Foods Corp. v. McDonald's Corp., 932 F.2d 1460, 1462-63 (Fed.Cir.1991), including the evidence of the development of this "family" occurring after Dove's entry into the market. See McDonald's Corp. v. Druck and Gerner DDS., P.C., 814 F.Supp. 1127, 1131 (N.D.N.Y.1993) (where plaintiff requested only injunctive relief, which is prospective, "the strength of the `family of marks' and the `likelihood of confusion' [should be measured] from a present standpoint," rather than retroactively); but see Marion Labs., Inc. v. Biochemical/Diagnostics, Inc., 6 U.S.P.Q.2d 1215, 1988 WL 252477 (1988) (proponent of a family of marks must prove that, prior to the junior user's entry, all or many of the marks in the alleged family were used and promoted in such a way as to create public perception of the family mark as an indicator of source). In the circumstances of this case, we do not believe it is necessary to rely on plaintiffs' evidence of the development of a family of marks after Dove's entry into the market, and we do not rely on any such evidence in assessing the likelihood of confusion.

[19] Taking counsel's advice "does not ipso facto and in all circumstances shield the actor from the consequences of his act." Cuisinarts, Inc. v. Robot-Coupe Int'l Corp., 580 F.Supp. 634, 638 (S.D.N.Y.1984). Inquiry must be made as to the facts disclosed to counsel and whether counsel's advice was "timely requested" and "honestly relied upon in shaping one's conduct. Otherwise counsel's advice is a sham, a smokescreen set up to mask the actor's real intent." Id.

Viner never testified squarely that he had been specifically advised by counsel that plaintiffs' mark "The Book of Virtues" was in the public domain and therefore not entitled to trademark protection. Also, no attorney testified in support of Viner's claim that he relied in good faith on the advice of counsel, as often occurs when a defendant asserts its good faith reliance on advice of counsel. See, e.g., Estee Lauder, Inc. v. The Gap, Inc., 932 F.Supp. 595, 615 (S.D.N.Y. 1996), rev'd on other grounds, 108 F.3d 1503 (2d Cir.1997).

[20] For example, in Cliffs Notes, the Second Circuit applied the Rogers balancing approach to a parody, "in which expression, and not commercial exploitation of another's trademark, is the primary intent, and in which there is a need to evoke the original work being parodied," Cliffs Notes, 886 F.2d at 495, and concluded that First Amendment concerns predominated over the likelihood of confusion. In Girls Scouts, the Court rejected an attempt by the national Girl Scouts and Boy Scouts to impose Lanham Act liability on a publisher of children's fiction for using the word "scouts" to describe a youth organization in which children engage in "scouting"-type activities. See Girl Scouts, 808 F.Supp. at 1120. Liability in both of these cases would have posed a far graver threat to free expression that liability does here. Dove's books need not evoke plaintiffs' work, as a parody must, nor must Dove use the phrase "book of virtues" in order to describe accurately its collections of classic stories and fables.

[21] Dove has not proposed a disclaimer as an alternative to injunctive relief. If it is so inclined, Dove may apply for relief from the injunction and propose a disclaimer, recognizing that it would bear a "heavy burden ... to come forward with evidence sufficient to demonstrate that any proposed [disclaimers] would significantly reduce the likelihood of confusion." Home Box Office, Inc. v. Showtime/the Movie Channel Inc., 832 F.2d 1311, 1316 (2d Cir.1987); see also Jim Beam Brands, 937 F.2d at 729.

[22] We decline to award profits under the unjust enrichment or damage theories. Plaintiffs produced no compelling evidence suggesting that Dove's infringement actually caused any sales diversion, a requirement for recovery of profits under the unjust enrichment theory. See George Basch, 968 F.2d at 1541. As for the damage theory of profits, plaintiffs produced no evidence of actual confusion and plaintiffs' consumer survey was only minimally probative.

[23] There is well reasoned authority in this district holding that the costs of performing consumer surveys are not recoverable under Section 35(a) of the Lanham Act. See Gillette Co. v. Wilkinson Sword, Inc., No. 89 Civ. 3586, 1992 WL 30938, at *10 (S.D.N.Y. Feb.3, 1992) (citing West Virginia University Hosps., Inc. v. Casey, 499 U.S. 83, 111 S.Ct. 1138, 113 L.Ed.2d 68 (1991), for principle that "items of cost should not be awarded unless a statute explicitly provides for those items"). Even if the costs of plaintiffs' survey were recoverable under Section 35(a), we do not believe an award is appropriate in this case because of the minimally probative nature of plaintiffs' survey.

[24] Dove argues that Section 368-d "excludes cases where infringement is claimed by a direct competitor selling a similar product," see Defendant's Trial Memorandum of Law at 12, but this view has been expressly rejected by the Second Circuit. See Nikon Inc. v. Ikon Corp., 987 F.2d 91, 96 (2d Cir.1993).