LEVAL, Circuit Judge:

Nabisco, Inc. and Nabisco Brands Company (collectively "Nabisco") appeal from the preliminary injunction entered by the United States District Court for the Southern District of New York (Shira A. Scheindlin, District Judge) upon the motion of Pepperidge Farm, Inc. and PF Brands, Inc. (collectively "Pepperidge Farm" or "Pepperidge"). The district court found that Nabisco's use of an orange, bite-sized, cheddar cheese-flavored, goldfish-shaped cracker (as part of a tie-in promotion of a Nickelodeon Television Network television production) would dilute the distinctive quality of Pepperidge Farm's mark consisting of an orange, bite-sized, cheddar cheese-flavored, goldfish-shaped cracker, in violation of the Federal Trademark Dilution Act ("FTDA"), section 43(c) of the Lanham Act, 15 U.S.C. § 1125(c), and New York's antidilution statute, N.Y. Gen. Bus. Law § 360- l. The court entered an order requiring Nabisco to recall and cease selling its goldfish cracker. We affirm.

BACKGROUND

1. Facts

Pepperidge Farm has produced small crackers in the shape of a goldfish (the "Goldfish" cracker) continuously since 1962. Although the Goldfish line of products includes crackers in various flavors and mixes, the primary product is the orange, cheddar cheese-flavored, fish-shaped cracker, sold in a bag or box under the trade name "Goldfish" and exhibiting a picture of the cracker on the exterior. The company has obtained numerous trademark registrations for the Goldfish design and name. In 1994, it launched an aggressive marketing campaign directed at children, who make up about half of Goldfish consumers, and between 1995 and 1998, it spent more than $120 million marketing the Goldfish line nationwide. The cracker has also been the subject of substantial media coverage, including a feature on "The Today Show" and an episode on "Friends." From 1995 to 1998, net sales of Goldfish crackers more than doubled, to $200 million per year. Measured by sales volume, Pepperidge Farm's Goldfish is the second-largest selling cheese snack cracker in America today. Measured in sales dollars, Goldfish ranks number one.

Occasionally, companies other than Pepperidge Farm have produced cheese crackers shaped like sea creatures, including "Guppies," "Dolphins Friends," and "Whales." Only one of these products has included a cracker shaped like a goldfish — Nabisco's "Snorkels." Snorkels obtained only a small market share, and it is no longer on the market.

In spring 1998, Nickelodeon Television Network approached Nabisco to explore a possible joint promotion for Nickelodeon's new cartoon program, "CatDog." In August 1998, Nabisco and Nickelodeon entered a Joint Promotion Agreement ("JPA"), giving Nabisco the right to produce cheese crackers in shapes based on the CatDog cartoon. The agreement required Nabisco to print on its packages that "CatDog and related titles, logos and characters are trademarks of" Nickelodeon's parent, Viacom International, Inc. Nabisco's CatDog product was intended to compete with other animal-shaped cheese crackers marketed to children.

The star of the CatDog cartoon program is the CatDog — a two-headed creature that is half cat and half dog. Each half of the CatDog has a distinct personality. The fish is the favorite food and the symbol for the cat half; the bone is the preferred meal and emblem for the dog half. Other characters that are featured on the cartoon include a mouse, a rabbit, a squirrel, and several dogs. In its first three months, the CatDog show garnered a 3.9 Nielsen rating, making it close to the most widely watched program for children.

Pursuant to its agreement with Nickelodeon, Nabisco developed a CatDog snack that consists of small orange crackers in three shapes: half the crackers in a package are in the shape of the two-headed CatDog character, one-quarter in the shape of a bone, and one-quarter in the shape of a fish. The fish-shaped cracker closely resembles Pepperidge Farm's Goldfish cracker in color, shape, and size, and taste, although the CatDog fish is somewhat larger and flatter, and has markings on one side. The CatDog product was to be sold in boxes featuring the CatDog and showing fish and bones in the background. The launch of the CatDog product was set for February 1, 1999.

2. Proceedings Below

In mid-December 1998, executives at Pepperidge Farm for the first time saw a sample of the CatDog product. On December 21, Pepperidge Farm wrote to Nabisco protesting the goldfish-shaped cracker and requesting that Nabisco cease and desist use of that cracker in its product and marketing. Nabisco responded by filing a complaint against Pepperidge Farm seeking a declaratory judgment under 28 U.S.C. § 2201 that the CatDog product did not violate any of Pepperidge Farm's rights in the Goldfish. Pepperidge Farm counterclaimed that Nabisco's goldfish constituted trademark infringement under § 43(a) of the Lanham Act, 15 U.S.C. § 1125(a), and dilution under the FTDA, § 43(c) of the Lanham Act, 15 U.S.C. § 1125(c), as well as dilution and unfair competition under New York law. Pepperidge Farm moved for a preliminary injunction barring Nabisco from marketing its product.

The district court found for Pepperidge and granted the preliminary injunction on the federal and state dilution claims, but not on the federal trademark infringement or state unfair competition claims. In a thorough opinion, the district judge concluded that the Pepperidge Farm Goldfish mark is nonfunctional, distinctive and famous and is protectable under the antidilution and infringement statutes. In deciding whether Nabisco's use of a fish in its CatDog product would dilute the Goldfish mark, the court applied the six-factor test proposed in a concurring opinion in Mead Data Central, Inc. v. Toyota Motor Sales, U.S.A., Inc., 875 F.2d 1026, 1035 (2d Cir. 1989) (Sweet, J., concurring), relating to the New York antidilution statute. The court found that each of the six Mead Data factors weighed in favor of a finding of dilution, and concluded that Pepperidge Farm had proven a likelihood of success on the merits of its dilution claims under both federal and state law.

In essence, Pepperidge Farm has taken a unique and fanciful idea — creating a cheese cracker in the shape of a goldfish — and turned this idea into its signature. Nabisco's inclusion of this signature element as part of the CatDog product strikes at the heart of what dilution law is intended to prevent: the "gradual diminution or whittling away of the value of the famous mark by blurring uses by others." Over time, the presence of Nabisco's goldfish-shaped cracker within the CatDog mix is likely to weaken the focus of consumers on the true source of the Goldfish.

Nabisco, Inc. v. PF Brands, Inc., No. 99 Civ. 0008, 1999 WL 47313, at *18 (S.D.N.Y. Feb. 3, 1999) (to be reported at 50 F. Supp.2d 188) (internal citation omitted). The court also concluded that likelihood of dilution "automatically" establishes irreparable harm, because "[d]ilution is itself an injury which [cannot] be recompensed by money damages." Id. at *6 (quoting Deere Co. v. MTD Prods., Inc., 860 F. Supp. 113, 122 (S.D.N.Y.), aff'd, 41 F.3d 39 (2d Cir. 1994)).

On the other hand the court found that Pepperidge Farm had failed to show a likelihood of success in establishing a claim of trademark infringement. This conclusion was primarily supported by Pepperidge Farm's failure to demonstrate that any actual confusion had occurred among consumers, and the fact that Nabisco's goldfish was part of a three-shape mixture and came in a package that featured most prominently the CatDog, rather than the fish.

Based on its finding of likely success in proving dilution, the court ordered Nabisco to recall and cease distributing its goldfish crackers. Nabisco appeals.

DISCUSSION

Nabisco's primary contentions on appeal are that: (1) Pepperidge Farm failed to show likelihood of success in proving dilution, because, in the market context (a) consumers would not associate the two products, and (b) Nabisco's mark was not substantially similar to Pepperidge Farm's; (2) the antidilution statutes were adopted to protect against dilution by the use of a similar mark on a non-competing product and do not apply to trademarks on competing products, which are governed instead by the infringement standard; (3) Nabisco's use of a fish is not a "trademark use" and is thus not actionable under the antidilution statutes; (4) dilution cannot be found without documentation of actual injury, consisting of an actual reduction in the senior mark's selling power.

1. Whether Pepperidge Farm showed likelihood of success in proving dilution.

We begin with the statutory text. The new federal antidilution act provides:

The owner of a famous mark shall be entitled, subject to the principles of equity and upon such terms as the court deems reasonable, to an injunction against another person's commercial use in commerce of a mark or trade name, if such use begins after the mark has become famous and causes dilution of the distinctive quality of the mark . . . .

15 U.S.C. § 1125(c)(1) (emphases added). The statute further provides a non-exclusive list of factors to be considered "[i]n determining whether the mark is distinctive and famous." Id. (emphasis added). The first of those factors invites courts to consider "the degree of inherent or acquired distinctiveness of the mark." Id. Dilution is defined as:

[T]he lessening of the capacity of a famous mark to identify and distinguish goods or services, regardless of the presence or absence of — (1) competition between the owner of the famous mark and other parties, or (2) likelihood of confusion, mistake, or deception.

15 U.S.C. § 1127.

We understand the FTDA to establish five necessary elements to a claim of dilution: (1) the senior mark must be famous; (2) it must be distinctive; (3) the junior use must be a commercial use in commerce; (4) it must begin after the senior mark has become famous; and (5) it must cause dilution of the distinctive quality of the senior mark.

The elements numbered (1) and (4) — fame of the senior mark, and junior use beginning after the senior mark has become famous use terms in their ordinary English language sense. Their meaning is clear. It is not disputed, furthermore, that Pepperidge Farm's Goldfish constitutes a famous mark, and that Nabisco's first use of its goldfish cracker would not occur until Pepperidge Farm's Goldfish had become famous. Nor is it disputed that Nabisco's sale of its goldfish cracker would involve a "commercial use in commerce." The third element is met.

The only elements that require discussion are numbers (2) and (5) — that the senior mark be "distinctive" and that the junior use "dilute its distinctive quality." Here the statute invokes a term of art in trademark law.

Distinctiveness in a mark is a characteristic quite different from fame. Distinctiveness is a crucial trademark concept, which places marks on a ladder reflecting their inherent strength or weakness. The degree of distinctiveness of a mark governs in part the breadth of the protection it can command. At the low end are generic words — words that name the species or object to which the mark applies. These are totally without distinctiveness and are ineligible for protection as marks because to give them protection would be to deprive competitors of the right to refer to their products by name. See Abercrombie Fitch Co. v. Hunting World, Inc., 537 F.2d 4, 9 (2d Cir. 1976) (Friendly, J.); Otokoyama Co. Ltd. v. Wine of Japan Import, Inc., 175 F.3d 266, 270 (2d Cir. 1999). Thus no one can claim the exclusive right to use the mark "CAR" for a car. One rung up the ladder are "descriptive" marks those that describe the product or its attributes or claims. These also have little distinctiveness and accordingly are ineligible for protection unless they have acquired "secondary meaning" — that is, unless the consuming public has come to associate the mark with the products or services of its user. See Abercrombie, 537 F.2d at 10. The next higher rung belongs to "suggestive" marks; these fall in an in-between category. See id. They do not name or describe the product for which they are used, but they suggest the qualities or claims of that product. They are more distinctive than descriptive marks, and thus are accorded trademark rights without need to demonstrate that consumers have come to associate them with the user of the mark. See id. at 11. Nonetheless, because they seek to suggest qualities of the product, they possess a low level of distinctiveness. They are given less protection than is reserved for more distinctive marks those that are "arbitrary" or "fanciful." Id. A mark is arbitrary or fanciful if there is no logical relationship whatsoever between the mark and the product on which it is used. However, even within the category of arbitrary or fanciful marks, there is still a substantial range of distinctiveness. Some marks may qualify as arbitrary because they have no logical relationship to the product, but nonetheless have a low level of distinctiveness because they are common. The most distinctive are marks that are entirely the product of the imagination and evoke no associations with human experience that relate intrinsically to the product. The arbitrary or fanciful quality is what renders the mark distinctive; another seller of the same product or service would have no justification for using the same or a similar mark. The strongest protection of the trademark laws is reserved for these most highly distinctive marks. See Abercrombie, 537 F.2d at 11.

It is quite clear that the statute intends distinctiveness, in addition to fame, as an essential element. The operative language defining the tort requires that "the [junior] person's . . . use . . . cause dilution of the distinctive quality of the [senior] mark." 15 U.S.C. § 1225(c)(1). There can be no dilution of a mark's distinctive quality unless the mark is distinctive. Furthermore, the statute lists factors to be considered in determining "whether the mark is distinctive and famous." Id. Clearly both qualities are required.

The requirement of distinctiveness is furthermore an important limitation. A mark that, notwithstanding its fame, has no distinctiveness is lacking the very attribute that the antidilution statute seeks to protect. The antidilution statute seeks to guarantee exclusivity not only in cases where confusion would occur but throughout the realms of commerce. Many famous marks are of the common or quality-claiming or prominence-claiming type — such as American, National, Federal, Federated, First, United, Acme, Merit or Ace. It seems most unlikely that the statute contemplates allowing the holders of such common, albeit famous, marks to exclude all new entrants. That is why the statute grants that privilege only to holders of distinctive marks. As explained below, we believe that the Pepperidge Farm Goldfish mark is neither near the top nor the bottom of the ladder of distinctiveness, but is reasonably distinctive — certainly sufficiently so to qualify for the statute's protection.

The fifth element, "dilution of the distinctive quality of the mark" is the key operative element of the statute. The statute explains that dilution is "the lessening of the capacity of a famous mark to identify and distinguish goods or services." 15 U.S.C. § 1127. We have likewise described dilution under the New York statute as the loss of the "ability to clearly and unmistakably distinguish one source." Hormel Foods Corp. v. Jim Henson Prods., Inc., 73 F.3d 497, 506 (2d Cir. 1996) (internal citations and quotation marks omitted).

The antidilution statutes rest on a judgment that the "stimulant effect" of a distinctive and well-known mark is a "powerful selling tool" that deserves legal protection. Restatement (Third) of Unfair Competition § 25 cmt. c (1995). This power derives not only from "the merit of the goods upon which [the mark] is used, but equally [from the mark's] own uniqueness and singularity." Frank I. Schechter, The Rational Basis of Trademark Protection, 40 Harv. L. Rev. 813, 831 (1927). Even when an unauthorized use of the mark does not cause consumer confusion, it can "reduce the public's perception that the mark signifies something unique, singular, or particular." H. Rep. 104-374, at 3 (1995), reprinted in 1995 U.S.C.C.A.N. 1029, 1030. The junior use thereby diminishes the "selling power that a distinctive mark or name with favorable associations has engendered for a product in the mind of the consuming public." Sally Gee, Inc. v. Myra Hogan, Inc., 699 F.2d 621, 624-25 (2d Cir. 1983).

It is not yet entirely clear how courts should determine whether a junior use causes a senior mark to suffer dilution. See, e.g., Sally Gee, 699 F.2d 621, 625 ("dilution remains a somewhat nebulous concept") (applying New York statute). In dealing with the related question of infringement by reason of likelihood of consumer confusion, this court through study of the individual facts of the cases, and the factors suggested by the facts of those cases, came to develop gradually over time a nonexclusive list of factors that can help courts in making this determination. See Polaroid Corp. v. Polarad Electronics Corp., 287 F.2d 492, 495 (2d Cir. 1961) (Friendly, J.). We adopt a similarly cautious and gradual approach.

Considering the factors that appear pertinent on these particular facts, which we discuss below, we agree with the district court that Pepperidge Farm is likely to succeed in establishing that Nabisco's use of its goldfish shape in an orange, cheddar-cheese-flavored, bite-sized cracker dilutes the distinctive quality of Pepperidge Farm's previously famous mark, consisting of a goldfish-shaped orange, cheddar-cheese-flavored, bite-sized cracker.

(a) Distinctiveness. In our view, distinctiveness plays a dual role. First, as discussed above, it is a statutory element. A mark cannot qualify for protection unless it is distinctive. Second, the degree of distinctiveness of the senior mark has a considerable bearing on the question whether a junior use will have a diluting effect. As the distinctiveness of the mark is the quality that the statute endeavors to protect, the more distinctiveness the mark possesses, the greater the interest to be protected. And conversely, the more the senior mark tends toward the weak, common, quality-claiming, or prominence-claiming type, the more strongly that weakness would argue against a finding of dilution, especially if the senior use is in a distinctly different field. See Restatement § 25 cmt. f ("The degree of distinctiveness of the prior user's mark is also relevant because it measures the likelihood that the mark evokes the type of exclusive images and associations protected under the antidilution statutes. A very distinctive mark is thus more likely to suffer dilution of distinctiveness than is a less distinctive mark.").

We believe the goldfish shape of Pepperidge Farm's Goldfish crackers exhibits a moderate degree of distinctiveness. The fish shape has no logical relationship to a cheese cracker. However, we recognize it is not in the highest category of distinctiveness that belongs to a purely fanciful made-up pattern or design, or a made-up word like Kodak. Merchants often use an animal's likeness as a trademark. Furthermore, we recognize that it is not uncommon for children's cookies or crackers to be made in animal shapes. Indeed, there have been other fish-shaped crackers on the market, including one goldfish-shaped cracker (Nabisco's "Snorkels"). But Nabisco's "Snorkels" were not successful and have been withdrawn, and the other fish-shaped crackers were not similar to Pepperidge Farm's Goldfish; these crackers have not significantly affected the distinctiveness of Pepperidge Farm's Goldfish. In sum, because the use of the goldfish shape has no logical relationship to a bite-sized cheese cracker and for the other reasons discussed above, we believe that Pepperidge Farm's senior mark is reasonably distinctive.

(b) Similarity of the marks. The degree of similarity of the junior mark to the senior is an obvious factor bearing on a finding of dilution. The marks must be of sufficient similarity so that, in the mind of the consumer, the junior mark will conjure an association with the senior. In that manner the junior mark will lessen the distinctiveness of the senior mark. See Mead Data, 875 F.2d at 1029 (majority opinion) ("We hold . . . that the marks must be `very' or `substantially' similar and that, absent such similarity, there can be no viable claim of dilution.").

Although we recognize that Nabisco's fish is not identical to Pepperidge Farm's, the similarity is sufficient to lessen the distinctive aspect of the Pepperidge Farm Goldfish in the eyes of consumers . Both fish are presented arbitrarily in the form of a cracker. Notwithstanding slight differences in shape, size and marking, Nabisco's crackers are essentially the same color, shape, size, and taste. The markings that appear on Nabisco's but not on Pepperidge Farm's crackers are quite small and faint so that they would not be easily noticed, except on a close inspection of a sort that is not likely to be performed by one who is intent on popping the crackers into his mouth. Furthermore, the markings appear on only one side of Nabisco's cracker.

Nabisco contends that the court's finding of similarity was in error because of the "context" in which Nabisco is using the goldfish. Nabisco emphasizes that it has packaged the CatDog product in a box quite unlike the Goldfish package, and that the CatDog product is a three-shape mix that is only one-fourth goldfish. Nabisco asserts that in these contexts, Nabisco's goldfish cracker will evoke an association only with the CatDog cartoon, not with the Goldfish cracker.

Like the district court, we are not persuaded by Nabisco's argument. As for Nabisco's box, many consumers of its crackers will not see the box; they will find goldfish-shaped cheddar cheese crackers served in a dish at a bar or restaurant or friend's house, looking very much like the familiar Pepperidge Farm Goldfish product. Infringement cases have consistently held post-sale confusion as well point-of-sale confusion to be actionable under the Lanham Act. See, e.g., Lois Sportswear, U.S.A., Inc. v. Levi-Strauss Co., 799 F.2d 867, 872-73 (2d Cir. 1986); I.P. Lund Trading ApS v. Kohler Co., 163 F.3d 27, 44-45 (1st Cir. 1998). We recognize that dilution can occur as well in a post-sale as in a point-of-sale context.

The fact that only one quarter of the crackers in Nabisco's mix are fish and that the other two shapes reinforce the CatDog theme is helpful to Nabisco, but in our view insufficiently so. Many consumers seeing the crackers in a dish will not recognize the relationship of the fish to the Nickelodeon CatDog story. They will recognize a fish reminiscent of Pepperidge Farm's fish.

(c) Proximity of the products and likelihood of bridging the gap. The legislative history of the antidilution statutes shows that the legislatures were largely concerned with junior uses of famous marks on products unrelated to the senior area of commerce as in the hypothetical cases of Buick aspirin, Schlitz varnish, or Kodak pianos. See H. Rep. 104-374, at 3, reprinted in 1995 U.S.C.C.A.N. 1029, 1030; Mead Data, 875 F.2d at 1031. Some courts and commentators have questioned the relevance of similarity of products because the law's "primary purpose was to apply in cases of widely differing goods." I.P. Lund, 163 F.3d at 49; see also 3 McCarthy § 24:94.1, at 24-164. Nabisco also contends the statutes should not apply to use on competing products, which should be governed exclusively by the infringement laws.

We disagree with this view; or at least find it overstated. While the antidilution statutes aim at a different harm than the infringement statute and dilution undoubtedly can occur among non-competing products, we see no reason why dilution cannot occur as well where the products are competing. See Restatement § 25 cmt. f ("The nature of the respective goods is . . . relevant to dilution."). The closer the junior user comes to the senior's area of commerce, the more likely it is that dilution will result from the use of a similar mark.

Consumer confusion — the nub of an action for infringement — is, of course, unnecessary to show the actionable dilution of a famous mark. It does not follow, however, that dilution cannot be found in circumstances that would also support an action for infringement. Consumer confusion would undoubtedly dilute the distinctive selling power of a trademark. There can be little doubt that a junior use of Buick, Schlitz or Kodak on automotive equipment, drinks, and photographic products would lessen the distinctiveness of the senior mark at least as much as if the junior use were on unrelated products, such as aspirin, varnish and pianos. We can see no reason not to apply the antidilution statute to use on competing or closely related products, where likelihood of confusion, and thus infringement, might also be found.

Indeed, there may well be cases in which proximity may be necessary to dilution. It is easy to imagine instances where because of the low level of distinctiveness of the senior mark, or insufficient similarity between the two, the use of the junior mark in a remote area of commerce would have little tendency to remind consumers of the senior mark and thus little capacity to dilute its effectiveness, but where use of the same junior mark in a closely related area would bring about the harm the statute was designed to avoid. A junior use that confuses consumers as to which mark is which surely dilutes the distinctiveness of the senior mark.

For example, as noted above, a fish shape, while reasonably distinctive, is not as distinctive as a made-up word like Kodak. If Pepperidge Farm were seeking to enjoin the use of a fish-shape as the logo for an automobile, a newspaper, or a line of sports clothes, much less as the mark of a fish merchant or fish restaurant, it seems questionable whether dilution should be found. But here the junior user's area of commerce is identical to the senior's. A second major seller of goldfish-shaped, orange-colored, cheddar-flavored, bite-sized crackers can hardly fail, in our view, to dilute the distinctiveness in the eyes of the consumers of the senior mark in a goldfish-shaped, orange-colored, cheddar-flavored, bite-sized cracker.

Where the junior mark is used on a different product, courts examine whether the senior is likely to enter the junior's market, or "bridge the gap." Polaroid, 287 F.2d at 495. This factor recognizes "the senior's interest in preserving avenues of expansion and entering into related fields." Hormel, 73 F.3d at 504 (citations omitted). Here, because the junior use is in the same segment of commerce as the senior, there is no need to consider the likelihood that either might bridge the gap between them. There is no gap.

(d) Interrelationship among the distinctiveness of the senior mark, the similarity of the junior mark, and the proximity of the products. The foregoing discussion suggests that there is a close interdependent relationship among these factors. The weaker any of the three factors may be, the stronger the others must be to make a case of dilution. To choose one of many possible hypothetical examples to illustrate this interdependence: with a highly distinctive senior mark, like Chevrolet for cars, even an only moderately similar junior mark — such as Chevremont might dilute the distinctive quality of the senior if it were used in the automotive industry, but probably not if the junior were used in a distant area like perfumes.

Considering these three factors in relationship to one another, in this case we have a moderately distinctive senior mark, a substantial degree of similarity between the senior and junior mark, and the use of the junior mark in exactly the same area of commerce as the senior. In view of the substantial similarity of the marks and their use in precisely the same area of commerce, we believe that Pepperidge Farm made a strong case for the likelihood of dilution, notwithstanding that its mark is only moderately distinctive.

(e) Shared consumers and geographic limitations. Another relevant factor is the extent of overlap among consumers of the senior user's products and the junior user's products. This factor is meaningful because dilution requires that a mark become less distinctive to consumers. If the consumers who buy the products of the senior user never see the junior user's products or publicity, then those consumers will continue to perceive the senior user's mark as unique, notwithstanding the junior use. See Restatement § 25 cmt. f ("If the goods are marketed in different stores to different buyers . . . a connection between the prior and subsequent use may be unlikely.").

This factor strongly favors Pepperidge Farm. The two products will be in direct competition with one another, each being a substitute for the other. Both are to be marketed nationally in grocery stores. Children are also target audiences for both products. The consumers of Pepperidge Farm's Goldfish will see Nabisco's product.

(f) Sophistication of consumers. Courts examining questions of infringement, as well as dilution, have looked at the sophistication of consumers as a relevant factor. See Mead Data, 875 F.2d at 1031-32; Sally Gee, 699 F.2d at 626; Polaroid, 287 F.2d at 495. Consumers who are highly familiar with the particular market segment are less likely to be confused by similar marks and may discern quite subtle distinctions. Conversely, unsophisticated customers lack this discrimination and are more vulnerable to the confusion, mistake and misassociations against which the trademark law protects. "Purchasers of relatively inexpensive goods such as ordinary grocery store foods are held to a lesser standard of purchasing care." 3 McCarthy § 23:95, at 23-188.

Nabisco argues that this factor strongly supports its case, because children will have no difficulty recognizing the Nabisco product as a reference to the CatDog and will thus keep the two marks separate and distinct. Even if Nabisco is correct in that surmise, it seems to us to have only moderate importance, for two reasons. First, while children may be the primary ultimate consumers of the crackers, they are generally not the purchasers. Adult purchasers of crackers may be less sophisticated than children in recognizing the differences between the two fish. Even if, in the minds of children, the addition of Nabisco's CatDog family to the cheese cracker landscape does not lessen the distinctiveness of Pepperidge Farm's mark in its Goldfish, it is likely to do so among adults who will have less awareness of Nickelodeon's CatDog and of the differences between the two competing crackers.

Furthermore, even though children maybe the primary target of Pepperidge's marketing, they make up only one half the Goldfish market. Bite-sized cheese crackers are frequently served to adults as snacks accompanying cocktails. Once removed from their packaging, we think Nabisco's new goldfish cheese-crackers may well dilute the distinctiveness of Pepperidge Farm's mark in its goldfish-shaped, cheese cracker.

(g) Actual confusion. Consumers' actual confusion of the junior and the senior mark may also be a factor in finding dilution. While we recognize that neither actual confusion nor likelihood of confusion is necessary to sustain an action for dilution, it does not follow that actual confusion cannot be highly probative of dilution. Confusion lessens distinction. When consumers confuse the junior mark with the senior, blurring has occurred.

Except as surveys are used to test the likelihood of consumer confusion between existing and yet-to-be marketed products, actual confusion can arise only where the junior and the senior mark have coexisted in the marketplace. Here the CatDog product has not yet been launched on the market. There has been no opportunity for actual confusion to arise. The absence of evidence of confusion of consumers, therefore, has no probative value.

(h) Adjectival or referential quality of the junior use. Unlike the remedy of trademark infringement, which generally protects the senior user's mark only within the senior's areas of commerce, the antidilution statute guarantees exclusivity to the senior in unrelated areas of commerce — as exhibited by the examples from the legislative history of Buick aspirin, Schlitz varnish or Kodak pianos. This raises a problem. The senior's mark might be arbitrary or fanciful in the senior area of commerce but highly adjectival — or even generic — in the junior's area. For example, the shape of Pepperidge's Goldfish crackers is arbitrary as a cheese cracker, but would be adjectival if used on a sign by a fish market or fish restaurant. Cf. Abercrombie, 537 F.2d at 9 n. 6 (noting that "Ivory" is arbitrary as applied to soap, but generic when used to describe products made from elephant tusks).

The stronger the adjectival association between the junior use and the junior area of commerce, the less likelihood there is that the junior's use will dilute the strength of the senior's mark. The logical association between a fish image and a fish business would lead consumers to understand the fish sign as descriptive of the junior's business, regardless whether it is also being used as a mark. Consumers would be unlikely to draw a diluting association between the junior mark and the Goldfish cracker. Similarly, if the hypothetical "Chevremont" mark in our earlier example were used on cheese, rather than automobile equipment, it would be less likely to dilute the distinctive quality of Chevrolet's mark.

It is a generally accepted principle of the trademark law, furthermore, that a senior claim to a mark does not bar a junior from using the same words (or symbols) comprising the mark in their descriptive sense. See 15 U.S.C. § 1115(b)(4); Restatement § 28. While the principle has been recognized primarily in connection with infringement actions, see, e.g., Car-Freshner Corp. v. S.C. Johnson Son, Inc., 70 F.3d 267 (2d Cir. 1995), we see no reason why it should have any less application to actions for dilution.

Nabisco claims some protection from the fact that its fish shape is not arbitrary but acts as a reference to the fish in Nickelodeon's CatDog story. The weakness in its argument lies in the fact that when the Nabisco crackers are served in a bowl, consumers who are not familiar with the Nickelodeon entertainment and its cross licensing with Nabisco will see simply crackers very similar to Pepperidge Farm's fish (together with other shapes) and will not know that it celebrates Nickelodeon's CatDog entertainment.

(i) Harm to the junior user and delay by the senior user. A factor we have noted in considering infringement is whether the senior user's effort to enjoin the junior use was made with reasonable promptness and whether the junior user will suffer harm resulting from any such delay. See Chandon Champagne Corp. v. San Marino Wine Corp., 335 F.2d 531, 535-36 (2d Cir. 1964) (Friendly, J.). The factor seems at least as relevant to the question of dilution, perhaps more so. A senior user cannot have sued, at least under the federal statute, prior to its enactment. Furthermore, the senior user may have felt no need to sue a junior user who was in a remote area of commerce. The junior may already have accumulated substantial goodwill in its non-infringing mark in the years prior to suit.

These factors cannot help Nabisco in this case. Immediately on learning of Nabisco's contemplated junior use, Pepperidge Farm took steps to block it, and Nabisco then sought declaratory relief before launching its product. Nabisco has not yet built up a following in its use of the goldfish and will suffer no harm in the form of deprivation of existing goodwill if barred from using it.

(j) Effect of senior's prior laxity in protecting the mark. Nabisco argues that Pepperidge Farm has lost its right to claim dilution by reason of its prior laxity in allowing others (including Nabisco) to market fish-shaped crackers. As discussed above, the other fish-shaped crackers marketed were either too dissimilar or too insignificant to warrant our finding that Pepperidge Farm forfeited the right to sue to preserve the distinctiveness of its mark.

* * *

Considering the reasonable distinctiveness of the Goldfish mark, the very close proximity of the products, the degree of similarity between the two goldfish crackers, the low level of sophistication of many consumers, the occurrence of adjudication at the start of the junior use (and consequent absence of injury to the junior user's accumulated goodwill in its mark), we conclude that Pepperidge Farm has demonstrated a high likelihood of success in proving that Nabisco's commercial use of its goldfish shape will dilute the distinctiveness of Pepperidge Farm's nearly identical famous senior mark. We conclude the district court committed no error in granting a preliminary injunction.

2. Application of antidilution statutes to competing products.

Nabisco contends that the dilution statute has no application to competing products. We disagree. For the reasons discussed above, we believe that dilution can occur where the junior mark's use competes directly with the senior's as well as where the junior use is in a non-competing market. In general, the closer the products are to one another, the greater the likelihood of both confusion and dilution. The senior user has a right to the antidilution law's remedy in either case. Here we further note that the text of the federal statute directly contradicts Nabisco's position. The act expressly states that dilution may occur "regardless of the presence or absence of . . . competition between the owner of the famous mark and other parties." 15 U.S.C. § 1127(c) (emphasis added).

We are not persuaded by the argument that it is unnecessary to apply the antidilution statute to a junior use on a competitive product because such a competing use would already be forbidden by the infringement statutes. In the absence of contrary legislative command, the fact that other remedies may be available to prevent a perceived ill does not seem to be sufficient reason to construe a statute as not reaching circumstances that fall squarely within its words. The fact that injured senior users may thus be given a choice of remedies is not sufficient reason to read into the antidilution statute limitations that Congress did not write. Moreover, failure to construe the antidilution statutes as reaching competing products may lead to a gap in coverage; the products might be found too far apart to support a finding of likelihood of confusion — (and therefore an infringement action) — yet too close together to permit a finding of dilution.

We have already held that New York's antidilution statute applies to "competitors as well as noncompetitors," Nikon Inc. v. Ikon Corp., 987 F.2d 91, 96 (2d Cir. 1993), and we now so hold under the FTDA.

3. Whether Nabisco's use of the goldfish cracker is a trademark use.

Nabisco also contends that the antidilution statutes require that the junior use of a mark be a "trademark use" — that is, a use for the designation of a sponsor or identification of a product and that its use of the fish shape is neither. Whether the antidilution statute applies only to trademark uses by the junior user, as Nabisco contends, seems a complicated question. We need not resolve it, however, because we disagree with Nabisco's contention that its use of the goldfish cracker is a nontrademark use. Even if Nabisco's argument is correct, the statute would nonetheless apply to Nabisco's joint venture with Nickelodeon to promote Nickelodeon's trademarks through Nabisco's crackers.

Nabisco's contention overlooks its association with Nickelodeon. As Nabisco's package acknowledges, the CatDog, the bone and the fish are trademark characters of Nickelodeon's CatDog television show. Nickelodeon has chosen to publicize its show by causing dissemination of its trademarks on crackers, while Nabisco seeks to boost its sales of crackers by associating them with Nickelodeon's popular trademark characters. Through their exclusive licensing arrangement, Nabisco and Nickelodeon have made a joint venture for their mutual benefit with the purpose of promoting Nickelodeon's trademark characters. While the characters depicted on the crackers are not trademarks of Nabisco, Nabisco has undertaken in its joint venture to act as Nickelodeon's agent to promote Nickelodeon's marks. Nabisco's use of the CatDog, fish and bone characters declares Nickelodeon's sponsorship. It is a trademark use.

4. The need for proof of actual dilution.

Relying on a recent decision by the Fourth Circuit, Nabisco also asserts that proof of dilution under the FTDA requires proof of an "actual, consummated harm." Ringling Bros.-Barnum Bailey Combined Shows, Inc. v. Utah Division of Travel Dev., 170 F.3d 449, 464 (4th Cir. 1999). We reject the argument because we disagree with the Fourth Circuit's interpretation of the statute.

It is not clear which of two positions the Fourth Circuit adopted by its requirement of proof of "actual dilution." Id. The narrower position would be that courts may not infer dilution from "contextual factors (degree of mark and product similarity, etc.)," but must instead rely on evidence of "actual loss of revenues" or the "skillfully constructed consumer survey." Id. at 457, 464-65. This strikes us as an arbitrary and unwarranted limitation on the methods of proof.

To require proof of actual loss of revenue seems inappropriate. If the famous senior mark were being exploited with continually growing success, the senior user might never be able to show diminished revenues, no matter how obvious it was that the junior use diluted the distinctiveness of the senior. Even if diminished revenue could be shown, it would be extraordinarily speculative and difficult to prove that the loss was due to the dilution of the mark. And as to consumer surveys, they are quite subject to manipulation and are therefore not highly reliable. If a junior user began to market Buick aspirin or Schlitz shellac, we see no reason why the senior users could not rely on persuasive circumstantial evidence of dilution of the distinctiveness of their marks without being obligated to show lost revenue or engage in an expensive battle of unreliable surveys. Plaintiffs are ordinarily free to make their case through circumstantial evidence that will justify an ultimate inference of injury. "[C]ontextual factors" have long been used to establish infringement. We see no reason why they should not be used to prove dilution.

The broader reading of the Fourth Circuit's "actual, consummated" dilution element would require not only that dilution be proved by a showing of lost revenues or surveys but also that the junior be already established in the marketplace before the senior could seek an injunction. We recognize that the language of the statute gives some support to this reading, in that it uses the formulation, "causes dilution," rather than referring to "likelihood of dilution." Ringling Bros., 170 F.3d at 464. In our view, however, such a reading depends on excessive literalism to defeat the intent of the statute. Notwithstanding the use of the present tense in "causes dilution," it seems plausibly within Congress's meaning to understand the statute as intending to provide for an injunction to prevent the harm before it occurs.

To read the statute as suggested by the Ringling opinion would subject the senior user to uncompensable injury. The statute could not be invoked until injury had occurred. And, because the statute provides only for an injunction and no damages (absent willfulness), see 15 U.S.C. § 1125(c)(2), such injury would never be compensated. The Ringling reading is also disastrously disadvantageous for the junior user. In many instances the junior user would wish to know whether it will be permitted to use a newly contemplated mark before the mark is launched rather than after. That is why Nabisco sought declaratory relief as soon as Pepperidge Farm objected. If the statute is interpreted to mean that no adjudication can be made until the junior mark has been launched and has caused actual dilution, businesses in Nabisco's position will be unable to seek declaratory relief before going to market. They will be obligated to spend the huge sums involved in a product launch without the ability to seek prior judicial assurance that their mark will not be enjoined.

We are not at all sure that the Ringling opinion intends to limit the application of the statute to dilution that has actually occurred, as opposed to the narrower reading discussed above relating to manner of proof. In any event, we read the statute to permit adjudication granting or denying an injunction, whether at the instance of the senior user or the junior seeking declaratory relief, before the dilution has actually occurred.

5. The court's finding of predatory intent.

We agree with Nabisco's objections to the court's finding of predatory intent as a factor supporting its finding of dilution. The court noted at the outset that predatory intent of the kind that might support a claim of dilution refers only to a junior user's adoption of its mark "hoping to benefit commercially from association with the [famous] senior mark." 1999 WL 47313, at *14 (citing Mead Data, 875 F.2d at 1037). Nonetheless, the court proceeded to find predatory intent on the basis of facts that had little or no tendency to suggest that Nabisco adopted its fish shape in the hope of benefitting from association with Pepperidge Farm's famous Goldfish cracker.

Factors that supported the court's finding of predatory intent included that (i) Nabisco could have selected a shape other than the fish within the CatDog cast of characters; (ii) Nabisco did not reveal its plans to launch the CatDog product, including the goldfish, to Pepperidge Farm; (iii) Nabisco, claiming attorney-client privilege, declined to produce an opinion letter of its counsel; and (iv) Nabisco impeded Pepperidge Farm's discovery by failing to turn over sample boxes, from which the court inferred that Nabisco intended to corner the survey evidence. See id. at *14-17.

The fact that Nabisco could have chosen to use a shape other than a fish is generally true in every trademark dispute. In virtually every case of choice of a mark, a different mark could have been chosen. If the availability of other marks supported a finding of predatory intent, it would be found in virtually every case.

As to Nabisco's planning its product launch in secret without advising its competitor, we do not think this supports the conclusion that it intended to benefit improperly from a consumer association with Pepperidge Farm's Goldfish mark. We believe it is commonplace for business entities to keep their business plans secret from their competitors. Such behavior does not support the conclusion that Nabisco intended to free-ride on Pepperidge Farm's Goldfish trademark.

As to Nabisco's failure to produce the forty boxes of CatDog crackers Pepperidge Farm demanded, this may well have been a reprehensible effort to foil Pepperidge Farm's discovery and its ability to commission a survey. Under the Federal Rules of Civil Procedure, there were remedies available to Pepperidge Farm. If it obtained a court order requiring Nabisco to comply with the discovery demand and Nabisco continued to refuse to comply, Rule 37(b)(2) would have provided a range of sanctions the court might have imposed on Nabisco. These include that: matters might "be taken as established for purposes of the action;" Nabisco might have been prevented from supporting or offering designated claims or defenses; its pleadings might have been stricken; or default judgment might have been rendered. See Fed. R. Civ. P. 37(b)(2)(A)-(D). But those sanctions are available only after the court has ordered compliance and the party ordered has nonetheless failed to disclose. The predicate for the sanction did not exist in this case. The simple fact that Nabisco did not comply with the discovery demand did not show that it intended to benefit from consumer association of its product with Pepperidge Farm's Goldfish.

We are particularly troubled by the court's reliance on Nabisco's assertion of the attorney-client privilege. Nabisco asserted the privilege to justify its refusal to produce an opinion letter of counsel. To be sure, the invocation of certain privileges can properly support inferences against the party claiming the privilege. For example, a inference will be drawn against a party to a civil suit that invokes the Fifth Amendment privilege against self-incrimination. See Baxter v. Palmigiano, 425 U.S. 308, 318-320 (1976); United States v. 4003-4005 5th Ave., Brooklyn, NY, 55 F.3d 78, 82-83 (2d Cir. 1995). But we know of no precedent supporting such an inference based on the invocation of the attorney-client privilege. This privilege is designed to encourage persons to seek legal advice, and lawyers to give candid advice, all without adverse effect. See Upjohn Co. v. United States, 449 U.S. 383, 389 (1981); 8 Wigmore, Evidence § 2291 (McNaughton rev. 1961). If refusal to produce an attorney's opinion letter based on claim of the privilege supported an adverse inference, persons would be discouraged from seeking opinions, or lawyers would be discouraged from giving honest opinions. Such a penalty for invocation of the privilege would have seriously harmful consequences.

The cases on which the district court relied to draw an adverse inference are based on different circumstances. Those cases were patent disputes in which the party asserting the privilege had "an affirmative duty of due care to avoid infringement of the known patent rights of others," including a duty to "seek and obtain competent legal advice." Electro Med. Sys., S.A. v. Cooper Life Sciences, Inc., 34 F.3d 1048, 1056 (Fed. Cir. 1994). The Federal Circuit has reasoned that in such cases, the assertion of the privilege supports an inference that the alleged infringer either failed in the duty to obtain an opinion of counsel or failed in the duty to avoid infringement by ignoring the mandated opinion. See id.

Those cases are altogether distinguishable because Nabisco had no legal duty to obtain an attorney's opinion. It was free under law to obtain an opinion or not. The law affording the attorney-client privilege was designed to encourage Nabisco to opt in favor of seeking legal advice. But the imposition of the sanction will undermine that objective. We believe the district court erred in drawing an adverse inference against Nabisco by reason of its invocation of the attorney-client privilege.

We conclude there was no basis for the court's finding of predatory intent. But this does not affect our conclusion. The case for a preliminary injunction was sufficiently well supported by the factors we have reviewed above that no harm resulted from the court's reliance on an additional finding that was not warranted.

6. Irreparable Harm.

We reject Nabisco's contention that Pepperidge Farm has failed to show it would suffer irreparable harm if Nabisco's CatDog product were released into the market. Nabisco's argument depends primarily on the fact that other fish-shaped crackers have been sold in the past. Notwithstanding the occasional presence of other fish-shaped crackers in the marketplace, the district court reasonably found that Pepperidge Farm has a famous and distinctive mark in its Goldfish. The other fish-shaped crackers either are not similar to Pepperidge Farm's Goldfish, or had little success in the marketplace. Nabisco's goldfish cracker is sufficiently similar to Pepperidge Farm's that there is a high likelihood of dilution of the distinctive character of Pepperidge Farm's mark if the Nabisco product is released.

7. The standards employed by the District Court.

Because the same questions will recur if the district court proceeds to trial on the merits, we comment on certain aspects of the trial court's analysis with which we disagree.

First, in considering Pepperidge's claim under the new federal antidilution statute, rather than exploring all factors that might bear on the issue of dilution, the district court limited itself to six items that were identified as the pertinent factors in a concurring opinion in this court considering the question of dilution under the New York statute. See Mead Data, 875 F.2d at 1035 (Sweet, J., concurring). Those factors were: (1) similarity of the marks; (2) similarity of the products covered by the marks; (3) sophistication of consumers; (4) predatory intent; (5) renown of the senior mark; and (6) renown of the junior mark. We pass for the moment the usefulness of these six factors in determining a question of New York law, as to which the federal courts will defer to the analysis of New York courts.

We think it would be a serious mistake at the outset of our consideration of the new federal antidilution statute to limit ourselves to these six factors or to any other putatively definitive list. At this stage, even the best-designed test would likely omit some important factors (and perhaps include others that will at times be irrelevant). Rather, in considering a new federal statutory right, it seems to us that courts would do better to feel their way from case to case, setting forth in each those factors that seem to bear on the resolution of that case, and, only eventually to arrive at a consensus of relevant factors on the basis of this accumulated experience.

When Judge Friendly made his famous list of factors pertinent to an infringement analysis in Polaroid, his opinion drew on generations of analysis of the laws of trademark infringement. And even then his list was non-exclusive; the opinion took pains to insist that in other cases other factors would no doubt emerge as relevant. See Polaroid, 287 F.2d at 495. In contrast the Mead Data concurrence proposes a short closed-end list of six factors, which the district court applied to a new statute that creates a concept previously unknown in the federal law.

The promulgation of such a list has a tendency to quash open-minded, constructive thinking about a new statutory right. We believe it is by far premature for federal courts to declare and close the list of factors that will be deemed pertinent in cases under the new federal act. We therefore decline to adopt the Mead Data factors as a fixed test for dilution under the FTDA.

Furthermore, in our view, the Mead Data list, at least as applied to the federal statute, seems to have several deficiencies. First, it confusingly conflates fame and distinctiveness. The opinion tests the "renown" of the senior mark as one of the six pertinent factors. Renown means fame. In its discussion of the "renown" factor in Mead Data, the opinion suggests that this factor could be satisfied by either fame or distinctiveness. See 1999 WL 47313, at *11 ("The `distinctiveness' of the mark . . . is essentially synonymous for fame and not a term of art."). As discussed above, fame and distinctiveness are altogether different considerations. A mark can be famous without being at all distinctive, as in the cases of American Airlines, American Tobacco Company, British Airways, Federated Department Stores, Allied Stores or the First National Bank of whatever. At the same time a mark can be highly distinctive, meaning arbitrary and fanciful, and yet be completely unknown. In our view, the lumping together of those factors under the name "renown" confuses more than it clarifies.

The Mead Data test furthermore fails to include a number of the factors reviewed above that we believe to be pertinent. Those include actual confusion and likelihood of confusion, shared consumers and geographic isolation, the adjectival quality of the junior use, and the interrelated factors of duration of the junior use, harm to the junior user, and delay by the senior in bringing the action. We make no suggestion that the factors we have focused on exhaust the test of what is pertinent. New fact patterns will inevitably suggest additional pertinent factors.

In short, we think no court should, at least at this early stage, make or confine itself to a closed list of the factors pertinent to the analysis of rights under the new antidilution statute. Upon the final trial on the merits, the district court should not limit itself to consideration of the factors listed in the Mead Data concurrence.

8. Infringement.

Lastly, because this issue will recur if the district court proceeds to trial, we also note our disagreement with an aspect of the court's treatment of the question whether Pepperidge Farm had shown a likelihood of success in proving infringement. The court found that two factors were "problematic for Pepperidge Farm" — its failure to prove actual consumer confusion and insufficient similarity of the two marks to cause consumer confusion. 1999 WL 47313, at *19. We make no comment on the court's treatment of the latter factor. We note only our disagreement with the reasoning that counted the absence of "actual confusion" against Pepperidge Farm.

The presence or absence of actual confusion can be highly effective in showing a high, or a low, likelihood of confusion if there has been ample opportunity for consumer confusion. If consumers have been exposed to two allegedly similar trademarks in the marketplace for an adequate period of time and no actual confusion is detected either by survey or in actual reported instances of confusion, that can be powerful indication that the junior trademark does not cause a meaningful likelihood of confusion. See McGregor-Doniger, Inc. v. Drizzle, Inc., 599 F.2d 1126, 1136 (2d Cir. 1979). In contrast, if numerous instances of consumer confusion have occurred, that suggests a high likelihood of continuing confusion.

On the other hand, if the junior mark has not yet appeared on the market, there has been no opportunity for confusion to manifest itself in the marketplace. It is a logical certainty that no "specific incidents" of actual confusion will have occurred. 1999 WL 47313, at *19. Furthermore, because Nabisco declined to produce sample boxes of its product in response to Pepperidge Farm's discovery demand, Pepperidge Farm was unable to conduct its own consumer surveys that might have shown confusion. In these circumstances, no inference can be drawn from the absence of instances of confusion. In such a case, the "actual confusion" factor simply drops out of the picture because it can have no relevance. See, e.g., Lois Sportswear, U.S.A., Inc. v. Levi-Strauss Co., 799 F.2d 867, 875 (2d Cir. 1986). It cannot, therefore, be a factor against Pepperidge Farm that it was unable to offer proof of "specific incidents of confusion among customers." 1999 WL 47313, at *19.

CONCLUSION

Pepperidge Farm has demonstrated likelihood of success in proving that Nabisco's use of its goldfish-shaped cheddar cheese cracker will dilute Pepperidge Farm's mark in its similar, famous, goldfish-shaped cheddar cheese cracker. We affirm the court's order preliminarily enjoining the distribution of Nabisco's goldfish cracker.