Cumberland Packing Corp. v. Monsanto Co. , 32 F. Supp. 2d 561 ( 1999 )

32 F. Supp. 2d 561 (1999)

CUMBERLAND PACKING CORP. and Stadt Corporation, Plaintiffs,
v.
MONSANTO COMPANY, The Nutrasweet Company, The Nutrasweet Kelco Company, and Olympia Industries, Inc., Defendants.

No. 97 CV 6938.

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

January 12, 1999.

Steinberg & Raskin, P.C., New York City (Martin G. Raskin and Joshua L. Raskin, of counsel), for plaintiffs.

Pennie & Edmonds LLP, New York City (Joseph Diamante, Katharine E. Smith, and Carol M. Wilhelm, of counsel), for defendants.

MEMORANDUM AND ORDER

NICKERSON, District Judge.

Plaintiffs Cumberland Packing Corporation and Stadt Corporation (collectively "plaintiff") brought this action claiming trademark and trade dress infringement, trademark dilution, and false advertisement against defendants Monsanto Company, The NutraSweet Company, The NutraSweet Kelco Company, and Olympia Industries, Inc., (collectively "defendant"), pursuant to 15 U.S.C. §§ 1114, 1125(a), 1125(c), New York General Business Law §§ 349-350, and the New York common law of unfair competition.

Plaintiff is the maker of the sweeteners Sweet'N Low and NatraTaste. Defendant is the maker of the EQUAL, Sweetmate, and NutraSweet brands of sweeteners.

The court has jurisdiction under 28 U.S.C. § 1338(a), giving federal courts jurisdiction over, among other matters, trademark cases, and 28 U.S.C. § 1338(b), providing jurisdiction over a claim of unfair competition when joined with, among other claims, a substantial and related trademark claim.

Plaintiff moved for a preliminary injunction prohibiting defendant from infringing the Sweet'N Low and NatraTaste trade dresses, in violation of section 43(a) of the Lanham Act, 15 U.S.C. § 1125(a); from infringing and diluting the Sweet'N Low trademark, in violation of sections 32, 43(a) and 43(c) of the Lanham Act, 15 U.S.C. §§ 1114, 1125(a), 1125(c); and from falsely advertising its NutraSweet brand sweetener, in violation of section 43(a) of the Lanham Act, 15 U.S.C. § 1125(a). Defendant moved to dismiss plaintiff's false advertising claim, or, in the alternative, for summary judgment on the claim.

The court heard argument on the motions on October 28, 1998.

I

The record shows, in substance, the following. Plaintiff and defendant are long-time competitors in the sweetener industry. Currently the two most commonly used artificial sweetening ingredients are saccharin and aspartame. In terms of sales, plaintiff's Sweet'N Low is the leading saccharin brand sweetener, and defendant's EQUAL the leading aspartame brand sweetener. In the combined sweetener market, Sweet'N Low is the largest volume seller and EQUAL the highest grossing sweetener.

Plaintiff first marketed Sweet'N Low in 1958 and has been selling it at retail in boxes with the same trade dress since 1963. Since 1970 saccharin, a chemical compound about 300 times sweeter than sugar, has been the sole sweetening ingredient in Sweet'N Low.

Aspartame, discovered in 1965 by defendant's then parent company, is a natural protein compound that can be metabolized. It is about 200 times sweeter than sugar. Aspartame was patented in 1972 and approved by the Food and Drug Administration for use in or with foods in 1981 and in beverages in 1983.

Saccharin and aspartame have different tastes, aspartame tasting more like sugar than does saccharin.

In the 1980s, defendant marketed aspartame primarily through a so-called "branded ingredient strategy." Defendant 1) gave aspartame the brand name of NutraSweet, 2) sold aspartame as an ingredient to makers of foods and beverage and required them to display prominently the NutraSweet name and logo, a red and white swirl, and 3) heavily advertised the NutraSweet name and logo to attempt to create a strong association between NutraSweet and aspartame in the minds of consumers buying products containing aspartame. As a result of defendant's promotion much of the public came to recognize the NutraSweet brand name and logo. The chief emphasis in defendant's advertising was that, whereas like saccharin aspartame had no calories, it tasted more like sugar than did saccharin.

In 1982, defendant began selling a so-called "table-top" product containing aspartame as a sweetener. Defendant called this product EQUAL. Table-top sales are made typically to individual consumers, or to restaurants for patrons to add to already prepared foods and beverages. From 1988 to 1996 all EQUAL boxes and packets displayed the brand name NutraSweet and the NutraSweet logo swirl.

Before the introduction of EQUAL, the table-top sweetener market was comprised entirely of sweeteners containing saccharin and sold about 5.1 billion packets per year. EQUAL was an immediate success, selling over 2.09 billion packets in its first full year of sales. But EQUAL sales did not significantly affect sales trends in saccharin sweeteners. The industry soon discovered that saccharin users continued to use saccharin. EQUAL's new customers were mainly people who had previously not used any artificial sweeteners. Because aspartame and saccharin have, as noted, different tastes, the two markets continue to have almost no overlap in customer-base.

With the expiration of defendant's patent on aspartame in December 1992, a number of companies, including plaintiff, jumped into the aspartame market. In 1993, plaintiff introduced its own aspartame-based table-top sweetener called NatraTaste (a name with some similarity to NutraSweet). NatraTaste sold for about half the price of EQUAL. NatraTaste's retail distribution rate has been growing "slowly but steadily" since 1993. As of the beginning of 1998, it was sold in about 70 percent of supermarkets in the United States. In 1997, about 891 million NatraTaste packets were sold. Of the industry leader EQUAL, about 2.09 billion packets were sold.

In 1997 defendant introduced the present table-top sweeteners, NutraSweet and Sweetmate. NutraSweet has aspartame as its significant ingredient and is priced to compete directly against NatraTaste. Sweetmate has saccharin as its sweetening ingredient and is priced to compete with plaintiff's Sweet'N Low. Sweetmate is a repackaged version of defendant's product SweetMate, first sold in the market in 1992.

The market for the NutraSweet and Sweetmate products has expanded rapidly since their introduction. As of July 1998, Sweetmate's retailer authorization was 64 percent; NutraSweet's was above 80 percent, which defendant says is equivalent to full market penetration.

Sales have also soared. From January 1997 to January 1998, 6.3 million packets of Sweetmate were sold. But for the one year period ending in July 1998, sales had increased to 36 million packets. Similarly, during the year 1997 23.7 million packets of NutraSweet were sold. For the six month period ending July 12, 1998, the sales increased to 168 million packets.

The parties disagree somewhat as to the impact that defendant's new products have had on plaintiff's sales. The most recent sales data in the record suggest that the effect has not been significant. Since 1993, Sweet'N Low sales have been declining by 1 to 3 percent per year. The same trend held during 1997 and 1998, after Sweetmate made its appearance. NatraTaste sales have been increasing by about 2 percent per year since its introduction in 1993. Sales of NatraTaste for the eighteen month period ending July 1998 showed no marked change.

All of the table-top products named above are sold in rectangular cardboard boxes containing individually wrapped paper packets. Except in plaintiff's false advertising claim, only the outer appearance of the boxes is at issue.

Plaintiff's NatraTaste box is rectangular with an overall blue coloring. A significant portion of the background uses lighter blue tones. In the front and back panels, the product name, "NatraTaste", appears in large font across the top portion of the box. The name is spelled out in a cursive font with the "N" and the "T" capitalized. The letters are green with white lining and stand out against the blue background. There is a photograph of a coffee cup and a saucer in the center right. Resting on the saucer is a photograph of an individually wrapped paper packet marked "NatraTaste". Another coffee cup to the lower left and saucers to the upper left cast shadows in the background. In the center right is a bright pink burst containing a comparative advertisement stating "Same Sweetener AS EQUAL.... At A Sweeter Price." The top and side panels also say "NatraTaste" in the same style and color, but in slightly smaller font.

Defendant's NutraSweet box is also rectangular but has dimensions differing considerably from those of plaintiff's NatraTaste box. The NutraSweet box is not as wide along the front, and the side panels are thinner. The box is light blue overall, but there are less variations in the blue tones than on the NatraTaste box. At the top of the front panel of the box is the red and white NutraSweet swirl logo, which stands out against the blue background. Below the logo is the trademark "NutraSweet" in big, black block print. The bottom half of the front panel is largely taken up by a picture of a coffee cup on a saucer. An individual packet of NutraSweet is on the left side of the saucer, tilted against the cup so that the NutraSweet swirl logo and the NutraSweet trademark printed on the packet are visible. The back panel is identical to the front panel. The two side panels also display prominently the logo and NutraSweet trademark.

Plaintiff's Sweet'N Low box has a red musical staff slightly undulating across the front panel and set off against a white background. Superimposed on the staff is the Sweet'N Low name in large, dark blue, block capital letters. In the background is a photograph of a glass containing ice tea with lemon. Below is a coffee cup. There is a photograph of a woman's fingers pouring some Sweet'N Low from a pink packet into the coffee. The overall coloring of the front and top panels is red with the exception of a white rectangular space on the left side of the front and top panels, each under an undulating musical staff.

Defendant's Sweetmate box is largely pink, except for the red top one-fifth portion. On the front the name "Sweetmate" appears in blue, cursive font and in thinner and appreciably smaller print than the Sweet'N Low print on the Sweet'N Low box. Although the background has a coffee cup and a glass of ice tea with lemon, they are portrayed in a cartoonish drawing, not a photograph. On the top portion of the front panel of the box is a red, curving wave containing a comparative advertisement: on the left is the statement "Compare'N Save" in white roman type, and on the right is a bright yellow "sunshine burst" with the statement "same sweetener as SWEET'N LOW at a SWEETER PRICE." At the bottom of the right side panel is the statement in small print "Sweet'N Low is a registered trademark of [defendant]."

Defendant previously sold a sweetener called Sweetmate (spelled "SweetMate" from 1992-96 and "Sweetmate" from 1996-97) but it used aspartame, not saccharin, as its main sweetening ingredient. The product was repackaged a number of times between 1992, when it was first introduced, and 1997, when the trade dress version currently at issue began to be used. Plaintiff concedes that Sweetmate's trade dress in 1996, which is not a subject of this or any other suit, looked very similar to the 1997 version. But plaintiff points out the following differences: (1) the present Sweetmate packaging no longer advertises itself as a saccharin-free product and (2) whereas the comparative advertising section in the 1996 version spelled out the Sweet'N Low name in ordinary font, the present version uses the Sweet'N Low trademark itself. Plaintiff says that the Sweet'N Low name is now used in a more prominent way and that this is likely to cause confusion among consumers as to the source of Sweetmate.

II

To obtain a preliminary injunction, a plaintiff must show (1) likelihood of irreparable harm should the injunction be denied and (2) either (a) likely success by plaintiff on the merits or (b) sufficiently serious questions going to the merits and a balance of hardship tipping decidedly in the plaintiff's favor. Jackson Dairy, Inc. v. H.P. Hood & Sons, Inc., 596 F.2d 70, 72 (2d Cir.1979). A preliminary injunction is an extraordinary remedy and not granted routinely.

In a trademark or trade dress case, where there is a showing of a likelihood "that an appreciable number of ordinarily prudent purchasers are likely to be misled, or indeed simply confused, as to the source of the goods in question" the court may find irreparable injury. Tough Traveler, Ltd. v. Outbound Products, 60 F.3d 964, 967 (2d Cir. 1995) (citing cases discussing this standard).

Section 43(a) of the Lanham Act, 15 U.S.C. § 1125(a), provides, in pertinent part, for a private cause of action against any person who:

in connection with any goods ... or any container for goods, uses in commerce 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).

The Lanham Act, enacted to protect both consumers and trademark owners, is designed to ensure that consumers purchasing a product may be confident of getting the brand they think they are getting, and that when trademark owners expend resources promoting their products to consumers their reputation and goodwill will not be misappropriated by pirates. See Mana Products, Inc. v. Columbia Cosmetics Mfg., Inc. 65 F.3d 1063, 1068 (2d Cir.1995).

Section 43(a) applies equally to trademark and trade dress claims. See Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763, 773, 112 S. Ct. 2753, 2760, 120 L. Ed. 2d 615 (1992). A product's trade dress is "its total image ... includ[ing] features such as size, shape, color or color combinations, texture, [or] graphics." Paddington Corp. v. Attiki Importers & Distrib., Inc., 996 F.2d 577, 582 (2d Cir.1993).

In order to prove trade dress infringement, plaintiff must show first, that its dress is inherently distinctive or has acquired a secondary meaning, and second, that the requisite likelihood of confusion as to source or sponsorship exists between plaintiff's dress and defendant's. A trade dress need not have widespread recognition among consumers to be distinctive. It need only serve to identify a product as emanating from a particular, although possibly anonymous, source. See Two Pesos Inc., 505 U.S. at 773-76, 112 S. Ct. at 2760-62.

A

A trademark or dress is generally classified on a spectrum of increasing distinctiveness as (1) generic, (2) descriptive, (3) suggestive, and (4) arbitrary or fanciful. See Abercrombie & Fitch Co. v. Hunting World, Inc., 537 F.2d 4, 9 (2d Cir.1976). A mark is generic if it refers merely to "the genus of which the particular product is a species." Id. A descriptive mark describes the nature of the product. A suggestive mark uses names in a creative way to suggest the nature of the product. It requires purchasers to use "imagination, thought and perception to reach a conclusion as to the nature of the goods." Thompson Medical Co. v. Pfizer, Inc., 753 F.2d 208, 213 (2d Cir.1985). Fanciful or arbitrary marks are those invented solely for trademark usage. They are marks that do not describe or suggest the nature or qualities of the product and are given the greatest protection. See, e.g., Exxon Corp. v. Xoil Energy Resources, Inc., 552 F.Supp. 1008, 1014 (S.D.N.Y.1981) (holding that "Exxon" is an arbitrary trademark).

Trade dresses that are suggestive or arbitrary are inherently distinctive. A descriptive trade dress may be distinctive if it has acquired a "secondary meaning" giving it distinctiveness to the consumer. See Two Pesos, 505 U.S. at 769, 112 S. Ct. at 2757-58. But a specific trade dress must be evaluated "to determine whether it is so distinctive as to point to a single source of origin and thereby be entitled to Lanham Act protection." Fun-Damental Too, Ltd. v. Gemmy Indus. Corp., 111 F.3d 993, 1000 (2d Cir.1997). Finally, a generic trade dress is ineligible for protection.

Plaintiff says that the NatraTaste and Sweet'N Low trade dresses are arbitrary or, at the least, suggestive, and therefore, distinctive. A product's distinctiveness is based on the way the trade dress appears to the observer when viewed as a whole. Individual aspects of the dress may be generic, but as long as the overall combination of the design is likely to identify the particular source of a product, its total impression is inherently distinctive. Jeffrey Milstein, Inc. v. Greger, Lawlor, Roth, Inc., 58 F.3d 27, 32 (2d Cir.1995).

Both the NatraTaste and the Sweet'N Low boxes have elements that are generic when viewed in isolation. Sweeteners commonly feature on their boxes images of a coffee cup, glass of ice tea, or individually wrapped paper packets. Viewed as discrete elements, these do not help distinguish plaintiff's products. Furthermore, in this industry consumers associate the color blue with aspartame sweeteners and the color pink or red with saccharin sweeteners. Thus, the dominant coloring of the boxes serve a functional purpose and does not differentiate them from other products. See Qualitex Co. v. Jacobson Products Co., 514 U.S. 159, 164-65, 115 S. Ct. 1300, 1303-04, 131 L. Ed. 2d 248 (1995); Forschner Group Inc. v. Arrow Trading Co., 124 F.3d 402, 407 (2d Cir.1997).

But the combination of the elements in plaintiff's trade dresses creates a "suggestive" package capable of identifying each product with a particular source. The presence of the NatraTaste and Sweet'N Low trademarks prominently displayed as an integral part of the respective trade dresses is a significant indication of their distinctiveness.

B

The existence of the requisite likelihood of confusion as to source or sponsorship between two trade marks is determined by applying the factors described in Polaroid Corp. v. Polarad Electronics Corp., 287 F.2d 492, 495 (2d Cir.1961). Those factors are: (1) the strength of the prior owner's mark; (2) the similarity between the marks; (3) the competitive proximity of the products; (4) the likelihood that the prior user will bridge the gap; (5) actual confusion as to source or sponsorship; (6) the defendant's good faith; (7) the quality of defendant's product; and (8) the sophistication of the buyers. These factors are not exclusive and the test makes no single factor dispositive. The court "weigh[s] each factor in the context of the others to determine if, on balance, a likelihood of confusion exists." W.W.W. Pharm. Co. v. Gillette Co., 984 F.2d 567, 570-71 (2d Cir.1993). The same considerations are applied to determine likelihood of confusion as to source or sponsorship between trade dresses. See, e.g., Bristol-Myers Squibb Co. v. McNeil-P.P.C., Inc., 973 F.2d 1033, 1043 (2d Cir.1992) (applying Polaroid factors to assess likelihood of confusion in a trade dress case).

In this case where the products are in direct competition with each other, the third Polaroid factor, proximity of the products, is obvious. The fourth, likelihood that the gap will be bridged, is not relevant. See Bristol-Myers Squibb, 973 F.2d at 1044. The seventh factor, the quality of defendant's product, is not in issue. Plaintiff does not say defendant's products are of inferior quality. Plaintiff has not submitted evidence as to the eighth factor, the sophistication of the buyers. But the nature and low price of the products suggest that consumers typically will not spend an extraordinary amount of time evaluating them before making a purchase. The court analyzes the remaining factors in turn.

(i) The strength of plaintiff's NatraTaste and Sweet'N Low trade dresses

A trade dress's strength is measured by both its conceptual and commercial strength. See W.W.W. Pharm. Co., 984 F.2d at 572.

The conceptual strength is determined by the degree to which a trade dress is distinctive. Id. NatraTaste and Sweet'N Low have "suggestive" trade dresses and, as noted above, they are distinctive.

The commercial strength of a trade dress is measured by the degree to which it is distinctive in the marketplace. The commercial success of a product reinforces the dress's strength. See Charles of the Ritz Group Ltd. v. Quality King Distributors, Inc., 832 F.2d 1317, 1321 (2d Cir.1987).

The NatraTaste brand is relatively new, and plaintiff has not submitted any evidence indicating how much recognition the product commands in the market. According to defendant, NatraTaste has low brand recognition. But the NatraTaste mark, while somewhat similar in looks and sound to the older NutraSweet mark, is distinctive and the product has enjoyed considerable commercial success since its introduction in 1993. In terms of aspartame table-top sales volume, it is second only to the industry leader, EQUAL. Among value-brand aspartame sweeteners NatraTaste is the current leader. Its dress has considerable commercial strength.

Sweet'N Low has been marketed in the same trade dress since 1963 and is the top saccharin sweetener in terms of sales. Because of extensive advertising over the past years, it has achieved national brand recognition. The commercial strength of Sweet'N Low's trade dress is strong.

(ii) Comparison of the trade dresses of NatraTaste and NutraSweet

The count considers whether the similarities in the NatraTaste and NutraSweet trade dresses are likely to confuse an appreciable number of ordinarily prudent purchasers as to the source of the products.

Plaintiff says that the (1) blue lighting and the "pronounced shadowing and gradation of the blue shades" make the NatraTaste and NutraSweet boxes look "strikingly" similar; (2) both boxes show a cup of black coffee (although NutraSweet's coffee is a dark brown); (3) both have a packet resting on a saucer; and (4) both have a cup and saucer "resting on an undefined surface."

The relevant inquiry is not how many details the two boxes share. It is whether the similarities create the same general, overall impression such as to make likely the requisite confusion among the appropriate purchasers.

It seems unlikely that a reasonably prudent purchaser would find the overall image of the two boxes confusingly similar. They both display their different product names prominently in front of the box. The largest print on the NatraTaste box is its name, "NatraTaste." It is on the top portion of the box in large, light blue or green print with white-lining. The curvy-lettering jumps out because of the color contrasts.

The NutraSweet box also has its product name in large print on the front of the box. The thick, block letters are in black and give off a very different impression from NatraTaste's curvy, light-colored letters. The NutraSweet box conspicuously features the NutraSweet logo, the red and white swirl. The NutraSweet name and logo have a high recognition rate among consumers and are powerful source identifiers. The combination of the NutraSweet name, presented in this particular color and font, coupled with the red and white swirl logo, has appeared since the early 1980s on numerous products using NutraSweet as an ingredient. The NatraTaste box displays no similar looking logo.

The conspicuous use of a well-known logo and product name sufficiently differentiates defendant's trade dress from NatraTaste's. See Merriam-Webster, Inc. v. Random House, Inc., 35 F.3d 65, 71 (2d Cir.1994) ("the impressions conveyed by the combinations of angular and curving logos, and by quite different names and some different colors [used in the respective letterings of the trade names] ... preclude a finding of similarity."); Bristol-Myers Squibb Co. v. McNeil-P.P.C., Inc., 973 F.2d 1033, 1045-46 (2d Cir.1992) (the district court's finding that the two trade dresses are similar was clearly erroneous because "the prominence of the trade names on the two packages weighs heavily against a finding of consumer confusion resulting from the overall look of the packaging.").

The other noticeable distinctions between the two boxes are that the NatraTaste box is significantly wider and longer than the NutraSweet box and has a bright pink burst on the surface, something the NutraSweet box does not have.

The court finds that the total images of the NatraTaste and NutraSweet trade dresses are so different that there is no significant likelihood prudent purchasers will be confused as to source by similarities in their trade dresses.

(iii) Comparison of the trade dresses of Sweet'N Low and Sweetmate

The court concludes also that it is highly unlikely that consumers will confuse Sweet'N Low and Sweetmate due to a similarity in their trade dresses. Plaintiff points out various individual features that the two dresses share in common. But their overall image is not at all similar. First, each box prominently displays the names of the products. Sweet'N Low and Sweetmate, although they both contain the word "sweet," sound and look different. The Sweet'N Low box is largely red on its front and top panels. The Sweet'N Low mark uses block lettering with sharp edges, and is in all capital letters, while defendant's Sweetmate capitalizes only the letter "S" and uses cursive lettering that is significantly thinner.

Moreover, the Sweet'N Low name is superimposed on an undulating musical staff, and is set off against a white background. This combination — the name Sweet'N Low in dark blue and in bold font superimposed along a musical staff — is the well-known Sweet'N Low trademark. Plaintiff's trademark registration indicates that the mark was first used in commerce in 1958, and according to plaintiff, has since gained national recognition due to sustained advertising and numerous promotional efforts.

The Sweetmate box is largely pink and has little resemblance to the overall impression created by the Sweet'N Low trademark and dress. The Sweetmate name is spelled out along a curved but not undulating line and there is no musical staff. In addition, the name "Sweetmate" appears under a comparative advertisement reading, "Compare `N Save!". This is a phrase commonly found among value brands or less established private labels to invite comparison to the leading brand. Of course, Sweet'N Low, the leading brand, has no such comparative advertisement on its box.

Significantly, whereas the coffee cup, the pink packet, and glass of ice-tea in the Sweet'N Low box is represented by a stark-looking photograph, the Sweetmate box has cartoon-like drawings of these items. The drawings are tilted at "playful angles" to create a "fun and whimsical" effect. The scattering of leaves and berries adds to this effect. The coffee in the Sweetmate box has a brown hue, in contrast to the Sweet'N Low black coffee. The contrast in the overall image of plaintiff's and defendant's products is striking.

Additional differences distinguishing the general appearance of the boxes are that the Sweet'N Low box is somewhat less than twice as thick as the Sweetmate box, and about a third of the Sweet'N Low box is white, whereas the Sweetmate box is mostly pink with some red on the front.

In sum, the overall images of the two boxes are so different that it is highly unlikely that any individual similarities they share will cause confusion among an appreciable number of purchasers.

(iv) alleged actual confusion

To establish likelihood of consumer confusion, plaintiff relies on the opinions of its expert, Dr. Michael Rappeport, interpreting three consumer surveys he designed. To be probative on the issue of confusion, a survey must "have been fairly prepared and its results directed to the relevant issues." Sterling Drug, Inc. v. Bayer AG, 14 F.3d 733, 741 (2d Cir.1994). In Toys "R" Us, Inc. v. Canarsie Kiddie Shop, Inc., 559 F. Supp. 1189 (E.D.N.Y.1983), Judge I. Leo Glasser listed seven standards that surveys should meet to be trustworthy. They are that (1) the "universe" or product market is properly defined; (2) a representative sample of that universe is selected; (3) the questions to be asked of the interviewees are framed in a clear, precise and non-leading manner; (4) sound interview procedures are followed by competent interviewers with no knowledge of the litigation or the purpose for which the survey was conducted; (5) the data gathered is accurately reported; (6) the data is analyzed in accordance with accepted statistical principles; and (7) the objectivity of the entire process is assured.

Dr. Rappeport conducted three consumer surveys in January 1998 to test for consumer confusion arising from similarities in the NatraTaste and NutraSweet trade dresses, and the Sweet'N Low and Sweetmate trade dresses. He claimed to have found a high likelihood of confusion as to source in both comparisons.

The three studies were:

1) NatraTaste-NutraSweet "Two Room, Array Methodology" Study: Respondents were shown a box of NatraTaste in one room and then escorted into a second room in which five other boxes of sweeteners, including NutraSweet, were displayed in a row on a table. Respondents were then asked,

Q1. Do you think any of these sugar substitutes are made by the same company as the first box you saw?

Those who answered "yes" were asked,

Q2. Which one or more of the sugar substitutes do you think are made by the same company as the first box you saw?

Finally, respondents were asked to explain their answer:

Q3. Why do you say that?

Result: Dr. Rappeport found that 31 percent of those surveyed confused NutraSweet for NatraTaste due to similarities in their trade dresses.

2) NatraTaste-NutraSweet "One Room, Sequential Methodology" Study: Respondents were shown the same products as in the first study, but one by one in the same room instead of in two separate rooms, and asked the same above questions.

Result: Dr. Rappeport found that 14 percent of those surveyed confused NutraSweet for NatraTaste due to similarities in their trade dresses.

3) Sweet'N Low-Sweetmate Survey: Respondents were shown a Sweetmate box from a distance of three feet, and then the box was removed from view. Afterwards, respondents were asked:

Q1. What is the name of the product you saw?

Q2. Please tell me everything that you recall about the box you just saw.

Q3. If you know, what other brands, if any, are made by the same company that made the first product you saw?

Result: Dr. Rappeport found that 30 percent of those surveyed thought that Sweetmate was Sweet'N Low or that it emanated from the same source as Sweet'N Low due to a combination of i) similarities in the trade dresses and ii) the appearance of the "Sweet'N Low" name in the comparative advertisement on the Sweetmate box.

Reviewing Dr. Rappeport's reports, his testimony at depositions, and the survey methodologies he used in the three studies against the seven Toys "R" Us standards, the court finds the studies flawed and the survey results untrustworthy and unreliable.

First, Dr. Rappeport selected a universe for the surveys that was over-inclusive. The survey universe for all three studies was defined as "users or buyers of sugar substitutes within the past six months." To be meaningful, a survey must "rely upon responses by potential consumers of the products in question." Universal City Studios, Inc. v. Nintendo Co., 746 F.2d 112, 118 (2d Cir.1984). The relevant confusion in a trademark or dress case is that which affects "the purchasing and selling of the goods or services in question." Lang v. Retirement Living Publishing Co., 949 F.2d 576, 582-3 (2d Cir.1991). Thus, the survey universe must be composed of people having a present purchasing interest in the product being surveyed. See Universal City Studios, 746 F.2d at 118 (universe improper because it included only former purchasers of Donkey Kong); American Footwear Corp. v. General Footwear Co., 609 F.2d 655, 661 n. 4 (2d Cir.1979) (universe improper because although survey participants were former purchasers of hiking boots, they did not necessarily have any present purchasing interest).

Dr. Rappeport says that past users are good proxies for future purchasers. This may be true for some products. But in the case of sweeteners, the assumption is, to say the least, questionable. Plaintiff's pink-packets and the defendant's blue-packets are available at most diners and coffee shops and many restaurants. A high percentage of past users of sugar-substitutes may never have purchased sugar substitutes and have no intention to do so in the future. The survey did not distinguish between past users who intend to purchase sugar-substitutes in the future and those who do not.

It was also improper to use all past "sugar-substitute" users without distinguishing between saccharin and aspartame users. As noted earlier, there is little overlap between saccharin and aspartame users. Despite EQUAL's instant success upon introduction in 1982, Sweet'N Low sales remained relatively stable because demand for EQUAL came almost entirely from people who had previously never used sweeteners. Thus, past users of saccharin are unlikely to be potential purchasers of aspartame sweeteners and vice versa.

In addition, as will be discussed hereafter, Dr. Rappeport's choice of so-called "controls" in the first two surveys was inappropriate. Given the unreliability of his controls, a sounder analytical approach was to review the reasons respondents gave for thinking that NutraSweet was made by the same company that made NatraTaste and to disregard the people who said they were confused for reasons unrelated to trade dress. The court conducted such an analysis and found only minimal levels of confusion. The third survey was flawed because, among other reasons, it failed to resemble actual market conditions. The leading nature of the questions asked and the lack of instructions against guessing flawed all three surveys.

(a) The First NatraTaste-NutraSweet Survey

In the first survey, the "Two room, Array Methodology" study, each respondent was shown the NatraTaste box in one room, and then taken to a second room and shown an array of five boxes. The five products were Sweet Servings (Box F), Equal (Box I), Sweet One (Box M), Sweet Thing (Box R), (collectively "the controls"), and NutraSweet (Box O) (the "test box"). The order in which the five boxes were displayed was rotated.

Each respondent was instructed to stand in front of the five boxes. They could see only the front panel of the boxes. Question one asked whether they thought any of these boxes were made by the same company that made the box they saw in the first room, namely the NatraTaste box. Those who answered "yes" were then asked, "Which one or more of these sugar substitutes do you think are made by the same company as the first box you saw?", and the open-ended question, "Why do you say that?"

Dr. Rappeport reported that 43 percent of the 120 people surveyed picked NutraSweet. He also reported that 13 percent, 18 percent, 7 percent, and 8 percent, respectively, thought Sweet Servings, Equal, Sweet One, and Sweet Thing, was a product made by the same company as made NatraTaste. Dr. Rappeport did not analyze the answers given to the third question where respondents explained their choices. Instead, he took the average of the confusion in the "four control stimuli," ((13% + 18% + 7% + 8%) ÷ 4), which equals 12 percent, and subtracted it from 43 percent, to arrive at a confusion level of 31 percent. Dr. Rappeport's opinion was that the 31 percent represents confusion caused solely by similarities in NatraTaste's and NutraSweet's overall trade dresses.

To assess the reliability of Dr. Rappeport's choice of controls and the conclusions he drew from the data, the court analyzed the reasons respondents gave for thinking NutraSweet was made by the same company as made NatraTaste. Dr. Rappeport surveyed 120 people. One hundred and seventeen of the 120 surveys were submitted to the court. A re-tabulation of the results showed that thirty-five of the 117 survey respondents (30%) thought that NutraSweet, but none of the controls, emanated from the same source as NatraTaste. The following table shows the reasons why they picked NutraSweet.

Why do you say that [NutraSweet is made by the same company that made NatraTaste]?
Quotes or Paraphrases of Reasons Given, Organized by Category
Color similar             Design similar           Name similar                Other
1 similar color             similar shape                                        "I am familar with
this product"
2 "same color blue"         "because   box                                       "also shows coffee
doesn't look cheap                                   on the front"
like the others"
3 "coloring of the                                                               "writing of 100
blue is same"                                                                  packets is same"
4 "light blue color         boxes look similar
background"
5 "color of box is
same"
6 "same color box"
7 "color of box"
8 "same coloring"                                                                "same pitcher"
9 "just because of the
color of boxes"
10 "Boxes C + O have                                                              "same size I think"
same color"
11 "looks more like the
first box, in color"
12 "color of box/same                                 "names Nutrasweet
blue shading"                                      and Natrataste are
very similar"
13 "color of box is                                   "name is very similar"
same"
14 "both blue boxes"                                  "both say nutra"
15 "color of box"                                     "because of the nutra
on box"
16 "same coloring"                                    "same print as other
box I saw"
17 "just the coloring"                                "saw the words Nutra
Sweet on both
boxes"
18 "they both look similar                            "Both contain Nutrasweet"
in color"
19 "it's blue also"                                   "Its got the word
nutra-sweet"
20 "same color"                                       "same ... label"
21                           "similar in packaging"   "names are similar"
22                           "looks very similar"     "almost the same
name"
23                                                    "Because it says nutra
sweet"
24                                                    "it seems to have
the same name"
25                           "box is similar"         "thought it was the
same name"
26                                                    "name Nutrasweet
is similar to Natrataste"

27                           "Because it looks
similar to the other
packaging"
28 "coloring and everything  "because boxes are
is the same"              alike in everyway"
29                           "same pattern on                                     "has same ingredients"
box" "box looks
same"
30                           "just the appearance
of the box"
31                           both look similar                                    "Easy to read; nice
looking box"
32                           "because box is                                      "logos are same"
same"
33                                                                                "this is a popular
brand so they might
have more than one
kind of box"
34                                                                                "because NutraSweet
makes more
brands than other
companies"
35                                                                                "because it has less
sodium"

Five of the thirty-five people picked NutraSweet solely because they thought the coloring of the boxes was the same or similar. As already noted, both boxes are predominantly blue. Blue signifies aspartame products in this industry and is not a protected element of NatraTaste's trade dress. In addition, five people (respondents 1, 2, 3, 8, and 10) cited color in conjunction with reasons not relevant to trade dress. All ten of these responses should be disregarded.

Of the remaining twenty-five people, twelve cited a similarity in the names, or similarities in the names plus the color of the products, as their bases for thinking the same company made NatraTaste and NutraSweet. Although the names NatraTaste and NutraSweet are somewhat similar, plaintiff does not contend they are confusingly so. These twelve responses should also be disregarded.

Responses thirty-three and thirty-four do not show any confusion based on similarities in the trade dresses of NatraTaste and NutraSweet. These persons essentially guessed NutraSweet because they recognized that as the only product that was a famous brand. Response thirty-five also should be disregarded as irrelevant to a trade dress case. Finally, response thirty-two shows confusion of NatraTaste for NutraSweet, not the other way around, since NatraTaste does not have a logo. In other words, like the respondents who picked NutraSweet because they thought they had seen the word "nutra" on the first box (which was in fact the NatraTaste box), this respondent "confused" the two products because he thought he had seen the NutraSweet logo on the NatraTaste box.

Dr. Rappeport testified at his deposition that because of his "controls" there was no need to analyze the reasons why people picked NutraSweet. He explained that people who would confuse NutraSweet for NatraTaste for reasons unrelated to trade dress would also confuse one or more of the other control boxes with NatraTaste. By subtracting the "average" of the confusion in the "four control stimuli" he claimed he would be able to net out the "noise."

In a test of a causal proposition the appropriate use of controls is crucial. Here the proposition tested is whether consumers are misled by similarities in the trade dresses into believing that NatraTaste and NutraSweet emanate from the same source. Proper controls will approximate "background noise," the amount of confusion existing due to reasons unrelated to similarities in trade dresses. Those controls must sufficiently account for factors legally irrelevant to the requisite confusion, so that confusion "caused by" similarities in the trade dresses can be isolated.

The above table showed that the two most common reasons people gave for thinking NutraSweet was made by the same company that made NatraTaste was the overall blue coloring of the boxes and the similarities in the names. Since neither of these reasons are relevant to plaintiff's trade dress case under the Lanham Act, appropriate controls should have been designed to net out confusion based on these two variables.

For reasons discussed below, Dr. Rappeport's controls could not possibly have served this end. Thus, his statement is not credible that he did not analyze the reasons respondents gave for choosing NutraSweet because he believed his controls would measure fairly the amount of confusion due to matters unrelated to trade dress.

The four controls chosen by Dr. Rappeport had the following characteristics. Sweet One has a predominantly white background with its most prominent labels in bright yellow. The front panel of the EQUAL box is mostly white, red, and green. While the background of Sweet Thing is blue, there is no one dominant color — the box has a hodge podge of blue, red, and yellow, as well as some green, white, black, and brown. Sweet Servings has on its front panel a picture of a tall glass filled with colorful fruit and its background is a darker, indigo blue.

Moreover, not one of the controls has a name sounding remotely like NatraTaste. None of these controls would have picked up people who said NutraSweet is made by the same company that made NatraTaste for the reasons that the overall packaging is a lighter blue or the word "natra" looks and sounds like "nutra."

During a deposition held on May 21, 1998, Dr. Rappeport stated, "[w]e did ask people why they said these two boxes came from the same place. And the answers did indicate that the reason they saw [sic] was the overall look, and not the name." He added, "I was concerned whether [respondents] were confused over the name.... It is my opinion that no significant fraction of these people did this [i.e., answered NutraSweet] on the basis of names." The actual answers refute this response.

In fact, over 40 percent of those who chose only NutraSweet among the five choices cited similarities in the name as a reason for their answer. A combination of similarities in color and name appeared to be the single, most common cause of confusion among this group.

Given the inadequacy of the controls, one cannot determine from the data the extent of the relevant type of confusion by indirectly approximating the background noise. An alternative is to analyze the reasons people gave for choosing NutraSweet. Of course, this method is not perfect. Respondents may not identify accurately the source of their confusion. Indeed, to the extent that pre-existing impressions or guessing influenced people's choices, the level of confusion is likely to be over-estimated. But to focus on the respondents' answers is the only reasonable alternative realistically available to the court.

The court concludes that analysis of the responses in a light most favorable to plaintiff's case shows that nine of the thirty-five people who chose only NutraSweet "confused" NutraSweet with NatraTaste for trade dress reasons.

Seventeen additional people (14%) chose NutraSweet as emanating from the same source as NatraTaste but these people thought also one or more of the control boxes were made by the makers of NatraTaste. Dr. Rappeport counted the seventeen, net of his measure for noise, in deriving his final figure for trade dress confusion. As his controls were flawed, defendant's expert, Dr. Jacob Jacoby, said it makes more sense to eliminate entirely these seventeen. They represent guesses or answers based on reasons unrelated to trade dress confusion.

The remaining sixty-five people (56%) said either that none of the five boxes was made by the same company as made NatraTaste or that one or more of the controls, but not NutraSweet, was made by the same company as NatraTaste.

In short, based on a study of the reasons the respondents gave for their choices, at most nine of 117 people confused NutraSweet with NatraTaste for reasons related to trade dress.

Even this figure is likely to be inflated because the survey did not resemble actual market conditions and Dr. Rappeport asked leading questions without giving instructions against guessing. Respondents were made to look at NatraTaste in one room and then go to another room to look at the NutraSweet box and the four control boxes. In grocery stores NatraTaste and NutraSweet are displayed side by side or very close to each other. In the market a shopper would be able to eye the two products more or less simultaneously. Then, given the prominence of the different looking trademarks on the respective trade dresses, and the presence of a well-known logo on only one of the boxes, it is unlikely that confusion would arise.

Moreover, respondents were not explicitly instructed against guessing. Although the written forms have the "I don't know" choice printed on them, the surveyors asked the questions orally and filled out the forms. A respondent had to volunteer "I don't know." When the average person is taken from one room to another, shown an array of five boxes, and asked whether she thinks one or more of those boxes are made by the same company as the first box she saw, she has a motive to figure out the purpose of the survey and will feel some pressure to "answer correctly." The position in which the respondent is placed make it important explicitly to instruct the respondents not to guess.

(b) The Second NatraTaste-NutraSweet Survey

Dr. Rappeport found a 14 percent confusion rate, net of noise, in the second survey using a sequential method. This number is inflated and unreliable for much the same reasons discussed above. In the second survey, the respondents were shown the NatraTaste box, and immediately thereafter, each of the five other boxes (the NutraSweet box and the four controls) in sequence. The four controls were the same as those in the first study, and the order in which the boxes were shown was rotated.

One hundred and forty people were surveyed. The court received 138 of the original surveys and tabulated the results. Eighty-six people (62%) said that (1) none of the boxes emanated from the same source as NatraTaste or (2) one or more of the controls and not NutraSweet did so. Thirty-five people (25%) said they thought NutraSweet emanated from the same source as NatraTaste but also thought one or more of the other boxes was made by the makers of NatraTaste. For the reasons discussed above, these respondents were guessing or saying they were confused for reasons unrelated to trade dress, and their responses should be disregarded.

Seventeen people (13%) said they thought only NutraSweet was made by the same company as NatraTaste. The court analyzed the answers these seventeen respondents gave. The results showed the following:

Color similar       Design similar                     Name similar            Other
Why do you say that [NutraSweet is made by the same company that made NatraTaste]?
Quotes or Paraphrases of Reasons Given, Organized by Category
Color similar       Design similar                     Name similar            Other
1 "color is the same"                                   "You might also
mistaken the first
box for this because
of the name"
2                                                       "This says Nutrasweet
— the first one
said nutra sweet"

3 "Box color is similar"                                "The brand name
seems to be from
the same company."
4                                                       "I think the first
box said Nutrasweet
and not sacrin."
5                                                       "because the first
one was NutraSweet
and so is
this."
6                                                       "both began with
nutra — same name,
same company"
7                                                                                "this is the first box
that the Nutrasweet
company
came out with, and
the first box I saw
may be the new design."
8 "because of the
same color scheme"
9 "same light color;
background is same
soft look"
10 "color — less images
(simple)"
11 "because of the colors"      because of "artistic
set-up"
12 "the colors co-incide"       "very similar in
concept and advertising
idea"
13 "same color"                 "same idea on the
front"
14                              "same design on
box"
15                              "looks similar"           "because of the
name"
16                                                                                "because it didn't
say anything negative
about the other
brands"
17 "similar in color"           similar in style

Nine of the seventeen people said they thought NutraSweet was made by the same company that made NatraTaste because of similarities in color, the name, or the combination of the two factors. Moreover, at least four of the nine (respondents 2, 4, 5, and 6) and one additional person (respondent 7) confused NatraTaste for NutraSweet, not the other way around. These responses suggest that NatraTaste is benefitting from the goodwill built up in the NutraSweet name. This is hardly surprising. While the NutraSweet table-top product was introduced to the market later than NatraTaste, NutraSweet is the senior mark and by far the better known brand.

In addition, response sixteen should be disregarded. It has nothing to do with trade dress confusion. In short, no more than six people out of 138 people confused NutraSweet for NatraTaste because of similarities in their trade dress.

(c) Sweet'N Low-Sweetmate Survey

Dr. Rappeport conducted a third survey to assess the likelihood of confusion between the trade dresses of Sweet'N Low and Sweetmate. Plaintiff does not claim that the name Sweetmate infringes the Sweet'N Low trademark. Respondents were shown a Sweetmate package from a distance of three feet and not allowed to handle the box or to view it from a closer distance. Then the package was removed from view. Respondents were then asked three questions: (1) "What is the name of the product you saw?"; (2) "Tell me everything you recall about the box you just saw."; and (3) "If you know, what other brands, if any, are made by the same company that made the first product you saw?"

Of the 100 people surveyed, twelve people in answer to question one said the name of the product they saw was Sweet'N Low. In addition, twenty-two people answered Sweet'N Low to the third question. Dr. Rappeport added the responses to questions one and three, (12 + 22), discounted it by seven, his measure for noise, and arrived at the number thirty. He concluded that 30 percent of those surveyed confused Sweetmate for Sweet'N Low because of a combination of (1) similarities in their trade dresses and (2) the use of the Sweet'N Low name in the comparative advertisement on the Sweetmate box.

Dr. Rappeport's conclusion is unjustified. Question one asked simply "the name of the product you saw?" There were no instructions against guessing. The Sweetmate box was tucked away from the view of the respondents. Those who had forgotten the name were likely to guess, and when people guess, they tend to guess the name of the product they have previously heard of, in this case Sweet'N Low. In an actual market situation, the product would not disappear from the consumer's eye just as he or she is about to make a purchase. Given the prominence of the name Sweetmate on its trade dress, it is highly unlikely that customers would think that the name of the product is "Sweet'N Low." Question one was essentially a memory test. The question elicited very little or no useful information for a trade dress case.

During the Sweet'N Low-Sweetmate survey, Dr. Rappeport conducted what he called a "control" study. He had fifty respondents look at a NatraTaste box from three feet away. The box was then taken away and respondents were asked to recollect the name on the box. Twelve people (24%) remembered the name to be "NutraSweet," "NatraSweet," or "NatureSweet." Fourteen other people (28%) thought the name had the word "nutra" in it. Dr. Rappeport claims that the results of this control study tend to prove that the respondents who in the Sweet'N Low-Sweetmate study said they saw Sweet'N Low after seeing the Sweetmate box were not guessing but just confused by the trade dress.

The reason he gives for such a claim is that, after seeing the NatraTaste box, and being asked, "What is the name of the product you saw?", no one guessed EQUAL, even though EQUAL is just as famous a brand as Sweet'N Low.

Why would anyone think the name they saw on the NatraTaste box was EQUAL, which neither sounds nor looks like NatraTaste? Dr. Rappeport fails to point out that more than half mistook the name, NatraTaste, for the well-known name, NutraSweet, or recalled the name to be "nutra something". The control study has no relevance to the Sweet'N Low-Sweetmate issue.

Moreover, the design of the Sweet'N Low-Sweetmate survey was also untrustworthy because it made no attempt to resemble an actual market situation. Sweetmate was shown in isolation and not compared to Sweet'N Low or to any other brands. In a grocery store consumers would have the benefit of such comparisons. The trade dresses of four other saccharin brands submitted by the defendant to the court all have a predominantly pink or red background. All have the word "sweet" in their names. All have a picture of a coffee cup. All mention the "Sweet'N Low" name in the context of comparative advertising. Dr. Rappeport showed none of these boxes to the respondents.

In short, in the Sweet'N Low-Sweetmate test he conducted no surveys comparable to the two NatraTaste-NutraSweet surveys. His test made it certain that the Sweet'N Low-Sweetmate test would not even approximate market conditions.

He claimed that his instruction to keep the respondents three feet away from the box did not create a problem because a consumer initially "confused" from a distance of three feet is still confused even if the confusion is cleared up when the consumer comes closer to the product to buy it. But the Lanham Act forbids the creation of confusion affecting purchasing decisions.

A consumer would have to draw nearer than three feet to purchase the product. A prudent purchaser would then see the name of the product as Sweetmate, not Sweet'N Low, and would observe the name Sweet'N Low mentioned as part of a conspicuous comparative advertisement saying "Compare'N Save." The prudent purchaser would also see other products mentioning Sweet'N Low by name to contrast their products with the leading brand.

The survey evidence of significant actual confusion between Sweetmate and Sweet'N Low among consumers in the marketplace is unpersuasive.

(v) defendant's intent

Plaintiff claims that defendant in adopting its trade dresses intended to capitalize on plaintiff's reputation and goodwill.

Where a party acts in bad faith and intentionally copies a trademark or trade dress, the court can draw an inference that there is likelihood of confusion. See Paddington Corp. v. Attiki Importers & Distrib., Inc., 996 F.2d 577, 586-87 (2d Cir.1993). But defendant's knowledge of a plaintiff's trade dress need not give rise to an inference of bad faith. "[A]doption of a trademark with actual knowledge of ... a very similar mark may be consistent with good faith." Lang v. Retirement Living Publishing Co., 949 F.2d 576, 584 (2d Cir.1991). Even if defendant had copied plaintiff's trade dress, "bad faith should not be inferred simply from the act of copying." Fun-Damental Too, Ltd. v. Gemmy Indus. Corp., 111 F.3d 993, 1005 (2d Cir.1997) "[C]opying in order to market a functionally equivalent alternative product might well benefit consumers, which is one of the aims of the Lanham Act." Id. at 1005 (citing Andrew C. Finch, When Imitation is the Sincerest Form of Flattery: Private Label Products and the Role of Intention in Determining Trade Dress Infringement, Comment, 63 U.Chi.L.Rev. 1243, 1255 (1996)).

Plaintiff has offered no substantial evidence of defendant's bad faith other than the asserted similarities in the trade dresses. Those similarities are by no means striking. Defendant was entitled to position its products in direct competition with those of plaintiff's. Of course, defendant would wish to target NatraTaste and Sweet'N Low. One purpose of the Lanham Act was to enhance, not stifle, competition. It would have been pointless for plaintiff to compete against products posing no competitive threat.

The record shows that defendant had legitimate reasons for adopting the trade dresses it did. In the case of NutraSweet, it created a package that could reap the benefits of the good-will built around the powerful and respected NutraSweet name and logo. It was not obliged to abandon this mark when selling a table-top aspartame sweetener. Defendant used the color blue according to the established industry norm for aspartame products and highlighted the logo and the NutraSweet name by using different gradations of blue.

Plaintiff points to a statement made by defendant, instructing its design team to make the "front of box and back of box same facing — just like NatraTaste," as proof that defendant copied plaintiff's trade dress in bad faith. But the sentence preceding the quoted phrase reads, "large swirl; separate `the original' from no calorie sweetener so that it becomes closer alignment w/ NS brand." Obviously, "NS" refers to "NutraSweet." Plainly, defendant was instructing the design team to make the front and back of the NutraSweet box look the same, just as the front and back of the NatraTaste box uses the same "facing."

For defendant to make its front and back facings the same does not suggest bad faith. That idea is not a protected element of NatraTaste's trade dress. The fact that the overall appearance of the NutraSweet box is not similar to that of the NatraTaste box refutes the claim of bad faith.

In the case of Sweetmate, plaintiff urges that defendant was guilty of bad faith because it advised its design team that "the new Sweetmate will go head-on for the Sweet'N Low customer," that Sweetmate would be priced lower than Sweet'N Low, and that the color pink and Sweet'N Low red "may be introduced to put Sweetmate in combat with SNL more head-on." Defendant had a right to compete — the Lanham Act encouraged it to do so — "head on" with plaintiff's Sweet'N Low by selling a comparable product at a lower price. Defendant emphasized that its product was not that of plaintiff's Sweet'N Low by adopting language asking the consumer to "Compare'N Save."

For reasons already stated, the trade dresses by no means create confusion as to the sources of the goods. Plaintiff has failed to establish bad faith.

(vi) conclusion

After considering the pertinent factors and weighing each of them, the court concludes that plaintiff has failed to establish the requisite likelihood of confusion.

The prominent display of the trade names in conjunction with other contrasts in the trade dresses create an overall impression so different that such confusion is unlikely. The conspicuous display of a nationally recognized brand name and logo makes it improbable that consumers will associate the NutraSweet product with a lesser known brand called NatraTaste. Similarly, the Sweet'N Low name superimposed on a musical staff is a nationally recognized trademark. Consumers are unlikely to confuse Sweetmate with it. The Sweetmate box displays a different product name in a different font, against a different background, with no musical staff, and under a phrase that reads "Compare'N Save".

Plaintiff's motion preliminarily to enjoin defendant from its allegedly infringing use of the NatraTaste and Sweet'N Low trade dresses will be denied. Since there is no likelihood of confusion, plaintiff has not established likelihood of irreparable harm by the denial of an injunction. The court need not address the issue of laches.

III

Plaintiff also contends that defendant's use of the Sweet'N Low trademark on the Sweetmate box in the yellow "sunshine burst" constitutes infringement, in violation of 15 U.S.C. § 1114, and a false designation of origin, in violation of 15 U.S.C. § 1125. Plaintiff seeks preliminarily to enjoin this use.

Section 32 of the Lanham Act provides a private cause of action against any person who:

uses in commerce ... a registered mark in connection with the sale ... of any goods or services on or in connection with which such use is likely to cause confusion, or to cause mistake, or to deceive....

15 U.S.C. § 1114.

Plaintiff's Sweet'N Low is a registered trademark.

A truthful reference to the trademark of another is permissible as long as the reference does not cause confusion as to source. See Mattel, Inc. v. Azrak-Hamway Int'l, Inc., 724 F.2d 357, 361 (2d Cir.1983); Invicta Plastics (USA) Ltd. v. Mego Corp., 523 F. Supp. 619, 623 (S.D.N.Y.1981). Not only is the fair use of another's mark permissible, it may well be beneficial, since consumers can compare at a glance two competing products. The Federal Trade Commission's policy encourages comparative advertising, and "to make the comparison vivid the Commission `encourages the naming of, or reference to competitors.'" August Storck K.G. v. Nabisco, Inc., 59 F.3d 616, 618 (7th Cir.1995) (quoting 16 C.F.R. § 14.15(b)).

Plaintiff relies on its survey to establish that there is likelihood of confusion. For reasons already stated Dr. Rappeport's findings have little, if any, weight.

Plaintiff points out that the Sweet'N Low trademark on the Sweetmate box appears in a larger font than any other word on the package except for the Sweetmate mark itself. But the name Sweetmate appears in letters vastly larger and more prominent. The Sweet'N Low trademark appears with a comparative advertisement stating "Compare'N Save!" and then, "same sweetener as Sweet'N Low at a SWEETER PRICE." Plaintiff can hardly claim in good faith that the phrase "same sweetener as Sweet'N Low at a SWEETER PRICE" is itself confusing since the same comparative statement appears on its own product, NatraTaste: "Same Sweetener as EQUAL.... At A Sweeter Price" (ellipses in original).

In explaining the purpose of its own comparative statement on the NatraTaste box, plaintiff wrote to defendant in 1995, stating:

it is unreasonable to contend that consumers would be misled to believe that a single manufacturer of two similar products, both containing the same sweetener, would undermine the market of its more expensive product by suggesting it is overpriced. Rather, the consumer is led to believe that a competing alternative to EQUAL is available from a competitive manufacturer at a lower price.

There is logic to this argument. The same logic applies to defendant's comparative statement in Sweetmate. Consumers are not likely to be confused because the product name, Sweetmate, is much more prominent and the statement "Compare'N Save!" appears directly to the left of the sunshine burst in relatively large, white letters against a red background.

A preliminary injunction is an extraordinary remedy. Plaintiff has not met its burden to establish likelihood of irreparable harm and likelihood of success on the merits or sufficiently serious questions going to the merits and a balance of hardship tipping decidedly in the plaintiff's favor.

IV

Plaintiff also seeks a preliminary injunction against defendant's use of the Sweet'N Low trademark in a comparative advertisement, on grounds that it dilutes the trademark in violation of section 43(c) of the Lanham Act, which states in pertinent part:

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).

"Fair use of a famous mark ... in comparative commercial advertising" is specifically excluded from the section. 15 U.S.C. § 1125(c)(4)(A). This is not a case where defendant has caused dilution by altering plaintiff's logo. Cf. Deere & Company v. MTD Products, Inc., 41 F.3d 39 (2d Cir. 1994). Plaintiff does not claim that defendant has distorted plaintiff's mark. It says only that the "physical association" of the Sweet'N Low trademark with the Sweetmate trademark will cause blurring. "Physical association" in a comparative advertisement is hardly likely to cause dilution. "As long as the mark is not altered, such use serves the beneficial purpose of imparting factual information about the relative merits of competing products and poses no risk of diluting the selling power of the competitor's mark." Id. at 44.

V

Plaintiff says that defendant's NutraSweet sweetener misrepresents the product in violation of Section 43(a) of the Lanham Act, 15 U.S.C. § 1125(a), and asks for a preliminary injunction. Defendant cross-moves for dismissal of or summary judgement on the claim.

Section 43(a) of the Lanham Act prohibits any person from using in commerce "any word, term, name, ... or false or misleading description of fact, or false or misleading representation of fact, which, ... in commercial advertising or promotion, misrepresents the nature, characteristics, qualities, or geographic origin of his or her ... goods...." To establish a false advertising claim under that section, plaintiff must show that the advertisement is either (1) literally false as a factual matter, or (2) although literally true, it is likely to deceive or confuse consumers. See National Basketball Ass'n v. Motorola, Inc., 105 F.3d 841, 855 (2d Cir. 1997). The misrepresentation must concern an inherent quality or characteristic that is likely to influence purchasing decisions. Id.

Plaintiff states that defendant is falsely advertising the product NutraSweet (1) by calling it "NutraSweet" (2) by calling it "100% NutraSweet", and (3) by calling it "The Original."

These arguments are without merit. The statement "100% NutraSweet" was apparently made in a mock-up coupon. Plaintiff does not contest that these coupons were never distributed to consumers. At the hearing held on October 28, 1998, the court asked of plaintiff's counsel whether the coupons were currently being circulated. He answered that he did not know. There would be little point in this court enjoining a coupon that defendant never used nor intended to use.

Plaintiff's third claim is also meritless. Plaintiff says that calling defendant's product "the Original" is a "false" statement because the term "original" means "the first in order of its particular class." According to plaintiff, only two products, Sweet'N Low and EQUAL, can claim that distinction because Sweet'N Low is the "original" sugar-substitute and EQUAL is the "original" aspartame-based sweetener. But there is no evidence that this term will lead people into believing that NutraSweet preceded EQUAL. It was legitimate for defendant to use the term "original" to show that the aspartame used in the NutraSweet sweetener is the same "original" aspartame patented by defendant and brand-named NutraSweet. The customer is thus assured of ingesting the same sweetener with the same "original" taste and effect as that the customer had enjoyed in food and beverage products that had featured the NutraSweet name and logo.

Plaintiff's claim that calling defendant's product "NutraSweet" is misleading to the consuming public is without the faintest merit. The trade name "NutraSweet" is admittedly extremely well-known. It is recognized as being associated with the aspartame sweetener. The critical ingredient in the NutraSweet product is aspartame. To say that defendant may not use its own trade name on such a product would be to deny to defendant what it spent years, and doubtless many millions of dollars, to create.

Plaintiff does not claim that the use of the name is literally false. Plaintiff thus bears the heavy burden of establishing that a "not insubstantial number of consumers" actually hold some false belief allegedly communicated by the advertisement. Johnson & Johnson * Merck Consumer Pharm. Co. v. Smithkline Beecham Corp., 960 F.2d 294, 298 (2d Cir.1992).

In implied falsehood cases, the court must determine (1) what the people that the advertisement targets find to be the message and (2) whether they are likely to be misled. See American Home Products Corp. v. Johnson & Johnson, 577 F.2d 160, 166 (2d Cir. 1978).

Plaintiff's theory is that calling the NutraSweet table-top sweetener "NutraSweet" is misleading because consumers will understand that term to mean a "sweetening ingredient" not a "sugar-substitute." As plaintiff's own witnesses conceded during depositions, in this industry the terms "sweetener," "sweetening ingredient," and "sugar-substitute" are used interchangeably.

There is no evidence that any customer would even ponder such a conceptual distinction. Not surprisingly, plaintiff did not present any evidence that any customer thought of "NutraSweet" as an "intense sweetening ingredient" made of pure aspartame and not a "sugar substitute" or sweetener. Instead, the evidence makes it clear that although many people have heard of the term "NutraSweet" they do not have a clear understanding of whether it is a brand, a product, or an ingredient. Presumably a customer has in mind primarily the taste, lack of calories, and the reliability of the producer. Nothing in the record even remotely suggests the consumer is concerned with what other minor ingredients make up the product.

Plaintiff argues that even if consumers do not clearly associate NutraSweet with an ingredient, the use of the name to refer to both an ingredient and a product is misleading. Specifically, plaintiff claims that consumers are somehow misled into believing that each packet of defendant's new product "NutraSweet" contains more aspartame than defendant's own aspartame sweetener, EQUAL.

To try to support this claim, Dr. Rappeport conducted another survey. Based on the results, he concluded that 29 percent of the sample thought that NutraSweet contained more of the "NutraSweet-brand aspartame" than EQUAL because of the words appearing on the packet.

For reasons similar to those discussed above, the survey was flawed and its results unpersuasive. In this study, 151 respondents were shown two packets, a NutraSweet packet and one of two versions of an EQUAL packet. Seventy-six people were shown an EQUAL packet with a picture of a coffee mug on it (Group I) and seventy-five people were shown an EQUAL packet with a picture of a strawberry on it (Group II). Each person in Groups I and II was asked, "Which of these, if either, do you think contains more NutraSweet per packet?"

This question is leading and ambiguous to say the least. Plaintiff concedes that some consumers think that NutraSweet is a brand. To them, this question would make little sense. This is only one example of the many flaws in the survey design.

But even if the court were to disregard the flaws and accept the survey, Dr. Rappeport's own results refute his claim that consumers are misled to believe that the NutraSweet table-top product contains more NutraSweet-brand aspartame than EQUAL because of the word NutraSweet. Dr. Rappeport's ambiguous question, not surprisingly, elicited random results. Of the 76 people in Group I, 36 percent said they thought the NutraSweet packet contained more NutraSweet-brand aspartame, 32 percent said EQUAL, and 33 percent answered, "the same" or "I don't know." In other words, each choice received about one-third of the total responses, the result one would expect if the responses had been made at random.

Dr. Rappeport showed a second group of people a packet of EQUAL, but this time with a picture of a strawberry on it instead of a coffee mug, and the same packet of NutraSweet. They were again asked, "Which of these, if either, do you think contains more NutraSweet per packet?" 55 percent said they thought NutraSweet contained more NutraSweet, 31 percent said EQUAL, and 15 percent answered, "the same" or "I don't know."

Dr. Rappeport stated during deposition that his survey showed that "clearly [it is] the name that is the most important" factor allegedly misleading consumers to believe that the NutraSweet table-top product contains more NutraSweet-brand aspartame than EQUAL. His claim is unsupported by his own survey results.

Holding everything else constant, including the name, NutraSweet, when one group was shown a picture of a strawberry and the other group a picture of a coffee mug, the results changed dramatically. At the least, the results suggest that the name is not the only statistically significant variable influencing which product people think has more NutraSweet. In the survey using the EQUAL packet showing a strawberry, it is impossible to ascertain exactly which factor or combination of factors made people think that the NutraSweet packet contains more NutraSweet-brand aspartame than EQUAL. But what is clear is that without the picture of a strawberry, Dr. Rappeport was unable to find any correlation between the name NutraSweet and people's perceptions of whether the NutraSweet or EQUAL brand contained more NutraSweet-brand aspartame per packet.

Even if one were to imagine that Dr. Rappeport's survey had shown that because of the name, NutraSweet, consumers are misled into believing that each packet of defendant's NutraSweet contains more NutraSweet-brand aspartame than each packet of defendant's EQUAL, one could not explain why such a misunderstanding would affect consumers' purchasing decisions about NatraTaste. Apparently the supposition is that if consumers believe that NutraSweet contains more aspartame than EQUAL, they must also believe that the former is of "a higher quality than it actually is." Why would consumers associate more aspartame per packet with higher quality? Even if they did, why would that affect NatraTaste sales?

Plaintiff says that because NutraSweet and NatraTaste are in direct competition with each other, "anything that results in increased sales of NutraSweet will take away sales from NatraTaste." The argument seems to be that the alleged false advertisement will cause aspartame consumers wishing to switch to a lower-priced brand to switch from EQUAL to NutraSweet rather than to NatraTaste. Plaintiff's theory is that consumers will choose NutraSweet over NatraTaste because they are misled into believing the former contains more NutraSweet-brand aspartame than EQUAL. This argument is illogical.

If it is NutraSweet-brand aspartame that the consumers wish, NatraTaste sales will be unaffected by defendant's alleged false advertising since NatraTaste does not contain any NutraSweet-brand aspartame. On the other hand, if consumers are interested simply in more aspartame, those consumers allegedly misled by the NutraSweet name into believing that the NutraSweet table-top product has more aspartame than EQUAL have no basis for inferring that NatraTaste contains either more or less aspartame than NutraSweet. Plaintiff has suggested no reason why consumers would think that because of its name the NutraSweet sweetener contains not only more aspartame than EQUAL but also more than NatraTaste.

To be entitled to summary judgment, the movant must demonstrate the absence of any material factual issue genuinely in dispute and that it is entitled to judgment as a matter of law. Fed.R.Civ.P. 56(c); Heyman v. Commerce & Industry Ins. Co., 524 F.2d 1317, 1320 (2d Cir.1975). The record must be construed in the light most favorable to the non-moving party and all reasonable inferences must be drawn to its advantage.

Defendant has met the burden of proving that there are no issues of material facts genuinely in dispute.

VI

Defendant's motion for summary judgment is granted with respect to plaintiff's false advertisement claim. Plaintiff's motion for a preliminary injunction is denied in all respects.

So ordered.