# The Charbucks Saga: How Starbucks Lost Its Dilution-by-Blurring Claim

> After more than a decade of litigation, the Second Circuit held that even a famous mark cannot prove dilution by blurring without meaningful similarity and real evidence of association, affirming judgment for a tiny New Hampshire roaster.

Topic: Trademarks  |  Author: Lidiia Levitska  |  Source: Intellectual Property Law (outsideipcounsel.com)
Canonical: https://outsideipcounsel.com/blog/starbucks-v-wolfes-borough-charbucks-blurring/


*Starbucks Corp. v. Wolfe's Borough Coffee, Inc.*, No. 12-364-cv, 736 F.3d 198 (2d Cir. 2013), decided November 15, 2013 in an opinion by Judge Lohier, is the capstone of a dilution dispute that ran for more than a decade. Starbucks sought to enjoin a small New Hampshire roaster, Black Bear Micro Roastery, from selling coffee under the names "Charbucks Blend," "Mister Charbucks," and "Mr. Charbucks." Despite the fame of its mark, Starbucks lost on dilution by blurring, and the opinion is now a standard reference for how the Trademark Dilution Revision Act's six blurring factors actually operate when applied to real evidence.

## At a glance

- **Case:** *Starbucks Corp. v. Wolfe's Borough Coffee, Inc.*, No. 12-364-cv, 736 F.3d 198 (2d Cir. 2013)
- **Court:** U.S. Court of Appeals for the Second Circuit
- **Decided:** November 15, 2013; opinion by Judge Lohier (joined by Judges Katzmann and Kearse)
- **Holding:** Starbucks failed to prove the "Charbucks" marks were likely to dilute its famous marks by blurring; minimal similarity and weak associative evidence outweighed the senior mark's fame
- **Disposition:** District court's judgment for the defendant affirmed

## A decade of litigation

The fight began in 2001 and produced multiple appellate decisions before reaching its 2013 conclusion. Wolfe's Borough Coffee, doing business as Black Bear Micro Roastery, is a family micro-roastery that named a dark roast "Charbucks," a play on the common gibe that Starbucks roasts its beans nearly to the point of being charred. Starbucks asserted dilution by blurring under both federal and New York law, along with infringement and tarnishment theories. An earlier Second Circuit decision in the same litigation (*Starbucks IV*, 588 F.3d 97 (2d Cir. 2009)) had vacated and remanded for the district court to reweigh the blurring factors under the TDRA, correcting earlier analytical missteps, including the mistaken notions that the marks had to be "substantially similar," that bad faith was required before the intent-to-associate factor could favor the senior mark, and that the absence of actual confusion could defeat the actual-association factor. By the time the case returned to the Second Circuit in 2013, the only question was whether, applying the statutory factors correctly, Starbucks had carried its burden.

## Working the six blurring factors

The TDRA, 15 U.S.C. § 1125(c)(2)(B), lists six non-exhaustive factors relevant to dilution by blurring: (1) the degree of similarity between the marks; (2) the degree of inherent or acquired distinctiveness of the famous mark; (3) the extent to which the owner is engaging in substantially exclusive use; (4) the degree of recognition of the famous mark; (5) whether the junior user intended to create an association; and (6) any actual association between the marks. The court evaluated each.

Factors two, three, and four favored Starbucks. The STARBUCKS mark is arbitrary as applied to coffee and so highly distinctive, it is used substantially exclusively, and 79 percent of the survey respondents were familiar with it. Fame was never in doubt. But the court said the more important factors in this case were similarity and actual association, and that distinctiveness, recognition, and exclusive use "do not overcome the weak evidence of actual association." Those three, it concluded, "weigh only weakly in Starbucks' favor." The case turned instead on the remaining factors, and above all on the first.

On similarity, the court agreed with the district court that the marks were only minimally similar. The names "Charbucks" and "Starbucks" share a syllable, but Black Bear never sold "Charbucks" in isolation; it sold "Charbucks Blend," "Mister Charbucks," and "Mr. Charbucks," in packaging and trade dress markedly different from Starbucks': a different logo, different color scheme, and prominent house branding. The court treated similarity as a factor of considerable weight in the blurring calculus and found the modest overlap insufficient to drive an inference of dilution.

## Intent and association: the decisive weaknesses

On intent (factor five), the court accepted that Black Bear had intended to evoke an *association* with Starbucks. Its founder testified that "Charbucks" came out of the "coffee wars in Boston" and that he "meant to evoke an image of dark-roasted coffee of the type offered by Starbucks." That was enough: the court agreed with the district court that the factor favored a finding of likely dilution, and on the de novo balance it attributed "a moderate amount of significance" to it. That was a change from the pre-remand posture, where the district court had put the intent factor in Black Bear's column on the theory that an intent to associate without bad faith did not count. On remand the district court accepted that factors two through five all favored Starbucks. What the court refused to do was let intent stand in for proof. An intent to create an association "is a separate factor under the TDRA and does not constitute per se evidence that the actual association factor weighs in favor of the owner of the famous mark." Starbucks had argued that Black Bear's admitted intent raised a "presumption of association." Adopting that, the court said, would effectively merge the fifth and sixth factors by making the former determinative of the latter.

The sixth factor, actual association, proved fatal. Starbucks' principal evidence was the Mitofsky telephone survey of six hundred participants. Asked what first came to mind on hearing "Charbucks," 30.5 percent said "Starbucks." But asked to name any company or store that might offer a product called "Charbucks," only 3.1 percent said Starbucks. The court found the survey fundamentally flawed because "the word 'Charbucks' was presented outside of its marketplace context," divorced from the actual names and packaging Black Bear used, and Starbucks, which bore the burden of proof, "failed to show that this flaw did not materially impact the survey results." Bare recall of a famous name is not the same thing as an association that impairs the famous mark's distinctiveness, which is what the statute requires. The court concluded that Starbucks had shown weak actual association at best. Balancing all six factors de novo, the strong showings on fame and distinctiveness could not overcome the minimal similarity and the thin associative proof.

## Open questions

The opinion sharpens, without fully resolving, a recurring problem in blurring litigation: how to evidence "association" in a way that proves likely impairment rather than mere recall. The court faulted Starbucks' survey for testing the junior term in a vacuum, but it did not prescribe what a sufficient survey would look like, leaving practitioners to design instruments that capture marketplace conditions and tie association to distinctiveness harm. The decision also underscores the outsized role of the similarity factor: although the TDRA does not require "substantial similarity," *Charbucks* shows that low similarity can be nearly dispositive even for an extraordinarily famous mark, raising the question of how much weight any single factor should carry in a test that is nominally a holistic balance.

## Implications

- **Fame is necessary but not sufficient.** A famous mark wins factors two through four almost automatically, yet can still lose on blurring if similarity is low and association evidence is weak, exactly what happened to Starbucks.
- **Survey design is decisive.** Evidence that tests a junior term stripped of its real-world packaging and house marks may be discounted; surveys should replicate marketplace conditions and connect association to impairment of distinctiveness.
- **Intent to associate is not proof of association.** An acknowledged intent to call a famous mark to mind does satisfy the fifth factor, and here it carried moderate weight for Starbucks. What it does not do is raise a presumption on the sixth factor: intent and actual association are distinct inquiries, and winning the first does not carry the second.

## Frequently asked questions

**Did Starbucks lose because its mark was not famous?** No. The court accepted that the Starbucks marks are famous, distinctive, and substantially exclusive. Starbucks lost because the marks were only minimally similar and its evidence of actual association was weak.

**Was the Charbucks survey enough to prove dilution?** No. The court found the survey of limited value because it tested "Charbucks" in isolation rather than as actually marketed ("Charbucks Blend," "Mister Charbucks") and measured mere mental recall rather than likely impairment of distinctiveness.

**Does the TDRA require the marks to be substantially similar?** No. An earlier decision in this litigation rejected a "substantial similarity" requirement for blurring. But this case shows that a low degree of similarity remains a heavily weighted factor that can defeat a claim even absent a strict threshold.

## Authorities and sources

- *Starbucks Corp. v. Wolfe's Borough Coffee, Inc.*, No. 12-364-cv, 736 F.3d 198 (2d Cir. Nov. 15, 2013). Full opinion text: [PDF hosted by Harvard's Berkman Klein Center](https://cyber.harvard.edu/people/tfisher/IP/2013_Starbucks.pdf); also at [FindLaw](https://caselaw.findlaw.com/court/us-2nd-circuit/1649664.html).
- Prior appeal establishing the governing framework (*Starbucks IV*): *Starbucks Corp. v. Wolfe's Borough Coffee, Inc.*, 588 F.3d 97 (2d Cir. Dec. 3, 2009), full text via [New York Law School Digital Commons](https://digitalcommons.nyls.edu/circuit_opinions/454/).
- Statute: 15 U.S.C. § 1125(c), including the six blurring factors at § 1125(c)(2)(B), via [Cornell Legal Information Institute](https://www.law.cornell.edu/uscode/text/15/1125).