Rogers v. Grimaldi: The Two-Part Test That Made Room for Art in Trademark Law

Rogers v. Grimaldi, Nos. 88-7826 & 88-7828, 875 F.2d 994 (2d Cir., decided May 5, 1989), is the rare appellate opinion whose name became a verb of sorts in trademark practice: lawyers speak of “Rogers-ing” an expressive-use defense. The case arose when the actress and dancer Ginger Rogers sued over Federico Fellini’s 1986 film Ginger and Fred, a melancholy story about two aging Italian cabaret performers, Pippo and Amelia, whose nightclub act had once imitated Fred Astaire and Ginger Rogers. Rogers claimed the title falsely implied her endorsement and violated her rights under Section 43(a) of the Lanham Act and state law. She lost — but the framework Judge Newman articulated to explain why has governed the intersection of trademarks and artistic expression ever since.

At a glance

• Case: Rogers v. Grimaldi, 875 F.2d 994 (2d Cir. 1989), Dockets 88-7826 & 88-7828
• Court: United States Court of Appeals for the Second Circuit
• Decided: May 5, 1989; opinion by Judge Jon O. Newman
• Holding: The Lanham Act does not bar a minimally artistically relevant use of a celebrity’s name (or a trademark) in the title of an expressive work unless the title has no artistic relevance to the work, or, having some relevance, explicitly misleads as to source or content
• Disposition: Summary judgment for the defendants affirmed; the false-endorsement and related claims failed under the new balancing test

The collision the court had to resolve

The opinion begins from a genuine tension. The Lanham Act exists to prevent consumer confusion about the source and sponsorship of goods and services, and a film is sold in commerce like any other product. But movie titles, song titles, and book titles are not merely commercial labels; they are part of the expressive work itself, chosen for meaning, allusion, and resonance. Applying the ordinary multi-factor likelihood-of-confusion analysis to titles, the court recognized, would let trademark law operate as a censor of artistic choices whenever a title evoked a recognizable name or brand. That risk implicated the First Amendment directly: titles “require more protection than the labeling of ordinary commercial products.”

At the same time, the court refused to grant artistic works blanket immunity. A title could still deceive — it could be used as a bait-and-switch to mislead consumers about who made or endorsed a work. The task was to calibrate a rule that protected expression without abandoning the public’s interest in not being deceived.

The test: artistic relevance and explicit misleadingness

The Second Circuit’s solution was a two-pronged screen that tips the balance toward expression. The Lanham Act applies to an expressive work’s title only if (1) the title has no artistic relevance to the underlying work whatsoever, or (2) if it has some artistic relevance, the title explicitly misleads as to the source or content of the work. Both prongs are deliberately speaker-friendly.

The first prong sets an exceptionally low bar. The use need only clear a threshold of relevance — “above zero,” as later courts would gloss it. Ginger and Fred easily passed: the title referred to the central characters and to the Astaire-Rogers iconography their act mimicked, and it was therefore relevant to the film’s content and themes. Fellini had not slapped a famous name on an unrelated product to trade on its fame.

The second prong is narrower still because of the word “explicitly.” It is not enough that a title might imply endorsement, or that some consumers could be confused; the title must expressly state a false affiliation — an overt claim that the celebrity sponsored or produced the work. The mere use of a real person’s name in an artistically relevant title, without an explicit misrepresentation of endorsement, does not satisfy the standard. Because Fellini’s title contained no such explicit claim, Rogers’s Lanham Act theory failed as a matter of law. The court applied parallel reasoning to her right-of-publicity claim under Oregon law, again finding the artistic relevance of the title decisive.

Why the framework endured

Rogers proved influential far beyond film titles. Courts extended it from titles to trademarks used within expressive works — in song lyrics, video games, art, and parody — wherever a Lanham Act claim threatened to suppress protected expression. The Ninth Circuit adopted the test in Mattel, Inc. v. MCA Records (2002) to protect the song “Barbie Girl,” and a generation of decisions built a robust expressive-use defense atop the two prongs. The test’s appeal was its administrability: it let courts dispose of weak claims early, often on summary judgment or a motion to dismiss, sparing artists ruinous, fact-intensive confusion trials.

That dominance was reshaped in 2023. In Jack Daniel’s Properties, Inc. v. VIP Products LLC, the Supreme Court held that the Rogers threshold does not apply when an alleged infringer uses another’s trademark as a designation of source for its own goods — that is, as a trademark in the trademark sense. In that posture, ordinary likelihood-of-confusion analysis governs and Rogers drops out. The Court was careful, however, not to repudiate Rogers across the board; it expressly declined to decide the framework’s continuing validity for genuinely expressive, non-source-identifying uses. The result is that Rogers survives, but its domain has been pruned to its core: titles and expressive uses that are not themselves functioning as brands.

Open questions

The post-Jack Daniel’s landscape leaves real uncertainty about where the source-identifying use ends and the protected expressive use begins. A title, a parody, or an artwork can simultaneously communicate a message and operate as a source signal, and courts are now sorting which characterization controls. Whether the “explicitly misleads” prong retains its narrow, speaker-protective bite in cases that fall outside Jack Daniel’s carve-out — and how circuits that had embraced Rogers most enthusiastically will recalibrate — remains to be worked out case by case.

Implications

• Titles get heightened protection. Using a real name or mark in the title of a film, book, song, or game is generally lawful if the title is artistically relevant and does not explicitly assert endorsement.
• The relevance bar is minimal. Plaintiffs rarely prevail on prong one; almost any genuine thematic connection suffices to clear it.
• “Explicitly” is doing the work. Implied or possible confusion is not enough; counsel should look for an overt, affirmative misstatement of sponsorship before predicting liability.
• Mind the Jack Daniel’s line. If the contested use functions as a source identifier for the defendant’s own goods, Rogers will not shield it and standard infringement analysis applies.

Frequently asked questions

Did Ginger Rogers lose? Yes. The Second Circuit affirmed summary judgment against her on the Lanham Act and right-of-publicity claims, holding the film’s title artistically relevant and not explicitly misleading.

Is the Rogers test still good law after Jack Daniel’s? Yes, but in a narrowed form. The Supreme Court held Rogers inapplicable when a mark is used as a source identifier for the defendant’s own products, while declining to decide its fate for purely expressive, non-source uses. For genuine titles and expressive references, Rogers continues to operate in the circuits that adopted it.

Does Rogers protect any use of a famous name? No. It protects uses that are artistically relevant and that do not explicitly mislead about source or endorsement. A title fabricated solely to exploit a name, or one that overtly claims false sponsorship, falls outside the shield.

Authorities and sources

• Rogers v. Grimaldi, 875 F.2d 994 (2d Cir. 1989), Dockets 88-7826 & 88-7828 (decided May 5, 1989): opinion (Public.Resource.Org reporter text); case summary (WIPO Lex).
• Subsequent treatment: Jack Daniel’s Properties, Inc. v. VIP Products LLC, 599 U.S. 140 (2023) (Cornell LII).
• Overview: Rogers v. Grimaldi (Wikipedia).