Tuesday, July 11, 2023

Professor McCarthy: "Some Thoughts on the Supreme Court Jack Daniel's Case"

Professor J. Thomas McCarthy shares his thoughts (below) on the Supreme Court's "Bad Spaniels" decision (pdf here), in the hope of stimulating discussion (in the comments) as to what readers think will be the impact of the case on the use of marks in "expressive" works. As you will recall, the Court vacated the decision of the U.S. Court of Appeals for the Ninth Circuit that had deemed the "Bad Spaniels" dog toy immune to claims of infringement and dilution of various Jack Daniel’s trademarks. The Supreme Court, however, ruled that the Ninth Circuit was wrong in concluding that the toy was entitled to First Amendment protection under Rogers v. Grimaldi because of the humorous messaging on the toy. The Court held that (1) the Rogers test does not apply when the defendant uses the challenged matter as a mark, and (2) the Ninth Circuit’s interpretation of the noncommercial exemption to dilution claims was overbroad.


© Tom McCarthy 2023

The Rogers Test for Infringement. On June 8 the Supreme Court decided the Jack Daniel’s case and restricted the scope of what is known as the “Rogers” test.  Under the Rogers two-part balancing test, a trademark used in an accused “expressive” work (such as a novel, movie or TV series) will not be a trademark infringement if the accused mark has  any possible “artistic relevance” to the message expressed and so long as the accused mark does not “explicitly mislead”  as to the source of the defendant’s work. The Rogers test  is a judge-created assessment to determine if there is trademark infringement, used instead of the statutory test of likelihood of confusion. It is a judge-made substitute for the Lanham Act test of infringement.

A Dog Toy is an “Expressive” Work? The Rogers test is intended to be a method for  balancing First Amendment free speech policies with trademark law’s policy of preventing the likelihood of confusion. The Ninth Circuit had held that JACK DANIEL’S whisky trademarks and trade dress were not infringed by VIP, which added to its product line of humorous chewable rubber dog toys a “Bad Spaniel’s” toy,  made to closely resemble the Jack Daniel’s trade dress. The Ninth Circuit said that the VIP dog toy was entitled to the special Rogers free speech analysis because it was an “expressive” work: the toy “communicates a humorous message.” I had thought that the Ninth Circuit was wrong because stretching “expressive works” to include ordinary consumer products potentially encompasses just about all commercial goods. If a jokey dog toy qualifies as an “expressive” work, then why not a box of breakfast cereal containing “expressive” images of a brand mascot, color graphics and  slogans? In the Ninth Circuit’s view, the Rogers test would replace the traditional test of likelihood of confusion in a huge number of trademark infringement cases.

 I’d thought that the Supreme Court would reverse and then redefine what qualifies as an “expressive” work suitable for application of the Roger’s rule. But the Supreme Court didn’t do this. Instead, it narrowed use of the Rogers rule in a different way. The Court in a unanimous decision held that the Rogers test does not apply when the accused use is itself a trademark use. In such cases, free speech policies are adequately accommodated by the traditional trademark infringement test of  likelihood of confusion.

The Rogers Test as an Off-Ramp to Dismissal of the Infringement Claim. As applied by the courts over a thirty year period, use of the Rogers test almost always resulted in dismissal of the infringement claim, often on a pre-trial Rule 12(b)(6) dismissal or on summary judgment. Some courts interpreted the Rogers test not as a balancing of rights, but as a slippery slope leading to a prompt dismissal. As government attorney Matthew Guarnieri described it in oral argument to the Supreme Court , the Rogers analysis provides “a special off-ramp at the beginning of the litigation.” Writing for the unanimous court, Justice Kagan agreed that the Rogers analysis is an extra-statutory diversion away from the traditional likelihood of confusion test for infringement: “[T]he Rogers test … offers an escape from the likelihood-of-confusion inquiry and a shortcut to dismissal….” 

The New Rule of the Jack Daniel’s Case. The government as amicus argued that the Rogers analysis is an extra-statutory test that is unnecessary to accommodate the First Amendment and should be eliminated by the Supreme Court. The government argued that the traditional likelihood of confusion standard can already factor in free speech policies: no judge-made extra-statutory test like Rogers is appropriate. But the Supreme Court took a narrower approach and limited the Rogers test to cases where the accused use is not itself a trademark use. Under this new bright line rule, attorneys and courts will be arguing over whether the accused use qualifies as a “trademark use.”

Is the Accused Use a “Trademark Use?” The Supreme Court said that VIP’s dog toy was a trademark use and not appropriate for the Rogers test. The court looked at the entire context in which the accused mark appeared to determine if the accused use was use as a trademark. The Court said that the first thing to consider is “the way the product is marketed.” In my view, there is a large body of precedent defining whether a designation has been used as a trademark. Both the U.S.P.T.O. and the Trademark Board use the terminology “failure to function” when referring to a use of a designation in a non-trademark manner.

What’s the Future of the Rogers Test? The future of the Rogers test is clouded and uncertain not only because of the cautious approach of the majority opinion but  because three Justices filed a concurring opinion that puts in doubt the future  of the Rogers test. Those Justices commented that “it is not obvious that Rogers is correct in all its particulars – certainly, the Solicitor General raises serious questions about the decision.” Remarking that the validity of the Rogers test “remains for resolution another day,” the three Justices  issued the ominous warning that “lower courts should be attuned to that fact.” I think this warning from the three concurring Justices should be interpreted by the lower courts as a clear caveat to apply the Rogers test with caution and to strictly interpret the bright line drawn by the unanimous Court.

Read comments and post your comment here.

Some text Copyright John L. Welch 2023.


At 7:21 AM, Blogger Joe Dreitler said...

Rogers was a horribly flawed and misguided decision when it was issued. It should have been a decision that used Polaroid factors and found no infringement of Ginger Rogers by a movie having nothing to do with her. Instead, it was used by virtually every circuit to allow certain infringements to escape liability. My personal “favorite” was a 6th Circuit decision that allowed an infringer to mass produce and sell posters of Tiger Woods (with his name on the poster) based upon an oil painting, solely on the basis of Rogers. Rogers deserves to die an ignominious death, it shielded simple infringements from liability for decades.

At 9:00 AM, Blogger John L. Welch said...

According to the "Bad Spaniels" decision, the first question is whether defendant is using the challenged "use" was a trademark use. If so, the standard likelihood-of-confusion test applies, but "parody" must be factored in. If the accused "use" is not use a trademark (or use as a source indicator), then one may look at the Rogers test. But if the "use" is not use as a trademark, what's the point? How can there be likely confusion if the accused use is not use as a source indicator? So, is the Rogers test now superfluous? a dead letter?

At 10:05 AM, Anonymous Matthew Anderson said...

To Mr. Welch's comment, if courts all along had properly applied "use" not in its plain meaning but in its "functions as a source mark" definition, then we would have had no Rogers test and Tiger Woods would have won in that poster's case that Mr. Dreitler cites. "Use" is the off ramp, and I guess the supreme court moved the needle a little bit in that direction.

At 12:31 PM, Anonymous Joseph Grasser said...

A little off topic from the other comments, but I'm curious what others think the impact of footnote 2 (re ability to decide likelihood of confusion at 12(b)(6) or MSJ stage). It seems to me that it should serve to embolden district judges to toss flimsy claims more easily than in the past. What do others think?

At 1:21 PM, Blogger Christine Farley said...

Thanks for sharing your thoughts on the decision, Tom! Two points on "trademark use." First, I cannot see how, other than by foolishly stating in its complaint that it owns a trademark in the design, it can be concluded that VIP made a trademark use of Jack Daniel's marks. The design is the parody; the design is the content of the product. And though the Bad Spaniels logo on the back of the hangtag that the Court finds "relevant" looks like a trademark use, it is NOT a use of Jack Daniel's marks because it id just a dog's and the words "Bad Spaniels." Second, I wonder if putting the two SCOTUS trademark cases together we can better understand the work the Court thinks "trademark use" should be doing more broadly. In Abitron v. Hetronic, the Court states that "use in commerce" means the bona fide use of a mark in the ordinary course of trade where the mark serves to identify and distinguish the mark user’s goods and to indicate the source of the goods. Since use in commerce by the defendant is required for all infringement actions, perhaps the Court thinks that all defendants must be making a trademark use of the plaintiff's mark. If so, then yes, the Rogers test won't be as determinative in future speech cases.

At 5:29 PM, Blogger John L. Welch said...

Christine: Bad Spaniels tells us that Rogers doesn't apply if the challenged work is being used as a trademark. Suppose a court finds that a challenged work is NOT being used as a mark (and therefore the likelihood of confusion test does not apply). Why does one then apply the Rogers test? We already know there is no likely confusion because the challenged work is not a source indicator. In other words, what claim remains to be avoided via the Rogers test? Is Rogers a dead letter?

At 9:44 PM, Blogger Tom McCarthy said...

The Lanham Act nowhere states that use as a trademark is required for infringement. While the Sixth Circuit once indicated that use as a trademark was a gateway requirement, it later admitted that this position is erroneous. The Second Circuit in the 2013 Oprah Winfrey case expressly rejected the Sixth Circuit view that plaintiff must plead and prove that the accused designation is a “trademark use.” The Second Circuit held that that plaintiff in a Lanham Act infringement suit has no burden to plead that defendant's accused designation qualifies as a “trademark use.” That said, a non-trademark use of another mark is highly unlikely to cause a likelihood of confusion. Likelihood of confusion over source, sponsorship, affiliation or approval usually happens when the viewer is confronted with two similar marks which deceptively and confusingly tell the viewer that they identify and distinguish a single or affiliated source. “Trademark use” is not a separate element of plaintiff's case, but should be one factor of the likelihood of confusion test. But I predict that most trial judges will not take that route if the accused use if not a trademark use. I think that in the interests of not being reversed they will take the safe and familiar road, apply the Rogers test , quickly march through the two Rogers factors and dismiss. But all that assumes away what I thought was the Achilles heel of the Rogers test: what exactly constitutes an “expressive” work?

At 6:57 AM, Blogger Christine Farley said...

John: Let's imagine the Gordon v. Drape (Honey Badger don't care) case being litigated after Jack Daniel's. Plaintiff would claim confusion and defendant would claim nonuse as a trademark and First Amendment defense. Step 1 would be to determine if the catch phrase that defendants used as a punchline in their humorous greeting cards was used as a mark. I say it was not. According to Jack Daniel's then, the court can apply Rogers rather than the LOC test and we get to the same result. The question I raised is whether, under the definition of use in commerce in Abitron, a defendant must now be making trademark use for an infringement claim to be viable. If so, then yes, you are correct, the case is dismissed with a finding of no trademark use. So the defendant id Gordon would win. An outstanding question for parodists and others that use marks not as marks, but for speech, is whether an off ramp for non-trademark use is preferable to the Rogers test. I doubt it is because it may require a more developed record including how consumers understand the use of the mark. This may begin to look a lot like the LOC test.

At 7:08 AM, Blogger Christine Farley said...

Tom: In a forthcoming guest blog on Abitron (on Eric Goldman's blog), I explain the Supreme Court just added a new definition of "use in commerce," which not only draws from both the definition of "use in commerce" AND the definition of trademark (both in sec. 1127), but also takes language that was intended by Congress to apply to the acquisition of trademarks, not the infringement of trademarks: "'use in commerce' means the bona fide use of a mark in the ordinary course of trade where the mark serves to identify and distinguish the mark user’s goods and to indicate the source of the goods." As a result, it could be that in order to satisfy the use in commerce requirement for an infringement claim, plaintiffs will now have to asset the defendant's trademark use. If I were representing a defendant, I'd make this argument.

At 7:28 AM, Blogger John L. Welch said...

Thank you for your comments, Tom and Christine. I think that the question of whether a defendant is making a trademark use is not a simple one. [Maybe it was too simple in the Bad Spaniels case in light of VIP's concession]. For example, suppose the packaging of Tom's cereal box parodies some other mark. How do you decide whether the packaging is trademark use? Defendant would say it's not. Plaintiff would say it's trade dress, and that it would be recognized as such by the public. Is any distinctive packaging a trademark use? Would the plaintiff conduct a survey?

If it's not a trademark use, and if (as Tom says) that is just one factor in the likelihood of confusion analysis, then we approach the "off-ramp" of the Rogers test - provided that the accused use is an "expressive work." That brings us right back to Tom's question - which the Supreme Court ducked: what is an "expressive work."

I hope I am making some sense in these comments. JLW

At 8:48 AM, Blogger John L. Welch said...

Another way to look at my comment: After "Bad Spaniels," if an accused use is not a source indicator (i.e., a trademark), how could it ever be "explicitly misleading" as to source under the Rogers v. Grimaldi test?

At 11:16 AM, Blogger Tom McCarthy said...

Excellent point John. It seems that the courts are going to have to re-formulate the elements of the Rogers test. Expect a period of uncertainty and confusion (in the courts).


Post a Comment

<< Home