Friday, September 28, 2007

IP Works in Progress, herein of the First Amendment

Works in Progress Intellectual Property Colloquium, panel 1

Lisa Ramsey, University of San Diego School of Law
Increasing First Amendment Scrutiny of Trademark Law
Abstract | Paper

Ramsey began with an interesting example of First Amendment concerns with TM: a company with a TM in LIFE IS GOOD that makes branded apparel has gone after parodists and competitors, including LIFE IS GAY and LIFE SUCKS. This is worrisome. She offers a framework for approaching 1A problems in TM.

Questions addressed: (1) Is this use of a TM actually speech? (2) Is this speech regulation state action? As to TM, she considers this an easy question: yes, the state is involved. (3) Is this a type of speech categorically excluded from 1A protection/entitled to lesser protection? (4) Is this content-based or content-neutral regulation? Also we need to consider vagueness, overbreadth, and prior restraints.

Benefits of this approach: it focuses us on the government’s regulatory purpose and requires carefully shaped laws. The 1A approach makes us decide whether confusion as to sponsorship should matter, and may offer a way out of the circularity of permissions leading to ever-more-expansive TM rights noted by many. It also has implications for dilution, initial interest confusion, cybersquatting, and other areas.

Ramsey considers TM to be a content-based, sometimes word-based or speaker-based, regulation, requiring a compelling interest and least restrictive means analysis. But this is different from regulation that stems from disagreeing with a word, like indecency regulation; the government isn’t really trying to control the communicative impact of the word used by any speaker in any context. So much TM law would survive.

Eric Goldman makes the interesting suggestion that she should apply her thesis to “para-trademark” laws like Utah’s online registry law. 1A challenges might work well there even if courts resist them in traditional TM law.

Rebecca Tushnet

It Depends on What the Meaning of “False” Is: Falsity and Misleadingness in Commercial Speech Doctrine
Abstract | Paper

My notes: The Lanham Act is about who is to be master of meaning in commercial advertising.

Even at the level of individual words – trademarks or other advertising claims – sorting true from false is extremely difficult. This is why I am deeply skeptical of proposals to equate commercial and political speech when they’re accompanied by promises that we’ll remain protected against consumer fraud.

True and false aren’t easy lines to draw. So just saying “false and misleading speech can be banned, which makes trademark law constitutionally acceptable” rushes over much of the real conflict between the Lanham Act and the First Amendment, which is who gets to decide what’s false and what’s misleading.

A fundamental thread running through much modern trademark scholarship and case law is that if a defendant’s use of a mark is truthful, it can’t cause confusion. The problem is that it’s not right. True statements can cause confusion, because of the multiple ways in which people interpret messages and because of the irreducible, even tragic, gap between speaker and listener.

False advertising law – also part of the Lanham Act – knows this, but we have adhered to this extremely useful fiction in trademark law because acknowledging that truth can be confusing would either destabilize much trademark law or expand trademark rights beyond acceptable boundaries.

Example: Ask doctors which of two products they prefer; 30% choose A, 10% choose B; 60% say there’s no difference. Is it false to say A is preferred by doctors 3-to-1? The Seventh Circuit said no, because it was true, even though the survey showed that it misled consumers about doctors’ overall preferences.

The obvious countermove is to say that if a statement is confusing, it can’t be true. But that hitches speaker’s truth to listener’s understanding, leaving us nothing independent against which to evaluate the listener’s understanding. It’s unlikely that we want to do that – or even that we can, as a matter of constitutional law.

To take an example from Mark McKenna, suppose the NYT runs a story truthfully saying “Nike Announces New Shoe Line,” and because Nike is so good at marketing consumers assume that Nike authorized that headline. The confusion is over what consumers perceive to be an implicit representation added to the explicit headline: Nike approved this story. The headline itself is true – standing alone -- and it is confusing – when the implicit message attached to it is added in.

The if-true-not-confusing position ignores implicit meaning. That’s fine here, since no one wants the NYT even potentially exposed to liability for such a headline – but it is an undertheorized thing, and it’s bound up with the problem of trademark use that is so consuming trademark scholarship these days.

Ignoring implicit meaning may also be an impossible thing as a practical matter, since speech depends so much on implicature for its success. In fact the presence of a claim in an ad is an implicit representation that it is or should be material to choice, and we routinely understand it that way – my favorite example comes from an ad for coffee that said “made with smaller grounds” – people liked it better, even though they were given no reason why smaller grounds were better and in fact there’s no objective evidence that they are.

But there are a lot of scarier things that can be said through implication – e.g., the recent informed consent case decided by New Jersey Supreme Court. And this is worth emphasizing: Regulating truth and falsity necessarily means shaping preferences. There is no way to limit regulation to “just the facts” because it’s the implications of the facts that aid or deceive consumers.

Another example, recombinant bovine growth hormone: What if a dairy advertises its milk as rBGH free? There’s no evidence of health or safety differences in the milk for humans. But there are concerns for the cows, the survival of family farming, and the general wisdom of genetic engineering.

The FDA has suggested that labeling foods as non-GMO has the potential to deceive consumers about health. If more consumers care about direct health and safety consequences than about other possible consequences of GMO food – which seems likely – and if consumers see labeling as a signal about health or safety, then the GMO label will help a small percentage of consumers, but only by deceiving a larger percentage. And that’s assuming that there are no preference-shaping effects, which complicate matters further.

Final point: trademark lawyers think TM is insulated from the First Amendment, as copyright lawyers once did, as defamation lawyers once did. But the Fifth Circuit last year decided a case about the statutory regulation of “Cajun”: the First Amendment precluded the legislature from banning the use of the term on non-Louisiana foods because there was no evidence it was inherently deceptive in a way that couldn’t be corrected by disclosure or consumer sophistication.

Especially as we consider expanding GI protection, it would be a serious mistake to think that our little corner of the law will continue to get a free pass when the Supreme Court has been increasing its scrutiny of other commercial speech regulations consistently over the past decade. Imagine the commercial speech standard – suppression allowed only when the speech is inherently deceptive – applied to private party trademark suits. That would be a big enough change, and it would only involve applying general commercial speech doctrine as it’s evolved under Central Hudson. Full constitutional protection for commercial speech would mark an even bigger change.

Many insightful comments, including a defense of laissez-faire (which I don’t think can be coherent because once you allow common-law protection against fraud into your laissez-faire state we are merely debating what counts as fraud – consumer protection is at least in one popular form about promoting autonomy).

Eric Goldman suggested important differences between preference-shaping by private actors and government. My reaction: Such differences might well justify certain procedural requirements on regulation, but – because consumers vary in the use they make of information -- government cannot get out of the business of preference-shaping; both banning the “non-GMO” label and requiring the “GMO” label are preference-shaping, and government might legitimately do either, depending on the evidence, because both could be justified as anti-fraud measures as to some significant number of consumers.

Goldman also pointed out the costs of labeling and mandates, since meaning evolves over time. The paper takes this up – I agree, but there are also costs of uncertainty and of confusion during the transition time; certain market participants might find it valuable to redefine “dolphin-free tuna” as tuna with no dolphin parts in the cans, but that would be costly to non-dolphin-killing tuna producers and consumers who want to buy their products. Finally, Goldman suggested that the internet is a good way to punish misbehaving sellers – consumers can communicate with each other. This is true, but information overload limits our ability to use all the information out there – and especially if the government cut back to common-law fraud, it would be even harder for us to figure out what we need to figure out.

David Olson, Boston College Law School
Towards a First Amendment Grounding for Copyright Misuse
Abstract | Paper

The paper grows out of Carol Schloss v. Estate of James Joyce: what can be done to make the world freer for this sort of scholarship? Fair use is too risky: indeterminate, expensive, leads courts to split the baby – though Schloss got attorneys’ fees, it’s too rare to rely on. Copyright misuse could be the proper sword against overreaching by copyright owners. But in order to induce courts to find misuse, which they won’t if the consequences for the copyright owner are too severe, he wouldn’t completely invalidate the copyright of an overreaching copyright owner – he’s considering various remedies, such as compulsory licensing until the misuse is cured, free licensing to the injured party, etc.

No comments: