Not quite God and Man at Yale, but you do what you can.
Moderated by Clarisa Long.
Rob Kasunic, AU: He reads Eldred to say that (1) the internal copyright safeguards of idea/expression and fair use prevent First Amendment conflicts, but (2) this could change – the traditional contours of copyright could be altered. What should courts do under such circumsntances?
What are the key characteristics of traditional contours? We know that duration isn’t one, nor subject matter, nor formalities. Congress would have to limit the internal safegaurds to create a constitutional conflict. Why are idea/expression and fair use so important? Because they’re judicially created and developed creatures of the common law. Limiting judicial discretion/breathing space through legislation that eliminates idea/expression and fair use but doesn’t create a comparable safeguard would be constitutionally shaky.
Section 1201 has the capacity to change the traditional contours. Most First Amendment attacks to date have been weak or meritless, but one can imagine factual situations that would present real problems, e.g., a corporate whistleblower who can’t provide the encrypted smoking gun without circumvention.
(1) The court in such a case could consider whether Congress satisfied the First Amendment within the statute, but no relevant exemptions help here. (2) The court could interpret the statute to avoid liability, as in Skylink and Storage Tech. The structure of 1201 distinguishes access from rights controls to maintain the traditional contours of copyright law, but when copyright owners don’t respect that theoretical distinction and merge access with rights controls, as Tony Reese has ably written, problems develop. Attempted distinctions, however, would result in rewriting the statute with negative effects on future cases. (3) The court could apply heightened First Amendment scrutiny, but that might result in invalidating the entire statute, which is too blunt a remedy. (Query: Why? This isn’t a new problem for First Amendment doctrine, which has standards for facial and as-applied challenges.) But if the statute changes the traditional contours of copyright, that might be the necessary result. (4) Courts have equitable authority to preserve First Amendment rights by limiting copyright. This is the optimal approach, and Congress has deferred to judicial doctrines and codified them before.
Some have suggested “anticircumvention misuse” as a limiting doctrine. This has many benefits, but carries its own baggage from antitrust. It looks at the problem from the plaintiff’s perspective – is the plaintiff overreaching? – when we should instead create a user’s perspective doctrine of “fair circumvention.”
Joe Liu, BC: As a copyright lawyer, he’s prone to copyright exceptionalism. The First Amendment may deal with other, similar issues routinely. (Comment: Amen!) He proposes that copyright is not different from other areas. Breathing space plays a key role in other areas, but virtually no role in copyright. This is a mistake.
Assume Eldred is right, idea/expression and fair use are all we need. What follows? Serious modifications to copyright law! These critical safety valves are uncertain and ill-defined. This can be expected to and does chill speech. Defamation, intentional infliction of emotional distress, etc. all consider the chilling effects of liability when formulating doctrine, even though they also have internal safety valves. In defamation, the Court recognized the need to modify doctrine to avoid chilling effects by changing the burden of proof and standard of liability.
This has a cost – unredressed harm to plaintiffs’ reputations – but we are not indifferent to the burden of uncertainty on speech.
How to create breathing space in copyright? Substantively, more sympathetic application of fair use in cases raising speech interests, as in Campbell. Procedurally, the burden of proof should be on plaintiffs when defendants establish a speech interest. (Comment: Say what? If copyright is nothing new, then treat it like defamation: the burden is on the plaintiff for everyone; the defendant needs to jump no hurdle to enter First Amendment-land. Maybe most copyright plaintiffs will easily meet that burden, but what’s wrong with that?) The burden should especially be on the plaintiff for market harm. Remedies could also be limited to damages.
These proposals require defining the set of cases raising speech interests. Why not downloaders? This shouldn’t be too difficult – courts distinguish between public and private figures, speech and nonspeech. Some categories – parody, satire, even some literal copying – fulfil speech interests. It would be easy to reject pretextual claims, like the 2000-song downloader.
The big objection to his proposal: copyright law is just different. Liu is not persuaded; his paper takes up the argument.
Paul Bender, Arizona State: Has the opposite caveat, in that he knows First Amendment law, not copyright. Everyone agrees that Eldred didn’t clarify the relationship between copyright and the First Amendment. The First Amendment bears “less heavily” on copyright – less heavily than what? Even worse is “traditional contours.”
The one fact everyone agrees on: the copyright clause and the First Amendment were adopted close in time; a limited monopoly is therefore compatible with free speech. But why did the Framers think so? Today, people say idea/expression and fair use, but did the Framers think about that? We don’t know what that means. (Comment: Consider, for example, that the newspapers Jefferson so loved weren’t copyrightable at the time of the Founding. Past compatibility is no guarantee of future performance.)
Does the idea that copyright promotes expression help? Usually, that doesn’t justify laws restricting speech. If copyright law violated a specific infringer’s First Amendment rights, it wouldn’t help that it promoted someone else’s speech.
So what were they thinking? 95% of infringements will involve no exercise of First Amendment rights because the First Amendment is about the freedom to make one’s own speech. Plagiarism isn’t a First Amendment activity. (Comment: Nor is plagiarism copyright infringement. Well, I wrote a whole piece about this – I will try not to interrupt too much.) Selling books is First Amendment activity, but stealing books to sell them isn’t.
It’s a mistake to take the 80% of infringement that’s just copying as First Amendment problems. We ought to look at those involving expression by the infringer. Is the copyist making his/her own point, and is copying expression important to that? The Zapruder case is an example of where copying is so important to the argument it must be allowed – incidentally, maybe the Zapruder film isn’t copyrightable at all because it’s accidental, not an intellectual production. Alan Cranston’s translation of Mein Kampf, showing the ugly parts left out of the official translation, is another example (Bender wasn’t aware of how this case came out – not fair use).
The best analogy: a sculptor wants to use a unique tree on my property for his sculpture. Normally he’d be liable for taking the tree even if the result is creative. People say that scarcity makes a difference, but assume I have a large marble quarry that isn’t exhaustible; the sculptor still has no right to take the marble. (Comment: So we’re asked to assume counterfactually that tangible property lacks one of its defining characteristics … and then simply expected to apply the tangible property rule to this imaginary situation? I am unpersuaded.)
My comments: I couldn’t resist saying a bit about Bender’s presentation. Bender provides us no reason to distinguish the Zapruder/Mein Kampf cases from the rock quarry. If property is property and that’s the end of the question, I don’t have any right to use either type no matter how helpful it would be to my message. But what that reveals is the deeper flaw in the claim: the label “property” cannot be dispositive; otherwise it would be trivial to legislate away First Amendment protections in any area (defamation as property in personhood – as the right of publicity threatens to do now; anti-flagburning laws as property in the flag; blasphemy laws as protecting religious property; etc.). The reasons for calling something property are relevant to whether the First Amendment permits particular unconsented “uses,” but Bender’s analogy is unhelpful because it doesn’t address whether the reasons for excluding people from rock quarries are the same as the reasons for excluding people from copying – or sampling – music. When pundits suggest that this is the first “YouTube election,” it’s hard to say that copying is serving no speech-promoting function.
The other big point is that, if you start to say that a copyist’s “mixing labor” with a creative work is what triggers at least a First Amendment inquiry, you are conceding far more ground to free speech concerns than most people think. Selection, coordination, and arrangement – the mix tape – are recognized types of creative labor in copyright law. Performances also involve the addition of creative labor to the underlying work. And of course any infringement of the derivative works right would trigger a First Amendment analysis under this theory. That’s a reasonable position to take, but let’s not pretend it largely preserves the stability of copyright law except in a few unusual cases.
The rest of my presentation was an attempt to draw some lessons for copyright by comparing concepts of the public domain to public forum doctrine. I’m still working on this; there is definitely meat there, but I haven’t quite figured out all the moving parts. At the very least, the difficulties of public forum doctrine promise a rough road ahead for useful theories of the public domain.
Jane Ginsburg: How do we know safety valves work, in the sense of protecting and promoting speech?
Liu: His proposal presumes they do, as the Court does in other areas. His concern is not so much about the baseline of defamation but for problems in copyright.
Bender: The way you know is to test them – get sued, create litigation.
Me: There is some empirical comparative evidence on defamation suggesting that the legal regimes here and the UK don’t have much differential effect on the practice of journalism. Of course, journalistic ethics – the desire to report truthfully – may be more important than defamation standards. What we do know is that the current copyright regime chills speech; publishers refuse to rely on fair use.
Kasunic: We need to create a record on this in other ways besides litigation. Best practices, along with records of how filmmakers are being silenced, are helpful. DMCA rulemaking can also acknowledge the effects of the law on particular uses.
Sprigman: His thought – how do we know that First Amendment safeguards are generally effective? Because we have a baseline, and when Congress departs from it we no longer know whether those limits are effective. But Sprigman’s preferred baseline is 1790; when copyright gets bigger than that, we should apply First Amendment scrutiny.
Liu: Liu is puzzled by what “traditional contours” could mean, and what the baseline is. The Court almost suggests that today’s copyright is the baseline, but why would that be true? Also, it’s hard to understand what would be sufficiently outside the contours to trigger constitutional scrutiny. And it’s hard to understand where “traditional contours” comes from – why does tradition matter (when it didn’t in defamation, blasphemy, obscenity, etc., etc.)? This is why he doesn’t want to tie “traditional contours” to his argument about breathing space for idea/expression and fair use.
Bender: It should be First Amendment values, not tradition, that guide us.
Graeme Austin: What about the reverse Mein Kampf situation – the Family Movie Act. Does the FMA support or harm First Amendment values?
Kasunic: Parents can make a choice not to hear speech, which means it supports their speech.
Bender: It violates kids’ rights!
Me: There are First Amendment values on both sides – the editors are asserting their right to make someone else’s speech but use different words, which makes it easier for them to speak but can interfere with the original speaker’s message.