A high-tech works-in-progress colloquy for intellectual property papers
View the existing comments and add your own comments by clicking the “comment” link above. Comments are moderated and may take some time to appear.
I thought I would get us started by offering Professors Hughes and Kwall my deep thanks for their extensive comments on my draft paper. I am humbled at their generosity. It is, I must say, characteristic of the giving spirit I have regularly encountered among my fellow i.p. professor; it is one thing I love about the field.
I also want to thank Professors Gibson, Cotropia, and Osenga for honoring me with the chance to serve as the first author in this workshop series. I hope they won’t regret it!
My plan, over the next day or two, is to offer some thoughts on some of the many engaging points Bobbi and Justin have made. Rather than corral those thoughts into one document, I’ll share my thoughts as I compose them (with what I hope are helpful headings).
Is Feist a question about constitutional originality, or statutory originality?
Justin expresses concern that I have overstated the degree to which the Supreme Court’s decision in Feist – the white pages case – is about constitutional originality, rather than statutory originality. Hughes at 6.
I don’t think I have, and here’s why:
It’s true, as Justin says, that the Court devotes considerable discussion to the fact that the Copyright Act tells us that originality can be found in the manner that a compilation work’s factual content has been selected, coordinated or arranged. Feist, 499 US 340, 356-59 (1991). And the Court tests the originality of Rural’s white pages by looking at just these issues – selection, coordination, arrangement. But that’s not the core outcome in the case.
The Court states its holding in constitutional terms: “The selection, coordination, and arrangement of Rural’s white pages do not satisfy the minimum constitutional standards for copyright protection.” 499 US at 362. See also 499 US at 346 (“Originality is a constitutional requirement.); 351 (“As we have explained, originality is a constitutionally mandated prerequisite for copyright protection.”); 362 (“As this Court has explained, the Constitution mandates some minimal degree of creativity ….”). In other words, Rural’s white pages fall below the constitutional minimum. The Court did not hold, so far as I can see, that the statutory originality standard was then, or must forever remain, exactly at the constitutional minimum. And if the case is centrally about statutory originality, all the talk about the constitutional constraint seems quite out of place.
For what it’s worth, others have reached the same conclusion about Feist’s constitutional (and not statutory) status. See, e.g., Heald, The Vices of Originality, 1991 Supreme Court Review 143, 144. That suggests, at the very least, that my view on this point isn’t unreasonable.
Finally, an important second-order point: Assume I’m wrong, and that Feist’s core holding is that Rural’s white pages fell below the statutory minimum for originality. It seems to me the Court still didn’t freeze or cap originality at a minimal level. Rather, it simply held that, however low originality can go, Rural’s white pages fell below its lowest point. On this view, hoisting originality remains an option, assuming it’s justified by the contemporary settings in which copyright is claimed and enforced.
I think Joe is right that his ideas on originality do not really depend on whether Feist is only a constitutional ruling. If it’s statutory as well, then adopting his approach would require legislative action, yes — but as a practical matter I don’t see courts doing it on their own even if the current statute gives them that discretion. The Bleistein warning rings too loudly in their ears for them to start messing about with a heightened originality standard without Congress’s prompting (even if the new standard turns Bleistein on its head, as Joe’s cleverly does).
I should not overstate courts’ devotion to Bleistein, however. They have certainly not shied away from making judgments about a work’s aesthetics in other copyright contexts: “qualitative” taking in substantial similarity analysis and in fair use factor three, the “nature” of the work in fair use factor two, whether a work is one of “recognized stature” for VARA purposes, useful article’s temporal displacement test, etc.
Like Justin Hughes, I was struck by the distinctly American nature of Miller’s argument. Europeans would never think about where to set the originality bar by thinking about it in relation to the patent standard.
There are, in fact, many other countries that have higher originality standards than U.S. law does, and because of that, I am less worried than Justin is about the consistency of Miller’s proposal with international treaty obligations. He’s not suggesting that courts impose either a novelty or a nonobviousness standard for judging originality, but I found the article somewhat unclear about just what the verbal formulation of the new originality standard would be. How unconventional would the work have to be to be protectable? More than trivially so, one would assume, but does that really help sort out the protectable wheat from the unprotectable chaff? I’m not convinced. Some effort at the end of the paper to show how this approach would work in practice would be helpful. I found Justin’s discussion of the Fairey poster to be more enlightening than Miller’s in terms of how copyright should approach this particular controversy.
Like Bobbi Kwall, I have some skepticism about whether we want to adopt a standard of originality that would require expert testimony on a routine basis as to how unconventional a work was or was not. This may add a new set of costs that will burden authors more than it will contribute to the social good. It is already quite expensive to bring infringement lawsuits, and unless one registered promptly–and most don’t–authors cannot recover attorney fees.
The most curious thing to me about Miller’s article was that it seemed to be concentrating on the initial protectability question, but originality plays a role in scope questions as well. Where I see originality (and lack thereof) doing the most work in the common law of (c) is in relation to how broad or narrow a scope of protection a work has. Highly original works tend to have a broader scope than modestly original works, and that seems like a better role for originality to play in mediating what others like Fairey can do with preexisting works. I haven’t yet read through the new Parchomovsky & Stein article, but they too focus on originality as a key scope issue.
Without a requirement that works must be registered in order to qualify for copyright protection and without an examination process at the Copyright Office to weed out the merely conventional works, I don’t see that one gains all that much by saying that works are unprotectable unless they are unconventional. It won’t stop the average person who thinks of himself as a copyright owner from thinking that his work qualifies for protection. Photographers who work within conventions seem to me to be entitled to earn a living, so I’d support their claims of copyright in news photos, photos of landscapes, and the like, but just narrow the scope of protection to that which is expressively original (which may not be much in photos operating within conventions and none in photos taken by automated processes).
That being said, I found Miller’s article stimulating and thoughtful.
Joe’s article, and the commentary that has followed, is fascinating to me. It exemplifies, at least in my mind, the following broader thoughts that his Article elicits.
First, there is a fundamental, systematic difference in how the two regimes (copyright and patent law) regulate the scope of protection (the core question in all IP areas). Copyright law, as most commentators have already identified, does its adjustments at the liability question phase. It is not a question of whether the work is protectable under copyright law (it almost always is in some way), but whether copyrightable expression was misappropriated by the alleged infringer. Patent law does much more on the front-end. The heavy doctrines come into play when determining whether the invention is even patentable, not when determining whether someone has infringed. This difference in when scope is adjusted also influences the conditions under which scope is adjusted€”for copyright law it is in the context of the alleged infringement (what can they capture specifically) while in patent law it is in the context of what has been done before (how far of an advance is the invention).
Second, this prompts the question as to why the regimes choose such different stages to do their scope adjustments. Many have opined on this before–its a product of the registration versus examination system; its because of what is essentially central claiming in copyright law versus peripheral claiming in patent law. Or, and perhaps justifying the differences prompts a more interesting inquiry–do the subject matters of the regimes justifies the difference in the approaches? or is one approach is simply better then the other, meaning that the IP area using the inferior method of measuring scope should change.
Third, and I think this is the neatest part of Joe’s Article in my mind, is that the uncertainty surrounding why there is a difference in approach justifies inquiries like Joe’s–whether one regime’s approach should be adopted in the other. Joe tests this out here–do more at the protectability stage–and others have thought about patent law borrowing from copyright law€”particularly with regards to the definition of scope (like Dan Burk and Mark Lemley and Jeanne Fromer most recently).
Professors Kwall and, here in comments, Samuelson both suggest that, were we to demand more creativity as a condition of copyright protection, the Copyright Office would need to begin examining copyright registration applications seriously (by analogy to the Patent Office).
I don’t see the need for such an examination system, and thus did not seek to describe one in the paper.
Current doctrine includes a creativity requirement (albeit one that’s too undemanding, in my view), and it tests originality in court, not at the Copyright Office. At least in the near term, as courts explore a more demanding creativity threshold, there doesn’t seem to me to be any need to examine every registration application for compliance. Call it “rational ignorance at the copyright office.”
I suppose that, in the longer term, years from now, an administrative examination process might be worth the trouble in copyright. Do we need to settle that question now?
Also, it’s worth pointing out that patent examination does not provide the sort of high-impact rigor that some may imagine. It’s more like a quick look review to weed out only the most plainly unpatentable inventions.
It seems I haven’t done as good a job as I might at putting my proposal in the context of our international obligations under the Berne Convention and TRIPS Agreement.
For example, Bobbi indicates that my suggestion for an attribution right – that we provide it to those whose works fall above the constitutional originality minimum but below the statutory originality minimum – “would be inconsistent not only with Berne but also with virtually every other country with moral rights protection.” Kwall at 1. I take Bobbi’s word for it, although the text of Berne Article 6-bis seems to leave at least a little play in the joints (beginning, as it does, with the statement, “Independently of the author’s economic rights, … “). At the same time, I’m happy to invite us to think more deeply about our current conception of moral rights. Is enough-creativity-to-merit-copyright-exclusion-rights really the best minimum test of whether someone has earned the right to the social respect that an attribution right embodies? Coming to the question without Bobbi’s expertise, I confess it’s not obvious to me that the answer is “yes.” Most important, I eagerly await the arrival of Bobbi’s forthcoming book on moral rights. I have a lot to learn in the moral rights area, and more interest in it than ever before (in no small part due to Bobbi’s work in the field).
Justin knocks me on a slightly different point. I failed, as he rightly says, to pay “even a footnote of attention to our international treaty obligations.” Hughes at 7. Guilty as charged. In my defense, here’s what the footnote I thought about writing, but decided not to, would say (and will say, in the final version of the piece when published):
My proposal that we condition copyright on a more demanding show of creativity doesn’t implicate our international treaty obligations at all. The Berne Convention does not define the originality or creativity required for copyright, and certainly doesn’t cap it at a particular level. See Daniel C.K. Chow & Edward Lee, International Intellectual Property: Problems, Cases, and Materials 130 (2006) (“Originality is often thought to be the touchstone of copyright requirements, although curiously it is not expressly mentioned in either TRIPs or the Berne Convention.”). Berne does refer to “literary and artistic works,” Art. 2(1), and “it is generally accepted that a work must be classified as fulfilling the criterion of originality or creativity in order to fall within the categories of production which are within the scope of the Convention.” JLA Sterling, World Copyright Law § 7.06 (3d ed. 2008). Berne also states that compilations are protectable, so long as they “constitute intellectual creations” “by reason of the selection and arrangement of their contents.” Indeed, at most Berne rules out the use of a “sweat of the brow” test for protection, the very standard the Supreme Court rejected in Feist. See Daniel J. Gervais, The Compatibility of the “Skill and Labour” Originality Standard with the Berne Convention and the TRIPs Agreement, 26 European Intell. Prop. Rev. 75, 80 (2004) (“the text and drafting history of the Berne Convention unequivocally demonstrate that the property test of originality is that the work must embody an author’s creative input”). My suggestion that we demand more creativity is, of course, plainly in keeping with the idea that we demand some creativity (rather than none) – as Feist holds, and Berne Art. 2(5) implies. TRIPS Article 10(2), on compilations, simply incorporates Berne Article 2(5). (Ditto for Article 5 of the WIPO Copyright Treaty.) More generally, TRIPS Article 9(1) incorporate Berne Articles 1-21 wholesale. Like Berne, TRIPS provides no definition of originality or creativity for determining copyright eligibility.
[ As the foregoing indicates, I’m a bit mystified that, according to Justin, “applying [my] heightened standard would almost certainly put us in violation of least two multilateral treaties.” Hughes at 7. ]
Chris, the other patent law peep here, is on my beam.
Patent and copyright regulate the scope of protection differently. So accustomed am I to wearing my patent law hat, it seems natural to me to ask, why not do more, in copyright law, to justify an exclusion entitlement at the front end?
Current copyright doctrine feels, to me at least, as far too ready to buy the claimant’s plea at the outset. We try to back out later (modulating “substantial similarity” according to quantum of creativity; adjusting fair use factors on the same ground). But why let the exclusion claim, with all its chilling effects, go forward in the first place? This question is all the more pressing, given the doctrine of unconscious copying. (I concede, as Justin points out (Hughes at 9), that there are few cases on the doctrine. But that is hardly an adequate measure of its expression-blunting effect. Lawyers’ warning letters, and worried private decisions simply to “skip it,” fly under the radar. And a few law professors tut-tutting that there’s no problem does not make the problem vanish.)
Chris’s deeper questions – why do the two regimes regulate scope at quite different stages? is one stage demonstrably better for both? – are powerful! I suppose I’ll be pondering them for quite some time …
Along these line, Dan Burk & Mark Lemley’s new paper on returning to a central claiming regime certainly rewards a close read!
Justin cogently encapsulates, and illustrates, that “American judges strive to reward what they perceive as investment and individual effort.” Hughes at 11.
Mike Madison, in blog post at madisonian.net makes a general version of the same point: “For all of the flaws of Lockean-style labor theories of value in copyright law, those theories are right to focus our attention on the ethical dimensions of labor. … Arguments over ‘creativity’ thresholds are really arguments over those ethical dimensions, transposed for historical, cultural, and doctrinal reasons into arguments about the ‘work of authorship.’ Recognizing the philosophical and cultural appeal of labor theories doesn't mean that they should be imported into IP law. There are good reasons not to do that. But we shouldn’t forget what we're really arguing about.”
One way to think about what I’m proposing is this – We should reward labor. We can all agree on that. But what kind of labor really counts as labor, for copyright purposes? The labor entailed in consciously copying, verbatim, the expression of another doesn’t cut it (notwithstanding the fact that, technically, it does take some effort). The labor entailed in breaking through to a lifeshaking expressive genre unimagined by all but you … that’s clearly enough labor. Where, between these, is the line we should adopt? I think we should reward the labor required in exploring, and pushing out – at least a bit – the boundaries of existing convention. Such labor is a bigger risk for the creator (in that it may bomb), a more valuable work for society (teaching us more than yet another conventional bit of expression), and a juicier target for imitators who were unwilling to take the creative risk the creator did (once the creator shows that the risk was worth it).
Pam and Justin are also right to say, “Joe, give us examples.” (Others have said it too!) I have not done so, to date, because I think an honest presentation of an example requires a factually thick description – a lengthy case study – that I have not yet pulled together. I will try to do so, likely as a follow-on project.
Raising the originality standard is a fascinating proposal. But I’m not sure it gets at the problem that is motivating the paper. The paper starts off citing concerns over the expansion of copyright law into everyday experience. But that expansion is most keenly felt with respect to films, television shows, and music, not the news photographs that Miller cites as particularly problematic in Part I. Although photograph infringement cases are intrinsically interesting for copyright lawyers, they are also relatively rare. The works that ordinary individuals have the most concern about are ones that are well above the current originality threshold. Either Miller’s change in the originality standard will have no effect on those disputes, or we are talking about a significant change to the standard.
The article doesn’t seem on its face to propose a significant change — only works that are “so utterly pedestrian, conventional, and obvious for its genre and time” would fail the modified originality standard. This could either be viewed as a change in the required level of originality, or perhaps a slight broadening of the scenes a faire doctrine. But this description of the revised standard seems inconsistent with the notion that this new standard would affect much more than news photographs. In particular, Miller suggests that it would make his own article “nonoriginal.” It seems to me you could only reach that result by applying the test at a fairly high level of generality — once one article on making originality more like novelty and nonobviousness is written, all subsequent articles on the same topic are nonoriginal. As Justin Hughes notes, this would seem to make large amounts of uncontroversially copyrightable stuff noncopyrightable.
I would also think that such a change would have inevitable spillovers into the test for substantial similarity. I.e., if you measure originality at a high level of abstraction, why not substantial similarity? It would seem to make sense that if all subsequent articles on making originality more like nonobviousness are nonoriginal because they are too similar to the first one, then they are also too similar for purposes of the substantial similarity analysis. I suppose you could simply declare the tests to be different, but that would seem to create a weird limbo of works too similar to another to be copyrightable, but not similar enough to be actionable.
Bruce is right – it’s hard to get a handle on just how significant a change I’m talking about.
From one perspective, the change is quite significant, in that it’s a reorientation of what the originality inquiry is actually about, picking up the Feist nugget and seeing where it takes us.
From another, the change seems minor, in that it could be that a marked percentage of what’s considered original now would be considered original under my approach too.
A few thoughts:
My approach could well spell “no copyright” for a fair bit of music, tv, film, etc., depending on what the evidence showed in a given case about just how pedestrian/conventional the expression in question was.
As for the article itself, it’s true that some of the ideas may be unconventional. But, of course, as ideas they’re not copyrightable even now. The expression itself, by comparison, is pretty conventional stuff. Isn’t that true for most reasonably well-composed (allow me to flatter myself for a moment) law review articles? As a professional group, we stay pretty close to some quite strong conventions about expression.
Finally, on the spillover to substantial similarity – Bruce could be right. Pam suggests something similar, in a sense. I suppose the aspect of substantial similarity about which I wondered more was this: If we make my suggested change, is there as much need to calibrate protection at the substantial similarity stage? In other words, perhaps the protective reach for a given work would/should be broader, for the smaller number of works that achieve copyrighted status. (We already have that, in a sense, with unconscious copying doctrine.) This is where Chris would chime in, as he did before, about making copyright over in patent law’s image.
I think readers’ uncertainty as to the significance of the change Joe proposes is another reason to consider including an example or two in the paper. I sympathize with Joe’s desire to give examples the time they merit, and that the sort of “factually thick description” he’d like to do might not fit well in the current paper. But examples would put a lot more meat on the the bones of a very interesting proposal.
To choose just one issue, for instance, how would prior art be identified and evaluated? Suppose it’s a case involving a film. Would attorneys have to watch all existing films to see if the disputed film uses innovative camera angles, lighting, focus, blocking, props — not to mention dialog, plot twists, jokes, characters, and other features that could derive from a wide variety of non-film media? (Part of this question is Justin-ish or Jamie-Boyle-ish: at what level of granularity does the analysis take place?)
I think if readers had a greater sense of the answer to this sort of question, they could better weigh Joe’s proposal against alternatives, like varying similarity analysis depending on the originality of the work or using other front-end devices like formalities to reduce the number of copyrighted works.
I’ve laid out the proposal as a common law, case-by-case process. That has implications for implementation, and I will try to be more clear about this in future drafts.
When we ask, “what level of granularity?”, the answer, at least in the near term, is “the net resulting level of (a) the level of granularity the defendant uses to try to persuade the court, and (b) the level the plaintiff urges in response.”
So, for example, consider a sort of film example: the recent documentary that captured a bit of the Simpsons show on a t.v. playing in the shot. (This example has been written about, I think, in some of the American University Law School or Duke Law School materials.)
Is the Simpsons clip copyrightable expression? To analyze that question using my approach, the documentarian would probably consider a range of contemporary pop-culture comedic animation. Fox, for the Simpsons show, might urge a much narrower slice of animation, in an effort to highlight what’s unconventional about the clip. If it came to litigation, a judge would have to decide what time slice and genre slice is a reasonable backdrop against which to judge the creativity reflected in the captured clip.
Messy? You bet. Especially early on in the process. But I’m confident, still, that this is at least asking the right question. (I’m also confident that the option Jim mentions – “all existing film” – probably won’t strike anyone as a meaningful genre. It is too big, and we can all see that it’s too big.)
So, accused infringers: Start making the arguments. Start building a strong fact record for your trial judge!
Well, folks, the time has come to close down this iteration of the Virtual Workshop. Our thanks to everyone who contributed, especially Justin, Bobbi, and of course Joe, whose willingness to be the guinea pig for this new initiative is very much appreciated.
If anyone has suggestions about how we might change the process next time around, please let one of us know. In the meantime, stay tuned for the next paper!
Mail (will not be published) (required)
Powered by WordPress and K2
Entries Feed and Comments Feed
57 queries. 0.365 seconds.