Balloon Dogs à Faire

February 2, 2011

Jeff Koons has been pursuing an interesting course of action lately.  Koons, who is primarily known for his appropriation artwork, created a sculpture that consists essentially of a solidified form in the shape of a balloon dog.  It’s actually a very cool sculpture, exhibited at the Met in NYC.  And now?  Now he’s suing people who make balloon dog bookends.

As always, there’s a few issues here.  First, it should be noted that Koons has some experience with being on the receiving end of several infringement lawsuits.  And losing, generally (though he did win a recent one).  While people have accused Koons of some level of hypocrisy in bringing this suit, I don’t blame him – if he’s going to be harmed by the law, he ought to take advantage of it where he can.  More importantly, there is some level of distinction (if you assume Koons genuinely believes that Park Life is copying HIS balloon dog, instead of, as I will address below, the concept of the balloon dog) here – Koons historically has engaged in transformation of his appropriated works, and this seems to just be direct infringement for profit.  His motives for bringing the suit may be slightly cynical, but because of these differences (and, perhaps, because of my own basic view on the appropriateness of cynicism) I don’t see it as hypocritical or unfair.

Second, and this is really the major issue: does Koons really have a workable copyright here?  His balloon dog sculpture meets that bare minimum of originality required, it’s fixed in tangible form, so there’s no real creation issue.  Ignoring the subject matter, we have to assume there’s a copyright here.

But.  There are a couple limiting doctrines in copyright needing some thought: merger and scenes a faire.

Scenes a faire is a fairly simple concept – there are certain elements that are so integral to a conception of certain things, and in a way so generic, that you can’t copyright them.  Although you don’t have to use them, they’re kind of core.  Nazis singing in a beer hall in a WW2 movie, numbered Swiss bank accounts and ridiculous gadgets in a spy movie, three-chord structures in pop music – these are just “the way things are expected to be”.  Scenes a faire.  Now, if you were to say “balloon dog” to just about anyone, it’s going to conjure up a very specific image – the exact image Koons relied on in creating his sculpture.  Under scenes a faire, you need some fairly exacting levels of copying of this type of work for it to be infringing.  If you examine the Koons sculpture linked above and the Park Life bookends you’ll probably see a few differences.  Just at a glance, the tail is quite different (a stub versus the extended sting of the Koons work), general shape (slightly more rounded), and the angle of specific components varies between the two.  Analogous to Ets-Hokin v. Skyy, the underlying item/idea is being expressed in ways that are different enough to not be infringing.

Merger is the other applicable doctrine.  Insofar as there is only one way (or a very limited number of ways) to create a balloon dog, merger prevents the copyrighting of those expressions in order to protect the ability to express the underlying idea (a dog made from a balloon).  You can’t copyright a representation of a poker hand or a chess board.  You can’t copyright the mathematical expression of the mass-energy equivalence implied by the Special Theory of Relativity.  These ideas have limited ways in which they can be expressed, and copyright in those expressions would effectively prevent any use of the underlying ideas.  (If you don’t understand why I keep using expression and idea, you should read about the idea/expression distinction in copyright law.)  Arguably, the idea of a balloon dog merges into the expression in such a way that you can’t copyright it, or can only copyright it in very limited ways – in this sense, merger and scenes a faire reflect different aspects of the same general concern, though merger may be less applicable here insofar as (as evidenced in Ets-Hokin or looking at the two dogs side by side) there are fine gradations of expression available.

In the end, Koons probably doesn’t have a very good case here; there’s enough difference, and enough hostility to the notion of the idea of a copyright in the balloon dog shape, that his claim has to fail.  Koon’s claim in video form?

(On a final, non-legal note – I like the idea of the judge or jurors taking a trip to the Met in order to get the best possible view of the Koons work.)

While I generally like Andrew’s journalism, I have to point something out as simply incoherent in his defense of Rand Paul’s appearance on Rachel Maddow.

Specifically, I’m referring to his statement “But was the Act in many respects an infringement of freedom? Of course it was.”

Well.  Yes, of course the CRA was an infringement of restaurant and hotel owner’s freedom to discriminate against their customers.  But not passing it would have been an infringement of freedoms too (or more correctly giving license to an ongoing infringement).

Let’s take it away from the ideologically loaded sphere of discrimination for a brief hypothetical.  Let’s say I own a piece of property.  My neighbor has an easement allowing him to walk across it without my permission.  Isn’t that easement an infringement of my freedom to do what I want with my property?  At the same time, were we to legislatively terminate his easement, isn’t that an infringement of his freedom to walk across my property?

Now, as a thought experiment, let’s say that the CRA set of rules barring such discrimination was our baseline.  Essentially, let’s assume we live in the world we live in today.

Let’s consider what would happen if we were to repeal the CRA.

Would it be any less an infringement of freedom to repeal it, infringing on private people’s freedom to choose the restaurant or hotel that they desire to eat in or stay at?  It’s an infringement of freedom to use coercive government force to prevent discriminatory choice by proprietors of institutions.  It’s equally an infringement of freedom to use coercive government force to enforce these discriminatory choices (by way of lending its authority to the choice of a property owner to expel such a customer).

I don’t pretend this is a particularly new insight; it isn’t.  It’s been written about for nearly a hundred years – see Robert Hale’s “Coercion and Distribution in a Supposedly Non-Coercive State” for an example (which was written in response to an early essentially libertarian viewpoint).  But to pretend that the CRA somehow increased the amount of government coercion going on simply ignores the baseline problem.  To make that assertion requires assuming that the infringement on freedoms that the “initial” state of affairs requires in order to operate are not, in fact, infringements.

Legally speaking (and this is coming from Campbell v. Acuff-Rose, the 2 Live Crew case, which is pretty much the case any parody issue involving copyright will turn to immediately for resolution): Parody is the use of portions of a copyrighted work to make a comment or criticism *reflecting on that work*. Satire is the use of the work to make a comment or criticism of a larger societal issue. Parody is near per-se protected under fair use; satire isn’t, having to fall back on the §107 factors.  The line can be hard to draw; here are some examples.

Campbell v. Acuff-Rose: Pretty Woman’s lyrics changed to criticize the notion of womanhood presented in the original song. Parody: reflects on the original.

OJ/Cat in the Hat case (I forget the cite): Dr. Seuss’s Cat in the Hat trade dress and patterns used in a book about the OJ Simpson trial. Satire: use of work to talk about an external issue.

Mattel v. MCA Records: Mattel sues Aqua/record label for use of their trademark “Barbie” in their song “Barbie Girl”. Parody: the song comments on the image surrounding Barbie and invokes it to criticize/comment on the unreality of it.

Parks v. LaFace Records: Rosa Parks sued Outkast for their use of her name/likeness rights in their song “Rosa Parks”. Not protected – the use of her name was simply to invoke her image for their own monetary purposes, there wasn’t any use of it to criticize or comment in any way. Not even a satire.

Essentially, the Downfall memes are *not* parody because they use the original copyrighted work as a vehicle to comment on some external issue (for example, DMCA takedown notices), not the original work. A hypothetical Downfall meme criticizing the concept of Hitler embedded in the movie would likely fit under the legal parody definition, but most of the ones I’ve seen wouldn’t. They may still be covered under fair use, but the analysis gets more complicated.

Essentially, fair use is a balancing test, weighing four factors against one another; the purpose and character of the use, the nature of the work, the amount taken, and the effect on the original. A quick fair use analysis example, using a hypothetical Downfall meme video talking about Ben Roethlisberger being a rapist. I’ll also note: fair use is widely recognized as an INCREDIBLY murky area of law. Other people may conclude differently than I have, quite reasonably. For example, the EFF thinks this is a clear case of fair use; I think it *should* be, but isn’t clearly so under the current state of the law.

Purpose and character of the use: this is also often known as the transformativeness factor, in that the degree to which the original has been used to contribute new expression or new aesthetic value is transforming the original into a new work. It also invokes commercial/noncommercial distinctions. Here, the uses are non-commercial (by the user, although you have to think about how Youtube stands to benefit from commercializing them, and whether that’s going to come into the analysis) and the work is transformed to some degree, although not extensively. This factor probably weighs weakly in favor of the defense.

The nature of the work: both works are expressive works of art. This factor is neutral; neither side wins. (If one work had been more factual, or if political speech was involved it might weigh weakly to one side or another, but generally fair use analysis ignores factor 2 to a pretty significant degree.)

Amount and substantiality of the piece taken: this is both a quantitative and qualitative evaluation. The quantitative amount taken is fairly low; 4 minutes or so out of 120 minutes, percentage wise, is not huge. However, it’s identified as one of the climactic scenes in the movie, so it does have some substantial amount of meaning in the original work. Probably a push, maybe weakly in favor of the plaintiff.

Finally, the effect on the original factor. Here is where (along with transformativeness) I feel existing law really has it wrong in terms of how widely they’ll look for an effect. That said: this is probably unlikely to have much effect on the market for the original. There’s no real reason to believe the filmmaker is going to want to license that four minute segment for these purposes, so no reason to believe this is going to harm a potential market. As such the real question is whether the clips will raise, lower, or not affect demand for the original. Historically I think this would have been looked at as harmful to the original – people will watch this and potentially be less likely to watch the original (looking at the video trailer cases, for example, though those are not perfectly on point) – but I think a modern court might be more receptive, if not perfectly so, to the argument that this is actually increasing demand for the original by making people aware of it. However, I’m not sure that that argument will be bought, and as such I’d say this is probably a push at best, possibly weakly in favor for the plaintiff. And that’s why I say it’s unclear/arguable what their real fair use status is – the weighting factors are going to come out near neutral (especially since courts tend to embrace factors 1 and 4 to a much greater extent than 2/3) and as such, an actual suit on the topic could go down either way.

(Complaints by users who’ve suffered takedown that YouTube isn’t protecting them ignore the fact that YouTube only receives their safe harbor from being sued because they do comply with takedown notices; the proper alternative is to file a counter-notification. After Universal v. Lenz, there’s a lot of incentive for companies not to file improper takedowns, and a decent incentive for users to counter-file.)

Rights-holders will tell you that a statutory license (also known as a compulsory license) is an infringement on their sacred property rights in their creative work.  I mean, of course they will – it hurts their economic interests to have one of their rights taken away from their own control and placed into federal government control, right?

Not exactly.  That viewpoint relies on the assumption that they have that right inviolate prior to the statutory license.  In the case of copyright, that viewpoint is often false as a matter of law – the rights in intellectual property explicitly being granted by the federal government (or, prior to 1972 for sound recordings, by the states), those rights often did not exist prior to grant, and as such if the grant is accompanied by a statutory regime for licensing, the characterization placed by rights holders of one of their rights being removed from their control is inaccurate.

However, when the right was a pre-existing grant, or an implied right that originates from some other explicitly recognized right, the characterization looks a lot better.  And when we cast it in the parlance of property, that some right of value to the right-holder is being taken away from them, our sympathy increases.

Should it?

It is not, I think, a new insight that property is no less a set of rights granted by the government than intellectual property.  I first ran across the notion in 1930s pieces by Robert Hale (Coercion and Distribution in a Supposedly Non-Coercive State), discussing the coercive nature of property, and Morris Cohen (Property and Sovereignty), discussing the idea that property is not about rights over things but rather rights against others regarding things.  In this sense, all property is intellectual property – the ability to make some use of your property does not convey you the right to use it in that way (copying a copyrighted work, or creating a nuisance on your land), nor does having the ability to prevent others from using your property allow you to do so legally (DMCA exceptions to technological measures, the defense of necessity in trespass).  All property rights flow not from ownership of the piece of property, but from recognition by the coercive authority that such rights exist.

So, if we have this notion, that property both real and intellectual is not inherent, and that the two are equally created notions flowing from coercive authority, why is it we can create a statutory licensing regime for intellectual property but not real property?

I can offer two reasons.  First, we do, in fact, create statutory regimes in real property.  We are willing to create zoning ordinances that necessitate allowing certain uses of ostensibly private property.  We have a Takings Clause that is, essentially, statutory licensing of the coercive sale of land and its valuation in cases of sale to the government, and a Due Process clause that denotes the procedure for modification and removal of property rights.  The primary reason we limit statutory licensing of real property is because of resource contention issues (analogous to the tragedy of the commons) – unlike intellectual property, there is one and only one piece of land at 1776 Sovereignty Lane.  However, as noted, some limited statutory license regimes against real property do exist.

Second, we lack guidance, if not authority, in the case of real property.  Unlike intellectual property, we lack some equivalent to the Intellectual Property Clause, to guide us in *why* we might limit these rights.  Arguably other clauses of the Constitution provide this guidance – if nothing else we must try to “promote the general Welfare” – but none in quite so clear a relation as that in the IP Clause gives us.  We are reluctant enough (albeit partially because of the absolutist perception of property we attach to intellectual property) to limit the rights in intellectual property based on promotion of progress; much less so when we don’t have that clear a goal with respect to real property.

A side note: we are less reluctant to consider such exemptions in the case of trademark law and patent law, though still reluctant, than we are in the case of copyright law.  A possible reason for the greater willingness to provide exemptions in trademark and patent is that these are explicitly based in economic incentives (although trademark is derived from consumer confusion rationales, those rationales are themselves derived from an economic basis – the desire to increase commerce by allowing trademarks to serve a valuable signaling function, a function which is eliminated in the event of widespread trademark infringement).  Where the economic rationale is the reason we protect the right, we readily accept economic rationalizations for limiting the rights.  Contrast that with copyright.  Although we say copyright is economically motivated, that’s only a first level analysis.  We often dress up moral rights for authors in economic terms, and the goal explicitly stated in the Copyright Clause is not to maximize value to the creators of “Science”, but rather to promote the “Progress of Science”.  We attempt to maximize creation (and yes, creation can also be categorized economically – but in doing so, we potentially run into a version of the utilitarian moral monstrousness problem).  As a result, explicitly economic motives often ring false, especially when they seem to cut against creation of new works.

Alan Siegel gave a lovely TED talk discussing the role of simplicity in legal agreements.  More specifically, he calls out the health care reform bill.  Why, when we have a 16 page Constitution with which we govern our country, do we need a 2100 page health care bill?

Well, we don’t.  We could have a 16 page health care bill, filled with generalities, and trust in courts or administrative agencies to fill in the blanks.  Just like with our Constitution.  We don’t have a 16 page Constitution, in practice.  We have a 16 hundred thousand page Constitution – and that number is probably conservative!  All of the decisions interpreting Constitutional provisions, the scholarship surrounding the Constitution, all of this goes into how our country is run, just with respect to the Constitution.  Ask a constitutional scholar – the crowning glory and the fatal flaw of the United States governing document is the same, that generality and flexibility.  Entire constitutional theories are generated primarily to provide better interpretative guidance, with less flexibility, out of those lovely generalities.

So, yes, we could write our health care bill that way.

Should we?  The complexity of the health care bill is partially an attempt to remove that kind of flexibility in order to retain the specificity the legislature intended.  Of course, some portion of it is a result of interest group negotiation, but that’s democracy.  In the end, the complexity of bills and legal language in general is not an attempt to avoid simplicity so much as an attempt to promote finality and clarity in a different way – not necessarily easy to understand initially, but with only one possible interpretation.

Whether that attempt works… is a topic for another time.

We find Pat Choate discussing patent infringement.

He objects to “efficient infringement”, a sub-species of efficient breach.  Essentially, like any corporation, various companies will add up the potential cost of infringing a patent (including an estimate of what they will be forced to pay and the probability they are forced to do so), total up the potential profits, and if the profits outweigh the costs, they breach (in this case, they breach the social contract of the patent system) – intending the whole while to pay if found liable in a court of law.  Where a patent holder’s valuation of a technology and a product creator’s valuation of the same technology vary widely, where (after accounting for the transaction costs of negotiation and litigation) the likely cost of infringement is lower than a patent holder’s asking price for use of a patent, efficient infringement is the result.

Now, you may personally feel that some moral opprobrium should attach to this.  But it’s not a new observation that the law does not feel the same.  Justice Holmes, in his influential turn of the century article “The Path of the Law”, noted that to the ‘bad man’ there is no difference between paying a fine for doing something he was not allowed to and paying a fee to be allowed to do something.  While the notion of efficient breach wasn’t truly refined until the law and economics movement picked it up three quarters of a century later, the essence of it is stated there.  To Intel, to Microsoft, to large corporations in general, infringement is often economically efficient.  To attach morality to this, as Choate does, reads into the law something that simply isn’t there – we don’t legislate morality.  It isn’t immoral to infringe a patent, simply illegal.

Beyond this, the assertion that companies are knowing infringers is at best questionable.  Though the problem of submarine patents has lessened with changes to the law, it still exists – often companies who infringe don’t know of the existence of the original patent.  One of the reasons those large companies are advocating for changes to the system that make willful infringement proof a more stringent requirement is because they’re so often subject to suits by non-practicing entities.  These companies are often some of the largest patent holders in the world – IBM is usually the top recipient of new patents in any given year and has an incredibly valuable patent portfolio.  They’re not interested in making infringement impossible to prove – they’re often the entity suing in infringement cases, and patent litigation typically represents a profit center for them.  They are interested in limiting the scope of willful infringement – for example, as represented by the modifications in the Seagate case, voiding the requirement of a company to conduct expensive patent searches before going ahead with use of a technology, and allowing good faith belief in the invalidity of a patent (as represented by opinion of counsel) to serve as a defense to willfulness.  The changes in law are intended to protect good faith actors.

Further, Choate misunderstands the remedy.  There’s already an extra penalty for efficient infringement.  If his assertion that a company knew of a patent and chose to go ahead and infringe it anyway (predicate assumptions to a situation of efficient infringement) is true in a given case (and not having followed the DataTreasury case, I offer no opinion on the validity of the statement with respect to it) there already is a remedy.  That’s willful infringement.  That’s treble damages (and, of course, the product owner must also pay the patent holder for their patent rights going forward or remove the patented technology from their product).

It may feel wrong to defend a corporation’s infringement, but it’s generally not a bad idea to try to limit the scope of willfulness in patent law – allowing it too much play will do anything but “promote the progress of . . . useful arts”.

The Pennsylvania school board case got me thinking – what’s the real meaning of Kyllo’s general public use prong?

For those of you who aren’t law students, or who don’t follow Fourth Amendment precedent, Kyllo was a search case which invalidated warrantless police usage of a thermal scanner to surveil a house.  The test used in Kyllo had two prongs – the traditional reasonable expectation of privacy, and the newer “technology not in general public use” prong.  That latter remains somewhat under-explored, and this case may provide an interesting way to do so.  (A note: I do not believe the school district plans to argue that Kyllo does apply here, though I could be wrong.  This is purely thought-experiment.)  The operative language from Kyllo is actions “obtaining by sense-enhancing technology any information regarding the interior of the home that could not otherwise have been obtained without physical ‘intrusion into a constitutionally protected area,’ Silverman, 365 U.S., at 512, constitutes a search–at least where (as here) the technology in question is not in general public use.”

Now, if we’re to assume that clause has function – that it was not simply a back door left for the Court to use to avoid ridiculous results where all humans have x-ray glasses but police are banned from using them – the obvious conclusion is that a technology in general public use that reaches into the previously constitutionally protected zone of the home will not be a Fourth Amendment violation if used by the police.  In a society already well on the way to being a private surveillance society (not a public surveillance society – it’s the difference between Orwell’s Big Brother and the BBC’s), societal structures leading to broader and broader usage of privacy-invading devices are common.  The zone of privacy has already begun to contract in the public sphere – look at London’s Ring of Steel, the LMSI in NYC, or the cameras in Dubai that provided video evidence of the recent assassination of Hamas militant Mabhouh.  These operate to remove the presumption of privacy when in public, and the usage of widespread public surveillance cameras is spreading.  That’s the first step.

Going beyond that, as cameras become more and more miniaturized and habits like lifelogging (continual recording of one’s life and surroundings), pervasive use of social media, and the continued intermingling of the private life with the public become more socially accepted, the notion of pervasive audiovisual surveillance in even semi-private areas becomes reasonable and it becomes hard to argue that cameras are not in general public use.

Which brings us to the point: look at your computer.

Is there a camera on it?  I’m staring at the bezel of my 3.5 year old MacBook.  Two small dots and a lens – one dot for a microphone, one dot for the “on” light.  I suspect you’re doing the same, and seeing much the same thing – laptop cameras, and even cameras on desktops, are becoming almost ubiquitous.  People use them to record their surroundings and to send their images to others.  Under any plain reading of the words “general public use”, computer-attached cameras have entered that sphere.  While police may be barred from activating the cameras for other reasons (unauthorized access to a computer, perhaps, or other remedies in statute or the privacy torts), it is hard to argue that the Kyllo standard allows us to consider actions like that of the Pennsylvania School Board as violating the Fourth Amendment.