Skip to content

Kirtsaeng: What is a “work”?

Does it advance the analytic ball re: the Kirtsaeng opinions to ask “what is a work?” in the first sale and importation contexts?

A friend and deep-thinking colleague asked me that question, knowing that I have a long-standing interest in the definition of “works,” and “inventions,” and “marks,” and “things” of all legal sorts. More precisely, the question was lobbed in my direction in this way:  If a “work” is reproduced in copies in the US for US consumers, and also reproduced in copies in another country for that country’s consumers (let us say, Thailand, just for argument’s sake), then is the Thai “work” the same work as the US “work,” or is it a separate work? If it’s a separate work, then perhaps Thai law should dictate the treatment of re-sales and exports and imports of the Thai copies, rather than US law. And where does that lead us?

Here’s my very provisional outline of an answer. Corrections and improvements are welcome!

In section 602, “work” in this context [Importation into the United States, without the authority of the owner of copyright under this title, of copies or phonorecords of a work that have been acquired outside the United States is an infringement of the exclusive right to distribute copies or phonorecords under section 106, actionable under section 501.”] has to mean “original work of authorship fixed in a tangible medium of expression,” I think, which means at least a “work” as defined according to US standards (standards in other countries are sometimes a little different and sometimes very different). Does it also mean a “work” that is protected under US law (US authors and Berne signatories)? I think that has to be right, too. What would section 602 really be intended to apply to in that case? Not a US work produced in the US; instead, section 602 is intended for a US author abroad who creates a work and publishes it simultaneously in the US and in a Berne country. (Or, a US author abroad creates an unpublished work.) Somehow, copies of that work are imported into the US without the author’s permission. 602 applies, right? Justice Breyer’s opinion for the majority tells me that this is more or less what the majority thinks that section 602 is for. The dissent, of course, starts from a different premise regarding section 602.

So change the facts so that we start with an export-from-the-US scenario rather than an import-into-the-US scenario. (That is, cousins of Quality King.) Does section 602 do anything in that scenario? Suppose a US publisher commissions a text that is prepared in the US (a “US work,” we assume, although we don’t really need to apply the statutory definition), then authorizes the printing of that work in the US (leading to zero problems jurisdictionally, for our purposes) and also in Thailand. Is that “work” now somehow a different “work” because it was reproduced in copies in Thailand? I don’t think so – I think that it is still the same work. But I am assuming that the work was first fixed in a copy in the US. It was merely transmitted to Thailand for printing.

But there’s more. I assume that Thai printing of this US work is merely a foundation for later sales. I assume that if the US work goes to Thailand for printing only, but the ultimate *and intended* destination of the copies is the US market, then we don’t have any cross-border problems if the copies appear in the US in non-endorsed ways; the publisher happened to pick Thailand for cost and convenience reasons but could just as easily have chosen Connecticut. The copyright analysis shouldn’t change. If, somehow, Thai-printed books escape the Thai factory and make their way from purchaser to purchaser to the US without having been “sold” along the way, then the publisher wins in the US on section 106(3) grounds. There is no need to resort to section 602; the copies were not authorized for sale in the first place. “Ownership” isn’t defined in the copyright statute, but it can’t be true that purchase of a purloined copy triggers exhaustion under 109. (The Kirtsaeng dissent appears to believe that section 602 is needed here because 106(3) would not be available by virtue of 109(a), but that strikes me as an overly broad interpretation of 109(a).) If the copies were “sold” somewhere, legitimately, before they made it to the US, then the publisher loses under  sections 106(3) plus 109(a) — says Quality King.

What if the publisher transmitted the US work to Thailand for Thai printing *and Thai distribution*? Is a Thai-printed copy of the US work, intended for Thai distribution, a “Thai work” that is not subject to US copyright law if that Thai work makes its way from (non-felonious) purchaser to (non-felonious) purchaser to the US? Now we are off in uncharted territory.

I think that’s a plausible way to go, though there’s nothing in the statute or, frankly, anywhere in copyright that I know of, that tells us to answer the question this way (or any other way). The usual baseline is to say that the author’s originality defines the work, and that originality is somehow eternal — once it arises, it pervades the work wherever it goes and however it is manufactured. A work that starts life as a “US work” remains a “US work” even if it is produced and/or marketed in another country. I think that’s the customary interpretation of international copyright norms. Itar-Tass Russian News Agency v. Russian Kurier has something to offer on this — a work becomes a “work” in its country of origin, but liability rules vary depending on the locus of the wrong. (More on Itar-Tass below.) So the publisher would sue in the US under US law and make the section 602 argument. The Kirtsaeng majority gives what I think is the right answer to this question on policy grounds, because section 602 was designed for a different scenario. But the text of section 602 is ambiguous; if you start with the dissent’s premise about market segmentation and price discrimination, then 602 gives you a different result.

If we go with the plausible, minority alternative — a US work becomes a Thai work if it’s manufactured in Thailand and intended for the Thai market — what does this lead to? Let’s assume an authorized sale as intended in Thailand.  If the publisher’s book intended for the Thai market is in fact a “Thai work” subject to Thai copyright law, then the next question is: Does Thailand have a doctrine of first sale? (I looked at this — — and I can’t really tell, but “first sale” or its equivalent does not leap out at me. However, the Thai equivalent of the distribution right appears to be “communication to the public,” and if the “public” limitation is meaningful, then maybe “personal” re-sales of copies are non-infringing and public re-sales need to be authorized.) If under Thai law the publisher’s rights in the Thai work are exhausted by virtue of an authorized sale in Thailand, then the publisher has no power in US courts to assert non-existent Thai rights against someone who bought copies in Thailand and brought them to the US for re-sale or re-sold them in the US. If the publisher’s Thai rights are not exhausted under Thai law, and if Thailand is a Berne signatory (as it appears to be), then the publisher can sue in the US — but under US law (106(3) as subject to 109(a), not 602), not Thai law. To be clear, this would be application of US liability rules to a Thai work, to which section 602 does not apply (by virtue of my assuming that section 602 only applies to US works). That strikes me as an Itar-Tass v. Russian Kurier problem of a second sort.

I fully acknowledge that this reading of copyright law is unconventional. The advantage of taking this second view, I think, is that it would preserve national exhaustion, if that’s an important policy goal, but at the price of some serious additional complexity in analyzing and applying the statute. My alternative reading reinforces the importance of some of the less attractive metaphysics of what makes a work a “work,” or a thing a “thing.” (I’ve skipped over an important complication, for example:  What if the publisher of the work initially fixed in the US intends to sell copies of that work in Thailand, but prints those copies in the US?  Or in the Philippines?  If we’re thinking of claiming that one creative product might become multiple copyright “works,” when and how, precisely, does that happen?)  Lots of people would prefer that these cases be decided purely on public policy / social welfare grounds, and they hate the distracting attention paid to what they regard as doctrinal niceties like “which work are we talking about?” But that distraction isn’t really a distraction at all; at the end of the day, copyright requires that attention be paid to its core concept. In patent, the name of the game is the claim, as the great judge once said.  In copyright, it’s the work.  In Kirtsaeng as elsewhere, we need a theory of the work that makes both the case and the statute make sense.  If mine is wrong — and it may well be wrong — then we shouldn’t simply assume that the alternative is acceptable by default.  What is “the work”?

2 thoughts on “Kirtsaeng: What is a “work”?”

  1. Mike, great post. I’m not sure I get why it matters whether there was a US publication though. You say “Section 602 is intended for a US author abroad who creates a work and publishes it simultaneously in the US and in a Berne country.” But why does it matter whether the author is American or whether it is published simultaneously in the US? It seems like 602 applies even to the foreign author where there’s no US publication, because the author is an “owner of copyright under this title.”

    “If, somehow, Thai-printed books escape the Thai factory and make their way from purchaser to purchaser to the US without having been ‘sold’ along the way, then the publisher wins in the US on section 106(3) grounds. There is no need to resort to section 602; the copies were not authorized for sale in the first place.” I think that’s right — under Quality King plus Kirtsaeng 602 only applies in the exceedingly narrow circumstances where pre-authorized distribution someone, like a foreign printer or a wholesaler, takes the copies, puts them on a boat, is there at the dock to receive their cargo, at which point the shipment is intercepted (by Customs or by the publisher with a TRO) before the first sale even occurs. There hasn’t been a (non-importation) distribution or infringing reproduction at that point; the only infringement is the act of crossing the US border with the copies under 602. That seems like an extremely arcane scenario.

  2. Re the “United States work” business, what I was trying to do was figure out what “works” section 602 applies to. The statute doesn’t say. I think that it’s a fair but not incontrovertible reading of the statute to infer that it applies to “works” as defined by US law, and I think that it’s a fair but not incontrovertible reading of that reading to infer that this means “US works,” as defined in Section 101. That’s why I plugged in the publication locales — because those are required by the definition of US works.

    I acknowledge up front that all of this is a bit speculative and even counterintuitive, though that’s true primarily, I think, because copyright lawyers (and scholars) so rarely bother to pause and ask themselves about the identity of the “work” that we’re talking about in any particular case. I know, for example, that the definition of a “US work” is designed to plug in to some very specific sections of the Copyright Act rather than to serve as an all-purpose definition of what works the American statute applies to.

    I also agree that my reading of 602 leaves us primarily with narrow and possibly arcane applications. Or a single, narrow application. Smuggling, in effect, would be one way to label it. I’m sure that such a narrow reading strikes many people as an implausible account of Congressional intent. I’m not so sure about that. In a lot of ways, we talk about IP piracy using metaphors that were suited initially to liquor regulation. Anti-bootlegging laws, for example. In the same vein, of course, in the alcohol context, anti-smuggling legislation has a long and distinguished history, and it’s not unreasonable to think that Congress might have framed border-management questions in IP more or less as it framed border-management questions for liquor. In any case, the Kirtsaeng opinions tell us, at the least, that there is no widespread agreement regarding what Congress thought it was doing. [And here I picture the bridge crossing scene in the Untouchables movie with Kevin Costner. Is Maria Pallante a modern Elliot Ness?]

Comments are closed.