I’ve moved part of this post from my earlier post on the ongoing Vermont divorce saga. This post has two parts: analysis of Judge Devine’s decision on the restraining order, and analysis of how strong a fair use defense Krasnansky could mount for his use of excerpts from Garrido’s journal / diary.
1. Judge Devine’s Decision: Or, A Little Copyright Knowledge Is a Dangerous Thing
Hat tip to Thomas O’Toole, blogging at TechLaw. He’s got a copy of the judge’s ruling (with filing date of 14 January 2008). The judge rules that Krasnansky’s “self-generated prose” and “intermittent leakage to the internet of real-world names and locations” is “‘interference with the personal liberty of [Garrido]‘ only in the most strained sense.” Thus, the harassment rationale for the initial restraining order fails.But the judge goes on to make some serious errors about Krasnansky’s use of excerpts from Garrido’s journal. He writes that Krasnansky’s “appropriation and unauthorized reproduction of [Garrido's] diary, accomplished for the announced purpose of publicizing her theretofore-private writings, is something distinct from speech. It is conduct, pure and simple.” This is just dead wrong as a matter of law. Garrido has a form of property rights in her journal’s writings, but they are statutory rights limited by copyright. Her ownership rights to the physical journal differ from her rights to the expression within its words – my Copyright Law class just finished covering this distinction. (The judge notes this distinction, stating that “Wife has a property and possessory interest in her books and papers as both tangible property and intellectual property,” but fails to grasp that these interests have very different legal entitlements attached to them.) This is an error of analogy: downloading one of Metallica’s songs from a peer-to-peer network may be copyright infringement, but it’s not theft – we have different rules for physical property versus intellectual property.
And the judge also misstates the relevant law governing the case: common law copyright doesn’t apply; the federal Copyright Act does. See 17 USC 301(a). (The judge cites a case from 1968, which is before the relevant federal law was adopted in 1976. He hedges a bit, noting the possibility of pre-emption, but seems then to be uncertain.) Once Garrido fixes her words by inking them on the journal’s pages, she has a copyright interest in them (but, note, she can’t file a lawsuit for infringement until she registers that copyright with the Copyright Office). For both these propositions, read what the Supreme Court has to say in Harper & Row v. Nation Enterprises, 471 U.S. 539, 552-55 (1985) (noting that the 1976 Copyright Act protects unpublished writings, and that the right of first publication under section 106 is subject to the fair use limits of section 107).
It’s also worth noting that once we’re in the zone of a copyright claim, the family law court clearly lacks subject matter jurisdiction; this might explain the judge’s efforts to paint possible infringement as “theft.” (See 28 USC 1338(a).) (Vermont’s rules for its family courts can’t trump Congress’s judgment about jurisdiction for copyright claims – that’s why the Constitution has the Supremacy Clause.)
The judge states, in a conclusory fashion, that posting the excerpts onto the blog crosses “the line from speech into conduct.” But it’s not that simple. All speech requires some conduct – even speaking requires a physical act. But we don’t treat information questions so formalistically. (See Eugene Volokh’s piece on speech as conduct, and generally U.S. v. O’Brien, 391 U.S. 367, 376-77 (1968) (stating that “when ‘speech’ and ‘nonspeech’ elements are combined in the same course of conduct, a sufficiently important governmental interest in regulating the nonspeech element can justify incidental limitations on First Amendment freedoms” but that “the quality of the governmental interest [must be] compelling”)). When we’re dealing with intellectual property in the form of copyright, we have a carefully crafted system to balance First Amendment interests (fair use) with possessory interests. See Eldred v. Ashcroft, 537 U.S. 186, 219-21 (2003) (noting that “copyright law contains built-in First Amendment accommodations” such as the idea-expression dichotomy and fair use and suggesting that “copyright’s built-in speech safeguards are generally adequate to address” the situation where “speakers assert the right to make other people’s speeches”).
In the end, the family court rightly acknowledges that copyright questions are beyond its ken (which, unfortunately, the opinion makes rather plain). But, if this is the case, what on earth is the judge doing issuing a restraining order based entirely on copyright issues?
2. Fair Use, Privacy, and Reluctant Authors
Let’s assume that Krasnansky appeals Judge Devine’s order and that at some point (either on appeal or, if the order is reversed, that Garrido files a claim for copyright infringement in federal district court) fair use comes up. Fair use is an affirmative defense to copyright infringement codified at 17 USC 107. It’s intended to draw upon the common law history of fair use and presents a non-exclusive, four-factor balancing test to determine whether a given use of copyrighted expression is fair. (Law students hate this type of standard; IP law is full of them.)
Again, the Salinger case is instructive here. A biographer quoted from J.D. Salinger’s unpublished letters (in violation of a contractual agreement that gave him access to the letters, but prohibited unauthorized quotations) in a forthcoming book. Salinger sued to block the book. The Second Circuit agreed with Salinger and enjoined publication, placing heavy emphasis on the fact the letters were unpublished and on the Supreme Court’s reasoning in Harper & Row v. The Nation. (The Supreme Court declined to hear an appeal by the biographer’s publisher, Random House.) For the Second Circuit, there was nearly an ironclad rule preventing fair use of unpublished material. While the court acknowledged that fair use applied even for unpublished stuff, it made clear that this fact tipped the analysis heavily in favor of Salinger. (Of course, the text of Section 107 points out that “The fact that a work is unpublished shall not itself bar a finding of fair use if such finding is made upon consideration of all the above factors.”)
This is bad for Krasnansky: Vermont is in the Second Circuit, and I’d guess Salinger would weigh heavily in the federal district court’s analysis.
Just for fun, let’s look at the four factors. First: purpose and character of the use. Krasnansky isn’t making commercial use of the excerpts, which helps him. (No Google ads on his blog.) He could argue he’s using the excerpts for “criticism,” though he’s actually criticizing the author and not the work. (Criticism is one of the Section 107 purposes that is more likely to be “fair.”) I think a court might, even if not overtly, view his use as somewhat unseemly – the blog seems mean-spirited – and that could hurt him. Overall, my analysis is that this factor tilts somewhat for Krasnansky.
Second: nature of the copyrighted work. It’s unpublished, which tilts against fair use, especially in the Second Circuit. The work is more factual than fictional (I presume – most diaries aren’t inventions, unless you’re dealing with the Clinton Whitewater scandal), which cuts for fair use. (Facts can’t be protected under copyright, though the particular expression of them can be.) On net, probably somewhat for Garrido.
Third: amount and substantiality of the excerpts relative to the entire work. We don’t know how long the journal is (though apparently it covers seven years of the marriage), or how vital the quoted passages are. If they’re the “heart of the work” (Harper & Row), that cuts against fair use. (Note that this is tricky: both Harper & Row and Salinger assess this factor based on commercial appeal. There’s no indication Ms. Garrido intends to sell her journal, nor what parts of it would be most attractive to consumers. So, we’re a bit in the dark here.) I skimmed Krasnansky’s blog briefly (hard to do, given the writing, which is, um, verbose) and the quotes are neither insignificant nor a substantial share of the content. I think this factor is neutral, but it could tip slightly either way.
Fourth: effect of the use on the value or potential market for the work. This is really hard. As Frank Pasquale notes, this analysis tends to be circular: whether there is a market depends on whether there’s a legal entitlement to license or assign, not vice-versa. Like Salinger, Ms. Garrido gives no evidence (that I know of) of wanting to make commercial use of her journal. Unlike with Salinger, there probably isn’t an active market for rights to her journal (though the NY Times piece might have launched some interest). We’re also a bit hamstrung here – Garrido likely wouldn’t license the right to quote from her journal to Krasnansky at any price. (As Wendy Gordon writes, fair use can be viewed as a remedy for market failures such as these. No one wants to license criticism of their work.) And to the (probably considerable) degree that Krasnansky’s blog makes Garrido seem less appealing, that harm isn’t cognizable under copyright law: criticism doesn’t count as harm, though damage to licensing potential through copying certainly does. I tend to think the effect on the value or market is minimal, because the market itself is minimal; thus, this factor is neutral or tips slightly for Krasnansky, but I suspect a court in the Second Circuit might disagree.
Overall, I think the fair use factors point to a close call, though like most academics, I’m pretty pro-fair-use, so I’d likely give a tie to Krasnansky. But two important factors cut the other way. One, Salinger looms in the background, and I think it would press both the district court and, if necessary, the Second Circuit to rule against fair use. Two, Krasnansky’s use seems unappealing. Judge Devine referred to it as “strange,” and that’s mild. Whatever the relative equities in the divorce, the excerpts’ incorporation in the blog looks a bit mean, and I think that would hurt Krasnansky’s fair use case – after all, the Section 107 analysis is a bit of a hodgepodge, and the court’s overall impression of the use matters a lot.
This case gets more fascinating by the hour. Stay tuned.