Courts, Injunctions, and WikiLeaks

As widely reported, Judge Jack Weinstein is holding a hearing tomorrow in his Brooklyn courtroom concerning Eli Lilly’s efforts to enjoin further distribution of internal documents related to its antipsychotic medication Zyprexa. The documents were originally obtained by an Alaska lawyer, who subpoeaned them from a doctor who possessed them in connection to his work as an expert witness in an unrelated class action case against Lilly. The documents were supposed to be sealed in the class action case, but thanks to the efforts of this Alaska lawyer they were soon in the hands of a New York Times reporter and a lot of bloggers. They spread fast over the internet, sometimes posted through anonymizing mechanisms such as Tor. The New York Times summarizes all the background here, and TortsProf Blog has a series of link-filled posts accessible through Technorati here. EFF is representing an anonymous blogger who subsequently linked to leaked documents that were, by then, available on the internet; the group has collected materials about the case here. This is fascinating on so many levels, it’s hard to know where to start. So I’ll just stick with two points, one legal and one technological.

First, the legal point. It is highly instructive to think about this dispute against the backdrop of the landmark case Seattle Times v. Rhinehart, which involved a defamation and privacy suit against the newspaper over critical reporting about a fringe religious group. In the course of normal civil discovery for the defamation case, the defendant newspaper obtained all sorts of internal documents on matters such as finances that would have helped its investigative reporting of the religious group. The Supreme Court unanimously upheld a court order preventing the newspaper from disclosing the details of material uncovered in discovery. The court found it was not an unconstitutional prior restraint. (The order there included the important qualification that the newspaper could publish the same information if it was learned through different channels; I see no indication of any such alternate sources of information in the Zyprexa flap.)

Seattle Times seems to me to stand for the simple proposition that just because information is produced in discovery it doesn’t necessarily belong to the public domain. (For non-lawyers: it is important to realize that most discovery materials are not filed with the court or even used by counsel in their arguments. Often huge volumes are produced and very little of it becomes relevant to the case.) So, some of the sweeping rhetoric we see in the Zyprexa controversy — about sunshine, open courts, the public right to know, and the like — is just plain wrong. This material was sealed by court order. The expert witness who leaked the documents (and whom Lilly alleges colluded with the Alaska lawyer to cook up the subpooena) could and probably should get in a lot of trouble. Yes, there is an argument that documents seeming to show misdeeds of a pharmaceutical giant are in the public interest, but change the facts and see how you feel: what if the court order protected embarrasing private information about an individual whose opponent leaked them onto the internet maliciously? How else can we judge the balance of public interest and private litigants’ interests if not by relying on the courts entering these protective orders?

Fine. That takes care of the initial leak. But what about the second, technological point? In our networked world, the court is basically impotent to contain the leak once it occurs. There is something faintly ridiculous about the order at issue here. Judge Weinstein (who is, btw, a very famous and highly respected federal judge) listed 17 people and entities that were banned from disseminating the leaked documents, and then concluded his list with Needless to say, an injunction against a wiki is a little like an injunction against the whole world. And nothing in his order reaches the multiple other sites, some of them wikis or otherwise peer-produced, that have since mushroomed up to post or link to the documents, often anonymized by Tor or otherwise. Once the cat is out of the bag in a networked world, there is little the court can do. Or, as TortsProf Blog memorably headlined it: “Judge Tries to Unring Bell Hanging Around Neck of Horse Already Out of Barn Being Carried on Ship That Has Sailed.”

If this strikes you as a very bad thing, then once again change the facts and see how you feel. It just so happens that the Washington Post reports this morning on Wikileaks, a new effort to create a central, peer-produced wiki for anonymously posting government documents, particularly those exposing corruption and human rights abuses in repressive regimes: is a Web-based way for people with damning, potentially helpful or just plain embarrassing government documents to make them public without leaving fingerprints. Modeled on the participatory, online encyclopedia Wikipedia, the site is expected to go live within the next two months.


The site relies on a worldwide web of volunteers and contributors to post and vet the information, and dodge any efforts to shut it down. To protect document donors and the site itself, Wikileaks uses its own coded software combined with, for the techies out there, modified versions of Freenet and PGP.

So, how can we have a world where the good of helping anonymous dissidents in China can coexist with the (arguably) bad of leaking confidential court documents or even private information the same way? How can we decide what information is in the public interest and what information rightfully remains sealed? That’s a big question. The general answer is that we must rely on sensible, nuanced, and rigorously checked systems at the disclosure points. More specifically, to go back to Zyprexa, courts should enter confidentiality orders with care and under well-considered standards to balance protection of the public interest, privacy, and the smooth functioning of discovery and other functions of the civil litigation system. Sealing orders should not be, as too often they are, routine ministerial matters. Finally, there should be real penalties, not wrist-slaps, for violating a conscientiously considered order.

And what does a judge do when the cat/bell/horse/ship is already out/rung/free/sailed? Hold a hearing. Punish the initial leakers if they acted in bad faith. And throw up his hands. I predict that is what will occur in Brooklyn tomorrow.

3 Responses to “Courts, Injunctions, and WikiLeaks”

  1. fascinating post. as for your final point–there’s some interesting case in the First Circuit where the judge essentially essentially tried to use Rule 65 to craft an injunction “against the world” prohibiting the distribution of certain code.

  2. I’m curious about “change the facts and see how you feel.” If the facts are about a company causing widespread harm to the public, that would make the leak more in the public interest than if the facts were embarassing personal information leaked by someone with a vendetta.

    If there is an exception (or legal protection) for things done in the public interest, how can these facts be exchangeable?

    moreover, I’d be wary of making a parallel between an individual and a corporation (as in embarrassing facts about an individual and facts from a corporation). There seem to be several different kinds of rights and harms implied depending on who is harmed. For one, are we talking a right of privacy or a trade secret? Aren’t those different?

    I think the most valuable point implied is that these technologies are not only good, and that more access to and ability to share information is not always harmless. But would seem a pity to equate all the kinds of harm, especially in terms of what should be more or less defended against.

    Must the law do so?

  3. Excellent questions, Larisa.

    Just to restate it, all I am defending in my first point is the proposition that the judge needs to make a careful balancing choice about whether to seal discovery documents, precisely because, as you rightly point out, not all situations are the same. There are many variable interests in different cases — sometimes privacy, sometimes trade secrets, usually the protection of the smooth functioning of the discovery process itself, always the public interest. There can’t be one simple rule.

    So when I say “change the facts” I just mean that folks opposed to Lilly on these issues may be fairly comfortable defending this leak, but deplore the next leak in a different case. That’s why my conclusions are that (1) judges should weigh the many interests at stake in sealing orders very carefully and (2) once those orders are entered, they should be enforced vigorously. If someone acts unilaterally to flout a carefully-considered confidentiality order and releases documents the court has sealed (rather than, for example, urging the court to change the order), then that person should get into serious trouble.