Last Thursday, the plaintiffs in the AutoAdmit case filed a motion for expedited discovery, seeking permission to serve subpoenas on a number of ISPs, universities, and websites demanding information about the identities of the the pseudonymous posters named in the lawsuit. (For background on this case and links to the court documents, see the CMLP database entry, Doe v. Ciolli.) The plaintiffs believe that these ISPs, institutions, and websites either assigned IP addresses to some of the defendants, were used by some defendants to send email, or have relevant IP addresses because some defendants visited their sites shortly before posting on AutoAdmit threads.
Before filing the motion, the plaintiffs informally contacted AutoAdmit and its administrators, several ISPs, hosting providers, and others, requesting that they voluntarily turn over identifying information about the pseudonymous posters. All of the recipients refused, and now the plaintiffs want the court's help to make them comply. Unfortunately for the plaintiffs, it looks like many of the recipients do not have the information that they're looking for anyway. Many indicated that they do not compile logs of the information sought or had already disposed of those logs before being contacted by the plaintiffs.
The crux of the plaintiffs' problem is that AutoAdmit itself apparently does not log the IP addresses of its users. So, not only does it look like a subpoena addressed to AutoAdmit will be futile, but the plaintiffs have no IP addresses to link to the offending pseudonyms. (The ordinary route in these kinds of cases is for the plaintiff to get an IP address for the poster from the website and then get a subpoena requesting the owner of the IP address to turn over identifying information about its subscriber.)
In their investigation, the plaintiffs' lawyers have had to adopt unorthodox techniques, including posting messages on AutoAdmit asking nicely for the defendants to identify themselves. Not surprisingly, this hasn't worked. They have also requested information from websites apparently visited by some of the defendants before posting to an AutoAdmit thread (as evidenced by links posted to a Washington Post article, a college scholarship announcement, and a court opinion). Having gotten negative responses, the plaintiffs now want subpoenas compelling these sites to produce logs of everyone who accessed the article, the announcement, and the court opinion (each on its respective website) during a week in early March 2007.
In their memorandum of law in support of the motion, the plaintiffs argue that they have shown "good cause" to allow them to take discovery before serving the defendants and holding the first discovery conference (quite reasonable, because without the subpoenas there will be no service or first meeting). They also argue that First Amendment protection for anonymous speech (and thus the heightened Cahill and Dendrite standards) is not applicable in this case because "the speech at issue in this case does not implicate the First Amendment." This is a circular argument. It seems to me that, in a defamation case or other lawsuit alleging a speech-based cause of action, you have to apply the Cahill and/or Dendrite standard in order to decide whether or not the First Amendment protects the speech at issue in the first place. The speech is protected unless it is defamatory or threatening (which basically is what Cahill and Dendrite require a plaintiff to establish through substantial evidence); you can't take away this protection by merely saying "Look how awful it is! It has no redeeming social value." That said, one can't blame the plaintiffs for calling attention to the offensive, abusive, and socially worthless character of the comments made about them.
We will be taking up the First Amendment protection for anonymous speech in the first sections of our legal guide, which will be going live later this week.