Today I was preparing a brief in opposition to a subpoena served on a client seeking information to identify certain anonymous third parties. The argument against such business is pretty well established, as Paul Alan Levy and I had the experience of arguing together a little while back. The up-to-date version of the argument, which of course also includes disclosure about anonymous bloggers, goes something like this:
The Internet is a public forum, and First Amendment rights fully apply to communications over the Internet. The Supreme Court has treated the Internet as a forum of preeminent importance because it places in the hands of any individual who wishes to express his views the opportunity to reach other members of the public who are hundreds or even thousands of miles away at virtually no cost. Concomitantly, courts have also recognized the right to express oneself online anonymously (even at considerable social cost, it can be argued — and I’ve argued that side of it, too).
Numerous courts have enunciated standards to meet these due process concerns and govern the identification of anonymous Internet speakers. In the leading case on the subject, Dendrite v. Doe, 342 N.J. Super. 134 (App. Div. 2001), a company sued four individuals who had criticized it on a Yahoo! bulletin board and sought discovery of third parties to unmask their indentities. The court in Dendrite set out a five-part standard for cases involving subpoenas to identify anonymous Internet speakers:
1. Give Notice: Require reasonable notice to the potential defendants and an opportunity for them to defend their anonymity before issuance of any subpoena;
2. Require Specificity: Require the petitioner to allege with specificity the speech or conduct that has allegedly violated its rights;
3. Ensure Facial Validity: Review each claim in the complaint to ensure that it states a cause of action upon which relief may be granted based on each statement and against each defendant;
4. Require An Evidentiary Showing: Require the petitioner to produce evidence supporting each element of its claims; and
5. Balance the Equities: Weigh the potential harm (if any) to the petitioner from being unable to proceed against the harm to the defendant from losing the First Amendment right to anonymity.
Similarly, in Doe v. Cahill, 884 A.2d 451 (Del. 2005), the Delaware Supreme Court ruled that a town councilman who sued over statements attacking his fitness to hold office could identify the anonymous posters only upon a showing that he was not proceeding in bad faith and could establish that the statements about him were actionable because they might have a defamatory meaning. Though it rejected the balancing component of Dendrite, the Cahill case on all elements of a defamation claim that ought to be within his control without discovery, including that the statements are false. Other appellate and trial courts that have addressed the issue of subpoenas to identify anonymous Internet speakers, as well as several federal district courts, have adopted variants of the Dendrite or Cahill tests.
I didn’t say they all did, though, although I might well have thought so — and I might have been right until April 4, 2013, when the Court of Appeals of Michigan said “no” in a case called Thomas M Cooley Law School v John Doe 1. The Westlaw version of the opinion, which does not appear yet to include any proprietary copyrightable material of West, is at the link.
And what did the Court of Appeals of Michigan have to say? The language that impressed me is in bold below, but you have to read the whole excerpt here to get the point. The case involves criticism of Thomas M. Cooley Law School as a “diploma mill,” for reasons well known to people who follow such things. The court, now:
This appeal concerns the complicated interplay between First Amendment protections of the freedom of speech and the Michigan Court Rules concerning discovery. Plaintiff Thomas M. Cooley Law School (Cooley) filed a complaint in Ingham County against Defendant John Doe 1 (Doe 1) alleging defamation arising from statements that Doe 1 made on a website that, under a pseudonym, criticized Cooley. Doe 1 moved the trial court to (1) quash a subpoena that Cooley obtained in California seeking his identity, and (2) issue a protective order. Doe 1 now appeals as on leave granted an order of the court denying his motion to quash the California subpoena. He argues that the First Amendment’s protections for anonymous free speech shield his identity. . . .
After extensive reasoning, the trial court determined that there was no Michigan law on point and looked to decisions from other jurisdictions, in Dendrite and Cahill. The trial court determined that, in order to adequately protect Doe 1’s interests in remaining anonymous, it must balance those interests against Cooley’s interests in holding Doe 1 accountable for defamation. . . .
In a lengthy ruling from the bench, the trial court ruled that Michigan law does not address the situation in this case. It therefore adopted the Dendrite standards. Applying those standards, it determined not to quash the subpoena. We disagree with the trial court’s conclusion that Michigan law does not adequately address this situation. We conclude that Michigan procedures for a protective order, when combined with Michigan procedures for summary disposition, adequately protect a defendant’s First Amendment interests in anonymity. . . .
Under Cahill, which the Ninth Circuit recognized is the “strictest test,”65 the plaintiff must (1) allege the exact defamatory statements, (2) show that the defendant had notice of the action, (3) show it could survive a motion for summary judgment on the prima facie case, and (4) show the balance of equities between the defendant’s interests and its interests weighed in its favor.
But under Michigan law, the plaintiff must allege the exact defamatory statements. The plaintiff will have to survive an actual motion for summary disposition on its claims under MCR 2.116(C)(8). And the trial court may consider the weight of the defendant’s First Amendment rights against the plaintiff’s discovery request when determining whether to issue a protective order. Thus, the Dendrite and Cahill standards largely overlap with Michigan’s combined safeguards of a protective order under MCR 2.302 and the summary disposition standards and procedures under MCR 2.116(C)(8).
To the extent that Doe 1 urges us to adopt Dendrite because it more adequately protects other interests or is better public policy, we decline to do so. Doe 1 argues that any less stringent standards may chill internet criticisms because of the fear of being required to defend against a lawsuit for long enough to have the trial court dismiss it. Doe 1 also argues that the plaintiff in a defamation case may sue the defendant solely to subpoena the defendant’s internet provider for identifying information in order to acquire leverage for extra-judicial retaliation. We have concluded that Michigan rules of civil procedure adequately protect Doe 1’s constitutional interests. We decline to reach beyond what is constitutionally necessary to judicially create anti-cyber-SLAPP legislation. Such decisions of public policy are the province of our Legislature. And the writing, or rewriting, of our discovery and summary disposition rules is the province of the Michigan Supreme Court.
Judicial restraint lives in Michigan! (Well, maybe not in federal court, or maybe yes. But I meant the state courts.) The opinion even cites Occam’s Razor in addressing an obvious concern of litigants seeking to prevent the enforcement of such subpoenas:
We recognize that this opinion does not address the extreme case, a case that Doe 1 would like us to consider. The extreme case is one in which a plaintiff in a defamation case sues the defendant solely to subpoena the defendant’s internet provider for identifying information, in order to retaliate against the plaintiff in some fashion outside a court action. . .
It is this extreme case that both Dendrite and Cahill, through their notice provisions, address by providing some protection to persons in [the hypothetical] situation. But, we emphasize, this is not the case before us. Here, Doe 1 knew relatively early on that Cooley had filed suit against him and was attempting to ascertain his real name through its subpoena to Weebly[.com]. And Doe 1 was successful, at least to date, in preventing a public disclosure of that name. We therefore decline, under the well-recognized concept of judicial restraint [fn: See Occam’s Razor and the Principle of Parsimony: Simpler explanations are, other things being equal, generally better than more complex ones] to go beyond the facts that are before us in this case. We do not issue advisory opinions, nor does the Supreme Court, except in very limited circumstances not present here. We believe that our legal system in Michigan is capable of responding, either retroactively through litigation or prospectively through Supreme Court rule-making, if and when the extreme case arises.
So in Michigan, the first party seeking to quash a subpoena under these circumstances who does argue that he the court is presented with the “extreme case” — that party loses, I think, big time, if anonymity is of any value to him — once that’s gone, after all, it’s gone, no matter how much “retroactive” litigating you do. John Doe #2, perhaps, gets the benefit of Supreme Court rule-making, according to this opinion; but first John Doe #1 has to have his cover blown.
Is that the outcome? Not necessarily. The court did say, after all, “We do not issue advisory opinions” and perhaps, when presented with that fact pattern — and an able demonstration of how important the balancing of interests under Dendrite and Cahill, where the facts are different — the Michigan courts will reconsider its rejection, in this case, of what can fairly be called a nationwide judicial trend toward such balancing before the harm is done.
Is doing so really something bey0nd the scope of what a judge can and should do in Michigan under the doctrine of judicial restraint? Is the management of discovery in light of First Amendment concerns really a “matter of public policy” that only the legislative branch can address?
I hope not. This case is actually rather complex procedurally and technically, as the ultimate ruling makes “clear” (i.e., I can’t understand without spending more time reading this opinion than I can justify):
We conclude that the trial court abused its discretion when it denied Doe 1’s motion for a protective order after it adopted and applied foreign law. Michigan law adequately protects Doe 1’s free speech interests in this case. On remand, the trial court should determine whether it has the power to quash a California subpoena. If not, or if it declines to do so, the trial court should apply Michigan law to determine whether Doe 1 is entitled to an order protecting his identity.
Reversed and remanded. We do not retain jurisdiction.
I’m not alone in having concern. Judge Jane M. Beckering, dissenting in part, writes as follows:
I agree with my colleagues in the majority that the only remedy available to Doe 1, because his identity is known to Cooley, is a protective order and that the trial court, on remand, must evaluate the necessity of a protective order. As noted by the majority, and contrary to Cooley’s argument, Cooley’s knowledge of Doe 1’s identity does not render Doe 1’s appeal moot. It is possible to fashion a remedy, a protective order, if merited, that will have a practical legal effect on the controversy. A protective order can prevent Doe 1’s identity from being disclosed to others. I also agree with the majority that we may, and should, review the issue whether Michigan law adequately protects the respective rights of plaintiffs and defendants in the complicated interplay between the First Amendment right of anonymous free speech and a person’s right to know the identity of his or her defamer. The issue is a matter “of public significance that is likely to recur, yet evade judicial review.
Where I diverge from the majority is in its conclusion that Michigan law adequately protects a defendant’s right to anonymous free speech except for the “extreme” case. Because an anonymous defendant cannot undertake any efforts to protect against disclosure of his or her identity until the defendant learns about the lawsuit—which may well be too late given that discovery is available to a plaintiff as soon as the action is commenced—we, like numerous appeal courts in other jurisdictions, must adopt a formal procedure that balances the rights of plaintiffs and defendants. The majority of jurisdictions that have addressed this issue have adopted either the Dendrite or Cahill standard. These standards require, in part, that a plaintiff alleging defamation present the trial court with prima facie evidence on the elements of its defamation claim before it is allowed to discover the anonymous defendant’s identity. I would adopt a modified version of the Dendrite standard.
Judge Beckering doesn’t just say that… she says that, in a dissent that appears to be about as long as, or longer than, the majority opinion. Perhaps her dissent will be the basis for an application for review of the decision to the Michigan Supreme Court at some point all, alternatively, a request by some interest persons or group — however this works — for the Supreme Court rule-making adverted to in the majority opinion.
Meanwhile, Michigan, by being the only court I know of to actively reject adoption of either Dendrite or Cahill, may just become the world capital of anonymity-busting subpoenas. It’s not the world capital of much else these days… but things do change. I anticipate that Michigan’s law on this matter will, too.