Dark Law: Published Michigan Precedents on Protection Orders Missing from Westlaw

Three cases, including the two leading Michigan precedents on the First Amendment and restraining orders, are invisible to lawyers who rely on Westlaw. Lexis is also missing two of the three, but it's changing its policies to include them. And a federal statute is behind this.

[UPDATE: The problem was corrected a week after this post, on Feb. 22, 2019.]

Patterson v. Beverwyk (2017), Buchanan v. Crisler (2018), and TM v. MZ (2018) are precedential decisions of the Michigan Court of Appeals. The first two are already in the Michigan Appellate Reports advance sheets; the third will presumably be there soon (it was only decided in October). Patterson sets an important precedent about whether businesses can get personal protection orders (known in Michigan as PPOs). Buchanan and TM set important precedents about First Amendment limits on protection orders that purport to limit unwanted public speech about the plaintiff.

But all three are missing from Westlaw. This means that, as a practical matter, those cases are likely invisible to those lawyers who rely on Westlaw rather than the print reporters. Judges and litigants as well as lawyers are unlikely to benefit from the existence of those precedents. And I suspect that judges who wrote the precedents thought they were making law that would be found by other appellate judges, by trial court judges, and by lawyers—but the effect of the precedents is now sharply limited. Indeed, perhaps this is why TM, though dealing with much the same topic as Buchanan, doesn't cite it.

Why does this happen? It's a complicated story, that starts with an overbroad and possibly unconstitutional federal statute, moves on to an overcautious state policy, and then to what strikes me as a misguided decision by Westlaw.

[1.] In 2006, Congress mandated (in 18 U.S.C. § 2265(d)(3)) that states may not "make available publicly on the Internet any information regarding the registration, filing of a petition for, or issuance of a protection order … if such publication would be likely to publicly reveal the identity or location of the party protected under such order." (Any information like this is supposed to be kept "in secure, governmental registries for protection order enforcement purposes.")

The impulse behind this, I take it, is

1. to protect the beneficiaries of the protection orders from the people against whom they got these orders (that's the "or location" part of the rule), and
2. perhaps to protect the beneficiaries' privacy in some cases where the orders arise from sexual abuse or similar behavior.

But the statute extends even to information that the restrained party already knows, such as the protected person's identity, and even in cases involving circumstances that are no more private than those present in many publicly accessible criminal or civil cases. Moreover, the law is a content-based restriction on states' speech, which might mean that it violates the First Amendment, depending on whether states have First Amendment protections from the federal government (a matter that is surprisingly unsettled).

[2.] Some states have concluded (plausibly, given the text) that this statute bars Internet posting of appellate court opinions, including the use of parties' names in those opinions. Several states therefore have a policy of using pseudonyms or initials to refer to any protected parties in court opinions; Massachusetts, for instance, is one example, and indeed Michigan sometimes does this as well.

Many other states, as best I can tell, just barrel on without any regard to this statute, at least when it comes to their appellate opinions; California is an example of that.

But the Michigan Court of Appeals goes the furthest of all: It completely declines to post on its standard opinion release pages any opinions in which someone had gotten a protective order—though of course, it still publishes such opinions in print, since they are precedents. (Some of these opinions are available on the Michigan government site, but not in the usual location, and I suspect only by accident.)

[3.] Now the last puzzle piece: Westlaw and Lexis. They of course are not bound by 18 U.S.C. § 2265(d)(3), since they aren't state governments. And their job, it seems to me, is to distribute at least all the binding precedents, since those are law that lawyers need to see.

To be sure, for nonbinding decisions, they may choose not to post certain orders that courts themselves decide not to post (for instance, certain Colorado Court of Appeals unpublished opinions, as I learned recently). Likewise, they may go along with court decisions to retroactively seal certain opinions, on the theory that if the court wants an opinion sealed, they ought to accommodate the court's preference.

But here the Michigan courts want these decisions to be published, as binding precedent. They are labeled "for publication." They are printed in the Michigan Appellate Reports (which are, as it happens, published by the same company that owns Westlaw). Michigan courts expect them to be findable by Michigan lawyers; they just read 18 U.S.C. § 2265(d)(3) as barring the Michigan government from itself posting them online. And many lawyers rely on Westlaw and Lexis, precisely because of the usual completeness of their collections, rather than separately searching (and separately subscribing to) the Michigan Appellate Reports.

Westlaw's explanation, then, doesn't really work. They told me in an e-mail,

Our policy at Thomson Reuters is to align with the policy of the courts. We include on Westlaw those opinions that are published by the court on its website. If the court does not post selected opinions due to privacy statutes or court rules, we adopt that same approach and do not publish the opinions on Westlaw.

Regarding your question about opinions in restraining order cases, the relevant court rule that guides the approach on dissemination of the opinions is Michigan Court Rules, Rule 3.705 (MIRSPECPMCR3.705). Specifically, Subrule 3.705 (C) provides: "(C) Pursuant to 18 U.S.C. 2265(d)(3), a court is prohibited from making available to the public on the Internet any information regarding the registration of, filing of a petition for, or issuance of an order under this rule if such publication would be likely to publicly reveal the identity or location of the party protected under the order."

I like Westlaw a lot, and have generally gotten excellent service from them; but here I think they are mistaken. The policy of the Michigan courts cannot be, in 2019, to keep their precedent off the online research services on which lawyers rely. When Michigan courts choose to make certain opinions precedential, and to print them in case reporters, it makes no sense for Westlaw to exclude them from its online collection.

Lexis's response about the two cases that it's missing (Buchanan and TM) strikes me as much more sound:

Our content acquisition team investigated these issues today and determined that these documents were not collected initially due to notices that came up on the court's website that indicated the opinions (and associated data) in these cases were not available online per Michigan Court Rules…. Previous court contacts confirmed the documents had been "suppressed," and were not available. However, one of our editors contacted the court today and the court confirmed that while the documents are not available via the website, they were published and available from the court via written request. We are adjusting our process to account for the court's policy regarding suppressed opinions…. We are working with the court to quickly identify any additional documents that may have been excluded on this basis and will add them online, as appropriate.

[* * *]

In any event, though, whether you think Westlaw is right on this, one thing is clear: Michigan lawyers and judges should know about this policy. My impressionistic reaction from chatting with a few Michigan lawyers is that this policy is not widely known, so I thought I'd note it.

And more broadly, this situation shows how privacy-based speech restrictions can have unexpected indirect effects. (For more examples, see here and here.) In this instance, I think that Michigan courts and Westlaw could and should have avoided those effects; but still it seems to have been the federal statute that got the ball rolling, even if others could have stopped the ball.

Westlaw tells me they're unaware of any other states in which recent precedents are similarly invisible, though if you are aware of some, please let me know.

Disclosure: I filed an amicus brief and argued on behalf of amici in TM v. MZ; indeed, I learned about the subject matter of this post because someone who saw the TM opinion sent my invaluable pro bono local counsel Michael F. Smith a copy of a brief in a related case, which cited Buchanan. I was surprised that I hadn't seen Buchanan in my research—and then figured out that I hadn't seen it because it was missing from Westlaw.