Vermont Supreme Court offers guidance on admissibility of social media content

By now it is not particularly controversial that a litigant’s social media content is discoverable.  See, e.g., Lewis v. Bellows Falls Congregation of Jehovah’s Witnesses, Bellows Falls, Vermont, Inc., 2016 WL 589867 (U.S.D.C. Vt. Feb. 11, 2016) (content of personal injury plaintiff’s “private” Facebook account, including “private messages” sent to Facebook “friends,” were discoverable); Forman v. Henkin, 30 N.Y.3d 656, 93 N.E.3d 157 (2018) (personal injury plaintiff’s Facebook content discoverable regardless of whether the plaintiff designates it as “private”); Vasquez-Santos v. Mathew, 168 A.D.3d 587 (N.Y. App. Div. 2019) (ordering disclosure of personal injury plaintiff’s post-accident social media content regardless of what device it was located on).

But what about admissibility at trial? The Vermont Supreme Court just offered some guidance on that subject in State of Vermont v. Allcock, 2020 VT 60 (July 10, 2020). In Allcock, the court in a 3-2 decision reversed the trial court’s admission of inculpatory Facebook messages authored by the defendant. The Court held that the message were insufficiently authenticated by the State.

Facts and Procedure: The defendant Christina Allcock was charged with assault on a police officer. Before trial, a witness who had received private Facebook messages from the defendant showed them to the police on his phone. The messages essentially inculpated the defendant in the assault charge. The police then executed a search warrant on Facebook, which produced information in the form of a “Facebook Business Record” verifying that the messages came from the account of “Christina Allcock.” The Record provided the account holder’s email addresses, telephone numbers, and the internet protocol (IP) addresses for her computers. At trial, the State offered the Facebook Business Record, over the defendant’s objection, as essentially self-authenticating evidence that the inculpatory messages came from the defendant.

Why the Supreme Court Held That The Messages Should Not Have Been Admitted: According to the Court, the State had offered no testimony at trial that the email addresses, telephone numbers, or the IP addresses belonged to the defendant. In other words, the State failed to properly authenticate the messages.

Analysis: As a preliminary matter, the Court addressed the defendant’s contention that there should be a heightened evidentiary standard for the admission of social media communications, and rejected that contention. The Court confirmed that in Vermont the admission of social media evidence should be analyzed the same way as any other evidence, and pointed out that, in fact, the standard for authenticating evidence “is not a demanding one,” and merely requires “evidence sufficient to support a finding that the matter in question is what its proponent claims.” 2020 VT 16, ¶¶ 10-11, 13-14.

Nevertheless, the Court held that in this case the State fell short of meeting this relatively low standard for authentication with respect to the Facebook messages in question, i.e., of establishing that the defendant was likely the author of the inculpatory messages. The Court pointed out that, like any evidence, false social media accounts can be, and often are, set up by impostors or fraudsters.

What the State Could Have Done to Authenticate:

  1. offer testimony that the IP addresses listed on the Facebook Business Record matched those of the defendant’s computers;
  2. offer testimony from the witness who received the inculpatory messages from the “Christina Allcock” account that provided a reasonable basis for believing that they actually were authored by the defendant;
  3. offer evidence that the information contained in the messages (which were sent the day after the incident) could only have come from someone who knew information about the incident that was not yet public, such as the defendant; or
  4. offer evidence that the Facebook page associated with the account in question contained information belonging to the defendant (e.g., personal photos) that was not in the public domain.

In other words, the State needed to offer something more than the unadorned Facebook Business Record, to establish the probability that the messages sent to the witness were actually sent by the defendant.

Conclusion: The Supreme Court does not say so explicitly, but implies that any of the simple steps mentioned above would have gotten the State over the authentication hurdle and justified the trial court in admitting the evidence and letting the jury assess it. Id., ¶ 25.

On the face of it, it seems like the messages obviously came from the defendant and that the Court is elevating form over substance in holding that they were improperly admitted. On the other hand, it is true that the State simply offered the Facebook Business Record as basically self-authenticating and apparently did nothing to show that the messages came from the defendant, as opposed to, say, a Facebook impostor trying to inculpate the defendant. It easily could have, and should have, done something more. The fact that this was a criminal case, and not merely a civil trial, may have motivated the Supreme Court to demand something more from the State in terms of authenticating the evidence than just “phoning it in,” so to speak. Nevertheless, the Court’s mandate that some level of authentication is required for the admission of social media evidence, and the Court’s suggestions on how to accomplish that authentication, should definitely be heeded by civil practitioners as well.

Also note that this decision has no bearing on the discoverability of social media evidence, and civil practitioners must continue to pursue the discovery of such evidence wherever appropriate.