Yes there are appeal issues with the Kohberger defense survey ... but they aren't what you think
The questions aren't the problem. Silencing them is.
Over the past weeks, we’ve learned that Bryan Kohberger’s defense team has been surveying residents of Latah County to evaluate whether media coverage in the county has created a biased jury pool. But Prosecutor Bill Thompson learned of the survey and obtained an emergency order halting it because he objects to the questions and argues that they violate the nondissemination order. In the course of its telephone survey of 400 residents of Latah County, the defense surveyor asked about specific pieces of information that had been reported in the local media:
Have you read, seen, or heard if Bryan Kohberger was arrested at his parents’ home in Pennsylvania?
Have you read, seen, or heard if police found a knife sheath on the bed next to one of the victims?
Have you read, seen, or heard that DNA found on the knife sheath was later matched to Bryan Kohberger?
Have you read, seen, or heard if Bryan Kohberger owned the same type of car reported on video driving in the neighborhood where the killings occurred?
Have you read, seen, or heard if the cell phone tower data showed that Bryan Kohberger made several trips near the victims’ home in the month before the killing?
Have you read, seen, or heard if the university students in Moscow and their parents lived in fear until Bryan Kohberger was arrested for the murders?
Have you read, seen, or heard if Bryan Kohberger said he was out driving alone on the night of the murders?
Have you read, seen, or heard if Bryan Kohberger stalked one of the victims?
Have you read, seen, or heard if Bryan Kohberger had followed one of the victims on social media?
(If you want to watch the April 4 hearing where this issue first came to the court, I livestreamed it on YouTube here - the survey questions are at 7:52)
In the State’s view, the survey questions are suggestive and prejudicial, violate the gag order’s restrictions on talking about potential evidence, and taint the jury pool by exposing the information to potential jurors who had not already heard it. From the defense perspective, they are necessary to evaluate the effect of media coverage on the jury pool, limited in reach (approximately 1% of the county population was surveyed), and the specific questions are derived from media reports that are already public record.
(We also got to hear a lot more about the survey and why it was constructed the way it was from the defense expert who conducted the survey at the follow-up hearing on April 10 - available here)
I have some thoughts about the decision the judge has to make based on how I’d be considering the problem as an appellate attorney if this case landed on my desk.
Restrictions on attorneys’ speech must comply with First Amendment standards
In the U.S., we presume that anybody gets to say anything they want to anybody else. There are exceptions to the First Amendment, but they are few and they are specifically identified in the case law - things like defamation, incitement, true threats, speech that is inherently criminal (like perjury or fraud), obscenity and child pornography, and fighting words. The First Amendment applies to lawyers speaking about their cases - “lawyers’ speech” is not a recognized exception to the First Amendment. So, like with all speech, we start from the presumption that lawyers can say what they want to whom they want, including asking any questions they deem appropriate, and the government cannot restrict their right to do so.
However, even the valid exercise of First Amendment rights can be limited - for example, the government can place time, place, and manner restrictions on exercises of free speech. In the case of lawyers, courts have recognized that their speech about pending cases has the potential to jeopardize the fairness of the trial process and, for that reason, can constitutionally be restricted until the trial is over. The standard for restricting attorney speech was adopted in Gentile v. State Bar of Nevada, 501 U.S. 1030 (1991) prohibiting attorneys from making:
an extrajudicial statement that a reasonable person would expect to be disseminated by means of public communication if the lawyer knows or reasonably should know that it will have a substantial likelihood of materially prejudicing an adjudicative proceeding.
You may recognize this language from the Revised Amended Nondissemination Order filed in the Kohberger case on June 23, 2023:
The order then goes on to spell out specific categories of information that are specifically prohibited under this restriction, including “The identity or nature of evidence expected to be presented at trial or any sentencing phase of the proceedings,” which Mr. Thompson has argued the survey questions violate. However, because the court’s power to gag the attorneys is limited by the First Amendment standard, the restrictions have to be interpreted as limiting the specified statements if and only if they will have a substantial likelihood of materially prejudicing Bryan Kohberger’s trial. If the survey questions don’t meet this standard, then it is a First Amendment violation to prevent the surveyor from asking them.
What does it mean for a statement to have a substantial likelihood of materially prejudicing the trial?
Gentile discusses the standard as a totality of the circumstances test that considers the content of the statements, their timing, and their reach. In that case, a criminal defense attorney called a press conference following the return of an indictment in a highly-publicized case to state that his client was innocent, the likely perpetrator was a police officer, several witnesses lacked credibility due to deals they cut with the prosecution, and the police officer suspect might have been videorecorded under the influence of cocaine. His client was later acquitted at trial, and the State Bar association disciplined him for the comments.
The facts that the information reported had already been published before, the population from which the jury panel would be drawn was large, and the statement was made about six months before trial, led the U.S. Supreme Court to conclude that there was no substantial likelihood of material prejudice resulting from the statements:
___________
Applying that standard here calls into substantial question whether the survey questions satisfy it. They are based on information that was already reported (indeed, that’s why the questions were asked); they were asked of only 1% of the county population, leaving a large pool from which prospective jurors can still be called; and the trial date has not even been set yet but will not be earlier than at least a year from now, at which time the jury pool will be different (as people move to and from the county) and the details of the questions are likely to be forgotten.
Again, if the survey questions don’t meet the substantial likelihood of causing material prejudice standard, the judge violates the First Amendment by prohibiting the questions or penalizing the defense for asking them. Further, in the context of this survey, the First Amendment violation is impairing Mr. Kohberger’s ability to present a defense - he can’t develop the factual record he needs to meet his burden in a change of venue motion if he’s prohibited from talking to jurors about their media exposure. A constitutional violation will unquestionably create a potential appeal issue, and under the general standard for constitutional errors, the burden would be on the State to prove beyond a reasonable doubt that the violation did not affect the outcome. In other words, the State would have to convince a reviewing court that limiting the survey questions had no effect on the defense’s ability to measure the extent to which potential jurors in Latah County had prejudged Bryan Kohberger’s guilt and demonstrate that a change of venue was necessary for a constitutionally fair trial.
OK, but why did they have to ask these questions? Why couldn’t they have asked more general questions about media exposure and ability to be fair?
Many critiques of the questions that I have seen are pointing out, with some merit, that the questions could be suggestive because they may be presenting potential jurors with information that they had not previously heard, leading them to incorporate that information into their ideas and preconceptions (conscious or unconscious) about the case. So why did the questions have to be framed this way? We heard some technical explanations from the defense expert in the April 10 hearing - you can’t measure exposure to and impact of false information without asking about the false information, and the risk of taint is low because the people answering the case-specific survey questions already knew many details reported about the case. But there is also a legal answer that lies in the standards Ann Taylor must satisfy as a death-qualified capital defense attorney.
Capital defenders in Idaho are required to comply with the ABA’s Guidelines for the Appointment and Performance of Defense Counsel in Death Penalty Cases. These standards include duties to investigate thoroughly and to assert meritorious claims on the client’s behalf.
In the context of pretrial publicity and its effects on juror bias, the ABA has long advocated for exactly the kind of “content-based” questioning that was conducted here. Their Criminal Justice Standards state:
The reason is simple - to ensure an unbiased jury, it is essential to know what information jurors have been exposed to about the case so that its effect on their attitudes can be explored. Jurors are not psychologically inclined to assert that they cannot be fair, their assessment of their own fairness may not be realistic or credible, and they may simply be unaware of the ways in which the information has colored their perspectives. Thus, whether a juror has been biased by pretrial publicity is a question the court should decide for itself on the basis of a factual record - courts should not simply take the juror’s word for it. The ABA articulated this position recently in an amicus curiae brief submitted in the U.S. Supreme Court case of U.S. v. Dzhokhar Tsarnaev, the Boston Marathon bomber:
Any general questioning that did not inquire about specific content exposure would fall short of these standards and invite arguments down the road that defense counsel’s assistance was ineffective because it did not satisfy the prevailing norms and standards for eliciting proof of juror bias. In other words, Bryan Kohberger could potentially argue that Ann Taylor was ineffective if she did *not* ask specific questions about the media coverage.
________________________________
What about the gag order?
At the April 10 hearing, Judge Judge clarified that when the nondissemination order says in subsection 2(b) that the attorneys may discuss information that’s in the public record, he just meant the public case file, not all information that is publicly reported and discussed. But this “clarification” may have made the situation worse.
Under the First Amendment, content-based restrictions on speech must be narrowly tailored - they must not be over-inclusive, meaning broader than necessary to serve the interest of protecting Bryan Kohberger’s right to a fair trial, and they must not be under-inclusive, meaning too narrow to reach speech that is contributing to the problem. So when a restriction uses language that is vague or ambiguous - susceptible to more than one interpretation of what it covers - that restriction can be challenged as overbroad, meaning it fails to give fair notice of what is allowed and what isn’t. Overbroad restrictions are problematic because they risk chilling speech that is allowed - people will self-censor and avoid saying things they are allowed to say because they’re afraid of being punished. To avoid this chilling effect, it is critical to define the restrictions clearly and precisely.
Judge Judge has effectively admitted that he didn’t do that because the language he used - “information in the public record” - means something narrower than what the words themselves appear to say. The term “information in the public record” doesn’t mean information that’s publicly disseminated and discussed, it means only a specific category of information that’s publicly disseminated and discussed, namely, that public information that is derived from the court file and proceedings. As such, the speech the attorneys are allowed to engage in under the order is narrower than the language of the restriction appears to permit. This makes the language of the order over-inclusive.
But that isn’t the only problem - the narrowed restriction on speech about matters that are being publicly reported and discussed would itself present a problem under the Gentile standard. Again, to restrict the attorneys’ speech, that speech must present a substantial likelihood of materially prejudicing an adjudicative proceeding. But as everyone seemed to agree in the April 10 hearing, the threat to a fair trial comes from false information being reported and disseminated without being corrected. In fact, the prosecutor and the judge explicitly stated that the reports that Mr. Kohberger stalked one or more of the victims and followed them on social media were not true in an effort to correct that misinformation. Indeed, the judge appears to be leaning toward permitting the survey to continue as constructed with the addition of language in the introduction to the content-specific questions clarifying language that the media items being discussed may not be true or accurate. Since the media can’t be gagged or prevented from reporting information that is not true, the corrective action to preserve Mr. Kohberger’s right to a fair trial is more speech, not less. Thus, for the attorneys to publicly address false media reports about the case is not only not likely to prejudice the trial, it is the only way to measure and mitigate the prejudice that flows from the reporting of false information about the case.
The bottom line
Gag orders are always troublesome. In this case, the gag order does not seem to be based on a clear, mutual understanding of exactly what speech is prejudicial and why. I understand why, from the defense perspective, it was necessary to stop the flow of false information attributed to law enforcement sources or other sources presumed to have inside information. I also understand why the State would not want the defense to play tit-for-tat with its own investigation. But the parties and the court have struggled from the inception of the case with the First Amendment problems posed by trying to shut people up, revising the order several times to narrow its application and specify its reach. On the one hand, Judge Judge has said that the order is simply co-extensive with the ethical rules the attorneys must already follow - in which case, you have to wonder why we are bothering with an order at all. On the other hand, that doesn’t appear to be the standard that’s actually being applied here - not once has the judge or anyone in the courtroom articulated the Gentile standard.
On my reading of the law, if the judge is going to fiddle with the survey or restrict the questioning in any way, that interference has to be justified under the Gentile standard or there is a potential appeal issue for a First Amendment violation. He will need to make specific factual findings justifying a legal conclusion that the language of the survey creates a substantial likelihood of material prejudice to the trial process. And those findings will have to account for the factors that Gentile told us matter - the length of time until trial, the limited number of people reached compared to the jury pool as a whole, and the repetition of information that has already been reported in the media. Because I don’t think this standard can be met, the only way I see the judge avoiding a potential appellate issue is to resist the temptation to micromanage the defense investigation and vacate his temporary order halting the survey.
Since the majority, if not all, of the questions derive from the charging affidavit that the prosecutor Bill Thompson publicly shared and told everyone to go forth and multiply, I would think they would meet Judge Judge's the gag order explanation that you mention here: "Judge Judge clarified that when the non-dissemination order says in subsection 2(b) that the attorneys may discuss information that’s in the public record, he just meant the public case file, not all information that is publicly reported and discussed." Isn't the charging affidavit a public case file? (oh no, not THAT public case file! 😆)
So, precedent implies that such surveys are almost necessary in order to assure a fair trial in some cases.