Technology companies and some legal scholars are relying heavily on the Supreme Court’s 1979 decision in Herbert v. Lando to argue that the First Amendment precludes legislatures from imposing transparency mandates on social media companies. This reliance is misplaced. As Genevieve Lakier and evelyn douek observe here, Herbert was a case in which the Court endorsed discovery into editorial decisions. While Herbert obviously didn’t endorse the kinds of legislative transparency mandates we’re debating today, the case can’t fairly be read to have foreclosed such mandates, either. What Herbert suggests is that the constitutionality of a legislative transparency mandate probably turns on the extent to which the mandate burdens speech, and on whether the government has a good reason for imposing that burden. No surprise here—this is how the First Amendment usually works.
I want to add just one thing to this, which is that not even the defendants in Herbert—including CBS News, Mike Wallace, and Barry Lando, ably represented by Floyd Abrams—asked the Court to hold that editorial judgment should be insulated from inquiry altogether. The oral argument, which is fascinating and worth listening to, makes this very clear. During the argument, many of the justices struggled with the question of how the Court could reconcile Miami Herald v. Tornillo, which held that the First Amendment protects newspaper editorial processes against certain kinds of government intrusions, with New York Times v. Sullivan, which required plaintiffs to establish, presumably through discovery, that newspapers that published defamatory falsehoods did so either knowing the material to be false, or recklessly discounting the possibility. Abrams’ response to this question during the oral arguments was complicated, but it was definitely not his view that the First Amendment protected his clients’ editorial process from any intrusion at all.
Indeed, one of the things Abrams says repeatedly during the argument is that his clients had already turned over many documents relating to the editorial process. At one point in the argument, Chief Justice Burger asks Abrams quite directly how a defamation plaintiff can be expected to establish a media defendant’s state of mind without inquiring into the editorial process. Abrams begins his answer by noting that there had already been “an extraordinary amount of production in this case”and that the plaintiff hadn’t suggested that the documentary production was selective or incomplete. The defendants had also answered all kinds of questions about the editorial process, Abrams says, and are willing to answer more—questions like, “what did you learn?” and “what did you know?” and “who did you talk to?” The defendants have no objection to answering questions about the objective facts, Abrams says; they object only to “a small number of questions” about their beliefs, opinions, intent, and conclusions. The defendants would be willing to answer even these kinds of questions, Abrams says, so long as they relate to what they actually broadcast, rather than to what they decided not to. Abrams also says at one point that the First Amendment might provide “gradations of protection” suggesting (again) that some inquiries implicating editorial judgment might be more constitutionally problematic than others. In response to a question from Justice White, he says his clients shouldn’t be required to say why they decided not to include a particular interview in their final broadcast. “I think that falls within the area of protected information,” he says. But then he goes on:
Let me say that I think, that one could make and rather easily make some gradations of protection here. To the extent that one takes as one’s [lode] star here a notion of—and I appreciate the fact that it is necessarily and amorphous, a developing concept—editorial process, what is most important is what was on the program and what was off the program, and it seems to me that the single most protected thing in this area are questions such as ... why didn’t he put the four people on [the program] who [said] good things about [the plaintiff].
It's not entirely clear to me exactly why Abrams believes that this particular question—why didn’t you write/broadcast/publish something other than you did?—must be “the single most protected thing in this area.” When Justice Brennan asks him to explain, Abrams says only that “to probe that deeply into the mind of the journalist is first of all not necessary to allow the plaintiff to prove his case,”which isn’t very satisfying to me. (Abrams may well have had more to say, but the Court didn’t give him a chance to say it.)
For present purposes, though, the important point is just that the media defendants thought it was appropriate to distinguish some intrusions into the editorial process from others. Abrams argued that the First Amendment should bar the plaintiff from asking his clients why they hadn’t broadcast interviews with people who would have cast the plaintiff in a more favorable light. But he didn’t think the First Amendment presented a bar to the plaintiff asking his clients whom they had interviewed, what materials they had considered, whether they believed the person whose assertions they broadcast, or whether they’d interviewed anyone who told them those assertions were untrue.
Abrams’ arguments didn’t convince the Court, which denied his clients even the narrow, presumptive protection they requested. Still, it’s worth thinking about how the lines that Abrams proposed (however tentatively, and for whatever principled or tactical reasons) in his Herbert argument would map onto the debate about the imposition of legislative transparency mandates on social media companies. Perhaps a mandate that requires platforms to explain why they adopted particular content-moderation policies should be viewed differently from one that requires them to disclose what their content-moderation policies are. Or perhaps a mandate that requires platforms to disclose what their content-moderation policies are should be viewed differently from one that requires them to disclose the downstream results of those policies (e.g., how many times they removed user content on the grounds that it violated particular community standards). It does seem that these different kinds of mandates would chill editorial decisions in different ways and to different extents.
What I take away from Abrams’ argument, and from the Justices’ response to it, is that we could use more nuance in our debate about the imposition of transparency mandates on social media companies. Recognizing that these companies sometimes exercise editorial judgment is an important first step. But to determine whether any particular transparency mandate would intrude unconstitutionally on that judgment requires a careful consideration of how that editorial judgment is being exercised, how and to what extent the proposed mandate would burden it, and whether the government has a good reason for imposing that burden. (The Knight Institute made a version of this argument in this brief, and Scott Wilkens and I summarized it here.) Again, no surprises here—except, maybe, for those who think Herbert means that the First Amendment forecloses legislatures from compelling transparency relating to editorial judgment categorically.
Jameel Jaffer is executive director of the Knight Institute.