Professor Cass Sunstein on Falsity and Free Speech
From a live Zoom conversation held January 12, 2021.
Dave Heller, MLRC Deputy Director
This is our first Scholars’ Hour call of the new year and we’re extremely honored to have Professor Cass Sunstein with us this afternoon. Professor Sunstein is the Robert Walmsley University Professor at Harvard Law School. And before that, he taught for several years at the University of Chicago Law School where I think he may have crossed paths with President Barack Obama before he was president. He is a legal scholar of extraordinary depth and scope in subjects like constitutional law, administrative law and behavioral economics. He was also in government during President Obama’s first term as the head of the Office of Information and Regulatory Affairs, which is the federal department in charge of reviewing government information and technology policies. He is a prolific author of law review articles, press columns, books – really some terrific books with Nobel-Prize-winning economists like Richard Thaler – and I know there’s a forthcoming book with Daniel Kahneman. Professor Sunstein clerked for Justice Thurgood Marshall and he was on President Obama’s list of potential supreme court nominees. And he’s the husband of former UN Ambassador Samantha Power. Your forthcoming book, which we’re going to talk a lot about today, is called Liars, Falsehoods and Free Speech in an Age of Deception. And it comes out in March from Oxford University Press. And it couldn’t be a more timely topic as it addresses the role of law and social norms in addressing falsehoods. So, to kick this off and start with current events and this subject of falsehoods and free speech, why do people believe that the election was stolen?
Prof. Cass Sunstein
First, human beings tend to believe what they want to believe, so that if people are pleased by thinking something, they’re more likely to think it. And that clearly plays a role in belief; a false belief in a stolen election. A second reason is that people believe trusted others. So if you hear from someone whom you generally believe to be reliable, you might trust them on something on which you have no information, even if it turns out to be false. And a third reason is something called “truth bias,” which is kind of hardwired into our heads, that when we hear something, there’s some part of our mind that believes it even if it’s not plausible to us, in fact. So, if I said something false, right now – I’m not going to do the exercise – the chances would be that those of you on the call would in some sense believe it, even in two-weeks’ time, because you heard it and it’s probably evolutionarily sensible that we tend to believe, on average, what fellow members of the human species think. If you put those three things together: trusted others, belief in what you want to believe, and truth bias, you can kind of get why a lot of people believe the election was rigged.
Dave Heller
In sports, to take an analogy, a game won’t go someone’s way and they say the umpires robbed them. Sometimes it’s a little bit of a mix of fact and opinion. Do you think people understand, when they hear Trump repeatedly say that the election was stolen – does it necessarily mean to them that there is some sort of underlying provable facts of theft or corruption as opposed to, say, systemic bias against their side in this election?
Prof. Cass Sunstein
Well, I’m a Red Sox fan. And it just is the case that the umpires have been systematically biased against the Red Sox in favor of the New York Yankees for decades and it’s corrupt and the Red Sox are just victims. Okay, maybe that’s not true. Some part of my mind, as a Red Sox fan, actually believes that. So if you’re a Trump fan, then the fact that he lost is not readily believed. And if he’s saying either that there were particular acts of malfeasance, or that there’s a general rigging, that fits with what you want to believe and it makes it credible, partly because of desire to believe and partly because if you’re a fan of a politician, you think on average they’re more likely to be truthful than the other side. It just is the case that people’s beliefs about the world … if you’re sick, and you’re told, actually, it’s going to be okay, you’re more likely to believe that than if you’re sick and you’re told, you’re in big, big trouble. And people will draw comfort from a word that makes them feel good. And that’s a lot of what we’re seeing right now in American politics.
Dave Heller
What do you think the legal consequences should be for Trump’s role in, let’s say, spreading the falsehood of a rigged election, particularly in light of what happened last week at the Capitol?
Prof. Cass Sunstein
The most obvious is that the argument for impeachment, even at this very late date, is really strong. And to justify that conclusion, we need first to identify the standard for an impeachable offense. And the standard is an egregious abuse of the powers of the office. That’s basically a shorthand summary. And to say the kinds of things he did at the time and with the, let’s say, passion that he said them, is an egregious misuse of the authority he had. And that is an impeachable offense. What exactly follows from that statement? Different people have different views. But that’s a clear legal consequence, whether there’s criminal liability of one or another kind, we’d want to investigate the relevant statutes. And I haven’t done that. But clearly, there’s something non-trivial that happened on January 6th. And tragically, the President was probably a but-for cause for it happening.
Dave Heller
I don’t think the Supreme Court has really ever accepted – or you tell me whether Supreme Court has ever accepted – the but-for cause as a basis for liability, because a lot of activity is inspired by ideas and ideology, which is perfectly acceptable to promulgate. I mean, for me, what I found very interesting in this case is the short time between Trump’s afternoon speech and the violence, and you often don’t have that in other cases, whether it’s civil rights protests, which can be very heated in their rhetoric, but removed from any near action. I’m wondering whether you had any thoughts on whether the short time between speech and violence pushes this closer to the incitement test?
Prof. Cass Sunstein
Okay, so the first question is whether there’s a statute making what the President did illegal. And as I say, I’m agnostic on that question, pending study of the relevant statutes. And the second question, as you rightly point out, if the statute makes it a crime, is that a constitutional statute. And that would be governed by the Brandenburg test, which is the modern incarnation of the clear and present danger test, which has two elements. It has to be directed to incite; and likely to incite imminent lawless action. And so there are two prongs: it has to be directed to, which suggests intentionality; and likely to, more probable than not to, produce imminent lawless action. Now, in terms of short, temporal separation, it’s a little trickier and more fun than it might seem at first glance because something might be – at time one – unlikely, meaning not more probable than not to produce imminent lawless action, but it happens too. Then the test wouldn’t be met, even though the bad thing followed the words. Or something might, in fact, be likely to produce imminent lawless action, but it didn’t come through. So it was 80% probable that it would happen, but it didn’t happen. That would be regulable speech under the Brandenburg test. And if it’s unlikely to, but it happened to, then it would not be regulable speech under the Brandenburg test.
Dave Heller
You have a passage in your book, “when things are going well, social norms do the work of law” and I’m wondering whether it was in some ways an inspiration for the book itself? I think there’s a reason why you write this book now. But could you talk a little bit more about the inspiration for the book, and really what it says about what we’re seeing now – that apparently social norms are not doing the work that we ordinarily expect them to do?
Prof. Cass Sunstein
Particular thanks for that because you’re right; that is part of the beating heart of the book. With respect to intentional falsehoods that injure people, whether it’s about your neighbor, or a disease or about democracy, the best safeguard, and the one we want, is norms, not law. So, you wouldn’t say your neighbor is a drug dealer even if you’re really mad at your neighbor because that’s not the sort of thing civilized people do – unless your neighbor actually is a drug dealer, in which case, it’s fair game. So lying about people is a transgression of a pretty serious sort and norms should be kicking in to prevent them from happening. You mentioned my wife, Ambassador Power. And just recently, and this is true since she’s been in the public domain, there have been vicious lies about her; palpable falsehoods. And sometimes they’re believed and sometimes not, but norms aren’t doing the relevant work. The fact is that norms aren’t doing the work. So that’s one motivation.
The other impetus for the book, and maybe the more immediate one, is a sense that lies are all around us. They are causing real harm, including to public safety and health. And the free speech gains, let’s say, of the last 40 years, which are a cherished part of our heritage, are often operating too rigidly, I think, as an obstacle to thinking hard both about what private institution should do – and here I’m thinking about Facebook and Twitter – and also about what government should do. So if you have certain kinds of lies, including libel, to call New York Times v. Sullivan, and put it in bold letters and a big font and exclamation point. It’s a little, I think in the end, cruel to people whose lives might be ended, or severely damaged, as a result of something that someone achieved through words. So that’s kind of the impetus for the book.
Dave Heller
It’s not so explicit throughout your book, although it is there, and that’s the role of social media. Because when you say we’re surrounded by lies, I think most people would point to where they’re coming from and how social media acts as an accelerant – where a lot can be shared in the quantity that was never possible before. It came up with Twitter’s ban of President Trump; what norms social media platforms should exercise, and whether there’s a role for law in there. This is a huge question. But to begin, is this a social media problem?
Prof. Cass Sunstein
In part. So there’s the internet “large,” the whole thing, which means that one can say something false that can spread quickly, just by virtue of how the internet works. Social media, of course, makes that much easier. And that’s a big problem. So for YouTube and Instagram and Twitter and Facebook they are, in some ways, misinformation machines. The good news is – and this is something I discovered late in the writing of the book, which is slightly embarrassing – that Twitter and Facebook have been generating, to their credit, a kind of common law of control of misinformation, which is very much a work in progress. But which is more, let’s say, nuanced and fine-grained than anything the actual common law was capable of. And here are the kinds of things I’m thinking of. They can have, now famously, a warning. So they don’t censor anything but they say, “warning.” The warning can have any number of features. It can say, “this claim is disputed,” or it can say, “this claim is false.” They can downgrade distribution. They can do things to make sure that the relevant material either spreads only after a little friction, or that it doesn’t show up prominently or as much in a newsfeed. That’s something Facebook does. Or they can do things to educate consumers by linking to sites that will provide a fuller or more accurate picture. Now, those are three things. They don’t, by any means, exhaust the universe of things that Facebook and Twitter, et al., are doing. But the common law never thought of, at least, these precise kinds of things. And what’s beautiful is this gives a kind of menu of options for the legal system, which it never had present before. Whether the law should encode any of these things is a completely fair question. But in a time when the common law had, let’s say, damage remedies or criminal fines or sentences, we now have a more expanded view screen for what the legal system might consider doing with respect to certain kinds of falsehoods.
Dave Heller
I watched a video on the internet of a conversation you had about 12 years ago. It’s actually a discussion about blogs, during the heyday of blogs, really before the take-off of social media. But even then, with the discussion of blogs, there was already the question of whether people would be siloed in the ideas that they already believe and how closed off they become from any kind of counter speech, which would underlie the whole marketplace theory of ideas approach to free speech. Doesn’t social media put people into even more hard silos than 10 years ago?
Prof. Cass Sunstein
This is a great question. So, let’s separate the two points you’ve rightly made, and then bring them together. So, point number one is misinformation. So even in an unsiloed world, people could hear, “the president says x, the opposing candidate says y, both x and y might be false,” and everyone’s hearing them, and some will believe one, some believe the other. And that might be a problem. It might be a problem for public health, let’s say, if some people believe something that is untrue that can lead to early death. In a siloed world, it might be that people aren’t hearing misinformation, in part, they’re just hearing opinions. So you might hear someone saying that to join the Paris Climate Agreement is a really terrible idea, and others saying it’s a really good idea. And as stated, these aren’t views about facts, they’re just views about what ought to be done. If people are living in different political universes or different universes with respect to culture and any number of things, it may or may not be okay. In a way that’s part of freedom. But with respect to a well-functioning democracy: it can create a failure of mutual understanding, it can contribute to an incapacity to solve problems, it can contribute to enmity, and as we saw on a day that will live in infamy, January 6, it can contribute to violence.
Now, if you take those two different problems: misinformation, let’s say, and information cocoons, and then we put them together; now we’ve got a toxic brew where people are siloed to some extent in cocoons in which they’re learning different facts. And a number of the facts that some cocoons are portraying are just false facts and that can be harmful in one way or another. The view of the social media providers has the following feature: that the echo chamber/siloing concern is greatly overstated. Their view is that there are a lot of people who are getting a lot of diverse views on their platforms and that echo chambers and information cocoons preceded the problem of social media platforms, if it is a problem, and they shouldn’t be held responsible for it; to some extent, they are a cure for it. The data is on both sides of this question. And the short way to put it is: a lot of people use social media to get information of multiple different kinds and they aren’t siloing themselves. But a lot of people are using it for exactly that reason. And the people who are using it for that reason or that purpose tend to be the most politically active. That is, the active people tend to be siloed in terms of what they consult and what they pay attention to. And from the standpoint of, let’s say, figuring out what’s true, or a mutual understanding to different problems, that part of the echo chamber problem – that is those most interested in active in politics – is a real problem.
Dave Heller
We had a couple of questions in the chat box, essentially about the role of Section 230 immunity for platforms. One thing that you point out in your book, which I think is an answer to the concerns raised in the chat, is that there’s a great deal of information: take false health information about what you should or shouldn’t do, say whether you should take or not take the COVID vaccine, or whether vaccines cause autism. I mean, whether Section 230 exists or doesn’t exist wouldn’t stop that because those are not defamatory statements. You point out in your book that there isn’t that sort of catch-all provision to hold publishers liable for that type of speech. Maybe you could tell us more of your thoughts on the role Section 230 plays and whether it’s good or bad here?
Prof. Cass Sunstein
Okay, so on balance is the world better off with the long-standing existence of 230, or not? And probably the most prudent answer is that it’s been better off with it because its existence has promoted, or made possible, the rise of things that on balance plausibly are achieving a lot of good for people. So that’s a reasonable view. With respect to Section 230, now there’s also, I think, a reasonable view, and I embrace this view, that it’s too broad. There are things that 230 protects that are very hard to say should be protected. Suppose someone is libeled under the standards of New York Times v. Sullivan, it’s put on Facebook, it’s drawn to Facebook’s attention, and Facebook doesn’t take it down. And the person says, I want to sue you for the damage that you caused by not taking it down. I tried to make the case easy in the sense that Facebook is on notice, there’s actually been an adjudication, the New York Times v. Sullivan standards have been met. And Facebook isn’t liable for portraying that. That I think is a very hard view to defend. To see Section 230 as deserving immediate and wholesale repeal deserves a lot of thought also, because a magazine has a kind of control and visibility into its own contents, but with Facebook or Twitter, it’s more challenging for them to know exactly what’s on their platform at all times. Maybe with machine learning and algorithms that will become increasingly possible. You’re right that there are some things that cause harm that just aren’t against the law. And if the New York Times or The Boston Globe prints them, even though they’re harmful, there’s no law that makes the Boston Globe or the New York Times liable. And that would, of course, transfer over to social media platforms even if they didn’t have 230.
Dave Heller
This is a probably a good time to transition to the portions of your book that discuss New York Times v. Sullivan and defamation law, and I kind of mean this partly as a joke, but I should give a trigger warning to the audience that Professor Sunstein would be open to changing New York Times v. Sullivan and developing different standards for defamation.
Prof. Cass Sunstein
Excuse me, I’m sorry. I’m apologizing for those things. I can’t deny them. [laughter]
Dave Heller
No need. This came to great attention the previous year when Justice Thomas wrote his lengthy concurrence to a denial of cert., setting out why he thought Sullivan was wrong. And I guess it’s a hard question to practicing lawyers when you have a rule that is in place for 50 years; you can always hear the argument that some other rule might be better. But maybe you could sketch out a why you think your alternatives are better?
Prof. Cass Sunstein
Okay, so let’s talk about it a little bit. So, it would be amazing – it would be a miracle – if in 1964, in a radically different information environment, the Supreme Court got it exactly right. That would be a stunning act of prescience and universality that we ought not to think would be likely, unless we were so used to the status quo or so committed to something that we were unmovable. So just to say, 1964 was, with respect to what we’re talking about, a completely different universe. Almost a different planet in terms of how communication worked. Let me give a few ideas about New York Times v. Sullivan. Suppose the idea is that if you put something online that is false and damaging to reputation, you have to accompany that online with a statement: “This is false.” That’s the remedy. That violates the First Amendment? It’s very hard to give a theory of the case by which a one-word addendum to the “libelous by common law standards” statement is added. So let’s suppose someone says that a prominent member of the business community was a drug addict in college and actually spent six months in jail. And let’s just stipulate that the New York Times v. Sullivan standard wasn’t met because it wasn’t knowingly false or there wasn’t reckless indifference. But it turns out it was negligent and false. The First Amendment forbids the company official from requiring it to be put online that its false. From my point of view, that is not acceptable. Or to say that there’s a right to a nominal damage remedy, in the case of let’s say negligent, rather than knowing falsehoods, where the nominal damage remedy is, let’s call it, one penny. The First Amendment of the Constitution forbids that, if it’s demonstrated to be false, injurious to reputation and harmful? That I don’t get either. So, New York Times v. Sullivan is a glorious moment in our constitutional history because it vindicates the right to a robust communications universe in which people sometimes make mistakes. We need breathing space for that, certainly, but an online correction, or a one penny damage award, wouldn’t offend the salutary goals, I’m suggesting, of the New York Times v. Sullivan framework.
Dave Heller
You mentioned damage awards. And it’s an issue that’s always puzzled me why a constitutional argument can’t be raised that large damage awards are unconstitutional. For example, European countries often have schemes where plaintiffs have a much easier chance of succeeding on their defamation claims, but damages awarded are staggeringly low compared to those in America. Is there any constitutional argument to pare down damage awards?
Prof. Cass Sunstein
I think so. I’m glad you mentioned Justice Thomas’s opinion because, of course, to get traction on this we need a theory of constitutional interpretation. Justice Thomas’s is an originalist theory, by which the meaning of the Constitution is settled by the original public meaning at the time of ratification. I think he’s correct that New York Times v. Sullivan is very hard to defend on originalist premises, and whether damage awards are controlled by the First Amendment would, based on Thomas’s view, be a historical question. And it’s a good historical question, and I don’t have a good answer to it. The court as a whole, going down the New York Times v. Sullivan track, has not been originalist, though Justice Brennan made a reference – a kind of passing reference almost – to the debate over the Sedition Act and Madison’s views there.
But if we’re thinking more general, long-standing evolving principles and hard won, let’s say, understandings rather than time-machine constitutionalism, then it would be fine to say something like, certain kinds of damage awards – and there’s a line-drawing problem here, which is not ideal – but certain kinds of damage awards are incompatible with a system of freedom of expression, at least if we don’t have a case of an intentional falsehood. And other kinds of damage awards, let’s say nominal or more corrective of the record than punitive; those don’t raise the same kind of First Amendment concerns. That’s fair. And it fits a bit with what the Supreme Court has done in the context of punitive damages generally, where it has said there’s a proportionality requirement there. And that certain kinds of punitive damage awards, if they’re a big multiple of what the actual harm was, are out of bounds. Now here, the discussion wouldn’t be exactly that because, of course, compensation for reputational damage is very speculative.
And what I’m suggesting is: in a time when, if we’re talking about reputations, I’ve seen in real time, over the last 10 years, the devastating damage that let’s say, not actionable libel does to real human beings, where the damage might be leading to suicide or to long-standing distress of the sort that human beings shouldn’t have to face. And when I say, “not actionable,” I mean it’s protected either by Gertz or New York Times v. Sullivan. Is that something to celebrate? It’s cruel to allow that and a damage award that is nominal, at least, does something to correct the devastation. And it doesn’t do something to devastate freedom of speech because the person who was responsible for this speech basically pays something nominal. I would want maybe something a little more than nominal, but not something – and here’s where I’m with New York Times v. Sullivan in spirit – not something that makes it hazardous to speak out on pain of bankruptcy. That would be inconsistent with a system of freedom of expression.
Dave Heller
Another issue that you write about is how New York Times v. Sullivan was protecting speech about public officials. But you write a little bit about how – both in the common law and in the thinking among others – is that it can be more damaging to democracy to denigrate and tell lies about public officials. And you’ve raised the question about whether we’re getting the balance right, leading to your suggestions for change. One question in the chat box is about the role of the marketplace theory of ideas. You write about the transition from New York Times v. Sullivan, plus 50 years, to the stolen valor case. It seems, in your opinion, it’s kind of become incoherent. But maybe you could talk a little bit about the evolution of the marketplace of ideas theory from Sullivan to the Alvarez case.
Prof. Cass Sunstein
Great. So, the idea of a marketplace of ideas: Justice Holmes talked about free trade in ideas, so he’s the originator, in law. The idea of a “marketplace of ideas,” I think is like the idea, “don’t lie,” which is discussed in detail in the book, the moral prohibition on lying. And on one view, let’s just say it’s a respectable view, the idea, “don’t lie,” is kind of the right way in, but it’s not the right way out. So if someone comes to your door and says, “I kind of want to kill your kid. Where is your kid?” You can say, “my kid’s not home,” even if that’s false. Or if you are under terrible circumstances where you’ve been trapped, and you’re in the military, and the people who’ve trapped you say, “disclose military secrets or you’re dead,” you’re allowed to say, here’s the military secret, and it turns out to be you’re lying. That’s okay; probably obligatory.
The idea of a marketplace of ideas is a good way in because it gives a big thumb on the scale in favor of freedom. And that’s where we should be. But it works often that, with respect to whether to buy one computer or another or what kind of shoes and clothing to wear, the statements you hear that make their way through the marketplace of ideas are more often than not true. So that’s a good rule of thumb. But it’s inadequate, as in the case of untruthful statements, which are sometimes morally acceptable or morally obligatory. So if someone says, when they’re running for office, I won the Congressional Medal of Honor. “Amazing. You should vote for me, I have a Congressional Medal of Honor.” Congress says you can’t say that because that’s called “stolen valor.” I’m with Justice Alito here – which is not always the case – in saying that Congress can forbid that statement. It’s not like it’s negligent or even reckless. It’s a lie, which is trying to undermine democracy by attracting your votes by saying something false and damaging. The court’s response, which isn’t crazy, is that there’s a marketplace of ideas here and there are ways to respond to false statements about your military heroism. And the court said: that’s a less restrictive alternative.
Whether or not the court is right or wrong on the stole valor case it was, I suggest, “so 2012” in its opinion. “So 2012,” meaning an era where you could talk somewhat blithely about the ability to correct false statements in a democratic process by just saying what’s true. Palpably, that’s not always so, where you say something really damaging about a political opponent. Let’s say you say something that is clearly false and you knew it, or you say something about yourself that’s clearly false, and you knew it. It may well be that these things should receive constitutional protection. Certainly, we want the government to stay away from censoring the speech of candidates for public office. But to invoke the marketplace of ideas – that wouldn’t be an adequate reason for that conclusion.
Dave Heller
One of the interesting questions that came in is about the Supreme Court’s interim period when they decided Rosenbloom and were looking at the content of the speech, rather than the nature of the speaker, and whether that would have been, or still might be, an alternative approach.
Prof. Cass Sunstein
Okay, so where I think we should all be going does not turn on whether Justice Marshall’s effort to turn speaker protection into topic protection was correct. Who was right, Marshall or Brennan? I’m not sure they disagreed on this particular, but Marshall was protective of public-affairs-related speech. And the Court and Brennan, in New York Times v. Sullivan, was protective of speakers rather than topics.
I think the debate we should be having now is: what can we do to protect against illness, death, democracy destruction, reputation ruination, core with respect to, let’s say, teenagers. How can we do that compatibly with a system of freedom of expression? And let’s say, 1960 to 1990, or maybe 1960 to 2015, there’s a kind of blitheness with respect to that. And the goal of my book is not to say that New York Times v. Sullivan was fundamentally wrong. But to say that, “Houston, we have a problem,” where people are thinking that COVID-19 isn’t real. I had COVID-19 I’ll say; I’ll come out of the closet here. I had it about nine months ago. And while I didn’t almost die, it’s really bad. And it’s real. And a lot of people don’t believe that. And it’s because of what they hear. The question is: what are we going to do about that? We all probably know people who’ve been in the public eye who’ve been subject to credible death threats and really horrible things because of knowing lies. And what are we going to do about that? To say, Rosenbloom or New York Times v. Sullivan is a little rotary-phone, I think. That doesn’t address the problem that Houston currently has.
Dave Heller
It seems as though you really are putting your finger on one aspect of the problem, which is yes, people are harmed by speech and a variety of other things that you mentioned in the book. At the same time, since Trump was campaigning and took office, we’ve had this kind of war on the press and his, at least, aspiration that libel laws would be changed so that he could vindictively pursue and punish the press. And we’re not there, but I think that is the background in which you could change the rules and then have a system like Turkey, or the Philippines, where public officials can unleash official terror on the press. And so, of course, it’s a question of balance. And I wonder if you could speak to the other half of the balance, which would be the absolute horror of public officials unleashing their power of the state against free speech?
Prof. Cass Sunstein
Okay, that’s a great point. So, two things. First, I hope that President-elect Biden will say that the press is the friend of the people, which is, A) true; and B) kind of essential to say now. The enemy of the people stuff is fascist talk though it wasn’t backed up, so far as I’m aware, by criminal penalties or anything like that. But “enemy of the people” – “horror” is the right word. To say that people should be allowed to get something like the word “false” online when there has been falsity is not to say that the press is the enemy of the people. It’s to say that something went wrong and this is a corrective. I agree with that. Also, you make a really good point, which is that the system I’m at least exploring could potentially give rise to lawsuits by public officials who are trying not to protect their reputation, but trying to protect themselves against criticism. And we need to have good tools to protect against that. And this is the glory of New York Times v. Sullivan, that it responds to the incentive effects of even the prospect of a lawsuit. So, I’m not sure whether the right approach to that is to have severe institutional restrictions on suits by sitting public officials, or whether to have substantive law of the New York Times v. Sullivan sort. I haven’t criticized the actual malice standard; I’ve criticized it’s use to forbid, let’s say, minimally intrusive and not economically damaging remedies. I’ll give one that’s a hard case for me because I’m basically in agreement with you, suppose someone says about President-elect Biden that it’s extremely unfortunate that he spent the last 20 years battling heroin addiction. That I’m very confident is false. And let’s suppose that it’s a knowing falsehood. To allow the President-elect to do something about that actually is compatible with the New York Times v. Sullivan case on the assumptions I’ve given. But even if it’s merely negligent, to allow him to get a correction, saying that didn’t happen – I’m not sure that’s so bad. I’m really not sure.
Dave Heller
Another issue is really the role of the press because a lot of these falsehoods do not originate with the press. But when they become a part of the public conversation, the press reports what other people are saying. And then you get into the dilemma of whether or not they should report what is being discussed. Because they might be agnostic as to the truth or falsity of it, but it certainly might be interesting to the public to know that those discussions are there.
Prof. Cass Sunstein
Okay, great. So, see what you think of this. If there’s a discussion in, let’s say, the Boston Globe, my hometown newspaper, of conspiracy theories in which the Globe is accurately reporting that people are saying X and Y and Z, then to say that the Globe would be held liable for that would be extremely hard to defend. I think this is consistent with the music of your question, that repeating it would not be, even under the common law libel, a source of liability, so long as it’s done in the context of not repeating a lie, but covering a genuine news story. On the other hand, let’s say, the Boston Globe – I’m saying this with pain because I love the Boston Globe and I like them about as much, and for the same reasons, that I love the Red Sox – this wouldn’t happen, but if the Boston Globe repeated something without giving any context, but giving credence to the thing that’s false, then to ask whether the New York Times v. Sullivan standard is met is a reasonable idea. It might be that even if the speaker had reckless disregard, the Globe itself didn’t. And then the question is: what do we do about that? Nothing would be too simple an answer. It might be that the Globe has to say it online: this isn’t true. And that again, the soft, in that case, an essentially costless remedy, would be a small step for man, if not a giant step for mankind.
Dave Heller
Some of the other interesting comments, in effect, go back to the Section 230 debate. Some of it goes to the question of burden of proof. In Europe, for example, it’s often the case that the defendant bears the burden of proving truth. So, the good reputation of a plaintiff is presumed, and you can have lengthy trials about whether the Holocaust occurred as in the David Irving case. Is there anything wrong there? Is the burden of proof, in your opinion, properly placed with plaintiffs so that if you assert that something libelous has been said about you, at least you have to bear the burden of proving that factually.
Prof. Cass Sunstein
I say that in the vast run of cases, not a whole lot turns on that. So, if it is said that my neighbor is a drug dealer – and one or the other side bears the burden of proof – I’m looking out the window at my neighbor, I’m very confident that he’s not a drug dealer, and burden of proof in that case doesn’t matter. So the burden of proof would matter only in cases which are right at the border line. And I’m thinking: what are we talking about exactly? Are we talking about – let’s take a realish case – whether Secretary Clinton violated federal law with respect to her email practices; let’s say that could be an issue. I know a little bit about federal law with respect to email practices and the burden of proof isn’t going to matter very much. It’s going to come out the same way. So, it has a lot more, I think, symbolic or expressive content – the debate over burden of proof – than practical reality. Which is also a way to say I don’t have a strong conviction about that either way.
Dave Heller
I would just take an opposite view because it’s often in my experience that actual malice is a kind of pressure relief valve. You never have to get to the issue of truth or falsity because you can say, going in, you’re not going to be able to prove either actual malice or even negligence. So, a court isn’t going to be burdened with the metaphysical task of finding out the truth. I think that’s why – I mentioned England as an example – they end up having these lengthy trials. You have the Johnny Depp celebrity defamation trial with weeks of testimony on who’s telling the truth because the defendant-newspaper had to prove that the ex-husband was a wife-beater to avoid a huge libel judgment. So at least in my experience, it actually makes quite a bit of difference.
Prof. Cass Sunstein
Okay, I might be mistaken on this. But let’s talk about the actual malice standard, which requires either knowledge of falsity or reckless indifference to the question of truth or falsity. If the defendant has to prove that she didn’t know she was lying. Or the plaintiff has to prove she knew she was lying. In the easy cases, which are a lot of them, that’s going to be straightforward. So, let’s just bracket my empirical hunch, which is not going to matter a lot, or assume it’s false, which is the correct way to go? I don’t have a strong conviction on that. I have a kind of initial inclination to be pro freedom of speech there. But it might matter whether it’s a libel case or a case involving, let’s say, some statement about a product: that it’s going to cure cancer and whether that should be treated in a way that puts the burden of proof on the doesn’t-cure-cancer side. Now, it might be that the context of the falsehood would matter with respect to who properly bears the burden of proof. One thing I’m quite worried about, and I’m sure many of you are also, is the sheer expense of defending libel actions. And the burden of proof is probably a good weapon for those speakers to have as a way of protecting themselves against the litigation expense.
Dave Heller
Going back to something you mentioned about the issue of fault, I’m wondering whether you think, almost as an alternative to your proposals, there might just be judicial wiggle room in the idea of reckless disregard of falsity. I mean, sometimes it seems that it might take quite a lot to prove reckless disregard. And the Supreme Court actually hasn’t spoken in a defamation case in, I think, nearly 30 years.
Prof. Cass Sunstein
Can I tell you something funny, that’s not unrelated in its specifics to our general topic, and that’s directly related to your specific question, because I wrote a little column on some of these issues, saying that if you were not merely negligent, but it was obvious that what you were saying was false, there would be reckless disregard. So you might not know it was false. But if it was obvious that it was false, that would be reckless, it wouldn’t be just negligent. So you might subjectively think it’s true, but if to a reasonable person it’s obvious, that would be reckless disregard. And then someone, who is a defamation law specialist, wrote on – I’m embarrassed to say Twitter – that my column was completely ignorant that the Supreme Court had settled this, and that to be held liable for reckless disregard, you must subjectively know that what you were saying was false. That it must have been your subjective belief. And I thought, “oh my gosh, I made a terrible mistake on a free speech issue in public, and that’s really not good.” So, I went back and read the case and it does say what my critics say, but it goes on to say what I say. It’s not unambiguous, but it does use the word, “obvious.” Which is a way of saying that, under the reckless disregard test, suppose you have a speaker online, or someone working for newspaper, who should have known that it was false, but it wasn’t obvious that it was false. It was just negligent, if it was a public figure, no liability. But in the other case, where you actually thought it was true, or you had no idea, but a reasonable person would have found it obvious, the best reading of the long-standing precedent is that you can be held liable. And I think the Supreme Court would be there today.
Dave Heller
We’re almost at the end of the hour. And I’m going to pose a question which probably could be an hour itself. And that is the work that you’ve done in behavioral economics, which kind of looks for empirical evidence and background to our decision-making. How much do you use that in the constitutional law side of your work? And is that partly why you open up constitutional theory to broader inquiry?
Prof. Cass Sunstein
Well, when I was in the US government, I was actually overseeing regulation broadly. Not just technology, but environmental, food safety, transportation. President Obama gave me a very wide set of tasks, and behavioral science and behavioral economics are obviously relevant to that. Like, what do you do to encourage people not to use their cell phones while they’re driving? Or how can you encourage people both to save money and to pollute less with their cars? What do you do about cigarette smoking from which 480,000 Americans die a year? So it’s obviously relevant on the regulatory side? On the speech side, it’s also relevant. And here, we’re really at a frontier and I make a little effort to start down that path in this book, which is why do people believe falsehoods? Your opening question. And what can be done to dislodge a false belief? Behavioral science actually tells us a lot about that. The problem is more challenging, I think, than free speech lawyers and law professors like to believe. That is, the problem of dislodging a falsehood, whether it’s injurious to your health or your reputation, is harder for behavioral reasons than intuition suggests. The good news is that the behavioral work also tells you something about how to overcome that. So we know both: that it’s hard, but we know it’s doable. And the behavioral types are giving us lessons there.
Dave Heller
That’s a great way to wrap this call up. I mentioned just the title of one of your previous books, which is called Nudge, which is to, without the hammer of the government, but with encouragement, get people to behave differently or to think differently. And certainly, you’ve given us all a nudge today with some of your ideas to make us think a little bit harder and deeper about some of the issues that we work with day-to-day. So, I just want to thank you very much for spending the hour with us. It was terrific talking to you.
Prof. Cass Sunstein
If I can say thanks to you for a terrific discussion. And also, thanks to everyone on the call for your work during this time. There hasn’t been a time in memory when legal activity protective of free speech and protective of journalists has been more essential. So whatever your religious faith or not, you’re doing the work of God.
Dave Heller
Thank you very much. A perfect way to end the call. Thank you so much, Professor Sunstein.