Wednesday, January 13, 2021

Incitement, Imminence, and Free Speech: The Internet is a Game Changer

 By Eric Segall

There is a great tension between constitutionally protected free speech and expression that incites others to commit violence or other illegal acts. For example, pure political speech that includes heinous content such as bigotry, sexism, or general expressions of hatred towards identifiable groups has been protected in America by the Supreme Court since the 1960's, while other democracies do not tolerate such ideas. Holocaust deniers cannot be punished in America while in some European countries they can be fined and/or imprisoned for such speech. There is a general consensus here that non-targeted expressions of hate should be met with more speech, not governmental censorship. Additionally, we are allowed to speak out against our government in ways that do not include specific instructions to commit crimes or acts of violence. But where is the line between calls for dramatic change and illegal incitement or what the Court has called true threats? The tragic events of last Wednesday raise these admittedly difficult questions.

The key Supreme Court decision in the area of incitement is Brandenburg v. Ohio. In that case, the Supreme Court overturned the conviction of a Ku Klux Klan leader for making hateful comments about Blacks and Jews. Among other things, he said, “I believe the n****r should be returned to Africa, the Jew returned to Israel." He was convicted under an Ohio law that made it a crime to “advocate or teach the duty, necessity, or propriety of violence as a means of accomplishing industrial or political reform." The Court, recounting prior cases, said that in order to be punished for incitement, the speech must be “directed to inciting or producing imminent lawless action and is likely to incite or produce such action." The Ohio law didn't satisfy that standard, and the Klan leader was set free. This test, unfortunately, is still the law today.

As I detailed in a prior article, lower courts have had an enormously difficult time applying this standard to the internet. The main difficulty has been the imminence requirement. As I've previously written, the imminence requirement:

makes sense in the context of fiery speech uttered during political rallies, but it makes much less sense when applied to speech posted on the Internet where the threat can last forever and be seen by millions of people around the world. For example, there is a significant difference between a person who at a pro-life rally, yells that a particular doctor should be “stopped” if a year from now he is still performing abortions, and a person posting on the Internet the doctor’s home address and suggesting the same sentiment. The imminence requirement makes sense for the former but hardly lessens the danger of the latter. The fact of permanence, as well as the great reach of the threat, makes Internet speech more dangerous than rallying cries at political protests, and the Brandenburg test is just not flexible enough to deal with that problem.

Can political speech on the internet be protected enough without the imminence requirement? Let's work through several hypotheticals raised by the events of last Wednesday. Please have patience for getting to Trump-we will get there.

1) David Dukish posts on the internet the following message on January 15th: "The election was stolen! Let's Take Back our Country." Without much more, this is and should be protected speech. Full stop.

2) David posts: "The Election was Stolen. I'm going to get Biden!" There is a Supreme Court case fairly similar to this hypo. In Watts v. United States, the defendant was arrested under a statute making it a federal crime to threaten the President. During an anti-war rally, the defendant said, "They always holler at us to get an education. And now I have already received my draft classification as 1-A and I have got to report for my physical this Monday coming. I am not going. If they ever make me carry a rifle the first man I want to get in my sights is L.B.J." 

The Court reversed the conviction saying that, although true threats are not protected speech, this statement was just "political hyperbole" and absent a more specific threat, our country's robust devotion to political speech required that the defendant's conviction be overturned. I agree that "Let's get Biden," without more is likely protected speech.

3) David posts: "The Election was stolen! Let's all try to kill Joe Biden before our country goes to rot. He's giving a speech on Jan 17th in D.C. Let's go and shoot him dead." Applying the Brandenburg and Watts' standards to this posting, the imminence requirement is satisfied. The issue would be was his posting likely to cause lawless action. I hope we all can agree this speech should not be protected given that he identifies a specific target and a time and place where people should kill that target. If you don't agree with that interpretation, you might want to stop reading.

4) David posts the following message: "The Election was stolen. Let's kill as many Democrats as possible."  Now this one is harder. Under Brandenburg and Watts, absent more evidence, this is likely protected speech. Should it be? If someone chants this once during a political rally, we might just chalk it up to an overly emotional moment that does not present any danger. But this speech placed on the internet and kept there is different. In these kinds of cases, context is everything. 

This one post with nothing more probably would be held by most courts to be protected expression. What could and should make it unprotected would be evidence that the speaker knew this would incite other specifically known individuals or groups to actually go out and kill Democrats once they received this signal. If we add names to this message, "Let's go kill Nancy Pelosi and Chuck Schumer and as many other Democrats as we can," this should not be protected speech unless the context is clear the speaker was kidding or obviously speaking metaphorically. 

But the key point here is that these kinds of cases can only be evaluated after the fact when all the evidence is in. No test is going to tell us ahead of time how to decide cases that come close to the blurry line between protected expression and illegal incitement or true threats. What does need to change, however, is the imminence requirement. 

Let's go back to hypothetical number 3, which was "The Election was stolen! Let's all try to kill Joe Biden before our country goes to rot. He's giving a speech Jan 17th in D.C. Let's go, and shoot him dead." Change January 17th to December 17th, and nothing should change. In fact, assuming the threat stays on the internet, the longer away the date is, the more dangerous is the speech for a myriad of reasons that should be obvious. For one thing, the threat is of greater duration and more people may act on it. For another, having the threat on the internet for long periods could make the target much more nervous than a threat that comes and goes quickly. In short, the internet makes the imminence requirement set forth in Brandenburg obsolete.

Now, let's apply all of this to Trump. As Mike observed in his excellent post yesterday, we do not know all the details yet of Trump's involvement with the illegal attacks on the Capitol. In the fiery speech he gave right before the march to the Capitol, he said the following: "these people are not going to take it any longer....We'll never give up. We will never concede...We will not take it anymore and that’s what this is all about...We fight like Hell and if you don’t fight like Hell, you’re not going to have a country anymore." In that speech, which was filled with debunked claims of voter fraud and other grievances, Trump repeatedly told the crowd--including twice in the last 30 seconds--that they should "walk down Pennsylvania Avenue" to the Capitol.

Standing alone, Trump's words are likely protected. But context matters. Before his twitter account was suspended, Trump said many things that could be considered signals to his followers to commit acts of violence. And whether those tweets were six months ago or a week ago should not matter if Trump's idea all along was to stay in power by force if he lost the election. He repeatedly tweeted before the election that the only way he could lose was if the election was "stolen." All of his conduct, and who he is, and what his knowledge of what his followers would do, are important to whether his pure words (as opposed to his deeds) are constitutionally protected. To figure that out in the criminal context, we would need to have a trial. I do agree with Mike that Trump's acts taken cumulatively are clearly impeachable offenses whether or not he engaged in criminal conduct.

I want to end by saying a few words about the "slippery slope." I believe both that political speech needs strong judicial protection and that true threats or incitement should not be protected. There will be hard cases and judges and juries should often err on the side of free speech. But, as I mentioned above, it will be difficult ahead of time to fashion rules to deal with these situations. To the extent that Brandenburg and Watts tried to do that, we now live in a different world.

In some circumstances, in-person provocations are more dangerous than those that occur via social media. The former can turn an unruly but mostly peaceful crowd into a mob, which is arguably what happened last week. Yes, there were agitators who went looking for a fight, but without Trump's incitement there may not have been enough of them to overwhelm the police. In that context, the imminence of the threat exacerbated the content.

But in other contexts and in other ways, online agitation can be worse, making imminence an exacerbating circumstance but hardly a necessary condition for speech to be unprotected. Permanence, no less than imminence, can make threats much more dangerous. Trump's tweets, before they were taken down, are quite obviously relevant to whether he was intentionally inciting violence and whether they were related to the violence that actually happened. Their permanence (he could just host a website, direct his followers there, and he wouldn't need Twitter) makes a constitutional difference.

Imminence in some cases will be relevant to the line between protected speech and incitement, but it should not be what it is today, a pre-requisite for a judge to find incitement. The law needs to catch up to that internet reality.


hardreaders said...

Sort of apropos to this, How Appealing has coverage of an interesting 1A decision out of CA9.

Greg said...

I'm not sure I understand how much the internet changes this, for 2 reasons.

1.) Are you saying that David could open a full-page ad in the New York Times on January 1st saying "The Election was stolen! Let's all try to kill Joe Biden before our country goes to rot. He's giving a speech December 17th in D.C. Let's go, and shoot him dead."

2.) If the answer to the above is yes, what about paying in advance for a full-page ad that would run every week for a year saying the same thing?

Isn't posting a web page equivalent to the weekly full-page ad?

Eric Segall said...

He can’t do 1 or 2 in my world.

Solomon Goldman said...

I think that you’re downplaying the role of the “likelihood” of producing violence. For example, in your fourth hypothetical (kill as many Democrats as possible v. kill Pelosi and Schumer), the different result would not be based on the imminence of the violence, but rather on how likely violence would be when there’s a named target.
Just as with imminence, likelihood needs to be based on context, and I agree with your main point that the medium of the internet is an important part of that context. Here are a few suggestions about other contextual questions to ask for both likelihood and imminence:
1) Are the words addressed to a particular person?
2) Is there a victim/target of the words?
3) If there is a victim/target, is it a specific victim/target, or a general target.
4) Is there a timetable for the proposed action?
5) Is there a specific method for carrying out the proposed action?

Greg said...

Prof. Segall, I'm asking if either of those scenarios would fail the current imminence rule as it stands today.

I wonder if it could be successfully argued that any clearly identified date meets the imminence requirement, no matter how far in the future, because the real purpose of the imminence requirement is about being in the near enough future to be an identifiable time rather than a hypothetical, not about being within some arbitrary time limit.