Freedom of speech, narrowly defined, exists if the government cannot punish people for what they say. That form of free speech is implied by the First Amendment to the US Constitution. Most countries, including most western countries, have no equivalent in their legal system.
Freedom of speech, broadly defined, is not implied by the First Amendment. It is legal to fire an employee or choose not to hire an applicant because you don’t like what he says. More broadly, if you are a private person and not a government agent, anything you are normally free to do or not do you are still free to do or not do because of something someone said. The narrow and broad senses of freedom of speech are different legally, different in some other ways, but many of the arguments for and against them are the same.
One reason to change how you interact with someone in response to something he says is that his saying it gives you information about him relevant to the interaction. If you are considering hiring someone as an elementary school teacher, the fact that he says he hates small children is a reason not to, and similarly if the doctor you were considering going to tells you that he treats his patients on the basis of the medieval theory of humors. Some of the arguments I will be discussing are relevant to such situations but what I am interested in are the cases where the motive for firing someone, not hiring someone, making some decision with regard to him that makes him worse off, is to punish him in order to deter people from saying the sort of thing he said. That is the subject of this essay.
Arguments Against Freedom of Speech
One reason to prevent someone from saying something is that you believe it is false and that it is better for people to believe true things. During the Covid pandemic the government made considerable efforts to suppress speech online, not by punishing it but by pressuring online media not to repeat it. Their argument was that false statements, such as claiming that the Covid vaccines did not work and/or had serious side effects or claiming that treatments that didn’t work did, would result in people dying who otherwise would live. In other societies at other times, the argument for banning and punishing certain religious views was that they were false, would result in people going to Hell who would otherwise be saved. In our society, one argument for suppressing arguments claiming innate racial differences is that they are false and will result in people making incorrect decisions, discriminating on the basis of race rather than judging each individual on his own merits. This class of arguments can be used to defend violations of free speech in both the narrow and the broad sense.
Another reason is that someone saying something may have bad effects that do not depend on whether it is true. An atheist might believe that religious faith is a useful fiction, inducing people to behave themselves for fear of divine punishment, and so want to prevent people from arguing for atheism. Someone concerned about race relations might believe that the true fact that intelligence is in part heritable would encourage the false belief in racial differences in the distribution of intelligence, or that the true belief in racial differences in the distribution of intelligence would be used to explain away racial differences in the distribution of outcomes such as income that were in fact due mostly to racial discrimination.
Expressions of or arguments for a belief can be markers for belief in an ideology, membership in a political faction. If you believe that increasing the size and strength of that faction makes the country or the world worse, as most on the right believe of the left and vice versa, making it costly to express those views or arguments has three effects, all to you desirable. It makes it costly to hold those views, pushing some who do out of that faction and, with luck, into yours. It makes it costly to spread those views, reducing recruitment into that faction and defection from yours. By suppressing the expression of those views it makes the faction look smaller, weaker, less worth joining, allying with, paying attention to.
In 1964 I was a Harvard student and a supporter of Barry Goldwater, the Republican candidate for president. The Crimson, the school newspaper, reported that nineteen percent of the students supported Goldwater.1 I was surprised; I would have said that only a dozen did and I knew all of them. That reflected in part the non-random set of students I knew. But it also reflected the fact that if you were a Goldwater supporter at Harvard and did not like arguing with people who thought you were stupid, evil, or both, you kept quiet about your politics.2 Social pressure is one way of repressing freedom of speech, broadly defined.
A somewhat different argument against freedom of speech, broadly or narrowly defined, was illustrated in the aftermath of the assassination of Charlie Kirk when a considerable number of people were attacked, in some cases fired, for saying negative things about the victim. The same criticisms made a month earlier, when he was alive and active, might have gotten pushback from his supporters but not of the same sort; probably nobody would have been fired for it.
The difference was that criticism of Kirk after his death could be taken as justification, perhaps veiled, of the assassination. Permitting such could be seen as normalizing political assassination, perhaps encouraging a repeat with a different victim. Many people other than admirers of Kirk were strongly opposed to the normalization of political assassination, willing to approve of slapping down anyone who could be interpreted as doing so.
So much for arguments in favor of repressing speech. What is the argument against?
The Argument for Free Speech
It starts, like the argument on the other side, with the belief that it is on the whole better for people to believe what is true. The question is what mechanism gets you there. To put it differently, what mechanism for filtering speech does the best job of letting truth through while blocking falsehood?
The ideal mechanism would be an omniscient censor committed to truth, someone who always knows whether a statement is true or false and only blocks the false ones, but we don’t have any of those. Actual censors, people who act to prevent speech they disapprove of, are not omniscient and may have reasons, such as those just described, to block true speech and permit false. The ability to block or punish speech depends on a variety of characteristics, none of which reliably select for either wisdom or honesty. The ability to violate freedom of speech narrowly defined depends on political power, easier to obtain for someone willing to lie to get it, and those with political power have an incentive to promote beliefs according not to whether they are true but whether they favor themselves and their political allies.
The ability to violate freedom of speech broadly defined depends on being a party to relationships that benefit other parties, giving you the ability to make those parties worse off by ending the relationship, the obvious example being an employer. That probably does not anti-correlate with honesty, correlates with knowledge of what is true, but for only the narrow range of questions relevant to being a valuable person in a particular sort of relationship.
I conclude that filtering speech by the violation of freedom of speech, in either sense, has no particular reason to promote truth.
There is a possible counter-argument — that the ability to violate freedom of speech in both senses depends on how many people, especially people in your niche, agree with you. If twenty percent of academic economists disagree with my views and want to suppress their expression by refusing to hire anyone who expresses them that has only a minor effect on me. If it is ninety-five percent I might have a hard time getting a job unless I conceal my views. If twenty percent of factory owners want to suppress my political views, after the one I work for fires me and announces the reason I can probably find another job. That is harder if it is ninety-five percent. And, in a democratic system, the chance that the government will want to suppress my speech is greater the greater the number that disagree with it. If the number of people who believe something is a good proxy for whether it is true, that is one reason why violations of free speech will tend to favor truth over falsity.
Not all false statements are equally bad, however. A falsehood believed by many does more damage than one believed by few. Hence even if free speech protects more false statements than true, as under some circumstances it might, it protects the most valuable true statements, ones that contradict widely believed falsehoods and so are particularly at risk of suppression.
Suppose that violation of free speech is prevented by some combination of laws and norms. Speech is then filtered by how many people are saying it and how convincing it is. That still gives some advantage to the more popular beliefs and those supported by whomever controls the government but it also gives an advantage to beliefs that can be better defended, which should, on average, give truth an advantage.
It does not work perfectly because sometimes the wrong argument is more persuasive. Most public discussions of tariffs, whether arguing for or against, take for granted the theory of absolute advantage, a theory of the economics of trade which is two hundred years out of date and logically incoherent. It is, however, easier to understand than the correct theory, as long as you don’t look at it too carefully. The result is that, insofar as most voters have a theory of trade behind their view of tariffs, it is absolute advantage, whereas most academic economists, having been required by their profession to look at the question carefully, argue in terms of comparative advantage.3
Free speech does not solve that problem — but neither does the alternative.
People on opposite sides of some important issue, left and right, Christian and atheist, climate catastrophist and climate skeptic, have a conflict of interest over whose speech is suppressed and whose promoted but a common interest in discovering what is true. I do not believe that climate change threatens catastrophe, am not even confident that it is a bad thing,4 but if it does threaten catastrophe I would like to know it and if the truth is some mix of my views and the views of those on the other side of that debate we would both gain by knowing it. I am an atheist but if I am wrong and Heaven and Hell exist I would prefer to find out in time to act on the information.
Learning from each other is easier if people do not have to worry that they will be punished if they offer arguments and evidence for what they believe.
The Case For A Peace Treaty
Some of the reasons to want to suppress speech, such as to weaken a rival faction, involve benefits to one side but costs to the other. Even if suppression is, from their standpoint, worth doing they may benefit by abstaining in exchange for a similar abstention by the other side. Even if each side can benefit by suppressing the other’s speech, both may be better off if neither can suppress the speech of the other.
Arranging and enforcing a treaty to permit free speech broadly defined is difficult; since suppression of speech is the result of independent decisions by multiple people there is nobody to sign the treaty. It is more practical to protect free speech narrowly defined, since government agents are embedded in a political and legal system in part hierarchical. Arguably the First Amendment is such a treaty, its broad outline accepted by each side in the belief that violation by their opponents will cost them more than violations by their supporters will gain them. It does not work perfectly but it works.
To enforce free speech broadly defined the best we can do is to encourage norms, on both sides, that treat punishing people for saying things you dislike as shameful behavior.
Past posts, sorted by topic
My web page, with the full text of multiple books and articles and much else
A search bar for text in past posts and much of my other writing
The original story said less, I think fourteen or fifteen percent, but that turned out to be an arithmetic error.
The last three paragraphs of this blog post describe a relevant conversation I had at the time.
I explain this in Ptolemaic Trade Theory.
For a sketch of my reasons see my first post. More details can be found in my later posts on climate.
When we use to ride motorcycle we had a rule "Better be alive than right". So instead of arguing with the car drivers about the rules they need to adhere to to ensure safety of bikers, we are supposed to take precautions from our side that even the dumb car driver can not put us in harms way.
One thing I realized listening to Milton Friedman's videos was that he had this amazing ability to put forward his views while totally appearing to be on the side of the other person. Countless videos of Mr. Friedman shows him arguing with socialists, communists, feminists, eco-activists, partisan politicians and so on and at no point Mr. Friedman would sound acrimonious. Hence no matter how badly anyone disagrees with him it is simply hard to **hate** him. Similarly, my interactions with you (David) in real life quickly gave away the vibe that you were inherently a kind person and genuine in your desire to be intellectually honest. Somehow that kindness when displayed makes other person much more open to your ideas fostering a better environment for exchanging ideas. Charlie Munger and Warren Buffet are other two people who fall in the same category as the Friedmans. I can not say the same about Thomas Sowell or Walter Williams. Sowell for example argues really well and is factually correct but comes across as someone who has disdain for the other side on certain occasions.
I think modern day politicians and activists should be more like Milton rather than acrimonious communicators they try to be. Perhaps social media and modern media rewards the latter behavior than former in short run and hence people like to use expletives in their speech, call other retards, mentally ill etc.
My above statement should not be seen as any kind of justification of violence what so ever. As a matter of principle I feel political violence is an evil and we must do whatever in our power to stop it.
PS. : I had not known anything about Mr. Kirk until this tragedy happened. I feel deeply sorry for his family and pray they find strength to go through this difficult times.
Fine and true in theory, yet missing the truth about the current norms on Free Speech in the USA—many Republicans/ conservatives have been punished by being fired for their speech. Like Amy Wax is fighting.
On Charlie Kirk, many Dems claim, without evidence, that he is hateful & promotes hate, then use this to justify celebrating his murder. It is ok to celebrate the death of Mussolini, as a genuine enemy.
It’s not ok to celebrate the death of a peaceful politician, even with policies you don’t like.
The murder celebrants should all be fired.
Firing them will be the most likely path towards a norm against assassinations, as well as a tit against the many Dem tats, which is the most likely path towards a broader Free Speech norm. One that’s in accord with Rule of Law, so that Dems are less willing to cancel Republicans for speech.
The partisan Harvard should lose its tax exemptions, for being non-partisan, until it has at least 30% Republican professors & 30% Dems.