Americans almost universally believe that freedom of speech is an unmitigated good — and also a nearly absolute right. Enshrined in the First Amendment, it is among our most fundamental rights, essential to a democratic system of government and in many ways the foundation upon which all other liberties rest.
A corollary to this belief is that any restrictions on speech are dangerous, and any attempt to regulate, limit, or control speech is a step toward tyranny. Ideas must be free to compete in an unregulated market. Even ideas many believe to be harmful — and maybe especially those ideas — must be aired and countered only by other, presumably more powerful, ideas.
Of course, there have always been a few exceptions. Nearly everyone has heard the cliché that you can’t yell fire in a crowded theatre. You can’t use speech to commit fraud. False advertising is illegal. Libel, slander, verbal harassment, threats of violence, incitement, and hate speech can also be crimes.
These limited exceptions aside, Americans tend to believe the government should place few if any restrictions on speech. And private institutions that exist to promote and publish ideas — universities and the press, in particular — are understood to have a duty to allow all views to be freely expressed and debated.
In recent years, however, two developments have challenged our thinking around free speech:
First, there has been growing acceptance that certain forms of speech or the expression of certain ideas can materially harm individuals or entire classes of people and should therefore be restricted. Speech that creates a hostile or unwelcoming environment for women, for racial, ethnic, or religious minorities, or for people whose gender identity or sexual preferences vary from what is conventionally expected is increasingly seen as unacceptable.
Second, the rise of social media has resulted in the spread of false and misleading information, disrupting our social and political order, and leading to calls for stricter regulation of speech on these platforms. The social media companies resist regulation, touting their platforms as enablers of free speech that beneficially empower ordinary people to connect with a broad audience directly. A debate rages between those who want to limit the spread of destructive information on these websites and those who see any limits as dangerous constraints on freedom of speech.
The dispute about whether to restrict some types of speech hinges in part on our conception of speech. Most of us think of speech as the expression of ideas. And we think of ideas as operating inside our minds, affecting the real world outside our minds only if we choose to act on them. We tend to believe there is a barrier between idea and action, one we cross only if we make the conscious decision to do so. Ideas, therefore, can cause no harm on their own. Only if we choose to act on them can they be harmful, and it is the action, not the idea behind the action nor the speech that disseminates the idea, that can be regulated.
This conception of speech may be naïve, however. Advertisers have long recognized that speech (defined broadly to include not just words and text, but also broadcast images and sound) directly influences our emotions, and our emotions drive action more than ideas. Much of what we do is determined subconsciously, not consciously, and speech can influence the subconscious as much as it influences the conscious. In politics, negative advertising is used widely because advertisers know that once a negative image of a person is disseminated — even if the image is transparently false — it creates real suspicions about the person that are not easily eradicated by rational discourse. The damage is done by the act of disseminating the negative image, and once done, cannot be argued away by demonstrating the image is false or misleading.
A more sophisticated view of speech recognizes that it is rarely pure expression of ideas and is maybe more often a way of manipulating other people to believe, feel, or do things we want them to believe, feel, or do. Of course, we’ve long recognized that speech can motivate people to act in certain ways, but in the traditional view speech is used to persuade, not manipulate. Persuasion still operates in the abstract world of the mind where different ideas compete for dominance and the listener consciously selects which idea is best and acts on it only if he or she freely chooses to do so. Manipulation, though, presumes that by stimulating emotions or affecting the subconscious, speech can largely take away choice and impel people to believe or act in a desired way often without conscious awareness of why they are acting or any conscious decision to act as they do. Persuasion works on the conscious mind. Manipulation works on the subconscious.
When we begin to see speech as a tool of manipulation — a force to drive people to feel certain things and act accordingly — then purely free and unrestricted speech can be problematic. If a particular type of free speech is driving people to act in highly damaging ways — and if counter arguments are ineffective in halting the damage — then do we simply resign ourselves to the consequences? Is the damage caused by free speech, like the damage caused by unregulated firearms, merely “the price of freedom”?
This leads us to another deeply held American belief: that rights are absolute and that one must be either for a right completely or against it completely. There is no middle ground. Balancing a right against other rights or asserting that its free exercise may depend on context is anathema. In the American world of rights, we tend toward all-or-nothing thinking. To protect the absolute right of free speech we must leave it completely unconstrained, even if doing so drives our society and our democracy to ruin.
Here is where we maybe should begin to consider our preconceptions about free speech. There is no doubt that freedom of speech is in general a civic good, essential to the preservation of liberty and the maintenance of a democratic system. The people must be able to express and to hear different, unpopular, and even seemingly dangerous ideas and evaluate them on their merits.
But there are times when speech does cause harm. And if the harm is significant, mitigating the harm may be as desirable as protecting the speech. Sometimes argument and persuasion are enough. But when a counter argument is not enough or when the harm is caused not by the idea that’s expressed but by the emotions the speech stimulates and the actions those emotions animate, what is to be done? Should the speech be restricted or banned?
Two real-world examples may help us answer this question. First, let’s consider the case of the protest that silenced Charles Murray when he was invited to speak at Middlebury College. This protest — while praised by some “woke” progressives — has widely been condemned even by many liberals as a dangerous attack on free speech. The argument is that Murray — even if his ideas are offensive and ill-supported by the facts — must be allowed to express them. The way to address Murray is not to silence him but to present a more powerful argument to counter his own.
Murray is best known for his works on IQ and race. Murray claims that IQ is a significant and reliable measure of human intelligence and that much of an individual’s success or failure in life is determined by the individual’s IQ. Furthermore, IQ is highly heritable and varies by race. Black people, on average have lower IQs, and therefore their lower average economic success is likely caused not by discrimination but by their lower average intelligence. Further, Murray implies that IQ is likely heavily genetically determined and mostly unchangeable. There is therefore little that can be done to improve the average economic success of Black people. Affirmative action and other programs intended to help Blacks achieve equality are misguided and doomed to failure. Black people are condemned by their genetics to lower average success and the best we can do is find roles for them in society suitable to their inferior intelligence.
Murray’s arguments are in many ways similar to antebellum arguments in favour of slavery, just bolstered by a lot of statistics about IQ and economic success. The antebellum arguments for slavery similarly claimed Blacks were inherently inferior — intellectually, behaviourally, morally — and would be unable to succeed in white society. Slavery was therefore justified as providing Blacks a form of security in a role suited to their inferior abilities.
The protestors at Middlebury wanted to deny Murray a platform for his ideas. The argument for silencing Murray is that his ideas, by giving credence to old racist stereotypes and wrapping them in a veneer of scientific credibility, materially hurt Black people. Stereotypes create a powerful negative image of Black people that becomes pervasive and persistent in society and influences the way Black people are perceived and treated — not just by those who are overtly racist but even by many who don’t intend to be racist. Even Black people themselves can absorb negative stereotypes about their own race and be demoralized by them, accepting that success in certain areas of life is not for them. In the protestors’ view silencing Murray is essential to preventing the reinforcement of damaging stereotypes and the harm those stereotypes do.
Those who opposed the protest may agree that Murray’s ideas promote racism and harm Black people but believe the only legitimate way to counter harmful speech is with more speech. These would allow Murray to make his case and then have someone challenge it, as Middlebury in fact intended to do by having another professor question and debate Murray during the lecture.
Those for the protest and those against it differ in how they look at Murray’s speech. The protestors see it primarily as an act — a direct and slanderous attack on the character of Black people. Allowing Murray to speak and debating his ideas gives those ideas both undeserved credibility and currency. In the protestors’ view, for Middlebury to allow Murray to speak is to be complicit in his act of defaming Black people. Preventing Murray from speaking, therefore, is a moral imperative.
Those against the protest see Murray’s speech as the expression of ideas, ideas which can be disproven by counter arguments. They don’t see Murray’s ideas as directly doing any damage. In their view, Murray’s ideas can be debated and disproven before they have significant real-world impact. In fact, many would argue that defeating the ideas definitively is the best way to limit their ability to affect the real world.
Murray’s ideas, however, have not been so easily defeated. Quite the opposite, Murray remains a popular figure on the right, enthusiastically embraced by white supremacists, opponents of affirmative action, opponents of social justice and welfare programs, and others who want to deny racism and racial discrimination are significant factors in our society today. This is true despite the fact that strong counter arguments to Murray’s arguments have been presented for decades. The protestors might well argue that the right-leaning student group that invited Murray to speak had no interest in actual debate of his ideas — but was really engaged in a far more political activity of giving Murray’s ideas increased currency and credibility regardless of their strengths or weaknesses.
In fact, the dispute in the protestors’ view wasn’t about Murray’s ideas at all, but about whether the college would be complicit in promoting those ideas. If a student group were to invite a speaker to Middlebury who advocated for genocide of the Jews, we can be almost certain that Middlebury would deny that speaker a platform. The denial of a platform, further, would almost certainly be uncontroversial. The fact that Middlebury did not deny a platform to Murray was itself a statement by the college that Murray’s ideas are not unacceptable. The protestors made an opposite statement — that Murray’s ideas should be considered completely unacceptable — and that by allowing Murray to speak, the university was complicit with Murray in promoting racism.
So which side is right in this dispute — those who want to silence Murray, those who want to debate him, or those who want to promote him? The correct answer, I’m afraid, is there is no clearly correct answer. Americans like to believe there is some absolute and universal principle that could be applied to give us a definitive answer. But reality is more complex. Allowing Murray to speak does indeed give legitimacy to racist ideas, promote racism, and thereby cause real harm to Black people. The protestors are correct — Murray is not just expressing ideas, he’s committing an act of defamation.
Yet Murray’s ideas can’t be so easily suppressed. They are already widely published, and debating them — ideally disproving them — is maybe necessary to help defuse them. In fact, suppressing them likely has the unintended consequence of making them even more attractive to those who want to believe them — and the protest at Middlebury may itself have publicized Murray’s work and helped expand Murray’s platform. Most significant, though, free expression is a powerful good in itself, and we should always weigh our decisions on regulating any speech heavily toward keeping speech free. Of course, the protest itself was a form of speech, so preventing the protestors from assembling and loudly declaiming their forceful objections to Murray’s presence could itself be seen as a restriction on speech.
Rather than trying to find a definitive answer to this dispute — or an absolute principle that can resolve it in all cases — what we should do is ensure there’s a good process for determining in each particular case what’s best. Ideally that process leans very heavily to allowing all speech without restriction. But if a particular act of speech can be shown to produce significant material harm, restricting it or even, in rare cases, prohibiting it may be justified. We just need a good process for coming to a decision in a complex world where right answers aren’t always clear. Indeed, this is the case for everything we recognize as rights. While we should lean heavily toward leaving them unrestricted, we can’t ignore the fact that in our complex world rights can conflict with each other and can at times, if left completely unrestricted, cause serious and unacceptable harm.
The second example — social media — maybe makes the complexities of free speech and the regulation of speech even more apparent. The most idealistic way to look at social media is as a tool, usually provided for free, that allows any individual to publish any digitalized information — text, images, sound, and video — to a global audience. This is, in fact, how companies that provide social media platforms describe their products: as tools that build connections between people by allowing them to share information easily with each other and that democratize publishing by stripping away the gatekeepers, editors, and curators and allowing any individual to create and disseminate content directly to an audience that potentially includes everyone in the world with internet access.
But social media is not simply a publishing tool or a tool to facilitate connection between individuals. It is also a tool to collect data about the publishers, the readers, and the published content. That data is then used for two interrelated purposes. First, the data tells us a lot about the people publishing and reading the content. Mostly it tells us what type of content they like the most, will be keen to share with others, and will prompt them to act in some desired way, such as clicking to another website, buying a product, or sending a donation. Second, the data tells us a lot about the published content itself — what types of content are most likely to get the strongest reactions from a specific targeted person or group. This data allows the published content to be continually refined so that it gets ever more effective at producing the desired response from its targets. In fact, all the social media websites make money by using the data they collect to help advertisers find the people most likely to buy their products, refine their ads for maximum impact, and successfully motivate people to purchase their products.
Viewed cynically, social media can be described as a tool to maximize the manipulative power of information — with advertising being one of its more benign uses and propaganda and the spread of misinformation among its more sinister uses. Its extraordinary power comes from three of its characteristics:
· Digital content — the internet makes it easy to produce, modify, and reproduce text, images, sound, and video almost infinitely. This dynamic content can be extremely powerful at stimulating an emotional response. And the content, though it looks “real” is easily manipulated digitally. “Deep fake” videos, for instance, can show film of a person doing or saying something he or she never did or said — and the film looks completely real. It is possible with digital media to get people to believe as real things that are completely fake and to get them to react to those fakes with powerful emotions.
· Social networks — the second characteristic of social media that contributes to its manipulative power is its use of social networks. Human nature is such that we tend to trust information coming from people we know or view as friends or allies. Advertisers have long known that endorsements or testimonials from trusted people are among the best ways to convince buyers to purchase their products. We also crave engagement and dialogue with others and tend to respond positively those who seem to like us or share our interests and backgrounds. Social media is designed to create networks of people who trust each other — and then spread information along those networks. This an extremely efficient way to target information to receptive users, to ensure the information is trusted when it is received, and to increase the likelihood that the information will be viewed, replicated, and disseminated further.
· Data — we’ve already described how data is used both to target messages to the most receptive recipients and to refine those messages for maximum impact.
We often use the term “viral” for information that spreads rapidly and broadly through social media. The three characteristics of social media described above — compelling digital content, trusted social networks, and abundant usage data — allow for this viral replication of information. In fact, taking the biological metaphor further, we can look at posts of digital information on social media as evolving organisms that mutate and adapt in ways that allow them to better survive and replicate within the social media ecosystem. Variations on information that stimulate a strong response from recipients will persist and be replicated. During replication multiple users will modify and expand on the information, creating new “mutations” even better fit to survive and replicate. Content can therefore evolve, becoming ever more effective at capturing the attention of its targeted users and manipulating their beliefs and actions.
The mutations themselves can be created by users with specific intent — for instance political advertisers may constantly put out new content to see what gets the best response, the most likes and reposts, and proves most likely to stimulate readers to contribute money to a campaign or sign up to support a candidate. But there is also the possibility that people will create new digital information without any clear intention other than to get likes and reposts. Conspiracy theories like QAnon may indeed have no clear purpose, but simply thrive and replicate because people like them and respond to them. In the social media ecosystem, the conspiracy theories evolve to be ever more effective at capturing attention and being replicated. Over time, complete nonsense may come to be accepted in some social networks as important and compelling truth. It may even motivate some to violent action — carrying a gun into a restaurant to free (imaginary) abused children as happened in the Pizzagate conspiracy, for example.
While a tool like social media may be used for good purposes, something this powerful at manipulating people’s beliefs, stimulating their emotions, and motivating their actions will also be used by malicious actors, often operating under fake identities designed to build trust among their targets. These actors may have a clear goal — for instance, a foreign power might be interested in helping a particular political candidate come to power in the United States because it sees that candidate as sympathetic to the foreign power’s interests. Or the same foreign power may have no goal other than creating chaos, division, and mistrust to help weaken the United States, disrupt its social cohesion, and make the country increasingly difficult to govern. Or a domestic political party might find it useful to create distrust of elections they lose, potentially getting people to demand that the results of the elections be overturned and the winning candidate replaced by the losing.
Again, we run into a dilemma. If we believe free speech is sacrosanct, then regulating speech on social media is something we strongly avoid. Yet, if various foreign and domestic actors with malicious intent are using speech on social media to turn the United States into a failed state, do we simply tolerate it because the right to free speech is absolute regardless of consequences?
Much as with our earlier example of Charles Murray and the Middlebury College protest, there isn’t a simple rule that can be applied in all cases. Social media is indeed a very positive tool to democratize speech and empower ordinary individuals to self-publish and reach a vast audience. But it also can be used as a weapon to destroy the social and political fabric that holds the nation together, to lead people to believe in falsehoods, and to stimulate harmful — even sociopathic — behavior. If we consider the right to free speech absolute and view any regulation of speech as an unacceptable threat to that right, we run the risk of allowing the viral spread of malicious and false information to break our society apart. Ultimately all our rights and freedoms depend on a healthy democratic society for their preservation. If in our attempt to protect free speech we allow our democratic society to be destroyed, we will do exactly the opposite of what we intend. Not only will we lose our freedom of speech, we will lose every other right and freedom too.
