The internet has made it easier than ever to speak to others. It has empowered individuals to publish our opinions without first convincing a media company of their commercial value; to find and share others' views without the fuss of photocopying and mailing newspaper clippings; and to respond to those views without the limitations of a newspaper letter page. In this sense the internet has been a great boon to the freedom of speech.
Yet that very ease of communication creates new limits to the freedom part of free speech: the ability to speak our mind to those we wish without fear of reprisal.
I
The first problem is that what was once a difficult endeavour – to bring our words to the attention of others – is becoming difficult to avoid. An increasing amount of speech and its proxies (such as 'likes') is subject to automatic publication to the world. If not by us, if we are very careful with all our privacy settings, then by the devices and apps of those around us. It is becoming impossible to guarantee a private conversation.
That matters because the way we express our ideas and opinions in conversation, to specific people, is not how we set out our thoughts to the world, to strangers in general. The practical difference in the labours required for speech and publication used to track and support that distinction.
Speech is improvised and ephemeral. On the one hand it has no substance in itself but dissolves into the air and fallible memories of those around you. On the other hand it matters greatly as a means of working out and testing what you should think about something by trying out different positions, reasons, and phrasings. That gives it a dialectical character - you engage in it with other people in an effort to make progress together. It is often part of an ongoing relationship in which the parties know each other, bear some good will, and have a common knowledge and context to relate to. What is said can be directly addressed to the other's perspective and concerns. Your mistakes and their misunderstandings can both be corrected as you go along, and without being held against you.
Publication in contrast is – or was – a special and daunting undertaking, requiring great diligence and prudence to compose a version of your ideas that might stand the scrutiny of all sorts of readers without you being able to step in to explain. Those readers might live thousands of miles or even years away from you. They might dislike what you have to say. As Plato had Socrates put in Phaedrus,
While speech is designed around deep positive engagement with particular people, publication is designed around the challenge of generic communication and its virtues tend to be more defensive. Because the same form of words must communicate the same thing to everyone, what is published must be written for no one in particular and yet be clear enough for all. Furthermore, one does not get to choose one's readers. Some will be looking for reasons to disagree with you. If you care about that then you must fill your text with preemptive clarifications and defenses against their possible objections. (Academic writing is so bad because it takes this defensiveness to extremes. An existential terror of being misjudged - or having one's mistakes found out! - leads first to diluting one's claims with endless caveats and then adding in pompous verbiage to make it look as if something is still being said.)
And when [speeches] have been once written down they are tumbled about anywhere among those who may or may not understand them, and know not to whom they should reply, to whom not: and, if they are maltreated or abused, they have no parent to protect them; and they cannot protect or defend themselves.
While speech is designed around deep positive engagement with particular people, publication is designed around the challenge of generic communication and its virtues tend to be more defensive. Because the same form of words must communicate the same thing to everyone, what is published must be written for no one in particular and yet be clear enough for all. Furthermore, one does not get to choose one's readers. Some will be looking for reasons to disagree with you. If you care about that then you must fill your text with preemptive clarifications and defenses against their possible objections. (Academic writing is so bad because it takes this defensiveness to extremes. An existential terror of being misjudged - or having one's mistakes found out! - leads first to diluting one's claims with endless caveats and then adding in pompous verbiage to make it look as if something is still being said.)
When all speech is published - fixed for all time and for all to see - all speech faces the problems Plato identified. It must do without you to explain and defend your meaning. It must survive the caustic skepticism of people who bear you no goodwill and who are free to judge it against whatever standards they please. Moreover, because publications are assumed to reflect your settled views, your character as well as your meaning is on trial. You will be blamed for your mistakes, not merely corrected. You will also be blamed for misunderstandings on the part of the reader - often unfamiliar with the original context and uninterested in learning it.
The problem is compounded by the global character of internet publication. Previously, even publication usually came with a particular kind of audience. If one was publishing in a magazine, for example, one would have some idea of the interests and views of its few thousands of readers and how to talk to them. Books were trickier, but at least one knew in which country they would be sold. Nowadays one cannot predict at all where one's words will end up. Even language doesn't seem to limit their reach.
American progressives for example called out a respectable Dutch broadsheet newspaper for not following the complicated rules they have set for how and by whom the word 'nigger' may be used (they had used a quote from the book as the title of the review). The poor editor could only grovel that "The n-word is an English word and has a less offensive meaning in Dutch. We hadn't thought that it would be read in the US."
These days it is not enough to consider how your words will appear to the people for whom they are intended. You must bear in mind that anyone at all might discover them, share them with like-minded souls via social media, and hold you answerable to their moral standards. Efficient search engines also play a role by making it easier for people to discover speech that offends them, even long after you said it. You must assume that the very people you would least like to read your words are the ones most likely to find them.
Over the last 20 years we have seen the effects of inescapable publication on public figures. It is a sad vision of our future. Politicians for example, have become increasingly careful about what they say because any gaffe or tweet may be stripped of its context and used to humiliate and destroy them, perhaps years later. The result has not been greater accountability but a closing down of political communication. The complaint everywhere is that politicians parrot their lines like robots and never say what what they really mean or engage fully with the people they are talking to. They dare not take the risk! The other kind of politician who succeeds in this environment are the shameless boors like Donald Trump and Boris Johnson, who don't accept any responsibility at all for what they say.
The problem is compounded by the global character of internet publication. Previously, even publication usually came with a particular kind of audience. If one was publishing in a magazine, for example, one would have some idea of the interests and views of its few thousands of readers and how to talk to them. Books were trickier, but at least one knew in which country they would be sold. Nowadays one cannot predict at all where one's words will end up. Even language doesn't seem to limit their reach.
American progressives for example called out a respectable Dutch broadsheet newspaper for not following the complicated rules they have set for how and by whom the word 'nigger' may be used (they had used a quote from the book as the title of the review). The poor editor could only grovel that "The n-word is an English word and has a less offensive meaning in Dutch. We hadn't thought that it would be read in the US."
These days it is not enough to consider how your words will appear to the people for whom they are intended. You must bear in mind that anyone at all might discover them, share them with like-minded souls via social media, and hold you answerable to their moral standards. Efficient search engines also play a role by making it easier for people to discover speech that offends them, even long after you said it. You must assume that the very people you would least like to read your words are the ones most likely to find them.
Over the last 20 years we have seen the effects of inescapable publication on public figures. It is a sad vision of our future. Politicians for example, have become increasingly careful about what they say because any gaffe or tweet may be stripped of its context and used to humiliate and destroy them, perhaps years later. The result has not been greater accountability but a closing down of political communication. The complaint everywhere is that politicians parrot their lines like robots and never say what what they really mean or engage fully with the people they are talking to. They dare not take the risk! The other kind of politician who succeeds in this environment are the shameless boors like Donald Trump and Boris Johnson, who don't accept any responsibility at all for what they say.
II
The first problem the internet poses to free speech is that we have all lost control over who gets to hear what we say and what they make of it. We must assume that our words will be published and available to anyone anywhere in the world.
The second problem concerns the response to automatic publication. The same technology that allows our words to follow us around whether we like it or not has also allowed counter-speech to be weaponised in new or cheaper ways by people who like shouting at others - self-righteous indignants and sadistic trolls. (See Noah Smith's excellent analysis of 'The Shouting Class'.) Although shouters have always existed, the structure and dynamics of social media have given them enormous new power to direct negative attention against speakers they don't like.
This power of cyber terrorism dramatically raises the costs of speaking out or speaking back. It allows a tiny shouting class to censor speakers and ideas that displease them, while refusing to be accountable for their own use of this power of speech. Thus the new regime of 'political correctness' we live under, which is neither a tyranny of government nor or of the social majority (such as JS Mill feared). Rather it is a tyranny of minorities, of many small tribes who each believe they have a veto right over what anyone in the whole world may say.
Sadistic trolls are hooligans. They set out to upset people and provoke strife because they enjoy making people suffer. They are the ones posting videos like the Innocence of Muslims and sending abusive messages to people with cancer. They don't require much moral analysis, only containment. (Although trolling can also be used strategically - see Southpark's Theory of Trolling.)
The second problem concerns the response to automatic publication. The same technology that allows our words to follow us around whether we like it or not has also allowed counter-speech to be weaponised in new or cheaper ways by people who like shouting at others - self-righteous indignants and sadistic trolls. (See Noah Smith's excellent analysis of 'The Shouting Class'.) Although shouters have always existed, the structure and dynamics of social media have given them enormous new power to direct negative attention against speakers they don't like.
- Although shouters are only a tiny fraction of social media users, they actually seek publicity and so their reactions often reach many more people than the original speaker did. That makes their complaints seem important.
- Most people who hear them will consider shouters' complaints mistaken or excessive and just ignore them. But those people don't matter because disinterest is invisible on social media. Only engagement matters and any engagement will do. Even those who take the trouble to actively disagree with the shouter just wind up generating more publicity about what is disputed.
- By spreading their message both widely and narrowly (through networks of already similarly minded people), shouters can reach the few who actually agree with them. This passionately unhappy few are enough for an online shame mob that can not only unleash a tsunami of abuse on the person who offended them, but also coordinate on specific projects like threatening the reputation of their target's employer to get them fired. (As happened to an unfortunate Nobel Prize winning biologist: 1 person in his audience chose to interpret a joke literally and used social media to find enough other offendable people to muster a shame mob and get him fired.)
- All this can be done almost effortlessly and often pseudonymously, at the cost of a few seconds of attention and some clicks.
This power of cyber terrorism dramatically raises the costs of speaking out or speaking back. It allows a tiny shouting class to censor speakers and ideas that displease them, while refusing to be accountable for their own use of this power of speech. Thus the new regime of 'political correctness' we live under, which is neither a tyranny of government nor or of the social majority (such as JS Mill feared). Rather it is a tyranny of minorities, of many small tribes who each believe they have a veto right over what anyone in the whole world may say.
Sadistic trolls are hooligans. They set out to upset people and provoke strife because they enjoy making people suffer. They are the ones posting videos like the Innocence of Muslims and sending abusive messages to people with cancer. They don't require much moral analysis, only containment. (Although trolling can also be used strategically - see Southpark's Theory of Trolling.)
The other group is more complicated because they think they are acting morally. Those trying to punish people for posting disrespectful photographs, tweeting a failed joke, or being on the wrong side of a controversy such as Israel or transrights are motivated by moral beliefs. Just as most physical violence is motivated to enforce moral codes regulating social relationships, the indignants use cyber terrorism to punish those whose speech transgresses their moral code. From the outside they look like vicious thugs, but from the inside they are moral vigilantes, selflessly keeping the rest of us safe by wading into the sewers of twitter to call out and destroy disease bearing vermin.
The indignants are confusing moralism with morality proper. Moralism consists of enforcing the social conventions ('mores') one has become acculturated to see as natural and proper against transgressors. It relies on the psychological apparatus of moral reaction (such as disgust and indignation) that seems to have evolved in our species' prehistory to police groups' internal order and external borders. In contrast real morality focuses on content rather than social function: certain moral beliefs deserve defending because they are correct, not because they happen to be ours.
The indigants are moralists because they are driven by their moral reactions rather than a concern to get the content of their moral beliefs right. Their feelings of anger and disgust are taken as sufficient proof that their targets are wrong and evil. Moralism contradicts and undermines genuine morality. It heats up the blood and numbs the brain.
First, indignation drives people to act against the universal moral principles they claim to believe in. They often claim to be defending others from hurt, on the grounds that hurting people is wrong. But in their frenzy of indignation these moralists forget that those they seek to punish are people too. Their righteousness dehumanises their targets and casts them as demons that have brought their destruction upon themselves. Honest mistakes are not possible. The worst possible interpretation is the one that must have been intended. (As Mary Beard has once again discovered.) Not surprisingly, the moralist easily loses all sense of proportion. Seemingly minor transgressions can result in histrionic nastiness: rape and death threats; doxxing; the harassment of people connected to the target - family members, neighbours, employers, customers.
Second, indignation precludes moral growth. Investigating what morality requires is a different project from enforcing current conventions. Investigation requires open-mindedness to alternative views and the freedom to make mistakes - because one can learn from them. Speech matters here because speech is how you learn about and test ideas, such as how to get morality right. In contrast, moralism requires closed-minded loyalty to the crudest idea of politics as tribalism. Speech matters here because it identifies the speaker as a friend or enemy. Naturally enemies cannot be tolerated. People can't be allowed to just be wrong. They are an existential threat, not another human being trying to make sense of things like we all are, let alone an alternative perspective to learn from.
Besides being morally wrong, the moralists are practically foolish. It is one thing to enforce a social code within a closed community, for example by punishing girls who dare to ask for an education. It is another thing to enforce it against the whole world. For here you come up against many other moral tribes busy trying to enforce social conventions of their own. The resulting crossfire may hurt many innocents - such as the poor store clerks screamed at for saying 'Happy holidays!' instead of 'Happy Christmas!' - but it doesn't eliminate the transgressive behaviour that triggered your indignation.
The new political correctness of the indignants is distinctly limited in its power. It can only suppress the free speech of individuals who are psychologically or economically vulnerable to a social media shame-mob.
First, humans are social animals. Few people can withstand the tsunami of personal abuse that a social media shame mob can unleash (and which Google will helpfully associate with your name so that it follows you for the rest of your life). University students, for example, have always frightened governments because they tend to idealism, can organise easily, and have so little to threaten to take away. But students seem especially cowed by the new online call out culture.
Yet not everyone cares so much about reputation. Some people are just shameless - the reaction to Trump's pussy grabbing tape would have broken anyone with a normal level of self-regard. Others are ideologically motivated and socially supported by a community that can offer an alternative view of themselves as heroes. This is presumably how members of the Westboro Baptist Church overcome the abuse they receive for picketing soldiers' funerals.
Second, people need to make a living. Most people who speak do so as amateurs, in their free time. They are easily dissuaded or punished by putting reputational pressure on their employer or customers. Very few organisations - capitalist, government, or charitable - are willing to uphold the out of working hours free speech of their employees if it comes at any cost to their operations or prestige. The organisations that do defend their employees' right to say things that offend people are those whose financial or ideological interests benefit from it. For example, Breitbart and other rightist organisations provided an economic ecosystem that not only allowed Milo Yiannopoulos to resist leftist efforts to censor him, but also amplified his voice. That worked fine until an anonymous Canadian high school student decided he needed to be stopped and dug through old interviews for a transgression against the right that would get him fired.
What this means is that the cyber terrorism of the morally indignant is only effective against those who are either isolated or else members of their own communities. They cannot deter or punish offensive speech by real assholes, ideologues, the economically self-sufficient, or those who hide behind pseudonyms. But these are the people who say the most horrible things and at industrial scale. The indignation of the moralists can never be satisfied. Worse, their zealotry drives them to search the internet for speech that will offend them and then publicise it to each other. The indignants do enormous harm to the idea of a free society and to individual lives. But they are nonetheless to be pitied as the agents of their own misery.
The indignants are confusing moralism with morality proper. Moralism consists of enforcing the social conventions ('mores') one has become acculturated to see as natural and proper against transgressors. It relies on the psychological apparatus of moral reaction (such as disgust and indignation) that seems to have evolved in our species' prehistory to police groups' internal order and external borders. In contrast real morality focuses on content rather than social function: certain moral beliefs deserve defending because they are correct, not because they happen to be ours.
The indigants are moralists because they are driven by their moral reactions rather than a concern to get the content of their moral beliefs right. Their feelings of anger and disgust are taken as sufficient proof that their targets are wrong and evil. Moralism contradicts and undermines genuine morality. It heats up the blood and numbs the brain.
First, indignation drives people to act against the universal moral principles they claim to believe in. They often claim to be defending others from hurt, on the grounds that hurting people is wrong. But in their frenzy of indignation these moralists forget that those they seek to punish are people too. Their righteousness dehumanises their targets and casts them as demons that have brought their destruction upon themselves. Honest mistakes are not possible. The worst possible interpretation is the one that must have been intended. (As Mary Beard has once again discovered.) Not surprisingly, the moralist easily loses all sense of proportion. Seemingly minor transgressions can result in histrionic nastiness: rape and death threats; doxxing; the harassment of people connected to the target - family members, neighbours, employers, customers.
Second, indignation precludes moral growth. Investigating what morality requires is a different project from enforcing current conventions. Investigation requires open-mindedness to alternative views and the freedom to make mistakes - because one can learn from them. Speech matters here because speech is how you learn about and test ideas, such as how to get morality right. In contrast, moralism requires closed-minded loyalty to the crudest idea of politics as tribalism. Speech matters here because it identifies the speaker as a friend or enemy. Naturally enemies cannot be tolerated. People can't be allowed to just be wrong. They are an existential threat, not another human being trying to make sense of things like we all are, let alone an alternative perspective to learn from.
Besides being morally wrong, the moralists are practically foolish. It is one thing to enforce a social code within a closed community, for example by punishing girls who dare to ask for an education. It is another thing to enforce it against the whole world. For here you come up against many other moral tribes busy trying to enforce social conventions of their own. The resulting crossfire may hurt many innocents - such as the poor store clerks screamed at for saying 'Happy holidays!' instead of 'Happy Christmas!' - but it doesn't eliminate the transgressive behaviour that triggered your indignation.
The new political correctness of the indignants is distinctly limited in its power. It can only suppress the free speech of individuals who are psychologically or economically vulnerable to a social media shame-mob.
First, humans are social animals. Few people can withstand the tsunami of personal abuse that a social media shame mob can unleash (and which Google will helpfully associate with your name so that it follows you for the rest of your life). University students, for example, have always frightened governments because they tend to idealism, can organise easily, and have so little to threaten to take away. But students seem especially cowed by the new online call out culture.
Yet not everyone cares so much about reputation. Some people are just shameless - the reaction to Trump's pussy grabbing tape would have broken anyone with a normal level of self-regard. Others are ideologically motivated and socially supported by a community that can offer an alternative view of themselves as heroes. This is presumably how members of the Westboro Baptist Church overcome the abuse they receive for picketing soldiers' funerals.
Second, people need to make a living. Most people who speak do so as amateurs, in their free time. They are easily dissuaded or punished by putting reputational pressure on their employer or customers. Very few organisations - capitalist, government, or charitable - are willing to uphold the out of working hours free speech of their employees if it comes at any cost to their operations or prestige. The organisations that do defend their employees' right to say things that offend people are those whose financial or ideological interests benefit from it. For example, Breitbart and other rightist organisations provided an economic ecosystem that not only allowed Milo Yiannopoulos to resist leftist efforts to censor him, but also amplified his voice. That worked fine until an anonymous Canadian high school student decided he needed to be stopped and dug through old interviews for a transgression against the right that would get him fired.
What this means is that the cyber terrorism of the morally indignant is only effective against those who are either isolated or else members of their own communities. They cannot deter or punish offensive speech by real assholes, ideologues, the economically self-sufficient, or those who hide behind pseudonyms. But these are the people who say the most horrible things and at industrial scale. The indignation of the moralists can never be satisfied. Worse, their zealotry drives them to search the internet for speech that will offend them and then publicise it to each other. The indignants do enormous harm to the idea of a free society and to individual lives. But they are nonetheless to be pitied as the agents of their own misery.
III
The internet has enhanced our free speech in an imbalanced way: greater ease of reaching others with our speech has come at the expense of our freedom to speak without fear of reprisal. I focused on the particular phenomenon of social media shame mobs, but that is not all we have to fear from universal publication combined with efficient search engines. Employers googling you before job interviews may write you off on the basis of some ill-considered outburst you made on twitter years ago, the kind of ugly remark that might once have existed for only a few moments between friends at a bar after work, before being kindly forgotten by those who knew it wasn't the real you talking. Likewise the guy you asked out on a date or the parents of the children you teach may judge you by the single worst thing you ever said in your life, in the mistaken belief that it represents who you are.
We are going to have to adapt ourselves psychologically and institutionally to our new powers of speech. There are several approaches we might take, not all equally attractive or feasible. Likely we will end up adopting elements from several.
First, we could learn to censor ourselves, and teach our children from a very early age, to never say anything that we wouldn't be happy for the whole world to read in the length of a tweet. From a liberal perspective this is a ghastly prospect. Certainly we should learn to be more careful of what we say, out of care for others as well as prudence. But internalising full responsibility for other people's possible responses to our thoughts would be the end of individual freedom. Free speech matters because it is intimately linked to thought – speaking is thinking together. Liberalism starts from respect for the autonomy of the individual to form their own opinions on the right and the good. If we fear to think aloud because we fear the wrath of some part of society then we are not free to form our own opinions anymore. The fact that it isn't a tyrannical government stopping us from thinking is irrelevant to the tyrannical effect.
Second, we could vaccinate ourselves against cyber censorship by reducing our susceptibility to shame. Shame connects our sense of identity and well-being to what others think of us. It evolved to maintain harmony in small groups but can also be triggered by the online vitriol of distant strangers. Perhaps we should all be more inspired by that nursery rhyme about sticks and stones. Yet giving up on the mechanism of shame altogether would make us vulnerable to moral narcissism, as displayed by the likes of Donald Trump. A society of Trumps is a rather depressing prospect, if it is even possible. In any case, just because you're not ashamed of an ambiguous twitter joke doesn't mean you shouldn't fear losing your job.
Third, we could call on the government to save us from the mob, for example by restricting how search engines deliver the results of personal name searches (which has already been partly achieved in Europe); installing fire-breaks in social media networks that make the formation of flash mobs harder; making it illegal to fire the target of a shame mob without cause and due process; regulating social media to make it easier to keep private conversations from being published; making social media companies legally accountable for slanders and threats by pseudonymous members; and so on. Such interventions are censorship in the name of freedom of speech. Careful judgement and debate is required about the right balance of rights. Given the international character of the internet, some good ideas might not be institutionally feasible.
Fourth, we all - but especially shouters - could reconfigure our thresholds for moral disgust and indignation. It is trivially easy to find something on the internet that will outrage you to your core. Perhaps we can reduce our sensitivity to the lower quality of people's thoughts now being published to the world. We could just let people be wrong. Perhaps we can come to recognise, as the stoics taught long ago, that existing in a constant state of outrage about what strangers on the other side of the world are doing isn't all that great a way to live and neither does it do anything to make our society better. Perhaps we can learn to use our new powers of speech to talk about that instead.
Note: A previous version of this essay appeared in 3 Quarks Daily
This work is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License. Attribution: Thomas R. Wells is a philosopher at at Tilburg University. He blogs on philosophy, politics, and economics at The Philosopher's Beard.