Free speech is not a game of who can shout the loudest

I realise I am coming back to a topic that was discussed getting on for three weeks ago (and I have covered more than once over the past year or so), but I have been mulling it over while trying to keep my head above water in the wake of the seasonable deluge of work for my day job.

Yascha Mounk, the German–American political scientist and commentator, recently hosted a discussion at the Global Free Speech Summit at Vanderbilt University, in Nashville, Tennessee, for The Good Fight Club, in which the central thesis was that, when it comes to free speech, everyone is a hypocrite.

Or, in other words, the trouble with free speech is that everyone wants it—but only on their own terms.

In theory, all the contributors to the discussion were speaking from a mutually held position: in defence of free speech; in opposition to censorship; and in favour of pluralism. But just beneath the surface, there were, naturally enough, cracks. Whose speech are we really defending? And at what cost?

The conversation, which featured Jonathan Rauch (a senior fellow in the Governance Studies program at the Brookings Institution), Jacob Mchangama (Executive Director of The Future of Free Speech and a research professor at Vanderbilt University) and Renée DiResta (an Associate Research Professor at the McCourt School of Public Policy at Georgetown), was pitched as a candid, even combative exchange between thinkers of different political stripes.

And yet what stood out was not the breadth of disagreement but the repetition of the same refrain: We must protect speech, but we must also control it. As ever, the devil in this plainly contradictory notion lies in the definition of “we”. Implicit in this concept is the idea of who constitutes “us” and, by extension, “them”, which cuts through to the perpetual human need to categorise and to divide.

It was striking to me that the participants began by invoking John Stuart Mill—a nod, perhaps, to the noble philosophical heritage of the free speech debate. But it also underlined just how old and intractable this discussion is. For all our technological advances and social evolution, we have not yet solved the core dilemma: how to allow maximum freedom without enabling the degradation of the public square.

Rauch, author of The Constitution of Knowledge, put it this way:

“Free speech is necessary but not sufficient if you want to live in a world that is knowledgeable, peaceful, and free. You also need a fairly elaborate structure of institutions, rules, and norms that shape how we talk to each other in the search for truth.”

I agree, and when looked at in the context of the real world and thinking about the law, and society’s checks and balances that allow us all to get on with our lives, it is all rather obvious.

Yet the longer I spend online, the more convinced I am that we are failing to build the virtual version of these structures. Social media, which is, at its core, entertainment-driven, algorithm-fed and controversy-powered, is not simply a mirror of society. It is a filter, a distorting lens, and increasingly, a manipulator.

We talk about speech online as if it were casual and/or inconsequential, as if a tweet or a TikTok video were the digital equivalent of shouting in the back room of a pub or in a public square. But it isn’t.

What we say online is recorded, amplified, and—depending on the whims of an opaque algorithm—delivered with surgical precision to audiences most likely to react. Crucially, it is never lost; it can be found at any time and served to the world in a decontextualised freshness with the greatest of ease.

And herein lies one of the core misconceptions we’re still grappling with: the persistent illusion that online speech is ephemeral; that it vanishes unless someone takes the trouble to dig it up, like some researcher in a library leafing through yellowing copies of a long-forgotten local newspaper.

In truth, what we say online lingers indefinitely, inscribed not only in databases but in the emotional and political landscape of our time through its incorporation into endlessly evolving algorithms that constantly mould the filtered version of the virtual world that we encounter on a daily basis.

We have grown up, culturally speaking, with the notion that speech is free, that more is better, that the best ideas will rise to the top. But that idea, rooted in the world of printed pamphlets and public debates, is ill-equipped for a reality in which bots, monetisation strategies and engagement hacks dominate the flow of information.

In this context, my own rule of thumb for online discussion is simple: Say only what you would say to someone’s face. It’s astonishing how much “free speech” melts away under that test.

Moreover, the idea that social media is a chaotic free-for-all misses the truth. Our feeds are curated—by code, yes, but code written by humans. There is intention behind what we see, even if it is abstracted; as much intention as there is behind the editorial direction of a newspaper. And that intention is not noble, as it might occasionally be in the printed press. It is purely commercial, no matter what some proprietors might insist.

That is why, despite what free speech maximalists argue, platforms do not exist outside of responsibility. If an algorithm amplifies hate, disinformation or extremism, that has consequences. These platforms are not like town halls. They are privately owned media empires, with all the power—and all the obligations—that that entails.

Rauch again makes a telling point:

“Social media is not part of the constitution of knowledge. It is an entertainment product, an advertising vehicle.”

That distinction matters. Facebook and X (formerly Twitter) are not truth-seeking instruments. They are content farms designed to maximise engagement. And outrage is highly engaging.

In my earlier column on this subject, I argued that we must understand speech not just as liberty, but as responsibility. We have countless examples—Nazi Germany, Rwanda, the Balkans—of how speech can precede violence. We have seen, most recently in the U.S., how January 6 was not spontaneous combustion but the slow burn of rhetoric and propaganda.

To pretend that speech does not have material consequences is to play with fire. It is not censorship to limit hate speech. It is self-preservation.

But what Mounk and others seem unable—or unwilling—to admit is that America’s rigid First Amendment absolutism is not the norm, and in many respects, not even desirable. In Europe, where the memory of catastrophic violence is fresher, there is more nuance. Holocaust denial, hate speech, incitement to violence; these are restricted not because Europe is anti-freedom, but because it remembers what happens when such freedoms are weaponised.

At one point, Mounk argues that everyone is a hypocrite about free speech: progressives, conservatives, centrists; that everyone is in favour of speech until it turns against them.

That may be true, but it misses the deeper point. The issue is not hypocrisy. It is that speech has real consequences, and different societies have made different decisions about where to draw the lines.

Pretending that one culture’s approach—namely America’s—is the only valid one is not the defence of freedom. It is its erasure.

One of the more revealing tensions in the roundtable came when participants discussed content moderation. Rauch initially defends it, saying he previously thought: “You cannot do without it”. But he quickly acknowledges that the public will not tolerate anything that appears “heavy-handed.”

“I think we should be looking at…middleware, which gives individuals more control over their algorithms, and transparency, which lets us see and evaluate what is going on.”

It’s a modest proposal. And one that tacitly acknowledges a deeper truth: there is no perfect solution. There is no switch to flip that will balance liberty, civility, truth and plurality. What we have instead is a collective responsibility.

The truth is, we don’t agree on what kind of society we want. And until we do, we cannot build a shared architecture for speech. For one person, a drag queen reading a storybook is an act of joyful inclusion. For another, it is an affront. Whose view wins? That is not a question technology can answer. And crucially, there is no ‘answer’, right or wrong, and no winning of one view over another. They are both opinions, viewpoints, perspectives. In the real world, we can agree to disagree. Free speech as practised in the virtual world does not allow for that pragmatic co-acceptance.

And that is precisely where the idea of free speech breaks down.

To find a way through the impasse, we must educate ourselves. We must speak with care. We must listen to others with respect. And we must accept that the freedom to say something does not mean it should or has to be said.

That’s because free speech is not a game of who can shout the loudest. It is a compact. A bargain. A fragile agreement to use words not as cudgels but as bridges.

And if we fail to honour that, we risk not censorship, but societal collapse.