What is Free Speech? A Review

“Freedom of speech” is never an absolute. Even in a country where people have a great amount of freedom of speech, such as the US, there will still be restrictions on the time, place, and manner of speech, laws regarding libel and defamation, and against fraud and blackmail, and against inciting violence. Freedom of speech will not give the right to take anyone else’s copyrighted works and use them and distribute them however you like without the permission of the author of the works. How these laws are interpreted matters a great deal: severe constraints on “time, place, and manner” can be used to significantly curtail freedom of speech. A state declaring that a social media post opposing the government is “inciting violence” will likewise chill speech. Governments can selectively enforce vague laws to suppress speech that is critical of the party in office or its allies, but be lenient regarding speech critical of the opposition. The constraints on free speech, and the discretion given to the state to apply those constraints, matter.

I had tended to look at the question of freedom of speech (and many other policy issues as well) as having, very broadly speaking, two ways of approaching it.

One is the economic / cost-benefit / utilitarian approach: what regulations on speech will, over the long run, maximize the sum of human welfare? Richard Posner’s “Free Speech in an Economic Perspective” (1986) applies the “Law & Economics” approach to the question, setting up a cost-benefit calculation of the expected benefits or harm from certain types of speech and the costs of suppressing it. The cost of allowing someone with a megaphone in front of an angry crowd to yell “it is time to smash some windows” is greater than the benefit of allowing such incitement, and so this particular bit of speech is not protected. The worry here - and this is a worry with the utilitarian approach to anything - is if this principle is applied, what if you are the person who draws the short straw in the cost-benefit calculation, the one who must suffer constraints or losses for the greater good? What if the state decides that all things considered society is better off not having access to the tracts of your tiny religious sect?

The other approach to free speech is from the perspective of rights: my freedom to post on my blog, and your freedom to read it, should not depend on a calculation as to whether this adds to or subtracts from the general public good. Call this a deontological approach. There are rights to which we are entitled as individuals that cannot be taken from us by the state, even if allowing me to exercise my rights would fail a cost-benefit test. The problem here is that it opens the door to some speech that really is harmful and nasty, that is meant to insult and to express hatred for no other purpose than to insult and hate, or speech that promotes medical cures that are in fact quite dangerous if applied. This is speech that by any account - sometimes by design - lowers aggregate well-being.

I don’t know anyone who follows entirely one approach or the other. I think it was Bertrand Russell (?) who said that any moral philosophy taken to its final implications leads to barbarity, and I think he was right. “It is wrong to tell a lie. Except for sometimes.” “Free speech is a natural right that we must defend. Except when…”

---

I’d been anticipating Fara Dabhoiwala’s What is Free Speech? The History of a Dangerous Idea since I first saw the reviews from the UK, where it was released a few months ago, as I lack both his historical and his international perspective. Maybe I had been looking at the question from the wrong angles entirely.

Dabhoiwala has two goals in this book. The first is to give a richer account of the evolution of our thinking and our laws regarding free speech than is typically presented. The second is to advocate, especially for the United States, a shift in balance away from the right to free speech - what he calls an absolutist approach - to something that weighs costs and benefits and is more willing to regulate bad speech - what I have called above a utilitarian approach and what he terms a balanced approach. I think he is quite successful in his first goal; it made the book worth reading. He is less successful in his second goal.

---

Dabhoiwala sets the origins of our modern conception of freedom of speech in the eighteenth century. Before that, it was seen as unremarkable that speech would be regulated, given its capacity to cause harm to individuals, and, in turn, to the society at large. Modern freedom of speech rests upon the foundation that there is a fundamental qualitative difference between harms caused by speech and physical harm: my saying insulting things about someone on this blog, or making false claims about things other bloggers have written, or about the nature of the world, is not a comparable harm to my meeting another blogger and poking him in the leg with a pointed stick. If you hear speech that you find insulting or offensive or false, you can respond with speech of your own, but otherwise you will just have to accept the rights of the speaker to say what they will. For physical harms you can call on the police, but not, for the most part, for harms from speech. (One of the things that made the Posner / Law & Economics approach to free speech controversial was that it did not take this modern line: for economists all goods, and all harms, are commensurable - reduceable to utility or disutility, dollar values gained or lost).

In the pre-modern world, injury by defamation or insult was a serious business, as serious as physical acts, not just in the harms it would cause the insulted individual, but on the social order as well. And so Hobbes, in Leviathan (1651, chapter 15):

And because all signes of hatred, or contempt, provoke to fight; insomuch as most men choose rather to hazard their life, than not to be revenged; we may … for a Law of Nature, set down this precept, That no man by deed, word, countenance, or gesture, declare Hatred, or Contempt of another. The breach of which Law, is commonly called Contumely.

So what is it that changed, that led to the adoption of modern freedoms? We can point to three things. The first was to allow, within limits, the expression of differing religious beliefs. The Protestant reformation, and the century of violence that followed across Europe, led to the realization that there had to be some scope for people to be able to believe, worship, and speak about their beliefs that would not subject them to persecution. John Locke’s “Letter concerning toleration” (1691) was a key document in this regard. The second was the retirement in England of laws requiring publishers to have pre-publication licensing. This expired in 1695 when parliament could not agree upon a way to extend it, which in turn led to an explosion in the publication of newspapers. The third was the publication in England, by the journalists Thomas Gordon and John Trenchard, writing under the pen name “Cato,” between 1720 and 1723, of “Letters” that took the arguments that had been applied to freedom of expression in religion to freedom of political speech. Parliamentarians already had the right to (mostly) speak freely in parliament, so that the crown could receive advice based upon the strongest arguments (a notion that continues to this day: “where there is freedom of speech, the truth will ultimately prevail”). But “Cato” would extend this right to citizens, and these Letters proved enormously influential; Dabhoiwala notes that the advocacy of free speech by Benjamin Franklin in these years was taken directly from these Letters. From one of the Letters:

Without freedom of thought, there can be no such thing as wisdom; and no such thing as public liberty, without freedom of speech, which is the right of every man, as far as by it, he does not hurt or control the right of another: And this is the only check it ought to suffer, and the only bounds it ought to know.

This sacred privilege is so essential to free governments, that the security of property, and the freedom of speech, always go together; and in those wretched countries where a man cannot call his tongue his own, he can scarce call anything else his own. Whoever would overthrow the liberty of a nation, must begin by subduing the freeness of speech; a thing terrible to public traitors.

Dabhoiwala here notes, first, that Cato’s Letters were far more influential in the formation of US free speech doctrine than in any other place, where freedom of speech is typically tempered by constraints where there would be significant harm to the public good. The United States is exceptional. For example, although it came into place much later, see the Canadian constitution, which guarantees “freedom of thought, belief, opinion and expression, including freedom of the press and other media of communication” but “subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society”. As a result, Canada has laws against “hate speech” (though prosecution of such is constrained by strict guidelines, now known as the “Oakes test”), while the US does not (Dadhoiwala does not discuss the Canadian case even though it is an excellent example of the “balanced approach” for which he advocates; he does refer in a footnote to Canada’s 1960 Bill of Rights, but the Constitution Act of 1982 is what is relevant now. Canada is right next door and provides many interesting examples for comparative policy - I will continue on with my mission of trying to get American scholars to realize this).

Second, he criticizes the Cato Letters as essentially hack-work, hasty and written in the interest of gaining readership for publications. It was not rigorous by any means, and the authors of the Cato Letters, and their followers, could not have cared less about the speech rights of women, or of Black people, free or enslaved. They ignore the many challenges one could make of a blanket right of freedom of speech: that truth does not always win out, that slander and misinformation come with very real costs, that a profit-maximizing newspaper will not necessarily prize reporting the truth above all else, and that the suppression of speech still occurs, but through private actors more than through the state (Dabhoiwala should have cited Elizabeth Anderson on this topic, whose book Private Government is excellent).

Still, in the Western world a great shift was taking place:

Instead of presuming that speech and print should be regulated, they proceeded from the opposite premise: all expression was permissible, unless specifically excepted. In the course of the nineteenth century, this principle would gradually become the norm everywhere.

Dabhoiwala is great at showing how different countries, experimenting with increased license for speech and the press, learned from each other, and his discussion of the importance of developments in Sweden, Denmark-Norway, and Poland-Lithuania was all new to me and very rewarding. The centerpiece of the book, on the regulation of speech and the press in British colonial and post-colonial India, is invaluable, especially so as he incorporates within that discussion the man who began his working life as a clerk in the East India Company (although he never made his way to the sub-continent), John Stuart Mill, whose On Liberty (1859) is the contemporary touchstone for individual liberties and freedom of speech.

When it comes to the US, he also provides valuable details and the timelines of state constitutions and, ultimately, the federal government’s Bill of Rights and its First Amendment, and the back-and-forth with liberal forces in France.

I caution though that I would have given more weight to social forces, although they are more difficult to document. Our constitutional rights are a civilizational achievement, and affirmation of rights in our laws at least to some degree follow rather than lead. As Michael Oakeshott put it:

Political activity may have given us Magna Carta and the Bill of Rights, but it did not give us the content of those documents, which came from a stratum of social thought far too deep to be influenced by the actions of politicians.

---

Dabhoiwala believes that speech is underregulated in the United States, and that there ought to be a shift away from what he calls the absolutist approach and towards something more balanced. Who could be against being balanced? I don’t think he succeeds.

The first problem is with how he characterizes the “absolutist” position. He sees it as arising from a series of contingencies in colonial America, and one can certainly find holes in the reasoning from that time. But he doesn’t address the rights-based approach to speech more generally. Consider the model from John Rawls’s A Theory of Justice, where Rawls imagines that in coming to agree on a social contract from an original position where we do not know what our personal advantages and disadvantages would be, we would not adopt rules that called for the application of utilitarianism to solve social issues, but would have extensive protection of liberal rights, that could not be over-ridden on the basis of greater aggregate social welfare. No one has to think Rawls got this right, but I do think this approach ought to at least be addressed in a contemporary book about rights. For better or worse it is one of the most influential books of moral and political philosophy in the past sixty years; Dabhoiwala will not even grant Rawls a footnote. What he calls “absolutism” actually does have a better pedigree than the Cato Letters, beginning with Kant. Missing this sense of the value of liberty where each person has different personal goals and values causes Dabhoiwala to mischaracterize the issue. Consider this passage:

…to conceive of free speech only in terms of more or less permissiveness about what one can say, or in terms of access to the public sphere, tends to a reductionist focus only on whether liberty of speech is ‘advancing’ or ‘retreating’. …

Above all, it ignores the central question that so engaged eighteenth- and nineteenth-century observers - what is freedom of speech for? [his emphasis]. Their answers were multiple: its purpose was to advance truth, to spread enlightenment, to discuss political questions, to pursue justice, and so on. Each of these purposes required a different set of speech rules and conditions - for example, the prohibition of slander, the avoidance of obscenity or the exclusion of lies. Without those conditions, the discussion wouldn’t be able to achieve its desired aim. The same holds true in our own time. Proper rules are constitutive of free expression: they give it meaning, they channel it towards its intended aim.

This is a fundamental reason why conflict over free speech is inevitable. Different aims require different constraints. If its purpose is to establish truth, it requires one set of conditions; if justice, a different set; if political legitimacy, yet another; if to create art or amusement, still others, and so on. It is inherently unstable, contradictory ideal, even before we get to our own differences. And if, more simplistically, you think of freedom of expression not as a means to an end, but simply as an end in itself, then the result is you elevate speech for its own sake to the supreme ideal: more important than truth, justice, equity, democracy or any other value. That is not only logically problematic, it also implies that any constraint is wrong.

But this is a false dilemma. What if freedom of speech is not for reaching any overarching societal goal, since in a free society people have their own goals to pursue. And neither is the speech valued for representing some supreme ideal in itself (I’m not even sure what that would mean). Rather, free speech is about an individual having the right to publish, as with this very blog, without fear of state officials knocking on his door. Speech doesn’t have to prove itself to be for anything. It can simply be a right to express oneself, and to read other’s expressions, without hindrance. That doesn’t make speech more important than other good things like truth and justice. Dabhoiwala is being far too narrow here. He laments that “American free-speech jurisprudence has gradually abandoned any conception of the common good”, but that presumes some definition of “the common good” ought to always rule legislation. It is a possible approach; Jeremy Bentham was adamant that it was the only sensible approach, that “rights” are nonsense on stilts. But Dabhoiwala is too quick to dismiss the rights of individuals as “dogma” and “libertarianism” and “absolutism.”

The second problem is with what I would call “hard cases,” and Dabhoiwala’s reluctance to look deeply into how his balanced approach would apply in practice. Consider Salman Rushdie. Dabhoiwala does mention him, in the context of the personification of an absolutist; Rushdie is quoted from 2015, “the moment you limit free speech, it is not free speech.” But that’s it - there is no discussion at all of why Rushdie might have taken this view. And so we have one of the most infamous cases of censorship and violence in recent times, with Rushdie suffering horribly for it, and the case is not pursued by Dabhoiwala, or even mentioned. In the United Kingdom, thousands of British Muslims took to the streets to call upon the government to ban Rushdie’s novel The Satanic Verses. Well, under Dabhoiwala’s “balanced approach”, what should the UK government have done? Was the harm to deeply offended Muslims greater than the benefits to literature of allowing the book to be made available? Catharine MacKinnon and Andrea Dworkin’s advocacy for much greater regulation of pornography based upon its being an action - a deeply harmful one - and not just speech, is also mentioned once but then dropped without follow-up. The prosecution of the Contemporary Arts Center in Cincinnati over its exhibition of photographs of Robert Mapplethorpe was perhaps the biggest art censorship case in recent times in the US. It is not mentioned at all. In his Afterword he does raise, but not pursue, the question “Is it antisemitic to criticize Zionism, or to accuse the state of Israel of apartheid?”, as one of the questions we ought to be debating. But while that’s a question I might discuss with friends over beers, it is not a debate I would enjoy seeing in the context of the federal government pursuing legislation over the matter, which leads to my next point…

Dabhoiwala’s third problem is political. I simply cannot be persuaded that departing from the so-called “absolutism” of the First Amendment and giving the knaves and fools that constitute our political class more power to regulate speech is a sound idea. It is fair to say that I live with what Judith Shklar (another author ignored by Dabhoiwala) called the “liberalism of fear”: we need firm constitutional protections as individuals out of a quite rational fear of what the state might get up to absent those protections. Donald Trump has publicly stated that he believes the laws of libel and defamation in the United States are far too weak. He sues newspapers and threatens to revoke the broadcast licenses of media that give him less-than loving coverage. He uses the power of the federal purse to coerce or defund institutions that in the past have worked at something of an arm’s length distance. How do we think things would turn out if speech protections in the United States were lessened, with the idea that then we could have more “balance”? Because more “balance” would require more legislation from the government - not an idealized government, but the one we actually have.

In his Afterword, Dabhoiwala claims “The more libertarian our free speech doctrines are, the less able they are to counter the toxic effects of blatant lying in public life.” We have a lot of blatant lying in public life, to be sure, and it is toxic. But I am unpersuaded by him: the free press I consume daily, including on an internet which he finds especially toxic, is the best remedy against the flood of lies we are given daily by public figures (starting with the one at the top). I just cannot see how encouraging our courts and legislators - not imaginary ones, but the ones we have on August 5, 2025 - to regulate speech in the interest of truth and the common good would end well. He suggests that the practical way forward would be the creation of sensible, balanced laws regarding speech, which would undoubtedly be challenged in courts, but where those courts could begin to adopt, over the long run, a more balanced approach that would benefit the social good. But this is a book that gives us hundreds of historical examples of governments abusing their powers to regulate speech in ways contrary to the public good.

I just don’t like the odds.

---

I don’t want to end on a negative note. Although I am unpersuaded by Dabhoiwala’s goals for how we might better handle the questions of free speech and a free press - even though he and I agree that our world is full of bad speech, lies, vulgarity, and insults to our morals, aesthetics, and intelligence - his historical research and ability to provide a coherent international story of the evolution of thinking and policies regarding free speech are tremendously valuable. So many contemporary works of nonfiction in public affairs are essentially magazine essays with padding and more padding: What is Free Speech? is not that at all - every page has something new. I learned a lot from this book, and I am glad to have read it, and I recommend it to anyone with an interest in this topic.