It’s been more than two weeks since I started this Substack, and I have yet to get to a second post. Right now I’m working on a lengthy piece on the concept of “misinformation.” I tend to write very long pieces, at a minimum a book chapter in length, because I feel the need to think through something comprehensively before I say anything at all.

My thoughts on the subject of misinformation fit into a larger book project defending freedom of discussion. But that defense can only be made by exploring its limits. This is something that Mill himself suggests in his famous defense of freedom of discussion, On Liberty. Paradoxically, to exercise freedom of opinion, we must have freedom from the “tyranny of opinion.” In fact, Mill tells us, government censorship should not be our primary concern. Social coercion is far more effective at producing conformity and stifling genuine discussion. Where government censorship actually preserves hidden pockets of dissident opinion but driving it underground, social coercion produces a more general climate of fear of ostracism and loss of reputation (not to mention loss of employment, harassment, and other elements we today associate with “cancel culture”)

But it can be very difficult to separate social coercion from the normative implications of expressing opinions at all. To express a view on what is ethical and just inevitably implies something about the ethics of those who hold the opposing view. Meanwhile social coercion is an inevitable part of daily life: I can’t just say anything around others and expect them to remain my friends or employers. If discussion is really free, then it seems people will be free to express the harsh condemnations that are building blocks of the “tyranny of opinion.”

That contradiction can only be solved only by making distinctions between different speech acts and discursive contexts. Where social coercion is illegitimate, it is because contexts have been collapsed (something abetted by social media), and speech acts misconstrued because interlocutors use wildly permissive interpretive frames that ignore basic linguistic pragmatics (including the intent of interlocutors). Theoretical contexts—what I call “contexts of inquiry”—in which freedom of discussion and truth-seeking ought to reign, are conflated with what I call “contexts of offense,” everyday social contexts in which we engage in a great deal of self-censorship to avoid upsetting others. Illiberal interpretive frames allow us to infer offense, harm, and even “misinformation” in virtually any case we like.

Some have claimed that today’s free speech debates are not the result of a falling out with the general principle of freedom of speech. They argue that the domain of social coercion and offense have always set boundaries for discourse, and that the debate today is the result of a cultural shift in those boundaries. And those boundaries are necessary for the establishment of a genuinely free discursive domain. If we’re concerned about the effects of “cancel culture” on freedom of discussion, why not be concerned about the “harm” caused by certain forms of speech, and their effect on the freedom of opinion of those who are underprivileged in one way or another? If the views of trans activists constitute a “tyranny of opinion” in relation to J.K. Rowling and those who agree with her, why isn’t the converse also true?

I’ll argue that it is not in fact the boundaries of discourse that are at issue here, but its more fundamental rules of engagement. The dispute is not really about content, but about form and process. Where the approach to discourse becomes illiberal, it is because discursive contexts are collapsed, linguistic pragmatics and distinctions between speech acts ignored, and rational and hermeneutic capacities impaired or abandoned (often with incredible maliciuosness). To argue for freedom of discussion is not to argue that social coercion be eliminated from all contexts, but that it not be used promiscuously. To argue for freedom of discussion is to argue that at least in more theoretical contexts, the possibility of intelligent discussion be preserved. It is not “free speech” that is the fundamental issue, but the preservation of society’s capacity for intelligent speech, something that requires we not treat any portion of discourse as so taboo that it cannot even be dignified with counterargument or engagement. In fact, to establish a taboo in contexts of inquiry is just to institute another model of engagement, one that far more vulnerable to defeat in the long run than open discussion.