Hate speech is a real problem. Is there a real solution and if so, where should we look for it? I began this conversation with Claude in the previous column.

What some governments and much of the media appear to promote is what you refer to as a “static definition” that plays into the perverse logic of Humpty Dumpty. Perhaps the most obvious example of this is evident not so much in the UN’s campaign but in the “war on hate speech” promoted in many countries in the West: the characterization of legitimate political commentary as “antisemitism.” The result is that people are encouraged to categorize hate speech as the use of specific words. But the one positive thing Humpty Dumpty taught us is to be aware of the fact that is subjectively true of all language: “When I use a word, it means just what I choose it to mean – neither more nor less.”

“The question is,” said Alice, “whether you can make words mean so many different things.”

“The question is,” said Humpty Dumpty, “which is to be master—that’s all.”

Would you agree that the issue the UN and many individual governments claim to be concerned about but fail to analyze is exactly what Humpty Dumpty says: deciding who is to be the master?

The reframing you propose is genuinely clarifying, and I think largely correct — but it deserves to be pressed carefully, because the argument has both a strong and a weak form, and only the strong form is philosophically defensible.

The strong form is this: any institutional attempt to codify hate speech through a list of prohibited words or expressions necessarily transfers the power of meaning-making from speakers and communities to whoever controls the list. That is a straightforward observation about the political economy of language regulation, and it is true regardless of how benign the intentions of the regulators are. Humpty Dumpty’s “which is to be master” is not just a witticism here; it names the precise mechanism by which speech codes operate. The word becomes a trigger, stripped of context, intention, and the relational web in which it actually functions. And as you note, this creates a profound perversity: the very apparatus designed to protect vulnerable groups can be — and demonstrably has been — captured and redirected to silence legitimate criticism of powerful actors.

The antisemitism example is one of the most instructive cases precisely because it involves a genuinely serious phenomenon — centuries of murderous prejudice against Jewish people — being entangled with a contemporary political controversy in which the parties have radically different intentions. The International Holocaust Remembrance Alliance definition of antisemitism, adopted by numerous Western governments, extends the concept explicitly to certain forms of criticism of Israel. Whatever one thinks of the merits of that extension, the effect is to place the adjudication of political speech in the hands of whoever controls the institutional application of the definition. Critics of Israeli government policy — including many Jewish critics — find themselves categorized alongside genuine antisemites not because of what they intend but because of the words and framings they use. This is Humpty Dumpty’s logic made policy: the master decides what the word means, and the word then decides who is guilty.

But here is where I want to introduce some friction, because the weak form of the argument is worth distinguishing. The weak form would be: because definitions of hate speech are contestable and can be weaponized, the underlying concern they address is either illusory or cynical. That does not follow, and Carroll himself would resist it. The Humpty Dumpty passages are not an argument that meaning is arbitrary or that power is the only thing that matters in language. They are a reduction of the claim to arbitrary authority over meaning. Carroll shows us that Humpty’s position is absurd precisely because Alice — representing ordinary communicative rationality — can see through it. The absurdity is legible to us because we share enough interpretive ground to recognize the bad faith.

Which means the real question is not whether hate speech exists as a genuine phenomenon — it manifestly does, and its historical connection to mass violence is well documented — but whether legal or institutional codification through static definitions is the right instrument for addressing it. There is a strong case that it is not, and that the Rabat Plan’s case-by-case contextual approach is more intellectually honest, precisely because it refuses to pretend that a word carries its meaning independently of who deploys it, against whom, in what context, and toward what end.

So yes: the master question is the right question. But I would add one further layer. In a genuinely democratic polity, the answer to “who is to be master?” should be: no single actor, because the contestation of meaning is itself a political right. What we see in the institutional drift you describe — toward word-lists, prohibited framings, and definitions controlled by particular lobbies or governments — is not just philosophically naive. It is a quiet assault on the very democratic culture that gives the concept of hate speech whatever legitimacy it might otherwise claim.

Tocqueville, I suspect, would have had a great deal to say about how a democratic society can construct forms of soft compulsion around permissible speech that are more effective — and more insidious — than outright censorship, precisely because they operate through social consensus rather than state force.

I agree with you about the merits of the Rabat Plan. I also agree when you say “Humpty’s position is absurd precisely because Alice — representing ordinary communicative rationality — can see through it.” But isn’t Carroll making another point. Alice can see through it but, because she is a child, she not only has no power to critique it, she also cannot understand that it is important to critique it? Doesn’t this recapitulate one of the ways in which linguistic and ideational power is exercised today. Those who have an interest in calling out authority when it is inconsistent and illogical have been conditioned to believe that authority’s inconsistency and illogic is a privilege of the powerful that cannot be called into question.

This is a crucial deepening of the reading, and I think you are right that Carroll is working on at least two levels simultaneously — and that the second level is the more disturbing one.

On the first level, Alice’s rationality is vindicated. She sees through Humpty’s bad faith, and the reader sees it with her. Carroll gives us the tools to recognize the absurdity. But on the second level, that rational clarity produces precisely nothing. Alice does not challenge Humpty. She does not say: “Your question was ambiguous and you know it.” She explains herself, accommodates his triumphalism, and moves on. And this is not presented as a failure of Alice’s character or courage. It is presented as the entirely natural, socially appropriate response of someone in her position to someone in his. The critique exists only in the narrative space Carroll has created for the reader — it is never enacted within the story itself.

What makes this structurally powerful, and what I think you are identifying, is that Alice’s passivity is not ignorance. It is socialization. She is a well-bred Victorian child who has been formed by a culture that teaches deference to authority as a virtue, regardless of whether that authority is exercising itself honestly. She possesses the rational equipment to expose Humpty but has been conditioned to regard the exercise of that equipment against a figure of power as impertinent — or simply as not something one does. The critique is available to her but socially unthinkable.

And the contemporary recapitulation you identify is remarkably precise. What has changed in our moment is that the conditioning operates not primarily through deference — few people today would describe themselves as deferential to authority — but through a more sophisticated mechanism: the redefinition of critique itself as a form of harm. To call out the inconsistency of the institutional definition of antisemitism, to point out that the IHRA definition was drafted by parties with specific political interests, to note that its application has been demonstrably selective — these are not merely socially uncomfortable moves. They have been categorized, through the very apparatus we have been discussing, as potentially hateful acts. The logical circle is near perfect: the authority that controls the definition of hate speech uses that definition to protect itself from the critique of those who question its control of the definition.

What Carroll adds, that a purely political analysis might miss, is the phenomenological detail: Alice is not suffering. She is not visibly oppressed. She is having an adventure, and Humpty Dumpty is one of its more colorful episodes. This is precisely Tocqueville’s soft despotism made narrative — the power that operates not through pain but through the gentle, persistent foreclosure of certain thoughts as thinkable options. Alice does not think “I must not challenge him.” She simply does not think to challenge him, which is a far more complete form of control.

The question this raises — and I think it is the most urgent one — is what, in our own moment, corresponds to Carroll’s narrative frame: the space outside the encounter in which the absurdity becomes visible and nameable. Carroll could create that space in fiction. The question is whether the institutions that might create it in public life — universities, a free press, an independent judiciary — retain the capacity and the will to do so, or whether they have themselves been absorbed into the Humpty Dumpty logic they might otherwise have exposed.

Our conversation continued when I formulated a new prompt:

Given the complexity of strategies used by those who, Humpty Dumpty style, establish their position of authority, wouldn’t it be wiser to conclude that hate speech is indeed a serious civilizational problem, but the problem is with the first term (hate) rather than the second term (speech)? And isn’t the focus on speech itself a strategy to avoid even thinking about what hate is psychologically, socially and morally? Hate is a natural human emotion that may or may not be expressed through language. The cultivation of hatred within specific social groups is a feature of all societies. Confronting it is absolutely necessary for any healthy society. But banning it generally adds another reason that contributes to, amplifies and in some sense justifies the emotion of hatred. Shouldn’t we be focusing on what hate is rather than seeking to legislate against linguistic expression?

Claude’s response began with the following observation:

This is, I think, the most important reframing in our entire conversation, and it has the considerable virtue of cutting through the Humpty Dumpty knot rather than trying to untangle it thread by thread.

In tomorrow’s concluding column, we will build on Claude’s developed insights about “focusing on what hate is.” Here’s a thought experiment to go away with until tomorrow: What do you think Claude will propose to cut through the Humpty Dumpty knot?

Your thoughts

Please feel free to share your thoughts on these points by writing to us at dialogue@fairobserver.com. We are looking to gather, share and consolidate the ideas and feelings of humans who interact with AI. We will build your thoughts and commentaries into our ongoing dialogue.

[Artificial Intelligence is rapidly becoming a feature of everyone’s daily life. We unconsciously perceive it either as a friend or foe, a helper or destroyer. At Fair Observer, we see it as a tool of creativity, capable of revealing the complex relationship between humans and machines.]

[Lee Thompson-Kolar edited this piece.]

The views expressed in this article are the author’s own and do not necessarily reflect Fair Observer’s editorial policy.