We need to talk about why our obsession with hate speech definitions is completely broken. Every year, international bodies hold forums, launch awareness days, and write endless guidelines aimed at stopping harmful language. It looks like progress. It feels like someone is doing something. But it’s mostly an illusion.
The real issue with managing hate speech has very little to do with the actual words people use. It’s about power, context, and the dangerous fantasy that we can build a linguistic filter to fix deep social friction. When we try to police the language itself, we mistake the symptom for the disease. Worse, we hand the keys of censorship over to institutions that historically use that power to protect themselves, not vulnerable groups. In other updates, read about: Why Keir Starmer Failed To Stop Britain's Prime Minister Carousel.
Let's look at what’s actually happening beneath the surface of the free speech debate and why the current approach is failing.
The Definition Trap
Here is a striking fact. The United Nations actively promotes an International Day for Countering Hate Speech, yet international law contains no official, binding definition of what hate speech actually is. The New York Times has also covered this critical subject in great detail.
Instead, the UN relies on a broad working definition: any communication that attacks or uses pejorative or discriminatory language toward a person or group based on identity factors like religion, ethnicity, or nationality.
If you look closely at those words, the entire strategy collapses. Phrases like "pejorative" or "attacks" aren't objective linguistic features. They depend entirely on who is listening, who is speaking, and the political climate of the moment. One person's urgent political critique is another person's incitement.
This isn't just an academic puzzle. It has real consequences. Because the language is inherently vague, the term becomes a weapon. Governments worldwide routinely rewrite domestic laws to ban "hate speech" or "spreading misinformation," only to immediately turn those laws against journalists, political dissidents, and activists. When you create a vague rule to police language, the entity with the most power gets to decide what the words mean.
What Grice and Lewis Carroll Taught Us About Power
Linguists and philosophers figured this out decades ago. Paul Grice, a philosopher of language, showed that human communication relies heavily on what he called conversational implicature. Essentially, the vast majority of meaning in any conversation happens outside the literal definitions of the words. We read between the lines using shared context, tone, timing, and unspoken social agreements.
If you look only at a transcript of words, you miss the actual message. Take a simple phrase like "Nice day, isn't it?" Said during a sunny afternoon, it's friendly. Said during a torrential downpour, it's sarcastic. Said by a landlord right before delivering an eviction notice, it feels threatening. The words don't change. The meaning changes completely.
This brings us to the famous dynamic in Through the Looking-Glass by Lewis Carroll. Humpty Dumpty famously tells Alice, "When I use a word, it means just what I choose it to mean — neither more nor less." When Alice questions whether you can actually make words mean so many different things, Humpty delivers the real truth: "The question is, which is to be master — that’s all."
Humpty Dumpty wasn't making a point about linguistics. He was making a point about raw power. The person who controls the interpretation of an encounter holds the authority. When modern institutions claim they can cleanly separate acceptable speech from hate speech based on text alone, they are acting like Humpty Dumpty. They are claiming the authority to be the master of meaning.
Why Technical Frameworks Beat Vague Rules
Because literal definitions fail us, standard text moderation doesn't work. True protection requires evaluating the entire ecosystem surrounding an utterance.
The most realistic tool we currently have for this is the Rabat Plan of Action, a framework adopted by experts to help courts determine whether speech genuinely incites violence or discrimination under international human rights law. Instead of looking at a list of banned words, the Rabat framework uses a strict six-part threshold test:
- Context: Is there an ongoing conflict or a history of violence in the region?
- Speaker: Does the person hold political influence or a high public position?
- Intent: Did the speaker deliberately mean to incite hostility, or were they speaking carelessly?
- Content: What was the direct message, and how provocative was it?
- Extent: How far did the speech travel, and who was the audience?
- Likelihood: Is the threat of real-world harm imminent and direct?
Notice how little this test cares about simple vocabulary. A phrase that is completely harmless when muttered by an anonymous person on a forum can become incredibly dangerous when broadcast by a prominent politician to an armed crowd during an election cycle.
The focus shifts from the speech itself to the real-world mechanics of harm. It acknowledges that speech is a tool used within a specific social structure. If you ignore the structure, you miss the danger entirely.
Moving Past the Speech Panic
If we want to address the toxic polarization dividing communities, we have to stop expecting tech platforms or government committees to solve cultural crises through censorship. The current path leads to a dead end where everyone accuses their opponents of hate speech while claiming their own rhetoric is pure.
Instead of hunting for a perfect legal definition that doesn't exist, focus on the structural realities of communication:
- Demand institutional transparency. Whenever a platform or government agency proposes a new content moderation policy, look past the noble phrasing. Ask exactly who sits on the review board, what specific appeals process exists, and how the rules apply to those currently in power.
- Apply threshold thinking. When evaluating controversial public statements, stop asking "Is this offensive?" Offensive speech is a permanent fixture of an open society. Start asking the Rabat questions: What is the speaker’s actual reach? What is the imminent likelihood of physical or economic harm?
- Invest in platform architecture over censorship. The ways modern algorithms amplify outrage matter far more than individual bad actors. Shifting our focus to design—like slowing down the speed of re-sharing or breaking up echo chambers—addresses the structural pipeline of toxicity without forcing tech executives to act as the world's moral police.
Words carry weight, but they don't exist in a vacuum. If we keep pretending the problem is just the speech, we will continue giving up our freedoms to institutions that have no intention of protecting us.
