Is There a Secret Rulebook Hidden Inside Every Word?
A Strange Sort of ‘Should’
Imagine you are learning English. You point to a dog and say “cat.” Someone corrects you: “That’s a dog.” You accept it. But why should you accept it? Why is calling a dog a “dog” more than just a habit? The moment you think about it, meaning starts to feel bossy. It doesn’t just describe how we use words—it seems to demand something. The French word chien refers to the same animal, but if you are speaking English, you ought to say “dog.”
For decades, philosophers have debated whether this “ought” is built right into the very idea of meaning itself. One side, the normativists, says yes: meaning comes with a built-in rule about how a word should be used. The other side says no: the feeling of “should” comes from outside—from wanting to be understood, or from social pressure—not from meaning alone. At stake is what words and thoughts really are, and whether we can explain them with science alone.
The Simple Argument: If a Word is Correct, Must You Do It?
Suppose the word “green” means the color green. Then, using it for a green object is semantically correct; using it for a red one is incorrect. That is hardly controversial: meaningful words have correctness conditions. The big question is whether this correctness is enough to create a normative “ought.” Many philosophers, following work by Paul Boghossian (born 1957), think the step is immediate. If “green” means green, then you ought to apply “green” only to green things. Meaning itself, they argue, generates a prescription—a rule for what you should do.
This is sometimes called meaning engendered normativity (or ME normativity), because the norm is produced by the meaning, not the other way around. The simple argument tries to move from “it is correct to apply the word so-and-so” all the way to “you ought to apply it so-and-so.” To many, this feels obvious. When a friend says “I know how arcane means—it means old, right?” and you reply “no, that’s wrong,” you are not just stating a fact; you are criticising. You seem to be saying she violated a standard.
But opponents, the anti-normativists, challenge the move. They point out that the word “correct” can be used in two very different ways. You can say a calculator’s answer is “correct” without thinking the calculator had a duty to get it right. “Correct” can just mean matches a standard, with no ought attached. The fact that “green” is true only of green objects sorts uses into two piles—the true ones and the false ones—but does that sorting, by itself, tell you what to do? Imagine sorting rocks by colour into two bins. Labelling one bin “correct” doesn’t command you to put a red rock into the “green” bin. The anti-normativist says semantic correctness is like that—a sorting, not an order. If I use “green” for a red balloon, I say something false. But did I break a purely semantic rule? Not unless we already assume meaning is bossy. The simple argument, they say, begs the question.
The Error Problem: Who Says It’s a Mistake?
The debate gets sharper when you ask: what makes it possible to be wrong? This question was thrust into the spotlight by the American philosopher Saul Kripke (1940–2022), who developed a skeptical argument inspired by Ludwig Wittgenstein (1889–1951). Kripke asked us to imagine a meaning skeptic who challenges every attempt to explain what determines what a word means.
One popular idea is that meaning is determined by how speakers are disposed to use words—dispositionalism. If you are disposed to call green things “green” and red things “red,” that seems to fix the meaning. But the skeptic pounces: what about the time you stumble and call a green object “red”? If meaning were simply a matter of your dispositions, then every actual use would just be what you are disposed to do. Any apparent mistake would look like a change in what the word means for you. You wouldn’t be mistaken; you would simply mean something else.
This is the famous problem of error. Notice that it does not, by itself, prove meaning is normative. But Kripke’s skeptic uses a normativity constraint: the facts that determine meaning must be such that they show how the word ought to be applied. If a theory cannot explain how a use is incorrect—a violation of a standard—then it cannot be the right theory of meaning. The challenge is not just about explaining error; it is about whether any purely descriptive fact (like your brain’s wiring) can ever tell you what you should do.
Many philosophers have tried to answer the skeptic by building a bridge from dispositions to correctness without adding a mysterious “ought.” But the debate remains live because the constraint feels powerful: meaning seems to have a direction, a difference between getting it right and getting it wrong that is not just a matter of what you happen to do.
Rules That Make Meaning (Or Try To)
If meaning cannot be reduced to mere dispositions, maybe rules are the answer. This is meaning determining normativity (MD normativism): words have meaning because rules for their use are in force. The rules come first, and meaning is determined by them. Wittgenstein himself once wrote in his middle period: “without these rules the word has as yet no meaning; and if we change the rules, it now has another meaning (or none).”
The idea appeals because language feels artificial and arbitrary. The sounds “dog” and “chien” both work because we accept different sets of rules. But what does it mean for a rule to be in force? For a speaker to be governed by a rule, must she follow it consciously? Consider the rules of a game like chess. You know the knight moves in an L-shape. If you deliberately move it otherwise, you break the rule—but the rule itself is still in force for you, because you are still playing chess. The same might apply to words: you can intentionally say something false, but that doesn’t erase the semantic rule.
Trouble starts when we ask how we ever come to follow a rule without an infinite spiral. If understanding the rule requires another rule for interpreting it, we face a regress—the rule-following paradox. Quine once argued that if rules are merely “implicit” in our behavior, they lose their explanatory power and become empty labels. Distinguishing between truly rule-guided behaviour and mere regularity—like the difference between a person playing chess and a wind-up toy that moves in L-shapes—becomes very hard. Wittgenstein himself concluded that at some point, following a rule must be “blind,” a basic part of our shared human behaviour that cannot be explained by more rules.
Some philosophers, like Hannah Ginsborg, propose that a primitive feeling of appropriateness—a tiny “ought” so basic it doesn’t require prior concepts—is what turns noise into meaningful language. But critics worry that this feeling would need to happen before any meaning exists, which is hard to picture.
Why It Matters: Meaning in a Scientific World
So why does this argument continue? Because it cuts to a much larger question: can everything about us be explained by the natural world? Many normativists think the essential “ought” in meaning and thought blocks a purely reductive naturalism—the view that all facts are just physical or biological facts. If meaning is essentially normative, then a sentence like “‘green’ ought to be applied only to green things” is not just a disguised description of brain states or social habits. It is a different kind of fact, one that seems to resist translation into the language of physics or biology.
Yet the anti-normativist can reply that this “ought” still comes from us: we want to communicate, we want to understand each other, so we adopt the practical rule that using words in settled ways is useful. That is an instrumental “ought,” like “you ought to wear a coat if you don’t want to get cold.” Such an ought doesn’t show that meaning itself is normative; it just shows we have goals. The normativist then needs to argue that not all semantic oughts are instrumental—that some seem to hold regardless of what you want. For instance, even if you don’t care at all about being understood, does calling a dog a “cat” still somehow go against your word’s meaning? That is the crux.
You might notice this tension in your own life. When a friend misuses a word, you might correct her even if everyone understands perfectly what she meant. You feel she got the word wrong—not just that she failed to achieve a goal. But is that feeling itself proof that meaning is normative, or just proof that you have absorbed social habits strongly? Philosophy rarely gives a final answer, but it teaches you to notice the strangeness under the surface of everything you say.
Think about it
- If everyone in your town suddenly started calling dogs “cats,” and they all understood each other, would the word “cat” still be wrong, in a deep sense, for a dog? Why or why not?
- A parrot can be trained to say “hello” when someone enters the room. Does the parrot’s squawk mean hello in the same way yours does? What kind of difference might matter?
- Suppose a scientist could predict exactly how you will use a new word. Would that prediction capture everything important about what the word means, or would something be missing?
