Do Words Just Describe, or Do They Make the World?
The Professor Who Invented a Card Game
In 1951, John Langshaw Austin (1911–1960), a philosophy professor at Oxford, invented a card game called CASE. He played it with friends and students, always paying close attention to how they described their moves. This was not just a hobby — it was a clue to his whole way of doing philosophy. Austin believed that if you want to understand big ideas like truth, knowledge, and freedom, you have to look closely at the ordinary words we use every day. He thought most philosophers rushed past those words and got into trouble. So he got out his magnifying glass and asked: what are we really doing when we speak? He turned out to be one of the most original thinkers of the twentieth century, even though he published very little before he died.
Before the war, Austin studied classics, especially Aristotle and Kant. During World War II, he helped plan the D‑Day invasion — his intelligence work was known for life‑saving accuracy. After the war, he returned to Oxford and became a professor. Instead of building a grand theory, he held Saturday‑morning discussion sessions where he and his colleagues would comb through single paragraphs of a book for hours. His own published work was tiny, but his ideas traveled through his students and his unforgettable lectures. At the heart of it all was a simple, powerful thought: words are tools, and if we don’t know how they work, we get stuck in fake problems.
Words That Do Things
Austin noticed something that often gets overlooked: not all sentences are used to describe things. When you say “I promise to be there,” you aren’t reporting on a promise that already exists — you are making the promise by saying those words. He called these performative utterances. In contrast, constative utterances state facts and can be true or false, like “The cat is on the mat.” Performatives do something in the world: “I name this ship the Queen Elizabeth” (as you smash the bottle), “I bet you sixpence it will rain,” “I take you to be my lawful wedded spouse.” None of these is simply true or false. Instead, they are assessed as felicitous or infelicitous — happy or unhappy. An attempt to marry can misfire if the person conducting the ceremony isn’t authorized, or be an abuse if the bride or groom is insincere.
Austin then showed that the neat boundary between performatives and constatives falls apart. Even a constative like “France is hexagonal” can misfire if it’s uttered with no clear purpose — you can’t tell whether it’s true or false. Performatives can also be factually unhappy: if you say “I warn you the bull is about to charge” but the bull is asleep, your warning is mistaken. So Austin replaced the simple split with a more careful model. Every time you speak you perform a locutionary act — you produce sounds and words with a certain meaning and reference. At the same time you perform an illocutionary act — you do something with those words, like stating, warning, promising, or betting. Finally, your act may have further effects on the audience: persuading, frightening, annoying. That last layer is the perlocutionary act. For Austin, the illocutionary act was the most important, because it carries the force that makes our words count as moves in a social world.
Is It True? It Depends on What You’re Doing
If words do things, then truth itself isn’t a simple matter of a sentence matching reality. Austin agreed that truth involves correspondence with facts, but he rethought what facts are and how correspondence works. He thought of facts as particular, concrete situations or events — not as ghostly propositions. When you make a statement, two kinds of convention link your words to the world. Descriptive conventions connect a sentence (like “The cat is on the mat”) to a type of situation — any situation in which a cat is on a mat. Demonstrative conventions tie your actual utterance on a specific occasion to a particular historic situation — this cat, this mat, right now. A statement is true, Austin said, if the situation singled out by the demonstrative conventions is of the same type as the situation associated with the sentence by the descriptive conventions.
But here’s the twist: deciding whether a particular situation is sufficiently like the standard cases to count as the same type always involves human judgment. The world doesn’t come pre‑labelled. We have to judge similarity, and that judgment depends on our intents and purposes. For a general planning a war, “France is hexagonal” might be close enough. For a geographer, it’s rough and misleading. Said out of the blue with no clear purpose, it’s neither true nor false — it’s just rough. In the same way, “Lord Raglan won the battle of Alma” might be acceptable in a schoolbook, but in a detailed historical study it would be an exaggeration. Truth, Austin concluded, is not a one‑size‑fits‑all property of sentences. It’s a dimension of being the right thing to say, to this audience, for these purposes.
Can You Really Know It’s a Goldfinch?
Austin’s thinking about knowledge begins with an everyday example. You’re in the garden and you spot a bird. “That’s a goldfinch,” you say. Someone challenges you: “How do you know?” You might answer that you can tell by the red patch on its head. The challenger could push back: “Could it be a bird with a similar patch? Or a stuffed specimen?” Austin thought such challenges must have some specific reason behind them — a real doubt, not a far‑fetched fantasy. You don’t need to rule out that it’s a stuffed goldfinch, because you already know enough about the situation to know it’s not. Enough is enough — it doesn’t mean absolutely everything.
For Austin, knowledge is not a beefed‑up version of belief. If you know, you have a guarantee: you cannot say “I know it’s a goldfinch, but I might be wrong.” If you might be wrong, you don’t know. At the same time, the human abilities that produce knowledge — our senses and judgment — are inherently fallible. They can go wrong even in good conditions, just as a reliable car can sometimes break down. But that does not mean we never know. Knowledge is the successful exercise of real, though imperfect, capacities in the right circumstances.
Austin also rejected the idea that we need an unshakeable foundation of private mental items (called sense-data) to support all our knowledge. A classic argument says: a straight stick in water looks bent, so you don’t directly see the stick — you see a bent sense-datum. From there it’s a short step to claiming we never directly see physical objects. Austin dismantled this argument from illusion. He pointed out that the bent look is a public feature of the stick-in-water situation, not a private sensation trapped inside your head. A camera would record the same bend. And you don’t automatically judge the stick is bent; experienced observers resist the illusion. Perception puts you directly in touch with the world, messy though it may be.
Why “Oops” Tells Us a Lot About Freedom
Austin thought you can’t understand free action by starting with the giant question “Are we free?” Instead, he looked at the small words we use when things go wrong: excuses. “I didn’t mean to.” “It was an accident.” “I was clumsy.” “I didn’t realize what was happening.” Each type of excuse targets a different piece of the machinery of action — your awareness of the situation, your planning, your control over your body, your knowledge of your own strength. By studying excuses, we learn what makes an action count as fully responsible.
Austin made a surprising claim: in normal, everyday actions, it makes no sense to add modifiers like “voluntarily” or “on purpose.” He called this rule no modification without aberration. When you sit down in a chair in the ordinary way, it’s not true that you sat voluntarily, and it’s not true that you sat involuntarily — the words just don’t belong. You simply sat. The language of voluntariness only shows up when something unusual is going on. This turned a common philosophical assumption upside down. Many thinkers had assumed that every action must be either free or unfree, and that words like “voluntary” name the special ingredient that makes an action free. Austin suggested instead that freedom is the unmarked, standard case, and excuses chip away at responsibility in specific respects.
He also suspected that our ordinary talk of what we “could have done” conflicts with determinism — the idea that everything is caused by prior events, leaving no alternative possibilities. If a golfer misses a short putt and says “I could have holed it,” she means she had the ability, even though she failed on that occasion. Austin thought this ordinary notion of ability was incompatible with the view that every failure was determined in advance. He didn’t resolve the ancient freedom‑versus‑determinism debate, but his close attention to excuses and abilities gave both sides new questions to answer.
The Talk We Talk Every Day
Austin’s ideas still matter because you use them every day, whether you know it or not. You promise, you apologize, you assure, you argue, you make excuses. Seeing that these are actions, not just strings of words, makes you more careful about what you say — and more aware of what others are doing when they speak. When a friend says “I promise to help,” they aren’t describing their mental state; they are giving you a guarantee. When someone says “I didn’t mean to,” they are singling out a particular part of the machinery of action that failed. Recognizing these moves helps you navigate friendships, classrooms, and family arguments with clearer eyes.
More than that, Austin’s method — pay close attention to ordinary language, treat words as tools, and refuse to be bullied by false either‑or choices — is a way of thinking you can carry into any subject. He reminds us that “fact is richer than diction”: the world is always more complicated than our labels. So the next time you hear a grand philosophical claim, ask yourself what people would actually say in concrete situations. You might find that the big, scary problem starts to dissolve. Austin’s legacy is the idea that philosophy begins with noticing what your words are really doing.
Think about it
- If a friend says “I promise to help you,” and then doesn’t help, did they lie? Or did they just break a promise? What’s the difference?
- Can you think of a situation where saying “I know” is more like giving your word to someone than simply describing what’s in your head?
- Your teacher says “New York is south of Montreal” in a geography class. That’s roughly true, but if you check a map, a small part of New York City is actually north of Montreal. Is the teacher’s statement true or false? Does it depend on what you’re trying to learn that day?
