If, Then, and What It Means

You’re standing in the kitchen with a friend. They say: “If you touch that wire, you’ll get a shock.” You don’t touch it. Was what they said true?

That’s a surprisingly hard question.

On one hand, nothing bad happened, so maybe they were wrong. On the other hand, the wire was live, and if you had touched it, you would have been shocked. So maybe they were right, even though nothing happened.

Philosophers have been arguing about this for over a hundred years. The strange thing is that smart people still disagree about something as simple as “if… then…” sentences. But that’s philosophy for you: the closer you look at something ordinary, the weirder it gets.

Three Ways to Think About “If”

Let’s say someone tells you: “If it’s raining, the ground is wet.”

You already know a few things about this kind of sentence. You know it’s false if it’s raining and the ground is dry. That much is obvious. The disagreement starts when we ask: what about the other possibilities? What if it’s not raining? What then?

Hook (the nickname philosophers use for the truth-functional theory) says: if it’s not raining, the sentence is automatically true. Period. No matter what. “If it’s raining, the ground is wet” is true when it’s sunny, when it’s snowing, when you’re standing on the moon — as long as it’s not both raining and dry.

That sounds weird at first. But here’s why some smart people believe it.

Imagine you know two things for sure: one, there’s a ball in a bag. Two, the ball is either red or blue. You don’t know which. Now someone says: “If the ball isn’t red, it’s blue.” Do you know that’s true?

Yes, you do. You’ve ruled out the only way it could be false (the ball being neither red nor blue). So “if not red, then blue” is true, even though you have no idea what color the ball actually is.

This works. It’s simple. It’s mathematically clean. And it’s part of the logical system that makes computers work.

But there’s a problem.

The Problem with Hook

Suppose your friend says: “If the Republicans win the election, they’ll double income tax.” You think the Republicans probably won’t win. According to Hook, you should think this conditional is probably true — because if the Republicans don’t win, it’s automatically true. The only way it could be false is if they win and don’t double taxes, which you think is unlikely.

But that’s not how you actually think. You might say: “I don’t believe that at all. If they won, they probably wouldn’t double taxes.” You can think the Republicans will lose and think the conditional is false.

Hook says you’re being irrational. But you don’t feel irrational. And that’s the problem.

Arrow (the non-truth-functional theory) tries to fix this. Arrow says: when the antecedent (the “if” part) is false, the conditional might be true or false, depending on the situation. If the wire was live and you would have been shocked, then “if you touched the wire, you’d get a shock” is true, even though you didn’t touch it. If the wire was dead, it’s false.

This makes sense of our intuitions. But Arrow has a different problem.

The Problem with Arrow

Remember the ball in the bag? You know the ball is either red or blue. Should you be certain that “if it’s not red, it’s blue”?

Arrow says: not necessarily. Maybe the ball is red — that’s fine. Maybe the ball is blue — that’s fine. But the possibility that it’s green has been ruled out. Still, Arrow says you can’t be certain of the conditional. There might be some weird situation where the ball is red, but the nearest possible world where it’s not red is a world where it’s green. And in that world, the conditional is false.

That seems wrong. You are certain. You know the ball is either red or blue. You know that if it’s not red, it’s blue.

The Suppositional View

Supp (the suppositional theory) takes a completely different approach. Supp says: stop trying to figure out whether “if A, then B” is true as a statement about the world. Instead, think about what you’re actually doing when you make a conditional judgment.

Here’s the idea, first suggested by the philosopher Frank Ramsey in 1929. When you’re wondering “if A, then B,” you do something simple: you suppose A is true, and then you see what you think about B, given that supposition. You temporarily add A to your stock of knowledge and ask: now how likely does B seem?

That’s all. You’re not predicting some special kind of fact about the world. You’re doing something with your mind: making a hypothetical judgment.

This is surprisingly powerful. It explains both of our puzzles.

Puzzle 1: You know the ball is either red or blue. You suppose it’s not red. Given that supposition, how likely is it that it’s blue? 100%. So you’re certain of the conditional. ✓

Puzzle 2: You think the Republicans probably won’t win. You suppose they do win. Given that supposition, you might think: “Now what? They’d probably cut taxes, actually.” So your conditional belief is low, even though you think the antecedent is unlikely. ✓

Supp gets both answers right.

But Then… What Are Conditionals?

Here’s where it gets genuinely strange. If Supp is right, then “if A, then B” doesn’t describe a fact about the world. It’s not a proposition that can be true or false in the normal way. It’s more like a rule for thinking: suppose A, then judge B.

The philosopher David Lewis proved something remarkable in 1976. He showed that there’s no way to assign truth conditions to conditionals that makes the probability of “if A, B” equal to the conditional probability of B given A, for all possible probability distributions. In other words: what you’re doing when you assess a conditional doesn’t match up with assessing the probability that some proposition is true.

This is a big deal. It means that if Supp is right, conditionals aren’t ordinary statements at all. They’re something else.

What We Lose

If conditionals aren’t ordinary statements, we lose something. We can’t easily combine them with “and,” “or,” and “not” in the normal way. What does “It’s not the case that if you touch the wire, you’ll get a shock” mean? It’s not just “if you touch the wire, you won’t get a shock” — those are different claims.

Philosophers are still working on this. Some have developed complicated theories using three truth values (true, false, and undefined) or using pairs of possible worlds instead of single worlds. Others have given up and gone back to Hook’s view, arguing that our intuitions are just wrong and we should learn to live with the weird consequences.

What This Tells Us

The debate about conditionals matters because it’s really a debate about what kind of creatures we are. When we say “if… then…,” are we describing the world or doing something with our minds? Are we tracking some special kind of fact about what would happen in other possible situations, or are we just running a mental simulation?

Different philosophers give different answers. Hook says it’s simple logic. Arrow says it’s about possible worlds. Supp says it’s about supposition and reasoning.

Nobody has won the argument. That’s part of what makes it interesting. The next time you say “if” — and you will, probably within the next hour — you’ll be doing something that the smartest people in the world still don’t fully understand.

Key Terms

Term	What it does in this debate
Antecedent	The “if” part of a conditional (“if A, then B” — A is the antecedent)
Consequent	The “then” part (“if A, then B” — B is the consequent)
Truth-functional	A way of understanding a logical operator where the truth of the whole is determined entirely by the truth of the parts
Conditional probability	The probability of B given that A is true; written P(B\|A)
Supposition	Temporarily assuming something is true, even if you don’t believe it actually is
Modus ponens	The rule: if A is true, and “if A then B” is true, then B is true
Ramsey Test	The idea that you assess “if A, B” by adding A to your beliefs and seeing what you think about B
Triviality results	Proofs showing that conditional probabilities can’t match probabilities of any proposition’s truth

Key People

Frank Ramsey (1903–1930) — A British philosopher who died at 26 but left behind ideas that still shape philosophy. He suggested that conditionals are assessed by adding the antecedent hypothetically to one’s beliefs.
David Lewis (1941–2001) — A brilliant and creative American philosopher. He proved that conditional probabilities can’t be probabilities of truth of any proposition, which shook up the whole debate.
H. P. Grice (1913–1988) — A British philosopher who argued that many puzzles about “if” could be explained by rules of conversation rather than logic itself.
Ernest Adams (born 1924) — An American philosopher who developed the suppositional theory into a full logical system showing which arguments with conditionals are valid.
Robert Stalnaker (born 1940) — An American philosopher who tried to give truth conditions for conditionals using “nearest possible worlds” while still respecting the suppositional view.

Things to Think About

Your friend says: “If you study, you’ll pass the test.” You study and pass. Was your friend right? What if you hadn’t studied — would the conditional have been true then too?
Someone says: “If you get 100 on every test, you’ll get an A in the class.” You get 100 on every test but still get a B because of a mysterious grading policy. Was the conditional false? Or was it true but the situation was weird?
Can you think of a situation where you’d accept “if A, B” but reject “if A and C, B”? For instance, “if you strike the match, it lights” but “if you strike the match while it’s underwater, it lights.” What does this tell us about what “if” means?
If conditionals don’t describe facts about the world, what are we doing when we say “if”? Are we just expressing our reasoning process? Giving advice? Making a promise? Could it be different things in different situations?

Where This Shows Up

Computer programming: Every “if… then…” in code uses the truth-functional conditional. Programmers rely on it working exactly as Hook describes.
Law: Legal documents are full of conditionals (“If the tenant fails to pay rent…”). Courts sometimes argue about what these conditionals mean.
Everyday arguments: When someone says “if you really cared, you would have called,” they’re making a conditional claim that might be true even if you do care and didn’t call. Understanding how conditionals work helps you spot when someone’s reasoning is shaky.
Science: Scientists use conditionals to state predictions and hypotheses. The logic of “if… then…” is built into how experiments are designed and interpreted.
Gambling and probability: Conditional bets (“I’ll pay you if it rains tomorrow”) work exactly like conditional probabilities. Understanding this helps you see when a bet is fair.