Philosophy
Hempel's Raven Paradox (The Paradox of Confirmation)
Quick fact
The paradox was introduced by philosopher Carl Hempel in the 1940s, and it shows that the statement 'All ravens are black' is logically the same as 'All non-black things are non-ravens'—which makes any non-black non-raven seem to confirm the raven hypothesis.
Why this is interesting
You're testing the claim 'All ravens are black,' so you go looking for black ravens. But could a red apple in your kitchen secretly be giving you evidence about ravens too?
Read the full explanation
Understanding Hempel's Raven Paradox (The Paradox of Confirmation)
Imagine you want to prove that every raven is black. The natural way is to examine ravens and check that each one is black. Every black raven you find supports your claim. But the same claim has a hidden twin: 'Everything that is not black is not a raven.' This is its contrapositive, and it means exactly the same thing. If you see a red apple, it is not black, and it is also not a raven. So it fits the twin statement perfectly. Since the twin is logically equivalent to 'All ravens are black,' any evidence for the twin should also be evidence for the original claim. But intuitively, a red apple tells you nothing about ravens. That clash—where logic seems to say 'yes' and intuition says 'no'—is Hempel's raven paradox.
A deeper explanation
At the heart of the paradox are two principles that each seem sensible on their own. The first is Nicod's criterion: a claim like 'All F are G' is confirmed by observing something that is both F and G (a black raven), and disconfirmed by observing something that is F but not G (a white raven). The second is Hempel's equivalence condition: if two hypotheses are logically equivalent, the same evidence must affect them equally. Since 'All ravens are black' and 'All non-black things are non-ravens' are equivalent—their contrapositives are mirror images—any non-black non-raven, such as a green leaf or a yellow banana, counts as confirming evidence. The paradox reveals that qualitative confirmation is too crude. Modern Bayesian confirmation theory resolves it by distinguishing between strength and probability of confirmation. An observation of a red apple does provide some tiny amount of confirmation for 'All ravens are black'—but in a world with vastly more non-black things than ravens, the effect is so small it feels like nothing. The paradox thus teaches us that the logic of evidence is subtle and that our intuitive judgments often mask probabilistic quantities.