← All topics
Lemma

Primes can't be fooled by products

If a prime divides a product a×b, it must divide a or divide b — it can't 'split itself' across the two.
Formally: Euclid's lemmanumber theoryTakes focusdepth 4 in the graph

Composite numbers CAN be fooled: 6 divides 4×9 = 36, yet 6 divides neither 4 nor 9 (its factors 2 and 3 hid in different places). A prime has no parts to distribute, so if it divides a product it must live wholly inside one factor. This indivisibility of 'blame' is the deep property that makes factorization unique.

A single key that opens a two-room house must open one of the rooms. A keyring (composite) can have its keys split between the rooms.

7 | 84 and 84 = 12 × 7 → 7 | 7 ✓. Try to break it: 7 | (14 × 15) = 210? Yes, and 7 | 14 ✓. Contrast: 6 | 36 = 4×9 but 6∤4 and 6∤9 — composites fail.

This is THE pivotal lemma: uniqueness of prime factorization stands or falls with it, and through that, essentially all of multiplicative number theory.

Level 1 The precise statement

If p is prime and p | (a*b), then p | a or p | b.

Level 2 The proof
If p does not divide a then gcd(p, a) = 1, so by Bezout 1 = p*x + a*y. Multiply by b: b = p*b*x + (a*b)*y. Since p | a*b, p divides the right-hand side, hence p | b.
Level 3 What it stands on (3 direct)
Level 4 The verified record

This page is generated from a machine-checked node. The kernel confirms its dependencies resolve, nothing is circular, and it grounds in axioms (foundation: peano). The content hash below makes tampering evident.

sha256:eb0639c7dff6a06400200d78e85d196350a4f4e2343bbfca48ddc248e4f67ebb

1 downstream result would collapse with it. See the blast radius on the graph →

🔬 Open in the workbench🎓 More topics