MathLabs
Step 2 of 7: Set up the plan: start with any finite list of primes
In plain words

Suppose someone claims they have written down the complete, final list of all prime numbers in the universe on a sheet of paper. To prove them wrong, we do not need to find infinitely many primes at once — we just need a reliable recipe that takes their sheet of paper and produces one single prime number that is missing from it.

L={p1,p2,…,pn}L = \{p_1, p_2, \dots, p_n\}
Detailed analysis

We let L={p1,p2,…,pn}L = \{p_1, p_2, \dots, p_n\} be any finite list of prime numbers. In modern expositions, this is usually framed as a proof by contradiction: assume for contradiction that there are only finitely many primes in total, so that LL is the complete list of every prime that exists.

In Euclid's original wording in Book IX, Proposition 20 ("Let A, B, and C be the assigned prime numbers; I say that there are more prime numbers than A, B, and C"), the argument is stated constructively for any assigned finite list LL, and uses a small contradiction only inside a later sub-step. Either way, the plan is identical: starting from the finite list L={p1,p2,…,pn}L = \{p_1, p_2, \dots, p_n\}, we will construct a new number NN and use it to uncover a prime q∉Lq \notin L.

Terms in this step
Proof by contradiction
A method of proof that establishes a statement is true by temporarily assuming the opposite is true and showing that this assumption leads to a logical impossibility.
Knowledge used in this step