Table of contents

Select a white background

Proving AB

There are two common approaches to proving that two statements are equivalent to one another. One approach is to treat ≡ similarly to =. That is commonly done for propositional logic, for example.

Theorem. ¬(pq) ≡ p ∧ ¬ q.

Proof.

¬(pq)  ≡  ¬(¬pq) (by the definition of pq)
 ≡  ¬(¬p) ∧ ¬ q (by DeMorgan's law)
 ≡  p ∧ ¬q (by the double-negation law)

But often, it is convenient to break the proof down into two major steps by using the propositional equivalence (pq) ≡ ((pq) ∧ (qp)). The following example uses the definitions from the preceding page. We also need one more definition.

Definition. For integers n and m, where m > 0, n mod m is the remainder when you divide n by m. More precisely, dividing n by m yields a quotient q and a remainder r where

n = qm + r
0 ≤ r < m
and n mod m is the remainder, r.

Theorem. The following two statements are equivalent.

(a) xy (mod m).
(b) x mod m = y mod m.

Proof. We break the proof into two parts.

((a) → (b)). Suppose that

(1) xy (mod m).
By definition, that means
(2) m | (xy)
which in turn means
(3) ∃k(xy = km)
So select an integer c such that
(4) xy = cm
(5) x = cm + y (by algebra from (4)
Now take both sides of equation (5) mod m.
(6) x mod m = (cm + y) mod m
But
(7) (cm + y) mod m = y mod m
since adding multiples of m to the dividend cannot change the remainder. Putting (6) and (7) together yields
(8) x mod m = y mod m
which is just what we are trying to prove.

((b) → (a)). Suppose that

(1) x mod m = y mod m.
By division and the definition of the mod operator, there must exist integers q1, r1, q2 and r2 so that
(2) x = q1m + r1
(3) 0 ≤ r1 < m
(4) y = q2m + r2
(5) 0 ≤ r2 < m
(6) r1 = r2 (restatement of fact (1))
By facts (2) and (4),
(7) xy = q1m + r1 − (q2m + r2)
(8) xy = (q1q2)m + (r1r2)
(9) xy = (q1q2)m (by facts (6) and (8))
But fact (9) tells us that
(10) ∃d(xy = dm)
since d = (q1q2) clearly works. But that is equivalent to
(11) xy (mod m)
by the definition of what xy (mod m) means.