JOVANA
Explore Library Glossary Getting Started Three Levels Fields How it works Mission
Join the mission
All guides

Your First Two-Column Proof

Everything in this rung was leading here. We will take the if-then reasoning you just met and lay it out in two neat columns — statement and reason — to prove a real theorem that you can see is true but had never been forced to justify.

What a two-column proof actually is

By now you have all the parts. You know the undefined terms point, line, and plane; you can tell a definition from a theorem; you have postulates and common notions to lean on; and from the last guide you know a conditional statement has a hypothesis and a conclusion. A two-column proof is just a way of writing that reasoning down so nobody — including future you — can poke a hole in it. The left column lists statements; the right column gives the reason each statement is allowed.

The discipline is brutal and that is the point: every line in the left column must be justified by exactly one thing in the right — a postulate, a definition, a previously proved theorem, a common notion, or something you were given. No line is allowed to rest on "it looks true in the picture." A diagram can suggest where to go, but it can never be a reason, because a sloppy drawing or a special case can fool the eye.

Setting up: from a claim to a plan

Let us prove a real theorem — the one about vertical angles. When two lines cross, they make four angles; the two that sit opposite each other (not sharing a side) are vertical angles. The claim is: vertical angles are equal in measure. You have surely seen this and believed it. Believing is not proving. We need to start from things already granted and force the conclusion out.

First, restate the claim as a conditional and pin down the picture. Lines AB and CD cross at point O. Going around O, label the angles so that angle AOC and angle BOD are the opposite (vertical) pair we care about, with angle AOD sitting between them on one side. The key actor is the angle they each share a side with. Notice that angle AOC and angle AOD together form a straight line — they are a linear pair — and so are angle AOD and angle BOD.

The bridge: a linear pair sums to 180

Here is the one fact we borrow. A linear pair of angles — two angles that share a side and whose other sides point in exactly opposite directions along a straight line — are supplementary: their measures add to 180 degrees. This is not magic; it follows from the angle addition postulate (the two angles' measures add to the measure of the whole angle they form) plus the fact that a straight angle measures 180. We will cite it as a known result so the spotlight stays on the vertical-angle argument.

Watch what this gives us. The pair AOC and AOD is a linear pair, so m(angle AOC) + m(angle AOD) = 180. The pair AOD and BOD is also a linear pair, so m(angle AOD) + m(angle BOD) = 180. Two quantities that each equal 180 must equal each other — that is a common notion, the one Euclid stated as "things equal to the same thing are equal to each other." The shared piece, angle AOD, is about to cancel out.

The proof, in two columns

Now we just transcribe the plan. Read it slowly: each statement on the left earns its place only because of the reason on its right. Nothing here appeals to the picture; every line is a postulate, a definition, a common notion, or the linear-pair result we agreed to borrow.

Given: lines AB and CD meet at O.     Prove: m(angle AOC) = m(angle BOD)

  STATEMENT                                 REASON
  1. AOC and AOD are a linear pair          1. Definition of a linear pair
  2. m(angle AOC) + m(angle AOD) = 180      2. Linear pair is supplementary
  3. AOD and BOD are a linear pair          3. Definition of a linear pair
  4. m(angle AOD) + m(angle BOD) = 180      4. Linear pair is supplementary
  5. m(angle AOC) + m(angle AOD) =          5. Both equal 180 (substitution /
     m(angle AOD) + m(angle BOD)               things equal to the same thing)
  6. m(angle AOC) = m(angle BOD)            6. Subtract m(angle AOD) from both
                                               sides  (Q.E.D.)
The Vertical Angle Theorem, proved line by line — angle AOD is the same in both equations, so cancelling it leaves the two vertical angles equal.

That is a complete, honest proof of the vertical angle theorem. Step 6 quietly uses another common notion — "equals subtracted from equals leave equals" — which is exactly the everyday rule that you may subtract the same number from both sides of an equation. The little "Q.E.D." (or a filled square) just announces that the conclusion we set out to reach has been reached.

A second proof, and the habits to keep

Try the same machinery on a tiny segment claim, to feel that the method transfers. Suppose M is the midpoint of segment AB and you want to show |AM| equals half of |AB|. The steps below are deliberately short; notice that the engine is identical — a given, a definition, and an algebra-as-common-notion finish.

  1. Given: M is the midpoint of AB. (This is what you may assume.)
  2. By the definition of midpoint, |AM| = |MB|. (Reason: definition, not a picture.)
  3. By the segment addition postulate, |AM| + |MB| = |AB|. (The whole is the sum of its parts.)
  4. Substitute |AM| for |MB| (they are equal), giving |AM| + |AM| = |AB|, i.e. 2|AM| = |AB|.
  5. Divide both sides by 2: |AM| = |AB| / 2. Done — every step named a reason.

Two honest cautions as you go on. First, a proof is only as trustworthy as the results it cites: we leaned on "a linear pair is supplementary," which itself rests on the angle addition postulate — a real proof would either prove that lemma too or point to where it was proved, never assume it because it feels obvious. Second, watch the direction of a conditional: proving "if vertical, then equal" does not prove the converse "if equal, then vertical" (two angles can be equal for all sorts of reasons). And if your draft ever rests on "the diagram shows it," you have not found a reason — you have found a gap waiting for a counterexample.