Ordinary Induction
To prove for all : prove (base case); then prove that for every , implies (inductive step). Together these knock over every domino.
Induction is the proof method built into the natural numbers: they are exactly the set you reach from by adding repeatedly, so a property that starts true and survives "" is true everywhere. The inductive step is not "assume what you are proving" — it is "assume it for one specific and prove it for the next one". Writing the inductive hypothesis out explicitly, with a specific , is what keeps that straight.
Claim: for all . Base: : . Step: assume for some . Adding the next odd number, , which is the claim for .
Claim: for all . Base: . Step: assume . Then . The first term is divisible by by the hypothesis and the second is , so divides the whole.
Claim: for all . Base: . Step: assume for some . Then .
Two uses of hide in the chain: needs , and needs . The base case is because the claim is false just below it: , , . An induction must start where the claim starts being true, and the step must be proved only for at or beyond that point.
Write the case; find the case inside it (peel off the last term, factor out one power); substitute the inductive hypothesis; finish with algebra. If you cannot find the case inside, you may need strong induction.
Forgetting the base case — "" is true of "" and proves nothing without a start. And proving from scratch without using , which usually means induction was not needed.
Prove that for all .
Show the answerHide the answer
Base: . Step: assume for some . Then , using in the last step.
Strong Induction
In strong induction the inductive hypothesis is " holds for ALL with ", and you prove . It is needed when the case depends on cases further back than — factorizations, Fibonacci-type recurrences, games. It is logically equivalent to ordinary induction.
Why is it allowed to assume so much? By the time the dominoes reach , every earlier domino has already fallen, not just the one before. Formally, strong induction on is ordinary induction on the statement : " holds for every from to ". Proving is exactly proving from all the earlier cases.
Assume every with is a product of primes. If is prime, it is a product of one prime. Otherwise with ; each is a product of primes by hypothesis, and so is their product. (Ordinary induction fails here: the factors of are not .)
Every is a sum of s and s. Bases: , , , . Step: for , is a sum of s and s by hypothesis; add a . (Four base cases because the step reaches back .)
Too few base cases. If the step uses , the first four values must be checked by hand or the chain has a gap. With only the base above, the step would produce and never reach .
Define , and for . Prove that for all .
Show the answerHide the answer
Bases: : ; : . Both match. Step: assume the formula for and , with . Then
which is the formula for . Two bases are needed because the step uses the two previous terms. (Checking the bases first is also how you catch a wrong guess: the tempting gives at , not .)
Where Inductions Go Wrong
An induction has two obligations and each fails in its own way. A false or missing base leaves a valid chain with nothing to start it. A step that fails for one value of breaks the chain at that link, and everything beyond is unproved. A circular step uses to prove .
The step must hold for EVERY , because the chain from to uses every link from to . A step that "clearly works" for large but has a hidden requirement such as proves nothing about the small cases, and the small cases are exactly where false claims hide. Test the step on the smallest it is supposed to cover.
"Claim": is even for every . The step is genuinely valid: , so if is even, so is the next value. But the base fails: gives . In fact is always odd. Flaw: no base case — the chain was never started.
"Claim": in any group of horses, all have the same color. Base: one horse matches itself. Step: in a group of , the first share a color and the last share a color; the horses in both groups link the two, so all match. Flaw: the step needs the two groups to overlap, and for (a group of two) the "first one" and "last one" are different horses with nothing in common. The step fails at exactly one link, from to , and that is enough to break the whole chain.
A write-up that starts "assume " and simplifies until it reaches , which is true, has proved — the wrong direction. It can be rescued only if every step is reversible and the proof says so; the safe habit is to start from the expression and end at the claim.
Find the flaw. "Let : if and are positive integers with , then . Base : then . Step: if , then , so by hypothesis, and ."
Show the answerHide the answer
The hypothesis is only about POSITIVE integers, and or may be . For , : , but is not a positive integer, so says nothing about the pair . The step fails already at the first link, from to .
Induction Beyond Formulas
Induction proves statements about sets, boards, polygons and games — anything with a size . The step: take an arbitrary object of size , find inside it one or more objects of size at most , apply the hypothesis to them, and rebuild.
The word "arbitrary" carries the weight. The step must start from ANY object of size and break it down, not start from an object of size and build up, because building up only reaches the objects your construction happens to produce. Breaking down guarantees that every object of size is covered.
Claim: a set with elements has subsets. Base: the empty set has exactly one subset, itself, and . Step: take any set with elements and pick one element . Subsets not containing are the subsets of the other elements: by hypothesis. Subsets containing are those same sets with added. Total .
Claim: for every , a board with any one square removed can be tiled by L-shaped pieces of three squares. Base : a board minus one square IS an L. Step: cut a board with one square missing into four quadrants. The missing square is in one quadrant. Place one L at the centre so that it covers the corner square nearest the centre in each of the other three quadrants. Now every quadrant is a board with one square missing, and the hypothesis tiles each.
The single central tromino is the whole idea: it converts three complete quadrants into three copies of the smaller problem. A quick consistency check is that the count works out — squares must be a multiple of , and it is, since gives .
Building up. "Take a tiled board and attach three more quadrants" only proves the result for the boards that construction produces — here, boards whose missing square sits where you chose. Start from an arbitrary board of the larger size, with the missing square anywhere, and break it down.
Prove by induction that a convex polygon with vertices has diagonals.
Show the answerHide the answer
Base : a triangle has diagonals. Step: take any convex polygon with vertices and remove one vertex ; what is left is a convex -gon with diagonals. The diagonals of the big polygon are: those, plus the diagonals from to its non-neighbours, plus the segment joining 's two neighbours (a side of the -gon, a diagonal of the big one). Total
the formula for .
Why It Works
The well-ordering principle says every non-empty set of positive integers (or of integers ) has a smallest element. Suppose the base case and the step have both been proved, but fails for some . The failures form a non-empty set; let be its smallest element. Then , because the base case holds. So , and is true because is smaller than the smallest failure. The step, applied with , gives — contradicting the choice of . So there are no failures.
Given , define : " holds for every with ". The strong-induction step, "all of imply ", is exactly "". So ordinary induction on proves every , and in particular every . The two methods prove the same theorems; strong induction just lets you write the hypothesis you actually use.