9.2 Proof Techniques & Mathematical Induction

9.2 Proof Techniques & Mathematical Induction

Mathematical proof is the definitive currency of mathematical knowledge. While natural sciences establish theories through empirical observation and probabilistic verification, mathematics demands deductive certainty established through valid deductive chains. For secondary mathematics candidates preparing for the NBPTS AYA assessment, understanding the structural mechanics of direct proof, indirect proof (contrapositive and contradiction), counterexamples, and the Principle of Mathematical Induction is foundational. Accomplished teachers must not only execute rigorous proofs, but also scaffold secondary students' developmental transition from empirical pattern recognition to formal deductive argumentation (NBPTS Standard II; CCSSM SMP 3).

---

1. The Anatomy of Mathematical Proof: Axioms, Definitions & Deductive Rules

A mathematical proof is a finite sequence of statements leading from accepted starting assumptions to a necessary conclusion, where each statement is justified by an axiom, a definition, an established theorem, or a valid rule of logical inference.

The Foundational Components of Deductive Systems

• Primitive Notions (Undefined Terms): Foundational entities accepted without formal definition to avoid infinite regress (e.g., point, line, plane in Euclidean geometry; set, element in set theory).
• Axioms / Postulates: Initial propositions assumed to be true without proof within a formal axiomatic system (e.g., Euclid's Parallel Postulate, the Peano Axioms of arithmetic, the Zermelo-Fraenkel axioms of set theory).
• Formal Definitions: Precise, reversible biconditional statements characterizing mathematical objects. When a definition states "an integer n is even if n = 2k for some k ∈ ℤ," it means n is even ⟺ ∃k ∈ ℤ, n = 2k.
• Theorems, Lemmas, and Corollaries: A theorem is a major mathematical statement established by proof. A lemma is a preliminary theorem proven primarily to facilitate the proof of a larger theorem. A corollary is an immediate, direct consequence of an established theorem.

Rules of Deductive Inference

Deductive proofs transition between lines using truth-preserving inference rules:

1. Modus Ponens (Law of Detachment): From P → Q and P, deduce Q.
2. Modus Tollens: From P → Q and ¬Q, deduce ¬P.
3. Hypothetical Syllogism (Transitivity of Implication): From P → Q and Q → R, deduce P → R.
4. Disjunctive Syllogism: From P ∨ Q and ¬P, deduce Q.

---

2. Direct Proof vs. Proof by Contrapositive

Direct Proof: P ⟹ Q

In a direct proof, one assumes the hypothesis P is true, unfolds algebraic or geometric definitions, applies deductive inference and known theorems, and arrives directly at the conclusion Q.

Theorem: The sum of any two odd integers is an even integer.

• Hypothesis: Let m and n be arbitrary odd integers.
• Unfolding Definitions: By definition of odd integers, there exist integers j, k ∈ ℤ such that m = 2j + 1 and n = 2k + 1.
• Algebraic Deduction: Adding the two equations: m + n = (2j + 1) + (2k + 1) = 2j + 2k + 2 = 2(j + k + 1)
• Closure Property: Since the integers ℤ are closed under addition, p = j + k + 1 is an integer (p ∈ ℤ).
• Conclusion: Thus, m + n = 2p for p ∈ ℤ, which satisfies the formal definition of an even integer. Q.E.D.

Proof by Contrapositive: ¬Q ⟹ ¬P

Because a conditional statement P → Q is logically equivalent to its contrapositive ¬Q → ¬P, one can prove P → Q by assuming ¬Q (the negation of the conclusion) and showing that ¬P (the negation of the hypothesis) necessarily follows.

[!TIP] When to Select Contrapositive over Direct Proof: Select proof by contrapositive when the negation of the conclusion ¬Q provides concrete, linear algebraic structure to manipulate, whereas the original hypothesis P involves non-linear, composite, or divisibility conditions that are cumbersome to decompose directly.

Theorem: For any integer n, if n² is even, then n is even.

• Direct Attempt Analysis: If one attempts a direct proof, one assumes n² = 2k. Taking the square root gives n = √(2k), which introduces radicals and fails to demonstrate algebraically that n is an integer multiple of 2.
• Contrapositive Formulation: We prove the logically equivalent statement: "If n is not even (i.e., n is odd), then n² is not even (i.e., n² is odd)."
• Proof:
  1. Assume n is odd. By definition, n = 2k + 1 for some integer k ∈ ℤ.
  2. Compute n² algebraically: n² = (2k + 1)² = 4k² + 4k + 1 = 2(2k² + 2k) + 1
  3. Let m = 2k² + 2k. Because ℤ is closed under multiplication and addition, m ∈ ℤ.
  4. Therefore, n² = 2m + 1, which by definition means n² is an odd integer.
  5. Having established ¬Q ⟹ ¬P, the original conditional P ⟹ Q is proven. Q.E.D.

---

3. Proof by Contradiction (Reductio ad Absurdum) & Counterexamples

Proof by Contradiction

In a proof by contradiction, to establish a mathematical statement S, one assumes that S is false (i.e., assumes ¬S). One then reasons deductively until deriving a logical contradiction—a statement of the form C ∧ ¬C (such as 0 = 1, or an integer being simultaneously even and odd). Because valid deductive reasoning cannot produce a false statement from true premises, the initial assumption ¬S must be false, forcing S to be true.

To prove a conditional P → Q by contradiction, one assumes both P and ¬Q: ¬(P → Q) ≡ P ∧ ¬Q

Classical Proof 1: The Irrationality of √2

• Claim: √2 is irrational (√2 ∉ ℚ).
• Assumption for Contradiction: Assume √2 is rational. Then there exist positive integers a, b ∈ ℤ⁺ such that: √2 = a / b, where gcd(a, b) = 1 (fraction is in irreducible lowest terms)
• Deduction:
  1. Squaring both sides: 2 = a² / b² ⟹ a² = 2b².
  2. Since a² is a multiple of 2, a² is even. By our contrapositive theorem, a² is even ⟹ a is even.
  3. Because a is even, a = 2k for some integer k ∈ ℤ⁺.
  4. Substitute a = 2k into the equation: (2k)² = 2b² ⟹ 4k² = 2b² ⟹ b² = 2k².
  5. This demonstrates that b² is even, which implies b must also be even.
  6. If both a and b are even, then 2 divides both a and b, meaning gcd(a, b) ≥ 2.
• Contradiction: We have deduced that gcd(a, b) ≥ 2 and gcd(a, b) = 1, an explicit contradiction (C ∧ ¬C). Thus, the initial assumption that √2 ∈ ℚ is false. Therefore, √2 is irrational. Q.E.D.

Classical Proof 2: Euclid's Proof of the Infinitude of Primes

• Claim: There are infinitely many prime numbers.
• Assumption for Contradiction: Assume the set of all prime numbers is finite, denoted by P = {p₁, p₂, p₃, …, pₙ}.
• Construction: Define the integer N = (p₁ · p₂ · p₃ ··· pₙ) + 1.
• Deduction:
  1. Because N > 1, by the Fundamental Theorem of Arithmetic, N must have at least one prime divisor q.
  2. Since P allegedly contains all prime numbers, q must be an element of P; thus q = pᵢ for some i ∈ {1, 2, …, n}.
  3. Therefore, q divides the product: q ∣ (p₁ p₂ ··· pₙ).
  4. Since q ∣ N and q ∣ (p₁ p₂ ··· pₙ), q must divide their difference: q ∣ [N - (p₁ p₂ ··· pₙ)] ⟹ q ∣ 1
  5. The only positive divisor of 1 is 1. However, by definition, all primes satisfy q ≥ 2.
• Contradiction: q ≥ 2 and q ∣ 1 is impossible. The assumption that primes are finite must be false. Q.E.D.

The Role of Counterexamples

A single counterexample definitively refutes a universally quantified statement ∀x P(x). While billions of confirming examples cannot prove a universal statement, a single instance c where ¬P(c) is true disproves it: ¬(∀x P(x)) ≡ ∃x ¬P(x)

• Historical Example (Fermat Primes): Pierre de Fermat conjectured in 1640 that all integers of the form Fₙ = 2^(2ⁿ) + 1 are prime. The conjecture held for n = 0, 1, 2, 3, 4 (3, 5, 17, 257, 65537). However, in 1732, Leonhard Euler disproved the conjecture by producing the single counterexample n = 5: F₅ = 2^(2⁵) + 1 = 2³² + 1 = 4,294,967,297 = 641 × 6,700,417
• Pedagogical Warning Against Empirical Pattern Generalization: High school students frequently mistake inductive pattern observation for proof. Consider f(n) = n² + n + 41. For n = 0, 1, 2, …, 39, f(n) yields prime numbers. However, at n = 40: f(40) = 40² + 40 + 41 = 40(40 + 1) + 41 = 40(41) + 41 = 41(40 + 1) = 41² which is composite (1681). Checking 40 consecutive cases fails to constitute a mathematical proof.

---

4. The Principle of Mathematical Induction & The Well-Ordering Principle

The Well-Ordering Principle (WOP)

The Well-Ordering Principle of ℤ⁺: Every non-empty subset S of positive integers has a least element (i.e., ∃m ∈ S such that ∀s ∈ S, m ≤ s).

The Well-Ordering Principle is logically equivalent to the Principle of Mathematical Induction and provides the axiomatic foundation for inductive proofs.

Principle of Mathematical Induction (PMI - Weak Induction)

To prove that a predicate P(n) is true for all integers n ≥ n₀ (typically n₀ = 1):

1. Base Step (Anchor): Prove that P(n₀) is true.
2. Inductive Step: Prove that for every integer k ≥ n₀, if P(k) is true, then P(k+1) is true: ∀k ≥ n₀, [P(k) ⟹ P(k+1)]
3. Conclusion: By the Principle of Mathematical Induction, P(n) is true for all integers n ≥ n₀.

Worked Example: Sum of the First n Integers

Prove that for all n ≥ 1, ∑(i=1 to n) i = n(n+1)/2.

• Base Step (n = 1): LHS = ∑(i=1 to 1) i = 1. RHS = 1(1+1)/2 = 1. Since LHS = RHS, P(1) is true.
• Inductive Hypothesis: Assume that for an arbitrary integer k ≥ 1, P(k) holds: ∑(i=1 to k) i = k(k+1)/2
• Inductive Step (P(k+1)): We must deduce that ∑(i=1 to k+1) i = (k+1)((k+1)+1)/2 = (k+1)(k+2)/2. Starting with the LHS of P(k+1): ∑(i=1 to k+1) i = [∑(i=1 to k) i] + (k+1) Substituting the inductive hypothesis into the sum of the first k terms: = k(k+1)/2 + (k+1) = (k+1)[k/2 + 1] = (k+1)[(k+2)/2] = (k+1)(k+2)/2 The LHS equals the RHS of P(k+1). Thus, P(k) ⟹ P(k+1).
• Conclusion: By PMI, ∑(i=1 to n) i = n(n+1)/2 holds for all integers n ≥ 1. Q.E.D.

Strong Induction (Complete Induction)

In Strong Mathematical Induction, the inductive hypothesis assumes that the proposition holds for all preceding integers from the base case up to k: ∀k ≥ n₀, [P(n₀) ∧ P(n₀ + 1) ∧ … ∧ P(k)] ⟹ P(k+1) Strong induction is logically equivalent to weak induction, but is necessary when the proof of P(k+1) relies on earlier values P(j) where j < k (e.g., proving that every integer n > 1 can be factored into primes, or analyzing Fibonacci recurrences Fₙ = Fₙ₋₁ + Fₙ₋₂).

---

5. Secondary Pedagogical Scenarios & Circularity Misconceptions

The Physical Analogies: Dominos and Ladders

Accomplished educators use dynamic physical analogies to anchor the formal structure of PMI:

• The Domino Analogy: The base case knocks down the first domino (P(1)). The inductive step ensures that dominos are spaced such that whenever any domino k falls, it knocks down domino k+1 (P(k) ⟹ P(k+1)). Both conditions guarantee that every domino in the infinite chain falls.
• The Infinite Ladder Analogy: The base step proves you can step onto the first rung (P(1)). The inductive step establishes that your physical ability allows you to step from any arbitrary rung k to rung k+1. Together, you can climb to any arbitrary height n.

Diagnosing and Dismantling Student Misconceptions in Induction

Misconception 1: Accusing the Inductive Hypothesis of Circular Reasoning

Students frequently voice an objection: "In the inductive hypothesis, we assume P(k) is true. Isn't that assuming the very thing we are trying to prove?"

• Teacher Diagnostic: The student is conflating the universal theorem ∀n P(n) with the local premise within a conditional sub-proof. In mathematical logic, to prove a conditional A ⟹ B, one temporarily assumes A as a hypothesis to deduce B.
• Pedagogical Intervention: Make the conditional nature of the inductive step explicit. We are not asserting P(k) is true for all k; we are proving the bridge: IF P(k) is true, THEN P(k+1) is true. The base case P(1) provides the factual spark that traverses the bridge.

Misconception 2: "Working Backwards" and Assuming P(k+1)

Students frequently write down the equation for P(k+1) at the top of their paper, manipulate both sides simultaneously, and arrive at 0 = 0, declaring the proof complete.

• Teacher Diagnostic: The student has committed the fallacy of assuming the conclusion. In deductive proof, starting with P(k+1) as a premise is invalid because one cannot assume what has yet to be established.
• Pedagogical Intervention: Enforce unidirectional derivation protocols. Instruct students to write down only the left-hand side of P(k+1), identify where the expression contains P(k), substitute the inductive hypothesis, and algebraically rewrite the expression until it transforms into the right-hand side of P(k+1).