The Timpson Lyceum seal

The Timpson Lyceum · Real Analysis (2020)

Lecture 2 (Real Analysis 2020) part 3

Overview

This part builds the natural numbers up from the notion of an inductive set, then uses that foundation to justify the Principle of Mathematical Induction as a theorem rather than an article of faith. With induction in hand we climb the number systems — integers, rationals — and discover that even though the rationals satisfy every field and positivity axiom, they are not enough: there is no rational number whose square is 22. Proving 2\sqrt{2} is irrational motivates the final ingredient of the real numbers, the Completeness Axiom, expressed through least upper bounds (suprema). The part closes by defining the square-root symbol via a uniqueness result and showing that completeness for upper bounds automatically delivers greatest lower bounds (infima).

Key Concepts

Inductive set. A set S⊆RS \subseteq \mathbb{R} is inductive if 1∈S1 \in S and whenever x∈Sx \in S then x+1∈Sx+1 \in S.

The natural numbers. N\mathbb{N} is defined as the intersection of all inductive sets. Consequently:

Principle of Mathematical Induction. Let S(n)S(n) be a statement about each natural number nn. If

then S(n)S(n) is true for all n∈Nn \in \mathbb{N}. The proof: form A={ n∈N:S(n) is true }A = \{\, n \in \mathbb{N} : S(n) \text{ is true} \,\}. The base case gives 1∈A1 \in A; the inductive step gives k∈A⇒k+1∈Ak \in A \Rightarrow k+1 \in A. So AA is an inductive subset of N\mathbb{N}, hence A=NA = \mathbb{N}. Induction only reaches the naturals (starting at 11 and stepping by +1+1 never lands on 12\tfrac12, say), which is exactly why the conclusion is restricted to N\mathbb{N}.

The number systems.

Divisibility, even, odd. For integers a,ba,b: a∣ba \mid b means there exists c∈Zc \in \mathbb{Z} with ac=bac = b. An integer is even if it equals 2x2x for some x∈Zx \in \mathbb{Z} (equivalently 2∣a2 \mid a); odd if it equals 2x+12x+1 for some x∈Zx \in \mathbb{Z}. Basic results proved by these definitions: a product of two evens is even; a product of two odds is odd; and (by induction) every integer is even or odd.

Why the rationals are not enough. The polynomial x2−2x^2 - 2 is negative at x=0x=0 and positive at x=2x=2, so it "should" cross zero — but no rational number does so. Hence analysis needs more than the field/positivity axioms.

Bounded sets and the Completeness Axiom. A nonempty set S⊆RS \subseteq \mathbb{R} is bounded above if there is c∈Rc \in \mathbb{R} with x≤cx \le c for all x∈Sx \in S (such a cc is an upper bound). The Completeness Axiom: every nonempty set that is bounded above has a least upper bound. This least upper bound is the supremum sup⁡S\sup S. Symmetrically, a set is bounded below if some c≤xc \le x for all x∈Sx \in S, and its greatest lower bound is the infimum inf⁡S\inf S. For example {x:x2<2}\{x : x^2 < 2\} is bounded above with supremum 2\sqrt{2} — completeness is what conjures irrational numbers into existence.

Uniqueness of positive square roots. For each c>0c > 0 there is exactly one x>0x > 0 with x2=cx^2 = c; this unique positive solution defines c\sqrt{c}. Existence comes from completeness; uniqueness is proved below.

Infima come for free. Completeness is stated only for upper bounds, but reflecting a set through 00 converts a lower bound into an upper bound, so every nonempty set bounded below has an infimum — no second axiom required.

Worked Example

Claim. ∑j=1nj=n(n+1)2\displaystyle \sum_{j=1}^{n} j = \frac{n(n+1)}{2} for every n∈Nn \in \mathbb{N} (proof by induction).

Let A={ n∈N:∑j=1nj=n(n+1)2 }A = \left\{\, n \in \mathbb{N} : \sum_{j=1}^{n} j = \frac{n(n+1)}{2} \,\right\}. Showing AA is inductive forces A=NA = \mathbb{N}.

Base case (n=1):∑j=11j=1=1⋅22=1(1+1)2,so 1∈A.Inductive step: assume ∑j=1kj=k(k+1)2(k∈A).∑j=1k+1j=(∑j=1kj)+(k+1)=k(k+1)2+(k+1)=k(k+1)+2(k+1)2=(k+1)(k+2)2=(k+1)((k+1)+1)2.\begin{aligned} \textbf{Base case }(n=1):\quad \sum_{j=1}^{1} j &= 1 = \frac{1\cdot 2}{2} = \frac{1(1+1)}{2}, \quad\text{so } 1 \in A. \\ \textbf{Inductive step: assume } \sum_{j=1}^{k} j &= \frac{k(k+1)}{2} \quad (k \in A). \\ \sum_{j=1}^{k+1} j &= \left(\sum_{j=1}^{k} j\right) + (k+1) = \frac{k(k+1)}{2} + (k+1) \\ &= \frac{k(k+1) + 2(k+1)}{2} = \frac{(k+1)(k+2)}{2} = \frac{(k+1)\big((k+1)+1\big)}{2}. \end{aligned}

So k+1∈Ak+1 \in A. Thus AA is inductive, A=NA = \mathbb{N}, and the formula holds for all nn.

A second, contradiction-style proof: 2∉Q\sqrt{2} \notin \mathbb{Q}. Suppose for contradiction that 2=mn\sqrt{2} = \tfrac{m}{n} with m,n∈Zm,n \in \mathbb{Z}, n≠0n \neq 0, where (reducing the fraction) at least one of m,nm, n is odd.

m2n2=2  ⇒  m2=2n2  ⇒  m2 even  ⇒  m even.m=2k  ⇒  m2=4k2  ⇒  4k2=2n2  ⇒  n2=2k2  ⇒  n2 even  ⇒  n even.\begin{aligned} \frac{m^2}{n^2} &= 2 \;\Rightarrow\; m^2 = 2n^2 \;\Rightarrow\; m^2 \text{ even} \;\Rightarrow\; m \text{ even.} \\ m = 2k \;\Rightarrow\; m^2 = 4k^2 \;\Rightarrow\; 4k^2 = 2n^2 \;&\Rightarrow\; n^2 = 2k^2 \;\Rightarrow\; n^2 \text{ even} \;\Rightarrow\; n \text{ even.} \end{aligned}

Now mm and nn are both even, contradicting that at least one was odd. Hence no such fraction exists and 2∉Q\sqrt{2} \notin \mathbb{Q}.

Summary