The Timpson Lyceum · Real Analysis (2020)
Lecture 13 (Real Analysis 2020)
Overview
This lecture introduces the derivative — the third of the four big ideas of calculus (limit, continuity, derivative, integral). Intuitively, the derivative is the slope of a function at a single point, made precise as a special kind of limit. We start from the familiar middle-school slope of a line (rise over run), generalize it to the slope of a curve by letting two points slide together, and arrive at the formal definition as the limit of a difference quotient. We then build the basic machinery: the tangent line, the fact that differentiability implies continuity, and the standard rules for differentiating sums, products, reciprocals, quotients, and power functions.
Key Concepts
Slope of a line (motivation). For a straight line through two points and , the slope is the rise over the run: The slope is the same no matter which two points you pick — the defining feature of a line.
The difference quotient. For a general function and a fixed point , the slope of the secant line through and is This expression is called the difference quotient (a difference of values in a quotient). As moves closer to , the secant slope better approximates the slope at .
Neighborhood. An open interval is the set of all points strictly between and . An open interval containing is called a neighborhood of . We typically write for a neighborhood of .
Definition of the derivative. Let be a neighborhood of and . We say is differentiable at provided the limit exists, in which case this limit is the derivative of at . (Because it is a limit, never actually equals , so the denominator is never zero in the limiting process.) If is differentiable at every point of , we call differentiable.
When the derivative fails to exist. A function has no derivative where the slope is not well defined — at a sharp corner (e.g. at ), or anywhere it is discontinuous (no derivative without continuity).
The tangent line. If is differentiable at , the tangent line to at is This mirrors the point–slope form of a line; is the special case . Once the derivative exists at a point, this tangent line is unique.
Differentiability implies continuity. If is differentiable at , then is continuous at (i.e. ). The converse fails: is continuous everywhere but not differentiable at , and there even exist functions continuous everywhere yet nowhere differentiable.
Differentiation rules. For functions differentiable at :
- Constant slope of a line: if , then .
- Power rule: if for an integer , then (proved first for natural numbers, then extended to all integers).
- Constant multiple: if , then .
- Sum rule: .
- Product rule: — not the product of the derivatives.
- Reciprocal rule: if , then .
- Quotient rule: if , then ("low d-high minus high d-low, square the bottom").
- Consequence: every polynomial is differentiable everywhere, and every rational function is differentiable wherever its denominator is nonzero.
Worked Example
Prove from the definition that if , then . The key trick is to factor the numerator so the troublesome in the denominator cancels, after which the limit is found by direct substitution.
As a check on the product rule, take , whose derivative is . Applying the rule gives the same result:
Summary
- The derivative is the slope of at , defined as the limit of the difference quotient .
- A function is differentiable if this limit exists at every point of its domain; it fails at corners (like at ) and at any discontinuity.
- The tangent line at is , and it is unique wherever the derivative exists.
- Differentiability implies continuity, but not the reverse — continuity is necessary, not sufficient, for differentiability.
- Core rules: the power rule (all integers ), plus the sum, product, reciprocal, and quotient rules. The product rule is not the product of the derivatives.
- Consequently, polynomials are differentiable everywhere and rational functions are differentiable wherever the denominator is nonzero.