First PrinciplesStart anywhere. Prove it, then move on.

Rung 19

Limits

What a value approaches, including where it never arrives.

Best attempted after 11. Polynomials and roots. Nothing stops you trying this now — the gate will tell you if you were right.

The gate

Evaluate a limit that direct substitution cannot reach, both numerically from each side and algebraically, and show the two agree. Then construct a function whose left and right limits differ at a point, and say precisely why the limit does not exist there even though both one-sided limits do.

Nobody checks this but you. Do it honestly and the rungs above hold; do it loosely and they will not, somewhere further up where the cause is much harder to find.

A limit answers a question that looks illegal: what does this function do at a point where it is not defined? Not "what is its value there" — it has none — but "what value is it heading for as you close in".

That distinction is the whole idea, and every bit of calculus above this rung rests on it.

Why this is on the ladder

Because the derivative at rung 20 is a limit and nothing else. It is the slope between two points as the gap between them closes to nothing — a calculation that would be 0/0 if you simply set the gap to zero, and which has a perfectly definite answer as you approach it.

Skipping limits and memorising derivative rules is the most common way to arrive at rung 45 unable to debug anything. The rules are consequences. This is the thing they are consequences of.

Do this

Take f(x) = (x² - 1)/(x - 1). At x = 1 it is 0/0 — genuinely undefined.

Approach numerically: evaluate at 0.9, 0.99, 0.999, then 1.1, 1.01, 1.001. Write the values down. Both sides march toward the same number.

Now algebraically: x² - 1 factors as (x - 1)(x + 1), so for every x except 1 the function equals x + 1. The limit is 2. The function still has no value at 1 — a single missing point — and the limit does not care, which is the point.

Then build a failure. A function defined one way below a point and another way above it, where the two sides head for different numbers. Both one-sided limits exist; the limit does not. Say why that is the right convention rather than a technicality.

Where people get stuck

Concluding that the limit is the value. Usually it coincides, which is exactly what "continuous" means — but the interesting cases are where it does not, and those are the cases the machinery was built for.

The other snag is trusting the numerical approach alone. Getting closer and closer to 2 is evidence, not proof; floating point will eventually lie to you. Both routes, agreeing, is the gate.

Reading