the residue as the coefficient a_{-1}

Of all the infinitely many Laurent coefficients, exactly one survives when you integrate the function around a small loop — and that one coefficient has a name: the residue. It is the single most important number attached to an isolated singularity, the bridge from the static picture of a Laurent series to the active business of evaluating integrals.

The definition is disarmingly simple: the residue of f at an isolated singularity z_0, written Res(f, z_0), is just the coefficient a_{-1} in the Laurent expansion of f on the punctured disk around z_0 — the coefficient of the 1/(z - z_0) term. Why this one and no other? Because of integration. When you integrate (z - z_0)^n around a small circle centred at z_0, you get 0 for every integer n except n = -1, where you get 2 pi i. So integrating the whole Laurent series term by term collapses to a single survivor: the integral over the small loop of f(z) dz = 2 pi i times a_{-1} = 2 pi i times Res(f, z_0). Every other term integrates to zero; only the 1/(z - z_0) term leaves a trace.

This identity is the seed of the residue theorem, which sums residues to evaluate integrals around any contour. Crucially the residue is local — it depends only on f near z_0 — yet it controls a global integral. In practice you rarely expand the full Laurent series just to read off a_{-1}; for a simple pole you use Res(f, z_0) = limit as z approaches z_0 of (z - z_0) f(z), and for a pole of order m a derivative formula. But the conceptual anchor never changes: the residue is a_{-1}, the one coefficient the contour integral can see, and the gateway to the next field.