CA01 — Cauchy Integral Theorem and Cauchy-Riemann Equations

Field Value
Domain Complex Analysis
System Holomorphic functions on ℂ
Group U(1) (phase rotations of ℂ)
H^k tier
ISA Origami (β → ∞)
Status Validated
Opcodes ORBIT · TWIST · LABEL
Papers Paper 543, Paper 477

What this entry is about

This is the cleanest possible H⁰/H¹/H² example — accessible from first-year university calculus, but carrying the full weight of the ISA framework. No physics required: just functions of a complex variable.

The punchline in one sentence: the Cauchy-Riemann equations are the H¹ flatness condition. A function that satisfies them is holomorphic — its contour integral around any loop in a simply-connected domain is zero. When the domain has a hole, the integral can be nonzero — and that nonzero value is the H¹ TWIST winding number. That is the entire ISA story, in one paragraph of calculus.


Physical system

Write z = x + iy and f(z) = u(x,y) + iv(x,y). The Cauchy-Riemann equations are:

∂u/∂x = ∂v/∂y and ∂u/∂y = −∂v/∂x

These are the condition that f is holomorphic — complex-differentiable at every point. They look like two real equations, but they encode one complex constraint: df/dz̄ = 0 (f does not depend on z̄ = x − iy, only on z = x + iy).

Cauchy’s integral theorem: if f is holomorphic on a simply-connected domain D, then for any closed loop C inside D:

∮_C f(z) dz = 0

If D has a hole — say D = ℂ{0}, the plane with the origin removed — then loops that wind around the hole can give nonzero integrals. The canonical example:

∮_{ z =1} dz/z = 2πi

The function 1/z is holomorphic everywhere except z = 0; the value 2πi is the H¹ winding number of the loop around the singularity. This is the TWIST.


Target category

Hol(D) — the category of holomorphic functions on a domain D ⊂ ℂ. Objects: pairs (D, f) where f: D → ℂ is holomorphic. Morphisms: conformal maps between domains. The simply-connected domains are the “H⁰-flat” objects — they have no holes, so every closed 1-form (every holomorphic 1-form) is exact. Domains with holes have H¹ ≠ 0; each hole contributes one generator of H¹(D, ℤ) = ℤ.

Interpretation functor

F: C → Hol(D) defined by:

Opcode F(opcode)
ORBIT Analytic continuation: extend f from one region of D to another along a path; the ORBIT fixed point is the value f(z) at a point, uniquely determined by continuity
TWIST Winding number: n(C, z₀) = (1/2πi) ∮_C dz/(z−z₀) ∈ ℤ; the H¹ count of how many times the loop C winds around the point z₀; nonzero only when z₀ ∉ D
LABEL Cauchy’s integral formula value: f(z₀) = (1/2πi) ∮_C f(z)/(z−z₀) dz; the eigenvalue extracted from the loop integral

ISA programme

DOMAIN:  LABEL[D subset C | domain of f]             -- specify domain (simply connected?)
C-R:     TWIST[df/dzbar = 0? | C-R equations]        -- check H1 flatness (is f holomorphic?)
LOOP:    ORBIT[C | closed loop in D]                  -- choose contour
WIND:    TWIST[n(C,z0) = (1/2pi i) oint dz/(z-z0)]  -- winding number (H1 invariant)
CAUCHY:  LABEL[oint_C f(z)dz = 0 if simply connected] -- Cauchy theorem (H1=0 case)
EXTRACT: LABEL[f(z0) = (1/2pi i) oint f(z)/(z-z0) dz] -- Cauchy formula (LABEL output)

The H⁰/H¹/H² ladder in one example

Take D = ℂ{0} (the plane minus the origin). Consider the question: does 1/z have an antiderivative on D?

H⁰ answer (ORBIT): locally, yes. Near any point z₀ ≠ 0, we can write ∫ dz/z = log z in a small disk around z₀. The function log z is the local antiderivative — the ORBIT fixed point.

H¹ answer (TWIST): globally, no. If we try to continue log z all the way around the origin (ORBIT along a loop), we return with value log z + 2πi — not the value we started with. The multi-valuedness is the H¹ obstruction: the winding number n(C, 0) = 1 for a loop around the origin, and

∮_{ z =1} dz/z = 2πi ≠ 0.

The TWIST opcode “sees” the hole at the origin. Simply-connected domains (no holes) have H¹ = 0: every holomorphic 1-form is exact, and every antiderivative is single-valued.

H² answer (BIND): the origin itself — the singularity — is a H² BIND obstruction. The function 1/z does not extend to z = 0 at all; it has a pole there. The residue Res(1/z, 0) = 1 is the H² invariant (→ CA02).

This is the ISA framework in three lines of calculus:

  • H⁰: local antiderivative exists (ORBIT)
  • H¹: global antiderivative fails to single-valued due to winding (TWIST)
  • H²: pole at the singularity; residue is the obstruction (BIND → CA02)

Why the Cauchy-Riemann equations are a flatness condition

In differential geometry language: the 1-form ω = f(z) dz is closed (dω = 0) if and only if the Cauchy-Riemann equations hold. Closed means the integral around any contractible loop is zero. Exact means there is a global antiderivative.

On a simply-connected domain: closed = exact (Poincaré lemma). Every holomorphic 1-form has an antiderivative.

On a domain with holes: closed ≠ exact. The failure of exactness is measured by H¹(D, ℂ) — one generator per hole. The generators are the winding numbers around each hole. In ISA language: the TWIST opcode counts the generators of H¹; one TWIST per hole.

This is why the Cauchy-Riemann equations are H¹ flatness: they say ω is closed. Whether ω is also exact — whether the antiderivative is single-valued — depends on H¹(D), which measures the topology of the domain. The C-R equations are local; the ISA tier is global.

Connection to the β-plane (Paper 543)

The imaginary axis of the β-plane (β = it) is precisely where complex analysis lives: the Meld ISA with β = it gives complex amplitudes e^{−itE}, and the Cauchy-Riemann equations are the condition that these amplitudes are holomorphic as functions of the complex parameter z = x + iy = β.

The Wick rotation (β real → β imaginary) rotates from statistical mechanics (Boltzmann weights, real exponentials) to quantum mechanics (complex amplitudes, holomorphic functions). The Cauchy-Riemann equations are the condition that this rotation is well-defined — that the function of β is holomorphic at the rotation point. When C-R fail (the function is not holomorphic), there is a singularity on the real axis: a phase transition.

Phase transitions are poles of the partition function in the complex β-plane. The Lee-Yang theorem (1952) proves that the zeros of the grand canonical partition function Z(z) (as a function of fugacity z = e^{βμ}) approach the real axis in the thermodynamic limit, and the phase transition is the accumulation point of these zeros. The poles are H² BIND events (CA02); the approach to the real axis is the Wick rotation from imaginary β (complex analysis) to real β (statistical mechanics).

Validation

  • Cauchy (1825): integral theorem for holomorphic functions. One of the foundational results of mathematics; taught universally in second-year analysis.
  • Cauchy-Riemann equations: independently derived by Cauchy (1814) and Riemann (1851); equivalent to complex differentiability.
  • Winding number: homotopy invariant of π₁(ℂ{0}) = ℤ; the generator of the first fundamental group. Classical topology.
  • Poincaré lemma: closed = exact on contractible domains. Standard differential topology; foundation of de Rham cohomology.

Part of the ISA Zoo. Categorical foundations: Paper 591.