Theorem 1.4.4 (Rothstein 1935) — no proper holomorphic map from the bidisc to the ball
ProvedLeblSCV.BallPolydisc.rothsteinLet be the unit bidisc and the Euclidean unit ball. There exists no proper holomorphic mapping
Since a biholomorphism is proper, the bidisc and the ball are not biholomorphically equivalent, although they are homeomorphic: in several variables the geometry of the boundary, not only the topology of the domain, decides equivalence. The inequivalence goes back to Poincaré; the first complete proof is H. Cartan's (1931), and the stronger statement about proper maps is Rothstein's (1935).
Formalization Note. is Fin 2 → ℂ. The bidisc is unitPolydisc 2 and the ball is unitBall 2, written with the explicit Euclidean condition (the Mathlib sup-norm unit ball of Fin 2 → ℂ is the bidisc itself, for which the identity would be a proper holomorphic map). Holomorphic is DifferentiableOn ℂ on the open bidisc, equivalent there to the book's Definition 1.1.2; proper is IsProperMapOn (Definition 1.4.3 for the restricted map ).
import Mathlib import Definitions.Def_LeblSCV_BallPolydisc_unitBall import Definitions.Def_LeblSCV_BallPolydisc_unitPolydisc import Definitions.Def_LeblSCV_BallPolydisc_IsProperMapOn
namespace LeblSCV.BallPolydisc
/-- Theorem 1.4.4 (Rothstein 1935; Lebl, p. 33). There exists no proper holomorphic mapping of
the unit bidisc `𝔻² ⊆ ℂ²` to the (Euclidean) unit ball `𝔹₂ ⊆ ℂ²`. -/
theorem rothstein :
¬ ∃ f : (Fin 2 → ℂ) → (Fin 2 → ℂ),
DifferentiableOn ℂ f (unitPolydisc 2) ∧
IsProperMapOn f (unitPolydisc 2) (unitBall 2) := by sorry
end LeblSCV.BallPolydisc
Read-back
What the Lean code literally says, in plain math · claude-opus-5-5
The statement asserts that a certain kind of map does not exist. Throughout, means the space of functions , i.e. pairs of complex numbers. In Mathlib this space carries the sup norm , not the Euclidean norm. Whether that matters depends on how the imported definitions below are written.
The statement uses three names from this bundle. Their definitions are imported from separate files and their code is not part of what I was given, so I cannot unfold them. They are reported here only by name:
- — a subset defined in
Def_LeblSCV_BallPolydisc_unitPolydisc. - — a subset defined in
Def_LeblSCV_BallPolydisc_unitBall. - — a predicate on a function and two subsets, defined in
Def_LeblSCV_BallPolydisc_IsProperMapOn. Which set plays the role of domain and which of target, and what "proper" requires, is fixed by that definition and is not visible here.
The docstring's descriptions ("unit bidisc", "Euclidean unit ball") are comments, not code, and are not confirmed by anything shown.
The theorem asserts that there is no function satisfying both of the following conditions:
- Differentiable on . At every point , has a complex-linear Fréchet derivative within the set . This means the limit is taken only along points of . If is open, this is the same as being holomorphic on . If is not open, it is a weaker, one-sided-type condition at boundary points of lying in .
- holds, in whatever sense that imported definition gives it.
Nothing else is required of :
- is a total function on all of , and its values outside are completely unconstrained.
- No separate hypothesis says that maps into . Any such requirement can come only from inside .
- No continuity or regularity is required outside .
There are no variable dimensions, counts, or horizons: the dimension is fixed at in both places, so or one-element cases do not arise. The statement involves no division, natural-number subtraction, integrals, or suprema at the level shown. The main degenerate possibilities come from the unseen definitions:
- If can never be satisfied by any (for example, if it demands something impossible, or if or is defined so as to rule every function out), the theorem is vacuously true and says nothing about holomorphic maps.
- If were empty, the differentiability condition would hold for every . The whole claim would then rest entirely on being unsatisfiable.
- If is a ball measured in Mathlib's default sup norm rather than the Euclidean norm, it is itself a polydisc. The statement would then concern maps between two polydiscs rather than from a polydisc to a Euclidean ball.
Which of these cases apply is decided by the three imported definitions, which are not shown.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.