A vector space over an infinite field is not a finite union of proper subspaces
ProvedFamousTheorems.vector_space_not_finite_union_proper_subspaces_6cA vector space over an infinite field is not a finite union of proper subspaces. Let be a vector space over an infinite division ring , and let be a finite set of subspaces of , none equal to . Then .
The result is used for "generic choice" arguments, for example to find a vector outside finitely many hyperplanes, and in the proof of the primitive element theorem for infinite fields. The hypothesis that is infinite is needed: over , the space is the union of its lines. It is a consequence of B. H. Neumann's lemma on coset covers.
Formalization note. Mathlib's Subspace.biUnion_ne_univ_of_top_notMem. The finite family is a Finset (Subspace k E), and properness of every member is ⊤ ∉ s.
Preamble
import Mathlib
Formal statement
namespace FamousTheorems
theorem vector_space_not_finite_union_proper_subspaces_6c {k E : Type*} [DivisionRing k] [Infinite k] [AddCommGroup E] [Module k E] {s : Finset (Subspace k E)}
(hs : ⊤ ∉ s) : ⋃ p ∈ s, (p : Set E) ≠ Set.univ := by sorry
end FamousTheoremsSource
Listed in Mathlib's curated theorem manifests (docs/1000.yaml); formalized in Mathlib as `Subspace.biUnion_ne_univ_of_top_notMem`. Proof here reduces to that Mathlib result.