Lower bound for the number of cosets in a union of cosets
ProvedSubnormalCover.card_cosets_biUnion_geLet be a group and a subgroup of finite index , and let be subgroups of that each contain and are subnormal in . Then for any ,
equivalently — this is eq. (3.1) of Sun's Theorem 3.1, restricted to the subnormal branch (the alternative branch, where a composition series from to has prime-order quotients, is the pyramidal case handled elsewhere). Specialised to it recovers C. A. Rogers' theorem on residue classes, eq. (3.6).
Formalization note. is not assumed normal, so is used purely as a type of left cosets and the left-hand side is rendered as Set.ncard of the image of under QuotientGroup.mk (this counts the -cosets that meet the union, which coincides with those contained in it exactly because for all ); Nat.card (G ⧸ H) = H.index holds by rfl. Finiteness of is spelled H.index ≠ 0 rather than as the Subgroup.FiniteIndex instance, and the right-hand side is the cardinality of (Finset.range H.index).filter (fun n => ∃ i, (K i).index ∣ n).
import Mathlib open Pointwise
namespace SubnormalCover
theorem card_cosets_biUnion_ge {G : Type*} [Group G] (H : Subgroup G) (hfin : H.index ≠ 0)
{k : ℕ} (K : Fin k → Subgroup G) (hHK : ∀ i, H ≤ K i)
(hsn : ∀ i, (K i).IsSubnormal) (a : Fin k → G) :
((Finset.range H.index).filter fun n => ∃ i, (K i).index ∣ n).card
≤ ((QuotientGroup.mk '' (⋃ i, (a i) • (K i : Set G)) : Set (G ⧸ H))).ncard := by
sorry
end SubnormalCoverRead-back
What the Lean code literally says, in plain math · claude-opus-5
Setting. Let be a type in an arbitrary universe equipped with a group structure (no finiteness, commutativity or other assumption on ). Fix:
- an explicit subgroup , together with the hypothesis . Here is Mathlib's
Subgroup.index, defined as , whose junk value is exactly when the set of left cosets of is infinite; so the hypothesis says precisely that has finite index, and in particular . It does not say is finite. - an implicit natural number (allowed to be );
- an explicit family , written ;
- the hypothesis that for every index ;
- the hypothesis that every is subnormal in in Mathlib's sense (
Subgroup.IsSubnormal): there exist and a monotone chain of subgroups with , , and normal in for all ; - an explicit family of group elements.
Assertion. Let be the canonical map onto the set of left cosets (Mathlib's G ⧸ H is the quotient by the relation ). Then
On the left, the counting runs over the initial segment of of length , which includes ; the condition is an existential over indices , and is again the index in the above sense. On the right, is the pointwise left translate of the underlying set of , the union is taken over all inside , its forward image under is a subset of , and is Set.ncard, the natural-number cardinality that returns for infinite sets (here the ambient set is finite by the index hypothesis, so this junk value cannot be triggered). The inequality is non-strict, in the direction count of multiples number of cosets covered.
Degenerate cases silently allowed.
- : the family is empty, so the existential is false and the left-hand count is ; the union is empty and its image has cardinality . The statement reads and the hypotheses on , are vacuous.
- : since every natural number divides , the element always satisfies the filter condition, and lies in because ; hence the left side is at least .
- The hypotheses and together force each to be finite and nonzero, so the divisibility is never the degenerate case; nothing in the statement rules out , (index , in which case every qualifies and the left side equals ), or repetitions among the .
- Nothing requires the cosets to be disjoint, to cover , or to be related to beyond ; the elements are entirely unconstrained. The subnormality hypothesis appears only as an assumption and plays no role in the expressions being compared.
Confirmed by the mission captain (proposal self-audit).