Alphabeta Math
TheoremStatement: AI-adaptedProof: AI-adapted
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

Urysohn's lemma, under the axiom of dependent choice: in a normal space two disjoint closed sets are separated by a continuous function into [0,1], and conversely such a space is normal

Statement

Assume the Axiom of Dependent Choice (The axiom of dependent choice: a relation in which every element is related to something admits an N-indexed chain). Let (X,T) be a topological space.

  1. If X is normal (Normal spaces and T4 spaces, with the source disagreement over whether normality includes T1 stated explicitly) and A,B⊆X are disjoint closed sets, there is a continuous f:X→[0,1] (Continuity of a map of topological spaces at a point and globally, Intervals of R: the nine order-convex forms, nondegeneracy, and length) with A⊆f−1({0}) and B⊆f−1({1}).
  2. Conversely, if every pair of disjoint closed subsets of X admits a continuous function into [0,1] separating them in the sense of clause 1, then X is normal. This direction uses no choice principle.

Where the choice principle of clause 1 is spent. At each dyadic level finitely many new open sets are selected by Every natural-number-indexed list of nonempty sets has a choice function on its family of values, a theorem of ZF. Dependent choice assembles a sequence of admissible levels, each extending its predecessor. This proves sufficiency of DC for this construction; it asserts no necessity or independence result.

Facts & Assumptions

Given: A topological space (X,T) and dependent choice.

[A1]

DC: for every nonempty set P, every relation R⊆P×P entire on P (every p∈P has some q∈P with pRq), and every a∈P, there is a sequence (pk)k∈N with p0=a and pkRpk+1 for every k (The axiom of dependent choice: a relation in which every element is related to something admits an N-indexed chain).

[L1]

Shrinking: if X is normal, C⊆X is closed and O⊆X is open with C⊆O, then there is open W with C⊆W⊆W‾⊆O (A space is normal if and only if every closed A inside an open U admits an open V with A⊆V⊆V‾⊆U).

[L2]

Finite choice: a function F with domain a natural number n, all of whose values are nonempty sets, admits a choice function for the family F[n] of its values (Every natural-number-indexed list of nonempty sets has a choice function on its family of values, Choice function), a theorem of ZF.

[L3]

The dyadic rationals D=⋃nDn of [0,1] are an increasing union of finite levels; for n∈N, Dn+1=Dn∪{ tj:0≤j<2n }, where tj is strictly between the Dn-consecutive pair rj:=j/2n and sj:=(j+1)/2n, the 2n points tj are pairwise distinct and disjoint from Dn, and every two elements of D lie together in some common Dn (The dyadic rationals of [0,1], their finite levels Dn, and their density in [0,1]).

[L4]

Chaining: if V0,…,Vk (k≥0) are subsets of X with Vi‾⊆Vi+1 for every i<k, then V0‾⊆Vk, since Vi⊆Vi‾⊆Vi+1 for each i (Interior, closure, boundary, exterior, derived set and isolated point in a topological space) makes V0‾⊆V1⊆V2⊆⋯⊆Vk a chain of inclusions.

[L5]

The generic construction: if (Ur)r∈D is a family of open subsets of X with Ur‾⊆Us whenever r<s in D and U1=X, then g(x):=inf⁡({r∈D:x∈Ur}∪{1}) is a continuous map X→[0,1] (If (Ur)r∈D are open with Ur‾⊆Us whenever r<s and U1=X, then x↦inf⁡{r∈D:x∈Ur} is a continuous map X→[0,1], and no choice principle is used).

[L6]

The order rays (−∞,12) and (12,∞) are open in the usual topology of R (The absolute value makes R a metric space: d(x,y)=∣x−y∣ is a metric, its open balls are the intervals (x−r,x+r), and it is unbounded, clause 3), so their traces [0,12) and (12,1] are open in the subspace topology of [0,1] (Subspace topology: the traces of the open sets, its closed sets and its bases, the continuity of the inclusion, and the characteristic property of a map into a subspace, Intervals of R: the nine order-convex forms, nondegeneracy, and length). They are disjoint and contain 0 and 1, respectively.

Proof

technique · constructive
1.1assume-hyp

Assume X is normal and A,B⊆X are disjoint closed sets (the hypothesis of clause 1).

1.2assume-hyp

Assume instead that every pair of disjoint closed subsets of X admits a continuous function into [0,1] separating them as in clause 1 (the hypothesis of clause 2).

2.1step 1.1L1chooseconstruct

Under step 1.1: A⊆X∖B, since A∩B=∅, and X∖B is open since B is closed; by [L1] applied to the closed set A and the open set X∖B, fix open Φ0(0) with A⊆Φ0(0)⊆Φ0(0)‾⊆X∖B, and put Φ0(1):=X∖B, defining Φ0:D0→T on D0={0,1}.

2.2step 1.2choose

Under step 1.2: let C,E⊆X be disjoint closed sets; fix a continuous h:X→[0,1] with C⊆h−1({0}) and E⊆h−1({1}).

3.1step 2.1

Under step 1.1: A⊆Φ0(0); Φ0(0)‾⊆Φ0(1); and Φ0(1)=X∖B.

3.2step 2.2L6L7

Under step 1.2, continuing: by [L6], [0,12) and (12,1] are open in [0,1], disjoint, with 0∈[0,12) and 1∈(12,1]; put O1:=h−1( [0,12) ) and O2:=h−1( (12,1] ), open in X by [L7].

4.1step 3.1construct

Under step 1.1: for n∈N, call Φ:Dn→T admissible at level n when (i) Φ(r)‾⊆Φ(s) for every r<s in Dn; (ii) A⊆Φ(0); (iii) Φ(1)=X∖B. Put P:={ (n,Φ):n∈N, Φ admissible at level n }, and for (n,Φ),(n′,Φ′)∈P say (n,Φ)R(n′,Φ′) when n′=n+1 and Φ′∣Dn=Φ. By step 3.1, (0,Φ0)∈P.

4.2step 2.2step 3.2L6

Under step 1.2: C⊆O1, since h≡0∈[0,12) on C; E⊆O2, since h≡1∈(12,1] on E; and O1∩O2=h−1( [0,12)∩(12,1] )=h−1(∅)=∅.

5.1step 4.1L1L3

Under step 1.1: let (n,Φ)∈P. For each j with 0≤j<2n, with rj,sj,tj as in [L3]: since rj<sj in Dn, admissibility (i) gives Φ(rj)‾⊆Φ(sj), so by [L1] the set of open W with Φ(rj)‾⊆W⊆W‾⊆Φ(sj) is nonempty.

5.2step 4.2

Under step 1.2: since C,E were an arbitrary disjoint closed pair, step 4.2 exhibits disjoint open supersets for every such pair, so X is normal by Normal spaces and T4 spaces, with the source disagreement over whether normality includes T1 stated explicitly; this is clause 2, and no step of it used [A1].

6.1step 5.1L2choose

Under step 1.1, continuing under step 5.1: by [L2] applied to the function assigning, to each j<2n, the nonempty set of open W with Φ(rj)‾⊆W⊆W‾⊆Φ(sj), fix a simultaneous choice, giving open Wj with Φ(rj)‾⊆Wj⊆Wj‾⊆Φ(sj) for every 0≤j<2n.

7.1step 6.1L3construct

Under step 1.1: define Φ′:Dn+1→T by Φ′∣Dn:=Φ and Φ′(tj):=Wj for 0≤j<2n; this is well defined since Dn+1=Dn∪{tj:0≤j<2n} with the tj pairwise distinct and disjoint from Dn by [L3]. Then (n,Φ)R(n+1,Φ′).

8.1step 7.1step 6.1

Under step 1.1, with Φ,Φ′ as in step 7.1: for the Dn+1-consecutive pair (rj,tj): Φ′(rj)‾=Φ(rj)‾⊆Wj=Φ′(tj) by step 6.1; for the pair (tj,sj): Φ′(tj)‾=Wj‾⊆Φ(sj)=Φ′(sj) by step 6.1.

9.1step 8.1L3L4

Under step 1.1: for x<y in Dn+1, the finitely many elements of Dn+1∩[x,y], listed increasingly as x=u0<u1<⋯<uk=y, are Dn+1-consecutive at each step ui<ui+1, and each such pair is one of the pairs of step 8.1 (every Dn+1-consecutive pair has at least one member among the new points tj, since a new point was inserted into every Dn-consecutive gap); so Φ′(ui)‾⊆Φ′(ui+1) at each step, and [L4] gives Φ′(x)‾=Φ′(u0)‾⊆Φ′(uk)=Φ′(y).

10.1step 7.1step 9.1L3

Under step 1.1: A⊆Φ′(0)=Φ(0), since 0∈Dn is unaffected by the extension; Φ′(1)=Φ(1)=X∖B, since 1∈Dn is likewise unaffected; with step 9.1 this is admissibility of Φ′ at level n+1, so (n+1,Φ′)∈P.

11.1step 7.1step 10.1

Under step 1.1: by steps 5.1, 6.1, 7.1 and 10.1, every (n,Φ)∈P has some (n+1,Φ′)∈P with (n,Φ)R(n+1,Φ′); so R is entire on P.

12.1step 4.1step 11.1A1construct

Under step 1.1: P is nonempty by step 4.1 and R is entire on P by step 11.1; by [A1] applied with a:=(0,Φ0), there is a sequence ((mk,Ψk))k∈N with (m0,Ψ0)=(0,Φ0) and (mk,Ψk)R(mk+1,Ψk+1) for every k.

13.1step 12.1

Under step 1.1: since (n,Φ)R(n′,Φ′) forces n′=n+1, and m0=0, induction on k gives mk=k for every k∈N; so each Ψk:Dk→T is admissible at level k, and Ψk+1∣Dk=Ψk for every k.

14.1step 13.1L3construct

Under step 1.1: for r∈D, fix n with r∈Dn [L3] and define Vr:=Ψn(r); by step 13.1, for n≤n′ with r∈Dn, Ψn′(r)=Ψn(r) (chaining Ψn′∣Dn=Ψn through the intermediate levels), so Vr does not depend on the level n chosen.

15.1step 14.1step 13.1L3

Under step 1.1: for r<s in D, fix n with r,s∈Dn [L3]; then Vr‾=Ψn(r)‾⊆Ψn(s)=Vs by admissibility (i) of Ψn. Also A⊆V0 and V1=X∖B, by admissibility (ii) and (iii) of Ψn for any n.

16.1step 15.1construct

Under step 1.1: define Ur:=Vr for r∈D with r<1, and U1:=X. For r<s in D: if s<1, Ur‾=Vr‾⊆Vs=Us by step 15.1; if s=1, Ur‾=Vr‾⊆V1=X∖B⊆X=U1 by step 15.1. So Ur‾⊆Us whenever r<s in D, and U1=X.

17.1step 16.1L5

Under step 1.1: by [L5] applied to (Ur)r∈D of step 16.1, f(x):=inf⁡({r∈D:x∈Ur}∪{1}) is a continuous map X→[0,1].

17.2step 14.1step 13.1step 16.1L3L8

Under step 1.1: for b∈B and r∈D with r<1: fix n with r∈Dn [L3]; since 1∈Dn also, admissibility (i) of Ψn applied to r<1 gives Ψn(r)‾⊆Ψn(1)=X∖B, that is Vr‾⊆X∖B; since Vr⊆Vr‾ by [L8] and Ur=Vr by step 16.1, Ur∩B=∅, so b∉Ur.

18.1step 17.1step 16.1step 15.1

Under step 1.1: for a∈A: a∈V0 by step 15.1, and U0=V0 by step 16.1 (as 0<1), so a∈U0 and 0∈{r∈D:a∈Ur}; hence f(a)≤0, and f(a)≥0 since f maps into [0,1] by step 17.1, so f(a)=0.

18.2step 17.2step 16.1

Under step 1.1: for b∈B: by step 17.2, b∉Ur for every r∈D with r<1, and b∈U1=X by step 16.1; so {r∈D:b∈Ur}∪{1}={1}, giving f(b)=inf⁡{1}=1.

19.1step 17.1step 18.1step 18.2

Steps 17.1, 18.1 and 18.2 show that, under the hypothesis of step 1.1, f is a continuous map X→[0,1] with A⊆f−1({0}) and B⊆f−1({1}), which is clause 1.

20.1step 19.1step 5.2discharge-construct∎

Steps 19.1 and 5.2 establish clauses 1 and 2 respectively.

Remarks

Depends on

Dependency tree · 2 levels

Sources