Alphabeta Math
TheoremStatement: AI-adaptedProof: AI-adaptedSession-authored (Fable 5 assisted)precheck passverified 2026-08-09 (gpt-5.6-terra-codex-subscription) rests on unproved material
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][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\mathbb{N}-indexed chain). Let (X,T)(X, \mathcal{T}) be a topological space.

  1. If XX is normal (Normal spaces and T4T_4 spaces, with the source disagreement over whether normality includes T1T_1 stated explicitly) and A,BXA, B \subseteq X are disjoint closed sets, there is a continuous f:X[0,1]f : X \to [0,1] (Continuity of a map of topological spaces at a point and globally, Intervals of R\mathbb{R}: the nine order-convex forms, nondegeneracy, and length) with Af1({0})A \subseteq f^{-1}(\{0\}) and Bf1({1})B \subseteq f^{-1}(\{1\}).
  2. Conversely, if every pair of disjoint closed subsets of XX admits a continuous function into [0,1][0,1] separating them in the sense of clause 1, then XX is normal. This direction uses no choice principle.

Where the choice principle of clause 1 is spent, and why not less. The construction below builds, for each nNn \in \mathbb{N}, an assignment of an open set to every dyadic rational of level nn, extending the level-(n1)(n-1) assignment; at each single level the finitely many new open sets are chosen at once by Every natural-number-indexed list of nonempty sets has a choice function on its family of values, a theorem of ZF, but stringing together infinitely many such levels, each depending on the one before, is exactly the situation dependent choice is for. The published Urysohn's lemma is not a theorem of ZF, nor of ZF plus countable choice records, with its sources, that ZF\mathrm{ZF} and even ZF\mathrm{ZF} together with the Axiom of Countable Choice do not suffice, and that dependent choice does; nothing here claims dependent choice is necessary for clause 1, only that the construction given is carried out in ZF+DC\mathrm{ZF} + \mathrm{DC}.

Facts & Assumptions

Given: A topological space (X,T)(X,\mathcal{T}) and dependent choice.

[A1]

DC\mathrm{DC}: for every nonempty set PP, every relation RP×PR \subseteq P \times P entire on PP (every pPp \in P has some qPq \in P with pRqp \mathbin{R} q), and every aPa \in P, there is a sequence (pk)kN(p_k)_{k \in \mathbb{N}} with p0=ap_0 = a and pkRpk+1p_k \mathbin{R} p_{k+1} for every kk (The axiom of dependent choice: a relation in which every element is related to something admits an N\mathbb{N}-indexed chain).

[L1]

Shrinking: if XX is normal, CXC \subseteq X is closed and OXO \subseteq X is open with COC \subseteq O, then there is open WW with CWWOC \subseteq W \subseteq \overline{W} \subseteq O (A space is normal if and only if every closed AA inside an open UU admits an open VV with AVVUA \subseteq V \subseteq \overline{V} \subseteq U).

[L2]

Finite choice: a function FF with domain a natural number nn, all of whose values are nonempty sets, admits a choice function for the family F[n]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=nDnD = \bigcup_{n} D_n of [0,1][0,1] are an increasing union of finite levels; for nNn \in \mathbb{N}, Dn+1=Dn{tj:0j<2n}D_{n+1} = D_n \cup \{\, t_j : 0 \le j < 2^n \,\}, where tjt_j is strictly between the DnD_n-consecutive pair rj:=j/2nr_j := j/2^n and sj:=(j+1)/2ns_j := (j+1)/2^n, the 2n2^n points tjt_j are pairwise distinct and disjoint from DnD_n, and every two elements of DD lie together in some common DnD_n (The dyadic rationals of [0,1][0,1], their finite levels DnD_n, and their density in [0,1][0,1]).

[L4]

Chaining: if V0,,VkV_0, \dots, V_k (k0k \ge 0) are subsets of XX with ViVi+1\overline{V_i} \subseteq V_{i+1} for every i<ki < k, then V0Vk\overline{V_0} \subseteq V_k, since ViViVi+1V_i \subseteq \overline{V_i} \subseteq V_{i+1} for each ii (Interior, closure, boundary, exterior, derived set and isolated point in a topological space) makes V0V1V2Vk\overline{V_0} \subseteq V_1 \subseteq V_2 \subseteq \cdots \subseteq V_k a chain of inclusions.

[L5]

The generic construction: if (Ur)rD(U_r)_{r \in D} is a family of open subsets of XX with UrUs\overline{U_r} \subseteq U_s whenever r<sr<s in DD and U1=XU_1 = X, then g(x):=inf({rD:xUr}{1})g(x) := \inf(\{r \in D : x \in U_r\} \cup \{1\}) is a continuous map X[0,1]X \to [0,1] (If (Ur)rD(U_r)_{r \in D} are open with UrUs\overline{U_r} \subseteq U_s whenever r<sr < s and U1=XU_1 = X, then xinf{rD:xUr}x \mapsto \inf\{ r \in D : x \in U_r \} is a continuous map X[0,1]X \to [0,1], and no choice principle is used).

[L6]

The order rays (,12)(-\infty,\tfrac12) and (12,)(\tfrac12,\infty) are open in the usual topology of R\mathbb R (The absolute value makes R\mathbb{R} a metric space: d(x,y)=xyd(x,y) = |x-y| is a metric, its open balls are the intervals (xr,x+r)(x-r, x+r), and it is unbounded, clause 3), so their traces [0,12)[0,\tfrac12) and (12,1](\tfrac12,1] are open in the subspace topology of [0,1][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\mathbb{R}: the nine order-convex forms, nondegeneracy, and length). They are disjoint and contain 00 and 11, respectively.

[L8]

AAA \subseteq \overline{A} for every AXA \subseteq X (Interior, closure, boundary, exterior, derived set and isolated point in a topological space).

Proof

technique · constructive
1.1

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

assume-hyp
1.2

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

assume-hyp
2.1

Under step 1.1: AXBA \subseteq X \setminus B, since AB=A \cap B = \varnothing, and XBX \setminus B is open since BB is closed; by [L1] applied to the closed set AA and the open set XBX \setminus B, fix open Φ0(0)\Phi_0(0) with AΦ0(0)Φ0(0)XBA \subseteq \Phi_0(0) \subseteq \overline{\Phi_0(0)} \subseteq X \setminus B, and put Φ0(1):=XB\Phi_0(1) := X \setminus B, defining Φ0:D0T\Phi_0 : D_0 \to \mathcal{T} on D0={0,1}D_0 = \{0,1\}.

step 1.1L1chooseconstruct
2.2

Under step 1.2: let C,EXC, E \subseteq X be disjoint closed sets; fix a continuous h:X[0,1]h : X \to [0,1] with Ch1({0})C \subseteq h^{-1}(\{0\}) and Eh1({1})E \subseteq h^{-1}(\{1\}).

step 1.2choose
3.1

Under step 1.1: AΦ0(0)A \subseteq \Phi_0(0); Φ0(0)Φ0(1)\overline{\Phi_0(0)} \subseteq \Phi_0(1); and Φ0(1)=XB\Phi_0(1) = X \setminus B.

step 2.1
3.2

Under step 1.2, continuing: by [L6], [0,12)[0,\tfrac12) and (12,1](\tfrac12,1] are open in [0,1][0,1], disjoint, with 0[0,12)0 \in [0,\tfrac12) and 1(12,1]1 \in (\tfrac12,1]; put O1:=h1([0,12))O_1 := h^{-1}(\,[0,\tfrac12)\,) and O2:=h1((12,1])O_2 := h^{-1}(\,(\tfrac12,1]\,), open in XX by [L7].

step 2.2L6L7
4.1

Under step 1.1: for nNn \in \mathbb{N}, call Φ:DnT\Phi : D_n \to \mathcal{T} admissible at level nn when (i) Φ(r)Φ(s)\overline{\Phi(r)} \subseteq \Phi(s) for every r<sr < s in DnD_n; (ii) AΦ(0)A \subseteq \Phi(0); (iii) Φ(1)=XB\Phi(1) = X \setminus B. Put P:={(n,Φ):nN, Φ admissible at level n}P := \{\, (n,\Phi) : n \in \mathbb{N},\ \Phi \text{ admissible at level } n \,\}, and for (n,Φ),(n,Φ)P(n,\Phi), (n',\Phi') \in P say (n,Φ)R(n,Φ)(n,\Phi) \mathbin{R} (n',\Phi') when n=n+1n' = n+1 and ΦDn=Φ\Phi'|_{D_n} = \Phi. By step 3.1, (0,Φ0)P(0,\Phi_0) \in P.

step 3.1construct
4.2

Under step 1.2: CO1C \subseteq O_1, since h0[0,12)h \equiv 0 \in [0,\tfrac12) on CC; EO2E \subseteq O_2, since h1(12,1]h \equiv 1 \in (\tfrac12,1] on EE; and O1O2=h1([0,12)(12,1])=h1()=O_1 \cap O_2 = h^{-1}\big(\,[0,\tfrac12) \cap (\tfrac12,1]\,\big) = h^{-1}(\varnothing) = \varnothing.

step 2.2step 3.2L6
5.1

Under step 1.1: let (n,Φ)P(n,\Phi) \in P. For each jj with 0j<2n0 \le j < 2^n, with rj,sj,tjr_j, s_j, t_j as in [L3]: since rj<sjr_j < s_j in DnD_n, admissibility (i) gives Φ(rj)Φ(sj)\overline{\Phi(r_j)} \subseteq \Phi(s_j), so by [L1] the set of open WW with Φ(rj)WWΦ(sj)\overline{\Phi(r_j)} \subseteq W \subseteq \overline{W} \subseteq \Phi(s_j) is nonempty.

step 4.1L1L3
5.2

Under step 1.2: since C,EC, E were an arbitrary disjoint closed pair, step 4.2 exhibits disjoint open supersets for every such pair, so XX is normal by Normal spaces and T4T_4 spaces, with the source disagreement over whether normality includes T1T_1 stated explicitly; this is clause 2, and no step of it used [A1].

step 4.2
6.1

Under step 1.1, continuing under step 5.1: by [L2] applied to the function assigning, to each j<2nj < 2^n, the nonempty set of open WW with Φ(rj)WWΦ(sj)\overline{\Phi(r_j)} \subseteq W \subseteq \overline{W} \subseteq \Phi(s_j), fix a simultaneous choice, giving open WjW_j with Φ(rj)WjWjΦ(sj)\overline{\Phi(r_j)} \subseteq W_j \subseteq \overline{W_j} \subseteq \Phi(s_j) for every 0j<2n0 \le j < 2^n.

step 5.1L2choose
7.1

Under step 1.1: define Φ:Dn+1T\Phi' : D_{n+1} \to \mathcal{T} by ΦDn:=Φ\Phi'|_{D_n} := \Phi and Φ(tj):=Wj\Phi'(t_j) := W_j for 0j<2n0 \le j < 2^n; this is well defined since Dn+1=Dn{tj:0j<2n}D_{n+1} = D_n \cup \{t_j : 0 \le j < 2^n\} with the tjt_j pairwise distinct and disjoint from DnD_n by [L3]. Then (n,Φ)R(n+1,Φ)(n,\Phi) \mathbin{R} (n+1,\Phi').

step 6.1L3construct
8.1

Under step 1.1, with Φ,Φ\Phi, \Phi' as in step 7.1: for the Dn+1D_{n+1}-consecutive pair (rj,tj)(r_j, t_j): Φ(rj)=Φ(rj)Wj=Φ(tj)\overline{\Phi'(r_j)} = \overline{\Phi(r_j)} \subseteq W_j = \Phi'(t_j) by step 6.1; for the pair (tj,sj)(t_j, s_j): Φ(tj)=WjΦ(sj)=Φ(sj)\overline{\Phi'(t_j)} = \overline{W_j} \subseteq \Phi(s_j) = \Phi'(s_j) by step 6.1.

step 7.1step 6.1
9.1

Under step 1.1: for x<yx < y in Dn+1D_{n+1}, the finitely many elements of Dn+1[x,y]D_{n+1} \cap [x,y], listed increasingly as x=u0<u1<<uk=yx = u_0 < u_1 < \cdots < u_k = y, are Dn+1D_{n+1}-consecutive at each step ui<ui+1u_i < u_{i+1}, and each such pair is one of the pairs of step 8.1 (every Dn+1D_{n+1}-consecutive pair has at least one member among the new points tjt_j, since a new point was inserted into every DnD_n-consecutive gap); so Φ(ui)Φ(ui+1)\overline{\Phi'(u_i)} \subseteq \Phi'(u_{i+1}) at each step, and [L4] gives Φ(x)=Φ(u0)Φ(uk)=Φ(y)\overline{\Phi'(x)} = \overline{\Phi'(u_0)} \subseteq \Phi'(u_k) = \Phi'(y).

step 8.1L3L4
10.1

Under step 1.1: AΦ(0)=Φ(0)A \subseteq \Phi'(0) = \Phi(0), since 0Dn0 \in D_n is unaffected by the extension; Φ(1)=Φ(1)=XB\Phi'(1) = \Phi(1) = X \setminus B, since 1Dn1 \in D_n is likewise unaffected; with step 9.1 this is admissibility of Φ\Phi' at level n+1n+1, so (n+1,Φ)P(n+1,\Phi') \in P.

step 7.1step 9.1L3
11.1

Under step 1.1: by steps 5.1, 6.1, 7.1 and 10.1, every (n,Φ)P(n,\Phi) \in P has some (n+1,Φ)P(n+1,\Phi') \in P with (n,Φ)R(n+1,Φ)(n,\Phi) \mathbin{R} (n+1,\Phi'); so RR is entire on PP.

step 7.1step 10.1
12.1

Under step 1.1: PP is nonempty by step 4.1 and RR is entire on PP by step 11.1; by [A1] applied with a:=(0,Φ0)a := (0,\Phi_0), there is a sequence ((mk,Ψk))kN\big((m_k,\Psi_k)\big)_{k \in \mathbb{N}} with (m0,Ψ0)=(0,Φ0)(m_0,\Psi_0) = (0,\Phi_0) and (mk,Ψk)R(mk+1,Ψk+1)(m_k,\Psi_k) \mathbin{R} (m_{k+1},\Psi_{k+1}) for every kk.

step 4.1step 11.1A1construct
13.1

Under step 1.1: since (n,Φ)R(n,Φ)(n,\Phi) \mathbin{R} (n',\Phi') forces n=n+1n' = n+1, and m0=0m_0 = 0, induction on kk gives mk=km_k = k for every kNk \in \mathbb{N}; so each Ψk:DkT\Psi_k : D_k \to \mathcal{T} is admissible at level kk, and Ψk+1Dk=Ψk\Psi_{k+1}|_{D_k} = \Psi_k for every kk.

step 12.1
14.1

Under step 1.1: for rDr \in D, fix nn with rDnr \in D_n [L3] and define Vr:=Ψn(r)V_r := \Psi_n(r); by step 13.1, for nnn \le n' with rDnr \in D_n, Ψn(r)=Ψn(r)\Psi_{n'}(r) = \Psi_n(r) (chaining ΨnDn=Ψn\Psi_{n'}|_{D_n} = \Psi_n through the intermediate levels), so VrV_r does not depend on the level nn chosen.

step 13.1L3construct
15.1

Under step 1.1: for r<sr < s in DD, fix nn with r,sDnr, s \in D_n [L3]; then Vr=Ψn(r)Ψn(s)=Vs\overline{V_r} = \overline{\Psi_n(r)} \subseteq \Psi_n(s) = V_s by admissibility (i) of Ψn\Psi_n. Also AV0A \subseteq V_0 and V1=XBV_1 = X \setminus B, by admissibility (ii) and (iii) of Ψn\Psi_n for any nn.

step 14.1step 13.1L3
16.1

Under step 1.1: define Ur:=VrU_r := V_r for rDr \in D with r<1r < 1, and U1:=XU_1 := X. For r<sr < s in DD: if s<1s < 1, Ur=VrVs=Us\overline{U_r} = \overline{V_r} \subseteq V_s = U_s by step 15.1; if s=1s = 1, Ur=VrV1=XBX=U1\overline{U_r} = \overline{V_r} \subseteq V_1 = X \setminus B \subseteq X = U_1 by step 15.1. So UrUs\overline{U_r} \subseteq U_s whenever r<sr < s in DD, and U1=XU_1 = X.

step 15.1construct
17.1

Under step 1.1: by [L5] applied to (Ur)rD(U_r)_{r \in D} of step 16.1, f(x):=inf({rD:xUr}{1})f(x) := \inf(\{r \in D : x \in U_r\} \cup \{1\}) is a continuous map X[0,1]X \to [0,1].

step 16.1L5
17.2

Under step 1.1: for bBb \in B and rDr \in D with r<1r < 1: fix nn with rDnr \in D_n [L3]; since 1Dn1 \in D_n also, admissibility (i) of Ψn\Psi_n applied to r<1r < 1 gives Ψn(r)Ψn(1)=XB\overline{\Psi_n(r)} \subseteq \Psi_n(1) = X \setminus B, that is VrXB\overline{V_r} \subseteq X \setminus B; since VrVrV_r \subseteq \overline{V_r} by [L8] and Ur=VrU_r = V_r by step 16.1, UrB=U_r \cap B = \varnothing, so bUrb \notin U_r.

step 14.1step 13.1step 16.1L3L8
18.1

Under step 1.1: for aAa \in A: aV0a \in V_0 by step 15.1, and U0=V0U_0 = V_0 by step 16.1 (as 0<10 < 1), so aU0a \in U_0 and 0{rD:aUr}0 \in \{r \in D : a \in U_r\}; hence f(a)0f(a) \le 0, and f(a)0f(a) \ge 0 since ff maps into [0,1][0,1] by step 17.1, so f(a)=0f(a) = 0.

step 17.1step 16.1step 15.1
18.2

Under step 1.1: for bBb \in B: by step 17.2, bUrb \notin U_r for every rDr \in D with r<1r < 1, and bU1=Xb \in U_1 = X by step 16.1; so {rD:bUr}{1}={1}\{r \in D : b \in U_r\} \cup \{1\} = \{1\}, giving f(b)=inf{1}=1f(b) = \inf\{1\} = 1.

step 17.2step 16.1
19.1

Steps 17.1, 18.1 and 18.2 show that, under the hypothesis of step 1.1, ff is a continuous map X[0,1]X \to [0,1] with Af1({0})A \subseteq f^{-1}(\{0\}) and Bf1({1})B \subseteq f^{-1}(\{1\}), which is clause 1.

step 17.1step 18.1step 18.2
20.1

Steps 19.1 and 5.2 establish clauses 1 and 2 respectively.

step 19.1step 5.2discharge-construct

Remarks

Depends on

Used by

Dependency tree · next 3 levels

Direct dependencies and their dependencies through the next three levels: 132 results over 28 levels. An arrow runs from a result to what uses it, and this result sits at the bottom with a heavier outline. Click the chart to enlarge it.

Sources