betaFeedback

WorksMathematics

Subset-system derived topologies and coreflective categories of T0T_0-spaces

Abstract

Given a subset system HH on T0T_0-spaces, we associate with each topology a closure-testing derived topology and study the fixed points of this operator. The HH-determined topologies above a given topology are closed under intersections and hence determine a least fixed-point hull. We prove that this hull is obtained by transfinite iteration, is functorial, and defines a right adjoint to the inclusion of HH-determined spaces into Top0\mathbf{Top}_0. Consequently, the corresponding full subcategory is coreflective and complete. We isolate restriction-stability and open-locality hypotheses under which determinedness is inherited by closed and open subspaces, and we establish a general non-reflectivity criterion. If HH is contained in the directed-subset system, the category of HH-determined spaces is Cartesian closed; its exponentials are obtained by coreflecting the pointwise convergence topology on continuous maps. The construction subsumes familiar monotone- and irreducible-determined cases while separating the formal consequences of the subset-system axioms from the additional hypotheses needed for heredity and exponentials. It therefore provides a uniform categorical mechanism for producing complete Cartesian closed categories of T0T_0-spaces with explicit higher-order function spaces, as required in denotational semantics.

Introduction

Domain theory provides order-theoretic models of computation in which directed suprema describe approximation and the Scott topology records observable information; see, for example, [1, 5, 6]. From the viewpoint of denotational semantics, a category of semantic spaces is particularly useful when it supports products and higher-order function spaces; Cartesian closed categories provide exactly this structure through exponentials. A parallel line of work develops the order-theoretic core of domain theory directly on T0T_0-spaces. Monotone determined spaces are a basic example: their topology is detected by directed subsets, and they form a Cartesian closed category extending the Scott-topological setting; see [4, 7, 14]. The question pursued here is whether this categorical infrastructure can be generated uniformly from a functorially chosen class of distinguished subsets, rather than constructed separately for each convergence or approximation notion.

Many constructions in this area describe a topology by specifying which distinguished subsets are allowed to detect open sets. Replacing directed subsets by irreducible subsets leads to irreducible determined spaces. In [13], the corresponding derived topology, its transfinite idempotent hull, hereditary properties and categorical structure were studied for the irreducible subset system. The purpose of the present paper is not to replace the irreducible system there by a formal parameter. Instead, we isolate the fixed-point mechanism behind these constructions and determine what follows for an arbitrary subset system

H:Top⁡0⟶Set⁡.H : \operatorname{Top}_{0} \longrightarrow\operatorname{Set}.

The notion of subset system, inspired by the uniform closure constructions of Wright, Wagner and Thatcher [10], was used in [11] to organize several sobriety-type notions. Chen and Xu [3] also studied topological spaces determined by subset systems through an induced-operator framework. Here the subset system is used in a different way: from a T0T_0-topology τ\tau we form a closure-testing operator dH(τ)d_H(\tau). This operator is monotone and inflationary on the poset of T0T_0-topologies on the fixed underlying set, but its idempotence is not part of the subset-system axioms. The central object is therefore its least fixed point above τ\tau.

A second recent uniform approach is due to Chen, Kou and Lyu [2]. Their construction starts from convergence classes and C-generated spaces; in particular, it recovers monotone determined spaces as a Cartesian closed coreflective category and treats directed, chain and monotone sequential convergence classes in a common framework. The present construction organizes the data differently. It begins with a closure-testing operator on the lattice of T0T_0-topologies: the distinguished family H(X)H(X) depends functorially on the current space, candidate open sets are tested against closures of HH-sets, and the one-step operator may require transfinite iteration before reaching a fixed point. Thus the two approaches overlap in the monotone-determined case, but the fixed-point completion, its ordinal length and the associated heredity questions are intrinsic to the closure-testing operator studied here.

Our main results form three categorical levels. First, the HH-determined topologies above a given T0T_0-topology are closed under intersections. Hence there is a least fixed-point hull HdH(τ)\mathrm{Hd}_{H}(\tau), obtained by transfinite iteration of dHd_{H}; the assignment τ↦HdH(τ)\tau\mapsto\mathrm{Hd}_{H}(\tau) is a functorial closure operator on the poset of T0T_{0}-topologies on a fixed set. Second, this hull defines a coreflector: the full category HD of HH-determined spaces is coreflective in Top0\mathbf{Top}_{0} and therefore complete. We also obtain retract stability, a general non-reflectivity criterion, and restriction and localization hypotheses that ensure closed- and open-heredity. Third, if H≤DH \leq D, then HD is Cartesian closed. The exponential is obtained by applying the coreflector to the pointwise convergence topology on the set of continuous maps. Thus the same fixed-point mechanism produces the completion, the adjunction and explicit higher-order function spaces, which is the categorical structure most directly relevant to denotational semantics.

The paper is organized around these structural points. Section 2 introduces subset-system derived topologies and records the basic order-theoretic relations among them. Section 3 develops the fixed-point completion, its transfinite description and functoriality. Section 4 turns the completion into a coreflector and derives completeness, retraction stability and non-reflectivity criteria. Section 5 treats subspaces under explicit stability assumptions on HH. Section 6 constructs exponentials and establishes Cartesian closedness when H≤DH \leq D. The final section records questions suggested by the general framework.

Subset-system derived topologies

We recall only the notation needed below. For a T0T_{0}-space XX, the specialization order is denoted by ≤X\leq_{X}; thus x≤Xyx \leq_{X} y iff x∈cl⁡X{y}x \in\operatorname{cl}_{X}\{y\}. We write α(X)\alpha(X) for the Alexandroff topology of upper sets of this order. A nonempty subset DD is directed if every finite subset of DD has an upper bound in DD. The family of directed subsets of XX is denoted by D(X)\mathcal{D}(X), and Irr⁡(X)\operatorname{Irr}(X) denotes the family of nonempty irreducible subsets of XX.

Definition 2.1 A covariant functor H:Top0→SetH : \mathbf{Top}_{0} \to\mathbf{Set} is a subset system if, for every T0T_{0}-space XX,

S(X)⊆H(X)⊆2XS(X) \subseteq H(X) \subseteq2^{X}

where S(X)={{x}:x∈X}S(X) = \{\{x\} : x \in X\}, and for every continuous map f:X→Yf : X \to Y one has

H(f)(A)=f(A)∈H(Y)(A∈H(X)).H(f)(A) = f(A) \in H(Y) \quad(A \in H(X)).

If H1(X)⊆H2(X)H_{1}(X) \subseteq H_{2}(X) for every XX, we write H1≤H2H_{1} \leq H_{2}.

Standard examples are the systems S,F,C,D,RD,WD,R\mathcal{S},\mathcal{F},\mathcal{C},\mathcal{D},\mathcal{RD},\mathcal{WD},\mathcal{R} and P\mathcal{P} of singletons, nonempty finite subsets, nonempty chains, directed subsets, Rudin subsets, well-filtered determined subsets, irreducible subsets and nonempty subsets, respectively. The familiar inclusions include

S≤C≤D≤RD≤WD≤R≤P,S≤F≤P;\mathcal{S} \leq\mathcal{C} \leq\mathcal{D} \leq\mathcal{RD} \leq\mathcal{WD} \leq\mathcal{R} \leq\mathcal{P}, \quad\mathcal{S} \leq\mathcal{F} \leq\mathcal{P};

see [12].

In what follows, the capital letter HH always denotes a subset system H:Top0→SetH : \mathbf{Top}_{0} \to\mathbf{Set}.

Definition 2.2. Let (X,τ)(X,\tau) be a T0T_0-space. A subset U⊆XU \subseteq X is HH-open relative to τ\tau if

U∩cl⁡τA≠∅⟹U∩A≠∅(A∈H((X,τ))).U \cap\operatorname{cl}_{\tau} A \ne\varnothing\mathrel{\Longrightarrow} U \cap A \ne\varnothing\quad(A \in H((X,\tau))).

The family of all such sets is denoted by τH\tau_H. The topology generated by τH\tau_H as a subbasis is denoted by dH(τ)d_H(\tau) and is called the HH-derived topology. The space (X,τ)(X,\tau) is HH-determined if

τ=dH(τ).\tau= d_H(\tau).

The full subcategory of Top⁡0\operatorname{Top}_0 formed by the HH-determined spaces is denoted by HD\mathbf{HD}.

Since every τ\tau-open set satisfies the defining condition, τ⊆τH\tau\subseteq\tau_H. Moreover every member of τH\tau_H is an upper set for the specialization order because S≤H\mathcal{S} \leq H.

Proposition 2.3. Let τ\tau be a T0T_0-topology on XX.

  1. τ⊆τH⊆dH(τ)⊆α(X)\tau\subseteq\tau_H \subseteq d_H(\tau) \subseteq\alpha(X). In particular, τ\tau and dH(τ)d_H(\tau) have the same specialization order.

  2. If H1≤H2H_1 \leq H_2, then τH2⊆τH1\tau_{H_2} \subseteq\tau_{H_1} and dH2(τ)⊆dH1(τ)d_{H_2}(\tau) \subseteq d_{H_1}(\tau).

  3. If μ⊆ν\mu\subseteq\nu are T0T_0-topologies on the same set, then dH(μ)⊆dH(ν)d_H(\mu) \subseteq d_H(\nu).

  4. dS(τ)=dF(τ)=α(X)d_{\mathcal{S}}(\tau) = d_{\mathcal{F}}(\tau) = \alpha(X) and dP(τ)=τd_{\mathcal{P}}(\tau) = \tau.

  5. The family τC\tau_{\mathcal{C}} is a topology; hence dC(τ)=τCd_{\mathcal{C}}(\tau) = \tau_{\mathcal{C}}.

  6. If H≤DH \leq D, then τH\tau_H is a topology and therefore dH(τ)=τHd_H(\tau) = \tau_H.

Proof. For (1), only the last inclusion needs comment. If x∈U∈τHx \in U \in\tau_H and x≤Xyx \leq_X y, then x∈U∩cl⁡τ{y}x \in U \cap\operatorname{cl}_{\tau}\{y\}; since {y}∈H(X)\{y\} \in H(X), we obtain y∈Uy \in U. Thus UU is upper. Since both τ\tau and α(X)\alpha(X) induce the same specialization order, every topology between them does as well.

For (2), increasing the family of test sets strengthens the defining condition. For (3), the identity (X,ν)→(X,μ)(X,\nu) \to(X,\mu) is continuous, so H((X,ν))⊆H((X,μ))H((X,\nu)) \subseteq H((X,\mu)), while cl⁡νA⊆cl⁡μA\operatorname{cl}_{\nu} A \subseteq\operatorname{cl}_{\mu} A.

For (4), the closure of a finite set is the finite union of the closures of its points; hence the singleton and finite tests both yield exactly the upper sets. The assertion for P\mathcal{P} is the usual closure characterization of open sets.

For (5), arbitrary unions cause no difficulty. If U,V∈τCU,V \in\tau_{\mathcal{C}} and a chain CC has closure meeting U∩VU \cap V, choose c1∈C∩Uc_1 \in C \cap U and c2∈C∩Vc_2 \in C \cap V. The two points are comparable, and U,VU,V are upper; hence one of c1,c2c_1,c_2 belongs to U∩VU \cap V.

The same argument proves (6): if U,V∈τHU,V \in\tau_H and A∈H(X)⊆D(X)A \in H(X) \subseteq D(X) has closure meeting U∩VU \cap V, choose a1∈A∩Ua_1 \in A \cap U and a2∈A∩Va_2 \in A \cap V. A common upper bound a∈Aa \in A belongs to U∩VU \cap V because both sets are upper. □\square

Corollary 2.4. If H≤H∗H \leq H^{*}, every HH-determined space is H∗H^{*}-determined. Thus

HD⊆H∗D\mathbf{HD} \subseteq\mathbf{H^{*}D}

as full subcategories of Top⁡0\operatorname{Top}_0.

Proof. If τ\tau is HH-determined, then

τ⊆τH⊆dH(τ)=τ,(1)\tau\subseteq\tau_H \subseteq d_H(\tau) = \tau, \tag*{(1)}

so τH=τ\tau_H=\tau. Since H≤H∗H\le H^*, Proposition 2.3(2) gives

τ⊆τH∗⊆τH=τ.\tau\subseteq\tau_{H^*}\subseteq\tau_H=\tau.

Hence τH∗=τ\tau_{H^*}=\tau, and therefore dH∗(τ)=τd_{H^*}(\tau)=\tau.

The fixed-point condition also has a useful closed-set formulation.

Proposition 2.5 A T0T_0-space (X,τ)(X,\tau) is HH-determined if and only if every subset C⊆XC\subseteq X satisfying

A∈H(X),A⊆C⟹cl⁡τA⊆CA\in H(X),\quad A\subseteq C\quad\Longrightarrow\quad\operatorname{cl}_{\tau}A\subseteq C

is closed.

Proof. Take complements in Definition 2.2.

The general construction contains several familiar cases. For H=DH=\mathcal{D} one recovers the monotone determined topology of [4, 7, 14]; for H=RH=\mathcal{R} one recovers the irreducible derived topology of [13]. At the two extremes, S\mathcal{S}- and F\mathcal{F}-determined spaces are precisely the Alexandroff T0T_0-spaces, while every T0T_0-space is P\mathcal{P}-determined. Notice already that categorical statements cannot be uniform in HH without additional hypotheses.

Fixed-point completion and transfinite iteration

The operator dHd_H is monotone and inflationary by Proposition 2.3, but it need not be idempotent for an arbitrary subset system. We therefore pass to its least fixed point above a given topology.

Lemma 3.1 Let {μi:i∈I}\{\mu_i:i\in I\} be a family of HH-determined topologies on a set XX, all containing a fixed T0T_0-topology τ\tau. Then μ=⋂i∈Iμi\mu=\bigcap_{i\in I}\mu_i is again an HH-determined T0T_0-topology containing τ\tau.

Proof. The intersection is T0T_0 because it contains the T0T_0-topology τ\tau. By monotonicity,

μ⊆dH(μ)⊆dH(μi)=μi\mu\subseteq d_H(\mu)\subseteq d_H(\mu_i)=\mu_i

for every ii. Hence dH(μ)⊆μd_H(\mu)\subseteq\mu, and equality follows.

Definition 3.2 For a T0T_0-topology τ\tau on XX, define

Hd⁡H(τ)=⋂{μ:τ⊆μ and μ is H-determined}.\operatorname{Hd}_H(\tau)=\bigcap\{\mu:\tau\subseteq\mu\text{ and }\mu\text{ is }H\text{-determined}\}.

This family is nonempty because α(X)\alpha(X) is HH-determined and τ⊆α(X)\tau\subseteq\alpha(X) (note that XX is equipped with the specialization order).

Theorem 3.3 For every T0T_0-space (X,τ)(X,\tau), we have the following conclusions:

  1. Hd⁡H(τ)\operatorname{Hd}_H(\tau) is the least HH-determined topology containing τ\tau;

(2) τ⊆dH(τ)⊆Hd⁡H(τ)⊆α(X)\tau\subseteq d_H(\tau) \subseteq\operatorname{Hd}_H(\tau) \subseteq\alpha(X), where XX is equipped with the order ≤τ\leq_\tau;

(3) τ\tau and Hd⁡H(τ)\operatorname{Hd}_H(\tau) have the same specialization order;

(4) the assignment τ↦Hd⁡H(τ)\tau\mapsto\operatorname{Hd}_H(\tau) is a closure operator on the poset of T0T_0-topologies on XX: it is inflationary, monotone and idempotent.

Proof. Parts (1) and (2) follow from Lemma 3.1. Part (3) follows from the two inclusions τ⊆Hd⁡H(τ)⊆α(X)\tau\subseteq\operatorname{Hd}_H(\tau) \subseteq\alpha(X). Monotonicity in (4) follows because the class of HH-determined topologies above a larger initial topology is smaller; idempotence follows from (1).

We next describe the hull by iteration. Put τ0=τ\tau_0=\tau and define, for ordinals α\alpha,

τα+1=dH(τα),τλ=⋁β<λτβ(λ a limit ordinal),\tau_{\alpha+1}=d_H(\tau_\alpha), \qquad\tau_\lambda=\bigvee_{\beta<\lambda}\tau_\beta\quad(\lambda\text{ a limit ordinal}),

where the join is the topology generated by ⋃β<λτβ\bigcup_{\beta<\lambda}\tau_\beta.

Theorem 3.4 There is a least ordinal γH(τ)\gamma_H(\tau) such that

τγH(τ)+1=τγH(τ).\tau_{\gamma_H(\tau)+1}=\tau_{\gamma_H(\tau)}.

For this ordinal,

Hd⁡H(τ)=τγH(τ).\operatorname{Hd}_H(\tau)=\tau_{\gamma_H(\tau)}.

Proof. By transfinite induction,

τ⊆τα⊆Hd⁡H(τ)⊆α(X)\tau\subseteq\tau_\alpha\subseteq\operatorname{Hd}_H(\tau) \subseteq\alpha(X)

for every α\alpha, and the chain (τα)(\tau_\alpha) is increasing. Put κ=∣P(X)∣\kappa=|\mathcal{P}(X)|. If the iteration did not stabilize before κ+\kappa^+, then for every α<κ+\alpha<\kappa^+ we could choose

Uα∈τα+1∖τα.U_\alpha\in\tau_{\alpha+1}\setminus\tau_\alpha.

Whenever α<β\alpha<\beta, monotonicity gives Uα∈τβU_\alpha\in\tau_\beta, whereas Uβ∉τβU_\beta\notin\tau_\beta; hence the sets UαU_\alpha are pairwise distinct. This would give an injection κ+→P(X)\kappa^+\to\mathcal{P}(X), a contradiction. Thus the iteration stabilizes.

Let γ\gamma be the first stationary stage. Then τγ\tau_\gamma is HH-determined and contains τ\tau, so Hd⁡H(τ)⊆τγ\operatorname{Hd}_H(\tau)\subseteq\tau_\gamma by minimality. The reverse inclusion follows from the displayed chain.

Remark 3.5 The distinction between dH(τ)d_H(\tau) and Hd⁡H(τ)\operatorname{Hd}_H(\tau) is essential in the general theory. For several standard systems the first step is already idempotent (for example S\mathcal{S}, F\mathcal{F}, C\mathcal{C} and D\mathcal{D}), while for H=RH=\mathcal{R} the transfinite construction gives the fixed-point hull considered in [13]. A useful general problem is to characterize those subset systems for which dHd_H is idempotent on every T0T_0-topology.

The construction is functorial.

Proposition 3.6 If f:(X,τ)→(Y,δ)f : (X,\tau) \to(Y,\delta) is continuous, then the same underlying map is continuous

f:(X,dH(τ))⟶(Y,dH(δ))f : (X,d_H(\tau)) \longrightarrow(Y,d_H(\delta))

and

f:(X,HdH(τ))⟶(Y,HdH(δ)).f : (X,Hd_H(\tau)) \longrightarrow(Y,Hd_H(\delta)).

Proof. Let V∈δHV \in\delta_H and A∈H((X,τ))A \in H((X,\tau)). If f−1(V)∩cl⁡τA≠∅f^{-1}(V) \cap\operatorname{cl}_{\tau} A \ne\varnothing, then

V∩f(cl⁡τA)≠∅,f(cl⁡τA)⊆cl⁡δf(A),V \cap f(\operatorname{cl}_{\tau} A) \ne\varnothing,\quad f(\operatorname{cl}_{\tau} A) \subseteq\operatorname{cl}_{\delta} f(A),

and f(A)∈H((Y,δ))f(A) \in H((Y,\delta)). Hence V∩f(A)≠∅V \cap f(A) \ne\varnothing and therefore f−1(V)∈τHf^{-1}(V) \in\tau_H. This proves the first assertion. Iterating the argument transfinitely, including the evident limit step, gives continuity at every stage of Theorem 3.4 and hence the second assertion.

For maps whose domain is already a fixed point, continuity has a particularly simple test.

Proposition 3.7 Let (X,τ)(X,\tau) be HH-determined, let (Y,δ)(Y,\delta) be a T0T_0-space, and let f:X→Yf : X \to Y be a map. Then the following are equivalent:

  1. f:(X,τ)→(Y,δ)f : (X,\tau) \to(Y,\delta) is continuous;

  2. f(cl⁡τA)⊆cl⁡δf(A)f(\operatorname{cl}_{\tau} A) \subseteq\operatorname{cl}_{\delta} f(A) for every A∈H((X,τ))A \in H((X,\tau));

  3. f:(X,τ)→(Y,HdH(δ))f : (X,\tau) \to(Y,Hd_H(\delta)) is continuous.

Proof. The implication (1) ⇒\Rightarrow (2) is standard, and (1) ⇒\Rightarrow (3) follows from Proposition 3.6 because HdH(τ)=τHd_H(\tau) = \tau. The implication (3) ⇒\Rightarrow (1) follows by composing with the identity (Y,HdH(δ))→(Y,δ)(Y,Hd_H(\delta)) \to(Y,\delta).

For completeness, assume (2). If V∈δV \in\delta and A∈H(X)A \in H(X) have f−1(V)∩cl⁡τA≠∅f^{-1}(V) \cap\operatorname{cl}_{\tau} A \ne\varnothing, then V∩cl⁡δf(A)≠∅V \cap\operatorname{cl}_{\delta} f(A) \ne\varnothing. Since VV is open, V∩f(A)≠∅V \cap f(A) \ne\varnothing. Hence f−1(V)∈τH=τf^{-1}(V) \in\tau_H = \tau, proving (1).

The coreflector and categorical consequences

We use standard categorical terminology as in, for example, [8]. Let I:HD↪Top0I : \mathrm{HD} \hookrightarrow\mathrm{Top}_0 be the inclusion. Define

RH(X,τ)=(X,HdH(τ)),RH(f)=f.R_H(X,\tau) = (X,Hd_H(\tau)),\quad R_H(f) = f.

Proposition 3.6 makes RHR_H a functor.

Theorem 4.1. The functor RH:Top0→HDR_H : \mathbf{Top}_0 \to\mathbf{HD} is right adjoint to the inclusion II. Equivalently, HD\mathbf{HD} is coreflective in Top0\mathbf{Top}_0, with coreflection map

rX:(X,HdH(τ))⟶(X,τ),rX(x)=x.r_X : (X,\mathrm{Hd}_H(\tau)) \longrightarrow(X,\tau), \quad r_X(x)=x.

Proof. For A∈HDA \in\mathbf{HD} and X∈Top0X \in\mathbf{Top}_0, Proposition 3.7 gives, naturally in both variables,

Top0(IA,X)=HD(A,RHX),\mathbf{Top}_0(IA,X)=\mathbf{HD}(A,R_HX),

with the same underlying set maps on the two sides.

This formulation makes the limit theory immediate.

Corollary 4.2. The category HD\mathbf{HD} is complete. If a diagram in HD\mathbf{HD} has limit (L,λ)(L,\lambda) in Top0\mathbf{Top}_0, then its limit in HD\mathbf{HD} is

(L,HdH(λ)).(L,\mathrm{Hd}_H(\lambda)).

In particular, for HH-determined spaces (Xj,τj)(X_j,\tau_j),

∏j∈JHDXj=RH(∏j∈JTop0Xj).\prod_{j\in J}^{\mathbf{HD}} X_j=R_H\left(\prod_{j\in J}^{\mathbf{Top}_0}X_j\right).

Proof. The category Top0\mathbf{Top}_0 is complete, and the right adjoint RHR_H preserves limits.

Proposition 4.3. Every retract of an HH-determined space is HH-determined.

Proof. Let p:X→Yp:X\to Y and s:Y→Xs:Y\to X be continuous with p∘s=id⁡Yp\circ s=\operatorname{id}_Y, and suppose XX is HH-determined. If V∈dH(O(Y))V\in d_H(\mathcal{O}(Y)), Proposition 3.6 gives p−1(V)∈dH(O(X))=O(X)p^{-1}(V)\in d_H(\mathcal{O}(X))=\mathcal{O}(X). Hence

V=s−1(p−1(V))∈O(Y).V=s^{-1}(p^{-1}(V))\in\mathcal{O}(Y).

Thus dH(O(Y))=O(Y)d_H(\mathcal{O}(Y))=\mathcal{O}(Y).

Reflectivity behaves differently. We use the criterion of Shen, Xi and Zhao [9]: if a full subcategory K\mathbf{K} of Top0\mathbf{Top}_0 is closed under homeomorphism and is not contained in Top1\mathbf{Top}_1, then reflectivity of K\mathbf{K} implies that every sober space belongs to K\mathbf{K}.

Proposition 4.4. For every subset system HH, the category HD\mathbf{HD} contains non-T1T_1 spaces. Consequently, if Sob⊈HD\mathbf{Sob}\not\subseteq\mathbf{HD}, then HD\mathbf{HD} is not reflective in Top0\mathbf{Top}_0.

Proof. First note that HD\mathbf{HD} is closed under homeomorphisms. Indeed, let f:(X,τ)→(Y,δ)f:(X,\tau)\to(Y,\delta) be a homeomorphism and suppose that (X,τ)(X,\tau) is HH-determined. Proposition 3.6 gives a continuous map

f:(X,τ)=(X,dH(τ))⟶(Y,dH(δ)).f:(X,\tau)=(X,d_H(\tau))\longrightarrow(Y,d_H(\delta)).

Hence V∈dH(δ)V\in d_H(\delta) implies f−1(V)∈τf^{-1}(V)\in\tau, and the original homeomorphism then gives V∈δV\in\delta. Thus dH(δ)⊆δd_H(\delta)\subseteq\delta, while the reverse inclusion is automatic, so (Y,δ)(Y,\delta) is HH-determined.

Every Alexandroff T0T_0-space is HH-determined by Proposition 2.3(1). In particular, the Sierpiński space belongs to HD and is not T1T_1. The reflectivity criterion of Shen, Xi and Zhao [9] now applies and proves the second statement.

Theorem 4.5 If HH is an irreducible subset system, i.e. H≤RH \leq R, then HD is not reflective in Top0\mathbf{Top}_0.

Proof. Let XX be any non-discrete Hausdorff space. It is sober, while Irr⁡(X)=S(X)\operatorname{Irr}(X)=\mathcal{S}(X). Since S≤H≤R\mathcal{S} \leq H \leq\mathcal{R}, we have H(X)=S(X)H(X)=\mathcal{S}(X), and therefore

dH(O(X))=α(X),d_H(\mathcal{O}(X))=\alpha(X),

which is the discrete topology because XX is T1T_1. Thus XX is not HH-determined, so Sob⊈HD\mathbf{Sob} \nsubseteq\mathbf{HD}. Proposition 4.4 applies.

Corollary 4.6 The categories of S\mathcal{S}-, C\mathcal{C}-, D\mathcal{D}- and R\mathcal{R}-determined spaces are not reflective in Top0\mathbf{Top}_0. The same holds for F\mathcal{F}-determined spaces because the S\mathcal{S}- and F\mathcal{F}-determined spaces coincide.

Remark 4.7 The hypothesis in Theorem 4.5 cannot simply be omitted. For H=PH=\mathcal{P}, Proposition 2.3(4) gives dH(τ)=τd_H(\tau)=\tau for every T0T_0-topology, so HD =Top0=\mathbf{Top}_0; in particular this category is reflective as well as coreflective. This also shows why a blanket non-reflectivity assertion for arbitrary subset systems is false.

Subspaces: stability hypotheses on the subset system

For the irreducible system R\mathcal{R}, irreducibility of a subset is unchanged when it is regarded inside a subspace containing it; this fact underlies the hereditary results in [13]. An arbitrary subset system need not have this property, and functoriality of HH only gives one direction. We therefore make the required hypotheses explicit.

Definition 5.1 A subset system HH is called

  1. restriction-stable if, whenever YY is a nonempty subspace of XX and A⊆YA \subseteq Y, the implication

A∈H(X)⟹A∈H(Y)A \in H(X) \Longrightarrow A \in H(Y)

holds;

  1. open-local if, whenever UU is open in XX and A∈H(X)A \in H(X) with A∩U≠∅A \cap U \ne\varnothing, one has

A∩U∈H(U).A \cap U \in H(U).

The reverse implication in (1) is automatic from the subset-system axioms because the inclusion Y↪XY \hookrightarrow X is continuous. The standard systems S\mathcal{S}, F\mathcal{F}, C\mathcal{C}, D\mathcal{D} and R\mathcal{R} are restriction-stable and open-local. For D\mathcal{D}, for example, a nonempty intersection of a directed set with an open upper set is again directed; for R\mathcal{R}, a nonempty intersection of an irreducible set with an open set is irreducible in the open subspace.

Theorem 5.2 If HH is restriction-stable, then every nonempty closed subspace of an HH-determined space is HH-determined.

Proof. Let XX be HH-determined, let C⊆XC \subseteq X be nonempty and closed, and let V⊆CV \subseteq C be HH-open relative to the subspace topology. Put U=V∪(X∖C)U = V \cup(X \setminus C). We claim that U∈O(X)U \in\mathcal{O}(X).

Suppose A∈H(X)A \in H(X) and U∩cl⁡XA≠∅U \cap\operatorname{cl}_{X} A \ne\varnothing. If A∩U≠∅A \cap U \ne\varnothing there is nothing to prove. Otherwise A⊆C∖VA \subseteq C \setminus V. Restriction-stability gives A∈H(C)A \in H(C), while CC is closed, so cl⁡XA=cl⁡CA⊆C\operatorname{cl}_{X} A = \operatorname{cl}_{C} A \subseteq C. Hence

∅≠U∩cl⁡XA=V∩cl⁡CA,\varnothing\ne U \cap\operatorname{cl}_{X} A = V \cap\operatorname{cl}_{C} A,

and the defining property of VV yields A∩V≠∅A \cap V \ne\varnothing, a contradiction. Thus UU is HH-open in XX and therefore open because XX is HH-determined. Consequently V=U∩CV = U \cap C is open in CC.

Theorem 5.3 If HH is open-local, then every nonempty open subspace of an HH-determined space is HH-determined.

Proof. Let XX be HH-determined, O⊆XO \subseteq X be nonempty and open, and let V⊆OV \subseteq O be HH-open relative to OO. We show directly that VV is HH-open in XX.

Take A∈H(X)A \in H(X) with V∩cl⁡XA≠∅V \cap\operatorname{cl}_{X} A \ne\varnothing. Then O∩cl⁡XA≠∅O \cap\operatorname{cl}_{X} A \ne\varnothing, and since OO is open, A∩O≠∅A \cap O \ne\varnothing. Put B=A∩OB = A \cap O. By open-locality, B∈H(O)B \in H(O). Moreover

A⊆B∪(X∖O)A \subseteq B \cup(X \setminus O)

and X∖OX \setminus O is closed, so

cl⁡XA⊆cl⁡XB∪(X∖O).\operatorname{cl}_{X} A \subseteq\operatorname{cl}_{X} B \cup(X \setminus O).

Intersecting with V⊆OV \subseteq O gives V∩cl⁡XB≠∅V \cap\operatorname{cl}_{X} B \ne\varnothing, equivalently V∩cl⁡OB≠∅V \cap\operatorname{cl}_{O} B \ne\varnothing. Hence V∩B≠∅V \cap B \ne\varnothing, and therefore V∩A≠∅V \cap A \ne\varnothing. Thus VV is HH-open in XX, so it is open. In particular it is open in OO.

Corollary 5.4 For H∈{S,F,C,D,R}H \in\{\mathcal{S},\mathcal{F},\mathcal{C},\mathcal{D},\mathcal{R}\}, the property of being HH-determined is both closed-hereditary and open-hereditary.

For H=RH = \mathcal{R}, Corollary 5.4 recovers the corresponding results of [13]; the point here is that the general argument identifies precisely the properties of the subset system that are used.

Cartesian closedness below the directed subset system

Assume throughout this section that

H≤D.H \le\mathcal{D}.

For X,Y∈HDX,Y \in\mathbf{HD}, write X×HYX \mathbin{\times_H} Y for their categorical product in HD\mathbf{HD}, namely the HH-determined hull of their product in Top0\mathbf{Top}_0 as in Corollary 4.2.

For T0T_0-spaces Y,ZY,Z, let C(Y,Z)C(Y,Z) be the set of continuous maps from YY to ZZ. Equip C(Y,Z)C(Y,Z) with the topology E(Y,Z)\mathcal{E}(Y,Z) of pointwise convergence, whose subbasic open sets are

[y,U]={f∈C(Y,Z):f(y)∈U},y∈Y, U∈O(Z).[y,U] = \{f \in C(Y,Z) : f(y) \in U\}, \quad y \in Y,\ U \in\mathcal{O}(Z).

The pointwise topology is T0T_0. Indeed, if f≠gf \ne g, choose y∈Yy \in Y with f(y)≠g(y)f(y) \ne g(y) and use the T0T_0 property of ZZ to find an open set that contains exactly one of f(y)f(y) and g(y)g(y); the corresponding subbasic set separates ff and gg. Moreover, its specialization order is the pointwise order: f≤gf \le g if and only if f(y)≤Zg(y)f(y) \le_Z g(y) for every y∈Yy \in Y. Define

[Y,Z]H:=RH(C(Y,Z),E(Y,Z)).[Y,Z]_H := R_H(C(Y,Z),\mathcal{E}(Y,Z)).

Thus [Y,Z]H[Y,Z]_H is automatically an object of HD\mathbf{HD}.

Lemma 6.1 For Y,Z∈HDY,Z \in\mathbf{HD}, the evaluation map

ev⁡:[Y,Z]H×HY⟶Z,ev⁡(f,y)=f(y),\operatorname{ev} : [Y,Z]_H \mathbin{\times_H} Y \longrightarrow Z,\quad\operatorname{ev}(f,y)=f(y),

is continuous.

Proof. We use Proposition 3.7. Let

A∈H([Y,Z]H×HY)A \in H([Y,Z]_H \mathbin{\times_H} Y)

and let (f,y)∈cl⁡[Y,Z]H×HYA(f,y) \in\operatorname{cl}_{[Y,Z]_H \times_H Y} A. The two projections are continuous, hence

f∈cl⁡[Y,Z]Hp1(A),y∈cl⁡Yp2(A).f \in\operatorname{cl}_{[Y,Z]_H} p_1(A),\quad y \in\operatorname{cl}_Y p_2(A).

The topology of [Y,Z]H[Y,Z]_H is finer than the pointwise topology, so ff also belongs to the pointwise closure of p1(A)p_1(A).

Let U∈O(Z)U \in\mathcal{O}(Z) contain f(y)f(y). Since f:Y⟶Zf:Y \longrightarrow Z is continuous, f−1(U)f^{-1}(U) is an open neighborhood of yy. Choose (g1,y1)∈A(g_1,y_1) \in A with y1∈f−1(U)y_1 \in f^{-1}(U). Then f(y1)∈Uf(y_1) \in U. Because ff lies in the pointwise closure of p1(A)p_1(A), the subbasic neighborhood [y1,U][y_1,U] meets p1(A)p_1(A); choose (g2,y2)∈A(g_2,y_2) \in A such that g2(y1)∈Ug_2(y_1) \in U.

Now A∈H([Y,Z]H×HY)⊆D([Y,Z]H×HY)A \in H([Y,Z]_H \mathbin{\times_H} Y) \subseteq D([Y,Z]_H \mathbin{\times_H} Y), so AA is directed in the specialization order. Choose (g0,y0)∈A(g_0,y_0) \in A above both (g1,y1)(g_1,y_1) and (g2,y2)(g_2,y_2). By Theorem 3.3(3), the categorical product has the same specialization order as the ordinary product, and the specialization order on [Y,Z]H[Y,Z]_H is the pointwise order. Since every continuous map is order preserving,

g2(y1)≤g2(y0)≤g0(y0).g_2(y_1) \leq g_2(y_0) \leq g_0(y_0).

Because UU is an upper set, g0(y0)∈Ug_0(y_0) \in U. Thus every open neighborhood of f(y)f(y) meets ev⁡(A)\operatorname{ev}(A), and hence

ev⁡(cl⁡[Y,Z]H×HYA)⊆cl⁡Zev⁡(A).\operatorname{ev}\bigl(\operatorname{cl}_{[Y,Z]_H \times_H Y} A\bigr) \subseteq\operatorname{cl}_Z \operatorname{ev}(A).

Proposition 3.7 now yields continuity of ev⁡\operatorname{ev}.

Theorem 6.2 If H≤DH \leq D, then HD\mathbf{HD} is Cartesian closed. For Y,Z∈HDY, Z \in\mathbf{HD}, the object [Y,Z]H[Y,Z]_H defined above is an exponential of ZZ by YY.

Proof. Let X,Y,Z∈HDX,Y,Z \in\mathbf{HD}. Given a continuous map

q:X×HY⟶Z,q : X \times_H Y \longrightarrow Z,

define q^:X→C(Y,Z)\widehat{q}: X \to C(Y,Z) by q^(x)(y)=q(x,y)\widehat{q}(x)(y) = q(x,y). For each fixed xx, the section y↦(x,y)y \mapsto(x,y) is the pairing of the constant map Y→XY \to X with id⁡Y\operatorname{id}_Y and is therefore continuous by the universal property of the categorical product. Thus q^(x):Y→Z\widehat{q}(x):Y \to Z is continuous. Similarly, for each fixed yy, the section x↦(x,y)x \mapsto(x,y) is continuous, so x↦q(x,y)x \mapsto q(x,y) is continuous. Hence q^:X→(C(Y,Z),E(Y,Z))\widehat{q}:X \to(C(Y,Z),\mathcal{E}(Y,Z)) is continuous by the defining initial property of the pointwise topology. Since XX is HH-determined, Theorem 4.1 (or Proposition 3.7) lifts this uniquely to a continuous map

q^:X⟶[Y,Z]H.\widehat{q}:X \longrightarrow[Y,Z]_H.

Conversely, if g:X→[Y,Z]Hg:X \to[Y,Z]_H is continuous, then

X×HY→g×id⁡Y[Y,Z]H×HY→ev⁡ZX \times_H Y \xrightarrow{g \times\operatorname{id}_Y} [Y,Z]_H \times_H Y \xrightarrow{\operatorname{ev}} Z

is continuous by Lemma 6.1. These two constructions are inverse to one another on underlying maps. Therefore there is a natural bijection

HD(X×HY,Z)≅HD(X,[Y,Z]H),\mathbf{HD}(X \times_H Y,Z) \cong\mathbf{HD}(X,[Y,Z]_H),

which is precisely the exponential universal property.

Corollary 6.3 The categories of SS-, FF- and CC-determined spaces are Cartesian closed.

Proof. The SS- and FF-determined spaces coincide by Proposition 2.3(4), while S,C≤DS,C \leq D. Apply Theorem 6.2.

Corollary 6.4 The category of monotone determined spaces is Cartesian closed; see also [7, 14].

Concluding remarks and questions

The fixed-point formulation separates three levels of the construction: the one-step topology dH(τ)d_H(\tau), its idempotent hull Hd⁡H(τ)\operatorname{Hd}_H(\tau), and the categorical coreflector RHR_H. This makes clear which conclusions are formal consequences of the subset-system axioms and which require additional structure on HH.

From the standpoint of programming semantics, the main categorical consequence is Theorem 6.2: for H≤DH \leq\mathcal{D}, the same fixed-point mechanism that produces the coreflective category also supplies its exponentials. The paper does not commit to a particular programming language or type system; rather, it isolates a reusable topological and categorical mechanism that can be tested in future against concrete domain-theoretic models and higher-order type interpretations.

The following questions arise naturally.

Question 7.1 Characterize the subset systems HH for which dHd_H is idempotent on every T0T_0-topology, equivalently, for which dH(τ)=Hd⁡H(τ)d_H(\tau)=\operatorname{Hd}_H(\tau) always holds.

Question 7.2 Can the restriction-stability and open-locality assumptions in § 5 be replaced by weaker intrinsic conditions that are still necessary and sufficient for closed- and open-heredity of HH-determined spaces?

Question 7.3 Characterize the subset systems HH for which HDHD is reflective in Top⁡0\operatorname{Top}_0. Theorem 4.5 gives a broad negative class, while H=PH=\mathcal{P} gives the opposite extreme.

Question 7.4 Can the hypothesis H≤DH \leq\mathcal{D} in Theorem 6.2 be weakened? More generally, which properties of HH guarantee continuity of the evaluation map for the coreflected pointwise function space?

Statements and Declarations

Funding. This research was supported by the National Natural Science Foundation of China (Nos. 12471070 and 12071199).

Author contributions. Xiaoquan Xu: Conceptualization, Investigation, methodology, formal analysis, writing-review and editing. Xinjie Huang: Conceptualization, investigation, formal analysis, writing-original draft. Yi Yang: Investigation, formal analysis, writing-review and editing All authors have reviewed and approved the final manuscript.

Competing interests. The authors have no relevant financial or non-financial interests to disclose.

Data availability. No datasets were generated or analysed in this theoretical study.

Use of AI-assisted tools. During the preparation of this work, the authors used ChatGPT (OpenAI) to assist with organization, language and readability. The authors reviewed and edited the resulting text, verified the mathematical content and references, and take full responsibility for the final manuscript.

References

  1. [1]Abramsky, S., Jung, A.: Domain theory. In: Abramsky, S., Gabbay, D.M., Maibaum, T.S.E. (eds.) Handbook of Logic in Computer Science, vol. 3, pp. 1–168. Clarendon Press, Oxford (1994)
  2. [2]Chen, Y., Kou, H., Lyu, Z.: Monotone determined spaces via C-generated spaces. Topology Appl. 393, 109973 (2026). https://doi.org/10.1016/j.topol.2026.109973
  3. [3]Chen, Y., Xu, X.: On topological spaces determined by subset systems. Appl. Math. J. Chinese Univ. 34, 487–491 (2019). https://doi.org/10.3785/amjcu-a.2019.04.0487
  4. [4]Erné, M.: Infinite distributive laws versus local connectedness and compactness properties. Topology Appl. 156, 2054–2069 (2009). https://doi.org/10.1016/j.topol.2009.03.029
  5. [5]Gierz, G., Hofmann, K.H., Keimel, K., Lawson, J.D., Mislove, M., Scott, D.S.: Continuous Lattices and Domains. Encyclopedia of Mathematics and its Applications, vol. 93. Cambridge University Press, Cambridge (2003). https://doi.org/10.1017/CBO9780511542725
  6. [6]Goubault-Larrecq, J.: Non-Hausdorff Topology and Domain Theory. New Mathematical Monographs, vol. 22. Cambridge University Press, Cambridge (2013). https://doi.org/10.1017/CBO9781139524438
  7. [7]Luo, S., Xu, X.: On monotone determined spaces. Electron. Notes Theor. Comput. Sci. 333, 63–72 (2017). https://doi.org/10.1016/j.entcs.2017.08.006
  8. [8]Mac Lane, S.: Categories for the Working Mathematician, 2nd edn. Springer, New York (1997). https://doi.org/10.1007/978-1-4757-4721-8
  9. [9]Shen, C., Xi, X., Zhao, D.: The reflectivity of some categories of T₀ spaces in domain theory. Rocky Mountain J. Math. 54, 1149–1166 (2024). https://doi.org/10.1216/rmj.2024.54.1149
  10. [10]Wright, J.B., Wagner, E.G., Thatcher, J.W.: A uniform approach to inductive posets and inductive closure. Theoret. Comput. Sci. 7, 57–77 (1978). https://doi.org/10.1016/0304-3975(78)90040-3
  11. [11]Xu, X.: On H-sober spaces and H-sobrifications of T₀ spaces. Topology Appl. 289, 107548 (2021). https://doi.org/10.1016/j.topol.2020.107548
  12. [12]Xu, X., Shen, C., Xi, X., Zhao, D.: On T₀ spaces determined by well-filtered spaces. Topology Appl. 282, 107323 (2020). https://doi.org/10.1016/j.topol.2020.107323
  13. [13]Xu, X., Yang, Y., Luo, S.: On irreducible determined T₀-spaces. Topology Appl. 382, 109762 (2026). https://doi.org/10.1016/j.topol.2026.109762
  14. [14]Yu, Y., Kou, H.: Directed spaces defined through T₀ spaces with specialization order. J. Sichuan Univ. Nat. Sci. Ed. 52, 217–222 (2015) (in Chinese)

Paper details

vlnc:W_h47yse3g
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 11

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

Not recorded for this paper

View statement
Looking for related papers…
Contents