Finite topological spaces

One of my favourite bits of point set topology is messing around with easy topological spaces. What could be easier than finite topological spaces? The main result (below) is that the category of finite topological spaces is equivalent to the category of finite preorders.

Recall (e.g. from algebraic geometry) the following definition:

Definition. Let X be a topological space. Then the specialisation preorder on (the underlying set of) X is the relation x \leq y if and only if x \in \overline{\{y\}}.

Note that it is indeed a preorder: clearly x \leq x, and if x \leq y and y \leq z, then \{y\} \subseteq \overline{\{z\}}, so x \in \overline{\{y\}} \subseteq \overline{\{z\}}, showing x \leq z. We denote this preorder by X^{\operatorname{sp}}.

Note that the relation x \leq y is usually denoted y \rightsquigarrow x in algebraic geometry, which is pronounced “y specialises to x”.

Definition. Given a preorder (X,\leq), the Alexandroff topology on X is the topology whose opens U \subseteq X are the cosieves, i.e. the upwards closed sets (meaning x \in U and x \leq y implies y \in U).

To see that this defines a topology, note that an arbitrary (possibly empty) union or intersection of cosieves is a cosieve. A subbase for the topology is given by the principal cosieves X_{\geq x} = \{y \in X\ |\ y \geq x\} for any x \in X. We denote the set X with its Alexandroff topology by X^{\operatorname{Alex}}.

Likewise, the closed sets in X are the sieves (or downwards closed sets); for instance the principal sieves X_{\leq x} = \{y \in X\ |\ y \leq x\}. The closure of S \subseteq X is the sieve X_{\leq S} = \bigcup_{s \in S} X_{\leq s} generated by S; for instance the closure of a singleton \{x\} is the principal sieve X_{\leq x}.

Theorem. Let F \colon \mathbf{PreOrd} \to \mathbf{Top} be the functor X \mapsto X^{\operatorname{Alex}}, and G \colon \mathbf{Top} \to \mathbf{PreOrd} the functor Y \mapsto Y^{\operatorname{sp}}.

  1. Let (X,\leq) be a preorder, Y a topological space, and f \colon X \to Y a function. Then f is a monotone function X \to Y^{\operatorname{sp}} if and only if f is a continuous function X^{\operatorname{Alex}} \to Y.
  2. The functors F and G are adjoint: F \dashv G.
  3. The composition GF \colon \mathbf{PreOrd} \to \mathbf{PreOrd} is equal (not just isomorphic!) to the identity functor.
  4. The restriction of FG \colon \mathbf{Top} \to \mathbf{Top} to the category \mathbf{Top}^{\operatorname{fin}} of finite topological spaces is equal to the identity functor.
  5. If Y is a topological space, then Y is T_0 if and only if Y^{\operatorname{sp}} is a poset.
  6. If (X,\leq) is a preorder, then X is a poset if and only if X^{\operatorname{Alex}} is T_0.
  7. The functors F and G give rise to adjoint equivalences

        \begin{align*}F\!:\mathbf{PreOrd}^{\operatorname{fin}} &\leftrightarrows \mathbf{Top}^{\operatorname{fin}}:\!G \\F\!:\mathbf{Pos}^{\operatorname{fin}} &\leftrightarrows \mathbf{Top}^{\operatorname{fin}}_{T_0}:\!G.\end{align*}

Proof. (1) Suppose f \colon X \to Y^{\operatorname{sp}} is monotone, let Z \subseteq Y be a closed subset, and let W = f^{-1}(Z). Suppose b \in W and a \leq b. Since f is monotone and Z is closed, we get f(a) \leq f(b), i.e. f(a) \in \overline{\{f(b)\}} \subseteq Z. We conclude that a \in W, so W is downward closed, hence closed in X^{\operatorname{Alex}}.

Conversely, suppose f \colon X^{\operatorname{Alex}} \to Y is continuous, and suppose a \leq b in X. Then a \in \overline{\{b\}}, so by continuity we get f(a) \in f\left(\overline{\{b\}}\right) \subseteq \overline{\{f(b)\}}, so f(a) \leq f(b).

(2) This is a restatement of (1): the map

    \begin{align*}\operatorname{Hom}_{\mathbf{PreOrd}}\big(X,G(Y)\big) &\stackrel\sim\to \operatorname{Hom}_{\mathbf{Top}}\big(F(X),Y\big) \\f &\mapsto f\end{align*}

is a bijection.

(3) Since \overline{\{x\}} = X_{\leq x}, we conclude that y \in \overline{\{x\}} if and only if y \leq x, so the specialisation preorder on X^{\operatorname{Alex}} is the original preorder on X.

(4) In general, the counit FG(Y) \to Y is a continuous map on the same underlying space, so FG(Y) is finer than Y. Conversely, suppose Z \subseteq FG(Y) is closed, i.e. Z is a sieve for the specialisation preorder on Y. This means that if y \in Z, then x \in \overline{\{y\}} implies x \in Z; in other words \overline{\{y\}} \subseteq Z. If Y and therefore Z is finite, there are finitely many such y, so Z is the finite union

    \[Z = \bigcup_{y \in Z} \overline{\{y\}}\]

of closed subsets of Y. Thus any closed subset of FG(Y) is closed in Y, so the topologies agree.

(5) The relations x \leq y and y \leq x mean x \in \overline{\{y\}} and y \in \overline{\{x\}}. This is equivalent to the statement that a closed subset Z \subseteq Y contains x if and only if it contains y. The result follows since a poset is a preorder where the first statement only happens if x = y, and a T_0 space is a space where the second statement only happens if x = y.

(6) Follows from (5) applied to Y = F(X) since X = G(Y) by (3).

(7) The equivalence \mathbf{PreOrd}^{\operatorname{fin}} \leftrightarrows \mathbf{Top}^{\operatorname{fin}} follows from (3) and (4), and the equivalence \mathbf{Pos}^{\operatorname{fin}} \leftrightarrows \mathbf{Top}^{\operatorname{fin}}_{T_0} then follows from (5) and (6). \qedsymbol

Example. The Alexandroff topology on the poset [1] = \{0 \leq 1\} is the Sierpiński space S = \{0,1\} with topology \{\varnothing, \{1\}, S\}. As explained in this post, continuous maps X \to S from a topological space X to S are in bijection with open subsets U \subseteq X, where f \colon X \to S is sent to f^{-1}(1) \subseteq X (and U \subseteq X to the indicator function \mathbf 1_U \colon X \to S).

Example. Let X = \{x,y\} be a set with two elements. There are 4 possible topologies on X, sitting in the following diagram (where vertical arrows indicate inclusion bottom to top):

    \[{\arraycolsep=-1em\begin{array}{ccccc} & & \{\varnothing,\{x\},\{y\},X\} & & \\ & \ \ / & & \backslash\ \  & \\ \{\varnothing,\{x\},X\} & & & & \{\varnothing,\{y\},X\} \\ & \ \ \backslash & & /\ \  & \\ & & \{\varnothing,X\}.\! & & \end{array}}\]

These correspond to 4 possible preorder relations \{(a,b)\ |\ a \leq b\} \subseteq X\times X, sitting in the following diagram (where vertical arrows indicate inclusion top to bottom):

    \[{\arraycolsep=-1.5em\begin{array}{ccccc} & & \{(x,x),(y,y)\} & & \\ & \ \ / & & \backslash\ \ & \\ \{(x,x),(x,y),(y,y)\} & & & & \{(x,x),(y,x),(y,y)\} \\ & \ \ \backslash & & /\ \ & \\ & & \{(x,x),(x,y),(y,x),(y,y)\}.\!\! & & \end{array}}\]

We see that finer topologies (more opens) have stronger relations (fewer inequalities).

Example. The statement in (4) is false for infinite topological spaces. For instance, if Y is the Zariski topology on a curve, then any set of closed points is downwards closed, but it is only closed if it’s finite. Or if Y is a Hausdorff space, then the specialisation preorder is just the equality relation \Delta_Y \subseteq Y \times Y, whose Alexandroff topology is the discrete topology.

I find the examples useful for remembering which way the adjunction goes: topological spaces generally have fewer opens than Alexandroff topologies on posets, so the continuous map should go X^{\operatorname{Alex}} \to Y.

Grothendieck topologies (topologies 4/6)

This post is the first goal in a series on sieves (subobjects of representable presheaves); I will give another generalisation in the next two posts. In the first post of the series, I defined sieves and gave basic examples, and last week I showed how the sheaf condition on a site can be stated in terms of sieves:

Corollary. Let \mathscr C be a (small) site. For a set of morphisms \mathscr U = \{U_i \to U\}_{i \in I} with the same target, write S_{\mathscr U} \subseteq h_U for the presheaf image of \coprod_{i\in I} h_{U_i} \to h_U. Then a presheaf \mathscr F \colon \mathscr C^{\operatorname{op}} \to \mathbf{Set} is a sheaf if and only if for every covering \mathscr U = \{U_i \to U\}_{i \in I} in \mathscr C, the inclusion S_{\mathscr U} \hookrightarrow h_U induces an isomorphism

    \[\operatorname{Hom}(h_U,\mathscr F) \stackrel\sim\to \operatorname{Hom}(S_{\mathscr U},\mathscr F).\]

Thus, if \mathscr C is a site (a small category with a Grothendieck pretopology), we should be able to obtain the category \mathbf{Sh}(\mathscr C) \subseteq \mathbf{PSh}(\mathscr C) of sheaves purely in terms of sieves. This is the notion of a Grothendieck topology that we describe at the end of this post.

Before giving the definition, note that any morphism f \colon Y \to X in \mathscr C gives a pullback \mathbf{Siv}(X) \to \mathbf{Siv}(Y) taking S \subseteq h_X to its inverse image under h_f \colon h_Y \to h_X (I avoid the word ‘pullback’ here to make sure this is truly a subpresheaf and not a presheaf with a monomorphism to h_Y defined uniquely up to unique isomorphism). Thus, \mathbf{Siv} is itself a presheaf \mathscr C^{\operatorname{op}} \to \mathbf{Set} (it takes values in \mathbf{Set} since \mathscr C is small).

Also note the following method for producing sieves: if \mathscr F is a presheaf, \mathscr G \subseteq \mathscr F a subpresheaf, and s \in \mathscr F(X) a section over some X \in \mathscr C, we get a sieve (s \in \mathscr G) \in \mathbf{Siv}(X) by

    \[(s \in \mathscr G)(Y) = \left\{f \colon Y \to X\ \big|\ f^*(s) \in \mathscr G(Y)\right\}.\]

By the Yoneda lemma, this is just the inverse image of \mathscr G \subseteq \mathscr F along the morphism h_X \to \mathscr F classifying s. Note that (s \in \mathscr G) is the maximal sieve h_X if and only if s \in \mathscr G(X).

Definition. Let \mathscr C be a small category. Then a Grothendieck topology on \mathscr C consists of a subpresheaf J \subseteq \mathbf{Siv} such that

  1. For all X \in \mathscr C, the maximal sieve h_X \subseteq h_X is in J(X).
  2. If S \in J(X) and S' \in \mathbf{Siv}(X) with S \subseteq S', then S' \in J(X).
  3. If S \in \mathbf{Siv}(X) is a sieve such that (S \in J) \in J(X), then S \in J(X) (equivalently, then (S \in J) is the maximal sieve h_X).

The sieves S \in J(X) are called covering sieves. Since J is a presheaf, we see that for any f \colon Y \to X and any covering sieve S \subseteq h_X, the pullback f^*S \subseteq h_Y is covering. Condition 2 says that any sieve containing a covering sieve is covering. In the presence of condition 1, conditions 2 and 3 together are equivalent to the local character found in SGA IV_1, Exp. II, Def. 1.1:

  • If S, S' \in \mathbf{Siv}(X) with S \in J(X), such that for every morphism h_Y \to S the inverse image of S' \subseteq h_X along h_Y \to S \to h_X is in J(Y), then S' \in J(X).

Indeed, applying this criterion when S \subseteq S' immedately shows S' \in J(X) if S \in J(X), since the inverse image of S' \subseteq h_X along h_Y \to S \to h_X is the maximal sieve h_Y. Thus the local character implies criterion 2. The local character says that if (S' \in J) contains a covering sieve S, then S' is covering. Assuming criterion 2, the sieve (S' \in J) contains a covering sieve if and only if (S' \in J) is itself covering, so the local character is equivalent to criterion 3.

Remark. One property that follows from the axioms is that J(X) is closed under binary intersection, i.e. if S, T \in J(X) then (S \cap T) \in J(X). Indeed, if f \in S(Y) for some f \colon Y \to X, then

    \[f^*(S \cap T) = f^*S \cap f^*T = h_Y \cap f^*T = f^*T \in J(Y),\]

so S \subseteq ((S \cap T) \in J). Axioms 2 and 3 give (S \cap T) \in J(X).

Example. Let \mathcal Cov(\mathscr C) be a pretopology on the (small) category \mathscr C; see Tag 00VH for a list of axioms. For each X \in \mathscr C, define the subset J(X) \subseteq \mathbf{Siv}(X) as those S \subseteq h_X that contain a sieve of the form S_{\mathscr U} for some covering \mathscr U = \{U_i \to X\} in \mathcal Cov(\mathscr C). (See the corollary at the top for the definition of S_{\mathscr U}.) Concretely, this means that there exists a covering \{f_i \colon U_i \to X\}_{i \in I} \in \mathcal Cov(\mathscr C) such that f_i \in S(U_i) for all i \in I, i.e. X is covered by morphisms f_i \colon U_i \to X that are in the given sieve S.

Lemma. The association X \mapsto J(X) is a topology. It is the coarsest topology on \mathscr C for which each S_{\mathscr U} for \mathscr U \in \mathcal Cov(\mathscr C) is a covering sieve.

Proof. We will use the criteria of Tag 00VH. If S \in J(X), then there exists \mathscr U = \{U_i \to X\}_{i \in I} \in \mathcal Cov(\mathscr C) with S_{\mathscr U} \subseteq S. If f \colon Y \to X is any morphism in \mathscr C, then f^*\mathscr U = \{U_i \times_X Y \to Y\}_{i \in I} \in \mathcal Cov(\mathscr C) by criterion 3 of Tag 00VH. But S_{f^*\mathscr U} = f^*S_{\mathscr U}, because a morphism g \colon U \to Y factors through U_i \times_X Y if and only if fg \colon U \to X factors through U_i. Thus, S_{f^*\mathscr U} = f^*S_{\mathscr U} \subseteq f^*S, so f^*S \in J(Y), and J is a subpresheaf of \mathbf{Siv}.

Condition 1 follows immediately from criterion 1 in Tag 00VH, and condition 2 is satisfied by definition. For condition 3, suppose S \in \mathbf{Siv}(X) satisfies (S \in J) \in J(X). Then there exists \mathscr U = \{f_i \colon U_i \to X\}_{i \in I} \in \mathcal Cov(\mathscr C) with S_{\mathscr U} \subseteq (S \in J). This means that f_i \in (S \in J)(U_i) for all i, i.e. f_i^*S \in J(U_i) for all i. Thus, for each i \in I there exists \mathscr V_i = \{g_{ij} \colon V_{ij} \to U_i\}_{j \in J_i} in \mathcal Cov(\mathscr C) such that S_{\mathscr V_i} \subseteq f_i^*S, i.e. f_ig_{ij} \in S(X) for all i \in I and all j \in J_i. Thus, if \mathscr V denotes \{f_ig_{ij} \colon V_{ij} \to X\}_{i \in I, j \in J_i}, then we get S_{\mathscr V} \subseteq S. But \mathscr V is a covering by criterion 2 of Tag 00VH, so S \in J(X).

If J' is any other Grothendieck topology for which each S_{\mathscr U} for \mathscr U \in \mathcal Cov(\mathscr C) is covering, then J' contains J by criterion 2. \qedsymbol

To state the obvious (hopefully), the notion of sheaf can therefore be defined on a Grothendieck topology in a way that coincides with the usual notion for a Grothendieck pretopology:

Definition. Let \mathscr C be a small category, and let J \subseteq \mathbf{Siv} be a Grothendieck topology. Then a presheaf \mathscr F \colon \mathscr C^{\operatorname{op}} \to \mathbf{Set} is a sheaf if for any X \in \mathscr C and any S \in J(X), the map S \hookrightarrow h_X induces an isomorphism

    \[\operatorname{Hom}(h_X,\mathscr F) \stackrel\sim\to \operatorname{Hom}(S,\mathscr F).\]

Thus, a Grothendieck topology is an internal characterisation (inside \mathbf{PSh}(\mathscr C)) of which morphisms S \to h_X one needs to localise to get \mathbf{Sh}(\mathscr C,J). In the last two posts, we will generalise this even further to a Lawvere–Tierney topology on an arbitrary topos.

Presheaves are colimits of representables (topologies 2/6)

Last week, I started a series on sieves and Grothendieck topoi. This is a short intermezzo on a well-known lemma from category theory that I will need for next week’s instalment.

Let \mathscr C be a small category. Recall that a presheaf on \mathscr C is a functor \mathscr C^{\operatorname{op}} \to \mathbf{Set}. Examples include the representable presheaves h_X for X \in \mathscr C, given by \operatorname{Hom}(-,X). The Yoneda lemma says that for any presheaf F \colon \mathscr C^{\operatorname{op}} \to \mathbf{Set} and any X \in \mathscr C, the map

    \begin{align*}\operatorname{Nat}(h_X,F) &\to F(X) \\\alpha &\mapsto \alpha_X(\operatorname{id}_X)\end{align*}

is an isomorphism. Applying this to F = h_Y shows that the Yoneda embedding

    \begin{align*}h \colon \mathscr C &\to [\mathscr C^{\operatorname{op}},\mathbf{Set}] \\X &\mapsto h_X\end{align*}

is fully faithful.

Given a functor F \colon \mathscr C^{\operatorname{op}} \to \mathbf{Set}, write (h \downarrow F) for the comma category whose objects are pairs (X,\alpha) where X \in \mathscr C and \alpha \colon h_X \to F is a morphism (natural transformation) in [\mathscr C^{\operatorname{op}},\mathbf{Set}]. A morphism f \colon (X,\alpha) \to (Y,\beta) is a morphism f \colon X \to Y such that the triangle

(1)   \[\begin{array}{ccccc}\!\!h_X\!\!\!\!\! & & \!\!\!\!\!\stackrel{h_f}\longrightarrow\!\!\!\!\! & & \!\!\!\!\!h_Y\!\! \\ & \!\!\!\!{\underset{\alpha}{}}\!\!\searrow\!\!\!\! & & \!\!\!\!\swarrow\!\!{\underset{\beta}{}}\!\!\!\! & \\[-.3em] & & F\! & & \end{array}\]

of natural transformations commutes, where h_f \colon \operatorname{Hom}(-,X) \to \operatorname{Hom}(-,Y) denotes postcomposition by f. Note that (h \downarrow F) is again a small category, and there is a forgetful functor U \colon (h \downarrow F) \to \mathscr C taking (X,\alpha) to X \in \mathscr C.

By the Yoneda lemma, the category (h \downarrow F) is isomorphic (not just equivalent!) to the category \int F of pairs (X,s) with X \in \mathscr C and s \in F(X) with morphisms f \colon (X,s) \to (Y,t) given by morphisms f \colon X \to Y in \mathscr C such that F(f)(t) = s. It’s convenient to keep both points of view.

Lemma. Let \mathscr C be a small category, and let F \colon \mathscr C^{\operatorname{op}} \to \mathbf{Set} be a functor.

  1. The object F \in [\mathscr C^{\operatorname{op}},\mathbf{Set}] is naturally a cocone under hU \colon (h \downarrow F) \to [\mathscr C^{\operatorname{op}},\mathbf{Set}] via the morphisms \alpha \colon h_X \to F.
  2. This cocone makes F the colimit of the diagram hU of representable functors.

In particular, any presheaf on a small category is a colimit of representable presheaves.

Proof. A cocone under hU is a presheaf G with a natural transformation \phi \colon hU \to G to the constant diagram (h \downarrow F) \to [\mathscr C^{\operatorname{op}},\mathbf{Set}] with value G. This means every (X,\alpha) \in (h \downarrow F) is taken to a natural transformation \phi_{(X,\alpha)} \colon h_X \to G, such that for any morphism f \colon (X,\alpha) \to (Y,\beta), the square

    \[\begin{array}{ccc} h_X & \stackrel{\phi_{(X,\alpha)}}\longrightarrow & G \\ \!\!\!\!\!\!\!\!\!{h_f}\downarrow & & |\!| \\ h_Y & \underset{\phi_{(Y,\beta)}}\longrightarrow & G\end{array}\]

commutes. By the Yoneda lemma, such a datum corresponds to an association of elements \phi_{(X,s)} \in G(X) for all (X,s) \in \int F such that for every f \colon (X,s) \to (Y,t) with F(f)(t) = s, we have G(f)(\phi_{(Y,t)}) = \phi_{(X,s)}.

(1) To make F a cocone under hU, simply take \phi_{(X,s)} = s \in F(X).

(2) Given any other cocone \phi \colon hU \to G under hU, define the natural transformation \eta \colon F \to G by

    \begin{align*}\eta_X \colon F(X) &\to G(X) \\s &\mapsto \phi_{(X,s)}.\end{align*}

Naturality follows since F(f)(t) = s implies G(f)(\phi_{(Y,t)}) = \phi_{(X,s)}. It is clear that \eta is the unique natural transformation of cocones F \to G under hU, showing that F is the colimit. \qedsymbol

One can also easily rewrite this argument in terms of natural transformations h_X \to G. For instance, the universal cocone hU \to F is the natural transformation \phi \colon hU \to F of functors (h \downarrow F) \to [\mathscr C^{\operatorname{op}},\mathbf{Set}] given on (X,\alpha) \in (h \downarrow F) by \alpha \colon h_X \to F. Naturality of \phi follows at once from (1). But checking that this thing is universal is a bit more tedious in this language.

Example. A standard example where this point of view is useful is simplicial sets. Let \Delta be the category of finite nonempty totally ordered sets, with (weakly) monotone increasing functions as morphisms. A simplicial set is a functor \Delta^{\operatorname{op}} \to \mathbf{Set}, and we often think of them as combinatorial models for topological spaces. The representable ones are the standard n-simplices \Delta^n = \operatorname{Hom}(-,[n]), where [n] is the totally ordered set \{0,\ldots,n\} for n \in \mathbf Z_{\geq 0}.

If X \colon \Delta^{\operatorname{op}} \to \mathbf{Set} is a simplicial set, its value at [n] is called the n-simplices X_n of X. By the Yoneda lemma, this is \operatorname{Hom}(\Delta^n,X). Then the story above is saying that a simplicial set is the colimit over all its n-simplices for n \in \mathbf Z_{\geq 0}. This is extremely useful, as many arguments proceed by attaching simplices one at a time.

A strange contractible space

Here’s a strange phenomenon that I ran into when writing a MathOverflow answer a few years ago.

Lemma. Let X be a set endowed with the cofinite topology, and assume X is path connected. Then X is contractible.

The assumption is for example satisfied when |X| \geq |\mathbf R|, for then any injection f \colon [0,1] \hookrightarrow X is a path from x_0 = f(0) to x_1 = f(1). Path connectedness of cofinite spaces is related to partitioning the interval into disjoint closed subsets; see the remark below for some bounds on the cardinalities.

Proof. The result is trivial if X is finite, for then both are equivalent to |X| = 1. Thus we may assume that X is infinite. Choose a path f \colon [0,1] \to X from some x_0 = f(0) to some x_1 = f(1) \neq x_0. This induces a continuous map [0,1] \times X \to X \times X. Choose a bijection

    \[g \colon (X \setminus \{x_0,x_1\}) \times X \stackrel \sim\to X,\]

and extend to a map \bar g \colon X \times X \to X by g(x_0,x) = x and g(x_1,x) = x_1 for all x \in X. Then \bar g is continuous: the preimage of x \in X is g^{-1}(x) \cup (x_0,x) if x \neq x_1, and g^{-1}(x) \cup (x_0,x) \cup x \times X if x = x_1, both of which are closed. Thus \bar g \circ (f \times \mathbf 1_X) is a homotopy from \mathbf 1_X to the constant map x_1, hence a contraction. \qedsymbol

I would love to see an animation of this contraction as t goes from 0 to 1… I find especially the slightly more direct argument for |X| \geq |\mathbf R| given here elusive yet somehow strangely visual.

Remark. If X is countable (still with the cofinite topology), then X is path connected if and only if |X| = 1. In the finite case this is clear (because then X is discrete), and in the infinite case this is a result of Sierpiński. See for example this MO answer of Timothy Gowers for an easy argument.

There’s also some study of path connectedness of cofinite topological spaces of cardinality strictly between \aleph_0 = |\mathbf N| and \mathfrak c = |\mathbf R|, if such cardinalities exist. See this MO question for some results. In particular, it is consistent with ZFC that the smallest cardinality for which X is path connected is strictly smaller than \mathfrak c.

Internal Hom


This is an introductory post about some easy examples of internal Hom.

Definition. Let (\mathscr C, \otimes) be a symmetric monoidal category, i.e. a category \mathscr C with a functor \otimes \colon \mathscr C \times \mathscr C \to \mathscr C that is associative, unital, and commutative up to natural isomorphism. Then an internal Hom in \mathscr C is a functor

    \[\mathbf{Hom}(-,-) \colon \mathscr C\op \times \mathscr C \to \mathscr C\]

such that -\otimes Y is a left adjoint to \mathbf{Hom}(Y,-) for any Y \in \mathscr C, i.e. there are functorial isomorphisms

    \[\operatorname{Hom}(X \otimes Y, Z) \stackrel\sim\to \operatorname{Hom}(X, \mathbf{Hom}(Y,Z)).\]

Remark. In the easiest examples, we typically think of \mathbf{Hom}(Y,Z) as ‘upgrading \operatorname{Hom}(Y,Z) to an object of \mathscr C‘:

Example. Let R be a commutative ring, and let \mathscr C = \mathbf{Mod}_R be the category of R-modules, with \otimes the tensor product. Then \mathbf{Hom}(M,N) = \operatorname{Hom}_R(M,N) with its natural R-module structure is an internal Hom, by the usual tensor-Hom adjunction:

    \[\operatorname{Hom}_R(M \otimes_R N, K) \cong \operatorname{Hom}_R(M, \mathbf{Hom}(N, K)).\]

The same is true when \mathscr C =\!\ _R\mathbf{Mod}_R is the category of (R,R)-bimodules for a not necessarily commutative ring R.

However, we cannot do this for left R-modules over a noncommutative ring, because there is no natural R-module structure on \operatorname{Hom}_R(M,N) for left R-modules M and N. In general, the tensor product takes an (A,B)-bimodule M and a (B,C)-bimodule N and produces an (A,C)-bimodule M \otimes_B N. Taking A = C = \mathbf Z gives a way to tensor a right R-module with a left R-module, but there is no standard way to tensor two left R-modules, let alone equip it with the structure of a left R-module.

Example. Let \mathscr C = \mathbf{Set}. Then \mathbf{Hom}(X,Y) = \operatorname{Hom}(X,Y) = Y^X is naturally a set, making it into an internal Hom for (\mathscr C, \times):

    \[\operatorname{Hom}(X \times Y, Z) \stackrel\sim\to \operatorname{Hom}(X, \mathbf{Hom}(Y,Z)).\]

When \otimes is the categorical product \times, the internal \mathbf{Hom}(X,Y) (if it exists) is usually called an exponential object, in analogy with the case \mathscr C = \mathbf{Set} above.

Example. Another example of exponential objects is from topology. Let \mathscr C = \mathbf{Haus} be the category of locally compact Hausdorff topological spaces. Then the compact-open topology makes \mathbf{Hom}(X,Y) := Y^X into an internal Hom of topological spaces. (There are mild generalisations of this beyond the compact Hausdorff case, but for an arbitrary topological space X the functor - \times X does not preserve colimits and hence cannot admit a right adjoint.)

Example. An example of a slightly different nature is chain complexes: let R be a commutative ring, and let \mathscr C = \mathbf{Ch}(\mathbf{Mod}_R) be the category of cochain complexes

    \[\ldots \to C^{i-1} \to C^i \to C^{i+1} \to \ldots\]

of R-modules (meaning each C^i is an R-module, and the d^i \colon C^i \to C^{i+1} are R-linear maps satisfying d \circ d = 0). Homomorphisms f \colon C \to D are commutative diagrams

    \[\begin{array}{ccccccc}\ldots & \to & C^i & \to & C^{i+1} & \to & \ldots \\ & & \!\!\!\!\! f^i\downarrow & & \downarrow f^{i+1}\!\!\!\!\!\!\! & & \\ \ldots & \to & D^i & \to & D^{i+1} & \to & \ldots,\!\!\end{array}\]

and the tensor product is given by the direct sum totalisation of the double complex of componentwise tensor products.

There isn’t a natural way to ‘endow \operatorname{Hom}(C, D) with the structure of a chain complex’, but there is an internal Hom given by

    \[\mathbf{Hom}(C, D)^i = \prod_{m \in \mathbf Z} \operatorname{Hom}(C_m, D_{m+i}),\]

with differentials given by

    \[d^if = d_D f - (-1)^i f d_C.\]

Then we get for example

    \[\operatorname{Hom}(R[0], \mathbf{Hom}(C, D)) \cong \operatorname{Hom}(C, D),\]

since a morphism R[0] \to \mathbf{Hom}(C, D) is given by an element f \in \mathbf{Hom}(C, D)^0 such that df = 0, i.e. d_Df = f d_C, meaning that f is a morphism of cochain complexes.

Example. The final example for today is presheaves and sheaves. If X is a topological space, then the category \mathbf{Ab}(X) of abelian sheaves on X has an internal Hom given by

    \[\mathbf{Hom}(\mathscr F, \mathscr G)(U) = \operatorname{Hom}(\mathscr F|_U, \mathscr G|_U),\]

with the obvious transition maps for inclusions V \subseteq U of open sets. This is usually called the sheaf Hom. A similar statement holds for presheaves.

Not every open immersion is an open immersion

An immersion (or locally closed immersion) of schemes is a morphism f \colon X \to Y that can be factored as X \to U \to Y, where X \to U is a closed immersion and U \to Y is an open immersion. If it is moreover an open morphism, it need not be an open immersion:

Example. Let X be a nonreduced scheme, and let X_{\text{red}} \to X be the reduction. This is a closed immersion, whose underlying set is the entire space. Thus, it is a homeomorphism, hence an open morphism. It is not an open immersion, for that would force it to be an isomorphism. \qedsymbol

Remark. However, every closed immersion is a closed immersion; see Tag 01IQ.

Higher pushforwards along finite morphisms

This post is about one of my favourite answers I have given on MathOverflow, although it seems to have gone by mostly unnoticed. In the post, Qixiao asks (essentially) the following:

Question. If f \colon X \to Y is a finite morphism of schemes, is the pushforward f_* \colon \Sh(X) \to \Sh(Y) exact?

Note that this is true on the subcategory of quasicoherent sheaves because affine morphisms have no quasicoherent higher pushforwards. Also, in the étale topology the pushforward along a finite morphism is exact on the category of all abelian sheaves; see e.g. Tag 03QP.

However, we show that the answer to the question above is negative.

Example. Let Y be the spectrum of a DVR (R,\mathfrak m), let R \to S be a finite extension of domains such that S has exactly two primes \mathfrak p, \mathfrak q above \mathfrak m, and let X = \Spec S. For example, R = \Z_{(5)} and S = \Z_{(5)}[i], or R = k[x]_{(x)} and S = k[x]_{(x)}[\sqrt{x+1}] if you prefer a more geometric example.

By my previous post, the global sections functor \Gamma \colon \Sh(Y) \to \Ab is exact. If the same were true for f_* \colon \Sh(X) \to \Sh(Y), then the global sections functor on X would be exact as well. Thus, it suffices to prove that this is not the case, i.e. to produce a surjection \mathscr F \to \mathscr G of sheaves on X such that the map on global sections is not surjective.

The topological space of X consists of closed points x,y and a generic point \eta. Let U = \{\eta\} and Z = U^{\operatorname{c}} = \{x,y\}; then U is open and Z is closed. Hence, for any sheaf \mathscr F on X, we have a short exact sequence (see e.g. Tag 02UT)

    \[0 \to j_! (\mathscr F|_U) \to \mathscr F \to i_* (\mathscr F|_Z) \to 0,\]

where j \colon U \to X and i \colon Z \to X are the inclusions. Let \mathscr F be the constant sheaf \Z; then the same goes for \mathscr F|_U and \mathscr F|_Z. Then the map

    \[H^0(X,\mathscr F) \to H^0(X,i_*(\mathscr F|_Z)) = H^0(Z,\mathscr F|_Z)\]

is given by the diagonal map \Z \to \Z \oplus \Z, since X is connected by Z has two connected components. This is visibly not surjective. \qedsymbol

Cohomology of a local scheme

This is a continuation of my previous post on local schemes. Here is a ridiculous lemma.

Lemma. Let (X,x) be a local scheme, and let \mathscr F be any abelian sheaf on X. Then H^i(X,\mathscr F) = 0 for all i > 0.

Proof. It suffices to show that the global sections functor \Gamma \colon \Sh(X) \to \Ab is exact. Let \mathscr F \to \mathscr G be a surjection of abelian sheaves on X, and let s \in \mathscr G(X) be a global section. Then s can be lifted to a section of \mathscr F in an open neighbourhood U of x. But the only open neighbourhood of x is X. Thus, s can be lifted to a section of \mathscr F(X). \qedsymbol

What’s going on is that the functors \mathscr F \mapsto \Gamma(X,\mathscr F) and \mathscr F \mapsto \mathscr F_x are naturally isomorphic, due to the absence of open neighbourhoods of x.

Remark. It seems believable that there are suitable site-theoretic versions of this lemma as well. For example, a strictly Henselian local ring has no higher cohomology in the étale topology. The argument is essentially the same: every open neighbourhood of the closed point has a section; see e.g. the proof of Tag 03QO.

Local schemes

Consider the following definition. It seems to be standard, although I have not found a place where it is actually spelled out in this way.

Definition. A pointed scheme (X,x) is local if x is contained in every nonempty closed subset of X.

Example. If (A,\mathfrak m) is a local ring, then (\Spec A,\mathfrak m) is a local scheme. Indeed, \mathfrak m is contained in every nonempty closed subset V(I) \subseteq X, because every strict ideal I \subsetneq A is contained in \mathfrak m.

We prove that this is actually the only example.

Lemma. Let (X,x) be a local scheme. Then X is affine, and A = \Gamma(X,\mathcal O_X) is a local ring whose maximal ideal corresponds to the point x \in X = \Spec A.

Proof. Let U be an affine open neighbourhood of x. Then the complement V is a closed set not containing x, hence V = \varnothing. Thus, X = U is affine. Let A = \Gamma(X,\mathcal O_X). Let \mathfrak m be a maximal ideal of A; then V(\mathfrak m) = \{\mathfrak m\}. Since this contains x, we must have x = \mathfrak m, i.e. x corresponds to the (necessarily unique) maximal ideal \mathfrak m \subseteq A. \qedsymbol

Classification of compact objects in Top

In my previous post, I showed that compact objects in the category of topological spaces have to be finite. Today we improve this to a full characterisation.

Lemma. Let X be a topological space. Then X is a compact object in \operatorname{\underline{Top}} if and only if X is finite discrete.

This result dates back to Gabriel and Ulmer [GU71, 6.4], as was pointed out to me by Jiří Rosický in reply to my MO question and answer of this account (of which this post is essentially a retelling). Our proof is different from the one given in [GU71], instead using a variant of an argument given in the n-Lab.

Before giving the proof, we construct an auxiliary space against which we will be testing compactness. It is essentially the colimit constructed in the n-Lab, except that we swapped the roles of 0 and 1 (the reason for this will become clear in the proof).

Definition. For all n \in \mathbb N, let X_n be the topological space \mathbb N_{\geq n} \times \{0,1\}, where the nonempty open sets are given by U_{n,m} = \mathbb N_{\geq m} \times \{0\} \cup \mathbb N_{\geq n} \times \{1\} for m \geq n. They form a topology since

    \begin{align*} U_{n,m_1} \cap U_{n,m_2} &= U_{n, \max(m_1,m_2)}, \\ \bigcup_i U_{n,m_i} &= U_{n,\min\{m_i\}}. \end{align*}

Define the map f_n \colon X_n \to X_{n+1} by

    \[(x,\varepsilon) \mapsto \left\{\begin{array}{ll} (x,\varepsilon), & x > n, \\ (n+1,\varepsilon), & x = n. \end{array}\right.\]

This is continuous since f_n^{-1}(U_{n+1,m}) equals U_{n,m} if m > n+1 and U_{n,n} if m = n+1. Let X_\infty be the colimit of this diagram.

Since the elements (x,\varepsilon), (y,\varepsilon) \in X_n map to the same element in X_{\max(x,y)}, we conclude that X_\infty is the two-point space \{0,1\}, where the map X_n \to X_\infty = \{0,1\} is the second coordinate projection. Moreover, the colimit topology on \{0,1\} is the indiscrete topology. Indeed, neither \mathbb N_{\geq n} \times \{0\} \subseteq X_n nor \mathbb N_{\geq n} \times \{1\} \subseteq X_n are open.

Proof of Lemma. If X is compact, then my previous post shows that X is finite. Let U \subseteq X be any subset, and let f \colon X \to X_\infty = \{0,1\} be the indicator function \mathbb I_U. It is continuous because X_\infty has the indiscrete topology. Since X is a compact object, f has to factor through some g \colon X \to X_n. Let h \colon X \to X_n \to \N_{\geq n} be the first coordinate projection, i.e.

    \[g(x) = \left\{\begin{array}{ll}(h(x),1), & x \in U, \\ (h(x),0), & x \not\in U. \end{array}\right.\]

Let m \in \N_{\geq n} be a number such that m > h(x) for all x \not\in U; this exists because X is finite. Then g^{-1}(U_{n,m}) = U, which shows that U is open. Since U was arbitrary, we conclude that X is discrete.

Conversely, every finite discrete space X is a compact object. Indeed, any map out of X is continuous, and finite sets are compact in \operatorname{\underline{Set}}. \qedsymbol

[GU71] Gabriel, Peter and Ulmer, Friedrich, Lokal präsentierbare Kategorien. Lecture Notes in Mathematics 221. Springer-Verlag, Berlin-New York, 1971. DOI: 10.1007/BFb0059396.