A little corner in the world of set-theoretic topology

This post puts a spot light on a little corner in the world of set-theoretic topology. There lies in this corner a simple topological statement that opens a door to the esoteric world of independence results. In this post, we give a proof of this basic fact and discuss its ramifications. This basic result is an excellent entry point to the study of S and L spaces.

The following paragraph is found in the paper called Gently killing S-spaces by Todd Eisworth, Peter Nyikos and Saharon Shelah [1]. The basic fact in question is highlighted in blue.

A simultaneous generalization of hereditarily separable and hereditarily Lindelof spaces is the class of spaces of countable spread – those spaces in which every discrete subspace is countable. One of the basic facts in this little corner of set-theoretic topology is that if a regular space of countable spread is not hereditarily separable, it contains an L-space, and if it is not hereditarily Lindelof, it contains an S-space. [1]

The same basic fact is also mentioned in the paper called The spread of regular spaces by Judith Roitman [2].

It is also well known that a regular space of countable spread which is not hereditarily separable contains an L-space and a regular space of countable spread which is not hereditarily Lindelof contains an S-space. Thus an absolute example of a space satisfying (Statement) A would contain a proof of the existence of S and L space – a consummation which some may devoutly wish, but which this paper does not attempt. [2]

Statement A in [2] is: There exists a 0-dimensional Hausdorff space of countable spread that is not the union of a hereditarily separable and a hereditarily Lindelof space. Statement A would mean the existence of a regular space of countable spread that is not hereditarily separable and that is also not hereditarily Lindelof. By the well known fact just mentioned, statement A would imply the existence of a space that is simultaneously an S-space and an L-space!

Let’s unpack the preceding section. First some basic definitions. A space X is of countable spread (has countable spread) if every discrete subspace of X is countable. A space X is hereditarily separable if every subspace of X is separable. A space X is hereditarily Lindelof if every subspace of X is Lindelof. A space is an S-space if it is hereditarily separable but not Lindelof. A space is an L-space if it is hereditarily Lindelof but not separable. See [3] for a basic discussion of S and L spaces.

Hereditarily separable but not Lindelof spaces as well as hereditarily Lindelof but not separable spaces can be easily defined in ZFC [3]. However, such examples are not regular. For the notions of S and L-spaces to be interesting, the definitions must include regularity. Thus in the discussion that follows, all spaces are assumed to be Hausdorff and regular.

One amazing aspect about set-theoretic topology is that one sometimes does not have to stray far from basic topological notions to encounter pathological objects such as S-spaces and L-spaces. The definition of a topological space is of course a basic definition. Separable spaces and Lindelof spaces are basic notions that are not far from the definition of topological spaces. The same can be said about hereditarily separable and hereditarily Lindelof spaces. Out of these basic ingredients come the notion of S-spaces and L-spaces, the existence of which is one of the key motivating questions in set-theoretic topology in the twentieth century. The study of S and L-spaces is a body of mathematics that had been developed for nearly a century. It is a fruitful area of research at the boundary of topology and axiomatic set theory.

The existence of an S-space is independent of ZFC (as a result of the work by Todorcevic in early 1980s). This means that there is a model of set theory in which an S-space exists and there is also a model of set theory in which S-spaces cannot exist. One half of the basic result mentioned in the preceding section is intimately tied to the existence of S-spaces and thus has interesting set-theoretic implications. The other half of the basic result involves the existence of L-spaces, which are shown to exist without using extra set theory axioms beyond ZFC by Justin Moore in 2005, which went against the common expectation that the existence of L-spaces would be independent of ZFC as well.

Let’s examine the basic notions in a little more details. The following diagram shows the properties surrounding the notion of countable spread.

Diagram 1 – Properties surrounding countable spread

The implications (the arrows) in Diagram 1 can be verified easily. Central to the discussion at hand, both hereditarily separable and hereditarily Lindelof imply countable spread. The best way to see this is that if a space has an uncountable discrete subspace, that subspace is simultaneously a non-separable subspace and a non-Lindelof subspace. A natural question is whether these implications can be reversed. Another question is whether the properties in Diagram 1 can be related in other ways. The following diagram attempts to ask these questions.

Diagram 2 – Reverse implications surrounding countable spread

Not shown in Diagram 2 are these four facts: separable \not \rightarrow hereditarily separable, Lindelof \not \rightarrow hereditarily Lindelof, separable \not \rightarrow countable spread and Lindelof \not \rightarrow countable spread. The examples supporting these facts are not set-theoretic in nature and are not discussed here.

Let’s focus on each question mark in Diagram 2. The two horizontal arrows with question marks at the top are about S-space and L-space. If X is hereditarily separable, then is X hereditarily Lindelof? A “no” answer would mean there is an S-space. A “yes” answer would mean there exists no S-space. So the top arrow from left to right is independent of ZFC. Since an L-space can be constructed within ZFC, the question mark in the top arrow in Diagram 2 from right to left has a “no” answer.

Now focus on the arrows emanating from countable spread in Diagram 2. These arrows are about the basic fact discussed earlier. From Diagram 1, we know that hereditarily separable implies countable spread. Can the implication be reversed? Any L-space would be an example showing that the implication cannot be reversed. Note that any L-space is of countable spread and is not separable and hence not hereditarily separable. Since L-space exists in ZFC, the question mark in the arrow from countable spread to hereditarily separable has a “no” answer. The same is true for the question mark in the arrow from countable spread to separable

We know that hereditarily Lindelof implies countable spread. Can the implication be reversed? According to the basic fact mentioned earlier, if the implication cannot be reversed, there exists an S-space. Thus if S-space does not exist, the implication can be reversed. Any S-space is an example showing that the implication cannot be reversed. Thus the question in the arrow from countable spread to hereditarily Lindelof cannot be answered without assuming axioms beyond ZFC. The same is true for the question mark for the arrow from countable spread to Lindelf.

Diagram 2 is set-theoretic in nature. The diagram is remarkable in that the properties in the diagram are basic notions that are only brief steps away from the definition of a topological space. Thus the basic highlighted here is a quick route to the world of independence results.

We now give a proof of the basic result, which is stated in the following theorem.

Theorem 1
Let X is regular and Hausdorff space. Then the following is true.

  • If X is of countable spread and is not a hereditarily separable space, then X contains an L-space.
  • If X is of countable spread and is not a hereditarily Lindelof space, then X contains an S-space.

To that end, we use the concepts of right separated space and left separated space. Recall that an initial segment of a well-ordered set (X,<) is a set of the form \{y \in X: y<x \} where x \in X. A space X is a right separated space if X can be well-ordered in such a way that every initial segment is open. A right separated space is in type \kappa if the well-ordering is of type \kappa. A space X is a left separated space if X can be well-ordered in such a way that every initial segment is closed. A left separated space is in type \kappa if the well-ordering is of type \kappa. The following results are used in proving Theorem 1.

Theorem A
Let X is regular and Hausdorff space. Then the following is true.

  • The space X is hereditarily separable space if and only if X has no uncountable left separated subspace.
  • The space X is hereditarily Lindelof space if and only if X has no uncountable right separated subspace.

Proof of Theorem A
\Longrightarrow of the first bullet point.
Suppose Y \subset X is an uncountable left separated subspace. Suppose that the well-ordering of Y is of type \kappa where \kappa>\omega. Further suppose that Y=\{ x_\alpha: \alpha<\kappa \} such that for each \alpha<\kappa, C_\alpha=\{ x_\beta: \beta<\alpha \} is a closed subset of Y. Since \kappa is uncountable, the well-ordering has an initial segment of type \omega_1. So we might as well assume \kappa=\omega_1. Note that for any countable A \subset Y, A \subset C_\alpha for some \alpha<\omega_1. It follows that Y is not separable. This means that X is not hereditarily separable.

\Longleftarrow of the first bullet point.
Suppose that X is not hereditarily separable. Let Y \subset X be a subspace that is not separable. We now inductively derive an uncountable left separated subspace of Y. Choose y_0 \in Y. For each \alpha<\omega_1, let A_\alpha=\{ y_\beta \in Y: \beta <\alpha \}. The set A_\alpha is the set of all the points of Y chosen before the step at \alpha<\omega_1. Since A_\alpha is countable, its closure in Y is not the entire space Y. Choose y_\alpha \in Y-\overline{A_\alpha}=O_\alpha.

Let Y_L=\{ y_\alpha: \alpha<\omega_1 \}. We claim that Y_L is a left separated space. To this end, we need to show that each initial segment A_\alpha is a closed subset of Y_L. Note that for each \gamma \ge \alpha, O_\gamma=Y-\overline{A_\gamma} is an open subset of Y with y_\gamma \in O_\gamma such that O_\gamma \cap \overline{A_\gamma}=\varnothing and thus O_\gamma \cap \overline{A_\alpha}=\varnothing (closure in Y). Then U_\gamma=O_\gamma \cap Y_L is an open subset of Y_L containing y_\gamma such that U_\gamma \cap A_\alpha=\varnothing. It follows that Y-A_\alpha is open in Y_L and that A_\alpha is a closed subset of Y_L.

\Longrightarrow of the second bullet point.
Suppose Y \subset X is an uncountable right separated subspace. Suppose that the well-ordering of Y is of type \kappa where \kappa>\omega. Further suppose that Y=\{ x_\alpha: \alpha<\kappa \} such that for each \alpha<\kappa, U_\alpha=\{ x_\beta: \beta<\alpha \} is an open subset of Y.

Since \kappa is uncountable, the well-ordering has an initial segment of type \omega_1. So we might as well assume \kappa=\omega_1. Note that \{ U_\alpha: \alpha<\omega_1 \} is an open cover of Y that has no countable subcover. It follows that Y is not Lindelof. This means that X is not hereditarily Lindelof.

\Longleftarrow of the second bullet point.
Suppose that X is not hereditarily Lindelof. Let Y \subset X be a subspace that is not Lindelof. Let \mathcal{U} be an open cover of Y that has no countable subcover. We now inductively derive a right separated subspace of Y of type \omega_1.

Choose U_0 \in \mathcal{U} and choose y_0 \in U_0. Choose y_1 \in Y-U_0 and choose U_1 \in \mathcal{U} such that y_1 \in U_1. Let \alpha<\omega_1. Suppose that points y_\beta and open sets U_\beta, \beta<\alpha, have been chosen such that y_\beta \in Y-\bigcup_{\delta<\beta} U_\delta and y_\beta \in U_\beta. The countably many chosen open sets U_\beta, \beta<\alpha, cannot cover Y. Choose y_\alpha \in Y-\bigcup_{\beta<\alpha} U_\beta. Choose U_\alpha \in \mathcal{U} such that y_\alpha \in U_\alpha.

Let Y_R=\{ y_\alpha: \alpha<\omega_1 \}. It follows that Y_R is a right separated space. Note that for each \alpha<\omega_1, \{ y_\beta: \beta<\alpha \} \subset \bigcup_{\beta<\alpha} U_\beta and the open set \bigcup_{\beta<\alpha} U_\beta does not contain y_\gamma for any \gamma \ge \alpha. This means that the initial segment \{ y_\beta: \beta<\alpha \} is open in Y_L. \square

Lemma B
Let X be a space that is a right separated space and also a left separated space based on the same well ordering. Then X is a discrete space.

Proof of Lemma B
Let X=\{ w_\alpha: \alpha<\kappa \} such that the well-ordering is given by the ordinals in the subscripts, i.e. w_\beta<w_\gamma if and only if \beta<\gamma. Suppose that X with this well-ordering is both a right separated space and a left separated space. We claim that every point is a discrete point, i.e. \{ x_\alpha \} is open for any \alpha<\kappa.

To see this, fix \alpha<\kappa. The initial segment A_\alpha=\{ w_\beta: \beta<\alpha \} is closed in X since X is a left separated space. On the other hand, the initial segment \{ w_\beta: \beta < \alpha+1  \} is open in X since X is a right separated space. Then B_{\alpha}=\{ w_\beta: \beta \ge \alpha+1  \} is closed in X. It follows that \{ x_\alpha \} must be open since X=A_\alpha \cup B_\alpha \cup \{ w_\alpha \}. \square

Theorem C
Let X is regular and Hausdorff space. Then the following is true.

  • Suppose the space X is right separated space of type \omega_1. Then if X has no uncountable discrete subspaces, then X is an S-space or X contains an S-space.
  • Suppose the space X is left separated space of type \omega_1. Then if X has no uncountable discrete subspaces, then X is an L-space or X contains an L-space.

Proof of Theorem C
For the first bullet point, suppose the space X is right separated space of type \omega_1. Then by Theorem A, X is not hereditarily Lindelof. If X is hereditarily separable, then X is an S-space (if X is not Lindelof) or X contains an S-space (a non-Lindelof subspace of X). Suppose X is not hereditarily separable. By Theorem A, X has an uncountable left separated subspace of type \omega_1.

Let X=\{ x_\alpha: \alpha<\omega_1 \} such that the well-ordering represented by the ordinals in the subscripts is a right separated space. Let <_R be the symbol for the right separated well-ordering, i.e. x_\beta <_R \ x_\delta if and only if \beta<\delta. As indicated in the preceding paragraph, X has an uncountable left separated subspace. Let Y=\{ y_\alpha \in X: \alpha<\omega_1 \} be this left separated subspace. Let <_L be the symbol for the left separated well-ordering. The well-ordering <_R may be different from the well-ordering <_L. However, we can obtain an uncountable subset of Y such that the two well-orderings coincide on this subset.

To start, pick any y_\gamma in Y and relabel it t_0. The final segment \{y_\beta \in Y: t_0 <_L \ y_\beta \} must intersect the final segment \{x_\beta \in X: t_0 <_R \ x_\beta \} in uncountably many points. Choose the least such point (according to <_R) and call it t_1. It is clear how t_{\delta+1} is chosen if t_\delta has been chosen.

Suppose \alpha<\omega_1 is a limit ordinal and that t_\beta has been chosen for all \beta<\alpha. Then the set \{y_\tau: \forall \ \beta<\alpha, t_\beta <_L \ y_\tau \} and the set \{x_\tau: \forall \ \beta<\alpha, t_\beta <_R \ x_\tau \} must intersect in uncountably many points. Choose the least such point and call it t_\alpha (according to <_R). As a result, we have obtained T=\{ t_\alpha: \alpha<\omega_1 \}. It follows that T with the well-ordering represented by the ordinals in the subscript is a subset of (X,<_R) and a subset of (Y,<_L). Thus T is both right separated and left separated.

By Lemma B, T is a discrete subspace of X. However, X is assumed to have no uncountable discrete subspace. Thus if X has no uncountable discrete subspace, then X must be hereditarily separable and as a result, must be an S-space or must contain an S-space.

The proof for the second bullet point is analogous to that of the first bullet point. \square

We are now ready to prove Theorem 1.

Proof of Theorem 1
Suppose that X is of countable spread and that X is not hereditarily separable. By Theorem A, X has an uncountable left separated subspace Y (assume it is of type \omega_1). The property of countable spread is hereditary. So Y is of countable spread. By Theorem C, Y is an L-space or Y contains an L-space. In either way, X contains an L-space.

Suppose that X is of countable spread and that X is not hereditarily Lindelof. By Theorem A, X has an uncountable right separated subspace Y (assume it is of type \omega_1). By Theorem C, Y is an S-space or Y contains an S-space. In either way, X contains an S-space.

Reference

  1. Eisworth T., Nyikos P., Shelah S., Gently killing S-spaces, Israel Journal of Mathmatics, 136, 189-220, 2003.
  2. Roitman J., The spread of regular spaces, General Topology and Its Applications, 8, 85-91, 1978.
  3. Roitman, J., Basic S and L, Handbook of Set-Theoretic Topology, (K. Kunen and J. E. Vaughan, eds), Elsevier Science Publishers B. V., Amsterdam, 295-326, 1984.
  4. Tatch-Moore J., A solution to the L space problem, Journal of the American Mathematical Society, 19, 717-736, 2006.

\text{ }

\text{ }

\text{ }

Dan Ma math

Daniel Ma mathematics

\copyright 2018 – Dan Ma

Advertisements

Every space is star discrete

The statement in the title is a folklore fact, though the term star discrete is usually not used whenever this well known fact is invoked in the literature. We present a proof to this well known fact. We also discuss some related concepts.

All spaces are assumed to be Hausdorff and regular.

First, let’s define the star notation. Let X be a space. Let \mathcal{U} be a collection of subsets of X. Let A \subset X. Define \text{St}(A,\mathcal{U}) to be the set \bigcup \{U \in \mathcal{U}: U \cap A \ne \varnothing \}. In other words, the set \text{St}(A,\mathcal{U}) is simply the union of all elements of \mathcal{U} that contains points of the set A. The set \text{St}(A,\mathcal{U}) is also called the star of the set A with respect to the collection \mathcal{U}. If A=\{ x \}, we use the notation \text{St}(x,\mathcal{U}) instead of \text{St}( \{ x \},\mathcal{U}). The following is the well known result in question.

Lemma 1
Let X be a space. For any open cover \mathcal{U} of X, there exists a discrete subspace A of X such that X=\text{St}(A,\mathcal{U}). Furthermore, the set A can be chosen in such a way that it is also a closed subset of the space X.

Any space that satisfies the condition in Lemma 1 is said to be a star discrete space. The proof shown below will work for any topological space. Hence every space is star discrete. We come across three references in which the lemma is stated or is used – Lemma IV.2.20 in page 135 of [3], page 137 of [2] and [1]. The first two references do not use the term star discrete. Star discrete is mentioned in [1] since that paper focuses on star properties. This property that is present in every topological space is at heart a covering property. Here’s a rewording of the lemma that makes it look like a covering property.

Lemma 1a
Let X be a space. For any open cover \mathcal{U} of X, there exists a discrete subspace A of X such that \{ \text{St}(x,\mathcal{U}): x \in A \} is a cover of X. Furthermore, the set A can be chosen in such a way that it is also a closed subset of the space X.

Lemma 1a is clearly identical to Lemma 1. However, Lemma 1a makes it extra clear that this is a covering property. For every open cover of a space, instead of finding a sub cover or an open refinement, we find a discrete subspace so that the stars of the points of the discrete subspace with respect to the given open cover also cover the space.

Lemma 1a naturally leads to other star covering properties. For example, a space X is said to be a star countable space if for any open cover \mathcal{U} of X, there exists a countable subspace A of X such that \{ \text{St}(x,\mathcal{U}): x \in A \} is a cover of X. A space X is said to be a star Lindelof space if for any open cover \mathcal{U} of X, there exists a Lindelof subspace A of X such that \{ \text{St}(x,\mathcal{U}): x \in A \} is a cover of X. In general, for any topological property \mathcal{P}, a space X is a star \mathcal{P} space if for any open cover \mathcal{U} of X, there exists a subspace A of X with property \mathcal{P} such that \{ \text{St}(x,\mathcal{U}): x \in A \} is a cover of X.

It follows that every Lindelof space is a star countable space. It is also clear that every star countable space is a star Lindelof space.

Lemma 1 or Lemma 1a, at first glance, may seem like a surprising result. However, one can argue that it is not a strong result at all since the property is possessed by every space. Indeed, the lemma has nothing to say about the size of the discrete set. It only says that there exists a star cover based on a discrete set for a given open cover. To derive more information about the given space, we may need to work with more information on the space in question.

Consider spaces such that every discrete subspace is countable (such a space is said to have countable spread or a space of countable spread). Also consider spaces such that every closed and discrete subspace is countable (such a space is said to have countable extent or a space of countable extent). Any space that has countable spread is also a space that has countable extent for the simple reason that if every discrete subspace is countable, then every closed and discrete subspace is countable.

Then it follows from Lemma 1 that any space X that has countable extent is star countable. Any star countable space is obviously a star Lindelof space. The following diagram displays these relationships.

Countable spread and Lindelof property

According to the diagram, the star countable and star Lindelof are both downstream from the countable spread property and the Lindelof property. The star properties being downstream from the Lindelof property is not surprising. What is interesting is that if a space has countable spread, then it is star countable and hence star Lindelof.

Do “countable spread” and “Lindelof” relate to each other? Lindelof spaces do not have to have countable spread. The simplest example is the one-point compactification of an uncountable discrete space. More specifically, let X be an uncountable discrete space. Let p be a point not in X. Then Y=X \cup \{ p \} is a compact space (hence Lindelof) where X is discrete and an open neighborhood of p is of the form \{ p \} \cup U where X-U is a finite subset of X. The space Y is not of countable spread since X is an uncountable discrete subspace.

Does “countable spread” imply “Lindelof”? Is there a non-Lindelof space that has countable spread? It turns out that the answers are independent of ZFC. The next post has more details.

We now give a proof to Lemma 1. Suppose that X is an infinite space (if it is finite, the lemma is true since the space is Hausdorff). Let \kappa=\lvert X \lvert. Let \kappa^+ be the next cardinal greater than \kappa. Let \mathcal{U} be an open cover of the space X. Choose x_0 \in X. We choose a sequence of points x_0,x_1,\cdots,x_\alpha,\cdots inductively. If \text{St}(\{x_\beta: \beta<\alpha \},\mathcal{U}) \ne X, we can choose a point x_\alpha \in X such that x_\alpha \notin \text{St}(\{x_\beta: \beta<\alpha \},\mathcal{U}).

We claim that the induction process must stop at some \alpha<\kappa^+. In other words, at some \alpha<\kappa^+, the star of the previous points must be the entire space and we run out of points to choose. Otherwise, we would have obtained a subset of X with cardinality \kappa^+, a contradiction. Choose the least \alpha<\kappa^+ such that \text{St}(\{x_\beta: \beta<\alpha \},\mathcal{U}) = X. Let A=\{x_\beta: \beta<\alpha \}.

Then it can be verified that the set A is a discrete subspace of X and that A is a closed subset of X. Note that x_\beta \in \text{St}(x_\beta, \mathcal{U}) while x_\gamma \notin \text{St}(x_\beta, \mathcal{U}) for all \gamma \ne \beta. This follows from the way the points are chosen in the induction process. On the other hand, for any x \in X-A, x \in \text{St}(x_\beta, \mathcal{U}) for some \beta<\alpha. As discussed, the open set \text{St}(x_\beta, \mathcal{U}) contains only one point of A, namely x_\beta.

Reference

  1. Alas O., Jumqueira L., van Mill J., Tkachuk V., Wilson R.On the extent of star countable spaces, Cent. Eur. J. Math., 9 (3), 603-615, 2011.
  2. Alster, K., Pol, R.,On function spaces of compact subspaces of \Sigma-products of the real line, Fund. Math., 107, 35-46, 1980.
  3. Arkhangelskii, A. V.,Topological Function Spaces, Mathematics and Its Applications Series, Kluwer Academic Publishers, Dordrecht, 1992.

\text{ }

\text{ }

\text{ }

Dan Ma math

Daniel Ma mathematics

\copyright 2018 – Dan Ma

Michael line and Morita’s conjectures

This post discusses Michael line from the point of view of the three conjectures of Kiiti Morita.

K. Morita defined the notion of P-spaces in [7]. The definition of P-spaces is discussed here in considerable details. K. Morita also proved that a space X is a normal P-space if and only if the product X \times Y is normal for every metrizable space Y. As a result of this characterization, the notion of normal P-space (a space that is a normal space and a P-space) is useful in the study of products of normal spaces. Just to be clear, we say a space is a non-normal P-space (i.e. a space that is not a normal P-space) if the space is a normal space that is not a P-space.

K. Morita formulated his three conjectures in 1976. The statements of the conjectures are given below. Here is a basic discussion of the three conjectures. The notion of normal P-spaces is a theme that runs through the three conjectures. The conjectures are actually theorems since 2001 [2].

Here’s where Michael line comes into the discussion. Based on the characterization of normal P-spaces mentioned above, to find a normal space that is not a P-space (a non-normal P-space), we would need to find a non-normal product X \times Y such that one of the factors is a metric space and the other factor is a normal space. The first such example in ZFC is from an article by E. Michael in 1963 (found here and here). In this example, the normal space is M, which came be known as the Michael line, and the metric space is \mathbb{P}, the space of irrational numbers (as a subspace of the real line). Their product M \times \mathbb{P} is not normal. A basic discussion of the Michael line is found here.

Because M \times \mathbb{P} is not normal, the Michael line M is not a normal P-space. Prior to E. Michael’s 1963 article, we have to reach back to 1955 to find an example of a non-normal product where one factor is a metric space. In 1955, M. E. Rudin used a Souslin line to construct a Dowker space, which is a normal space whose product with the closed unit interval is not normal. The existence of a Souslin line was shown to be independent of ZFC in the late 1960s. In 1971, Rudin constructed a Dowker space in ZFC. Thus finding a normal space that is not a normal P-space (finding a non-normal product X \times Y where one factor is a metric space and the other factor is a normal space) is not a trivial matter.

Morita’s Three Conjectures

We show that the Michael line illustrates perfectly the three conjectures of K. Morita. Here’s the statements.

Morita’s Conjecture I. Let X be a space. If the product X \times Y is normal for every normal space Y then X is a discrete space.

Morita’s Conjecture II. Let X be a space. If the product X \times Y is normal for every normal P-space Y then X is a metrizable space.

Morita’s Conjecture III. Let X be a space. If the product X \times Y is normal for every normal countably paracompact space Y then X is a metrizable \sigma-locally compact space.

The contrapositive statement of Morita’s conjecture I is that for any non-discrete space X, there exists a normal space Y such that X \times Y is not normal. Thus any non-discrete space is paired with a normal space for forming a non-normal product. The Michael line M is paired with the space of irrational numbers \mathbb{P}. Obviously, the space \mathbb{P} is paired with the Michael line M.

The contrapositive statement of Morita’s conjecture II is that for any non-metrizable space X, there exists a normal P-space Y such that X \times Y is not normal. The pairing is more specific than for conjecture I. Any non-metrizable space is paired with a normal P-space to form a non-normal product. As illustration, the Michael line M is not metrizable. The space \mathbb{P} of irrational numbers is a metric space and hence a normal P-space. Here, M is paired with \mathbb{P} to form a non-normal product.

The contrapositive statement of Morita’s conjecture III is that for any space X that is not both metrizable and \sigma-locally compact, there exists a normal countably paracompact space Y such that X \times Y is not normal. Note that the space \mathbb{P} is not \sigma-locally compact (see Theorem 4 here). The Michael line M is paracompact and hence normal and countably paracompact. Thus the metric non-\sigma-locally compact \mathbb{P} is paired with normal countably paracompact M to form a non-normal product. Here, the metric space \mathbb{P} is paired with the non-normal P-space M.

In each conjecture, each space in a certain class of spaces is paired with one space in another class to form a non-normal product. For Morita’s conjecture I, each non-discrete space is paired with a normal space. For conjecture II, each non-metrizable space is paired with a normal P-space. For conjecture III, each metrizable but non-\sigma-locally compact is paired with a normal countably paracompact space to form a non-normal product. Note that the paired normal countably paracompact space would be a non-normal P-space.

Michael line as an example of a non-normal P-space is a great tool to help us walk through the three conjectures of Morita. Are there other examples of non-normal P-spaces? Dowker spaces mentioned above (normal spaces whose products with the closed unit interval are not normal) are non-normal P-spaces. Note that conjecture II guarantees a normal P-space to match every non-metric space for forming a non-normal product. Conjecture III guarantees a non-normal P-space to match every metrizable non-\sigma-locally compact space for forming a non-normal product. Based on the conjectures, examples of normal P-spaces and non-normal P-spaces, though may be hard to find, are guaranteed to exist.

We give more examples below to further illustrate the pairings for conjecture II and conjecture III. As indicated above, non-normal P-spaces are hard to come by. Some of the examples below are constructed using additional axioms beyond ZFC. The additional examples still give an impression that the availability of non-normal P-spaces, though guaranteed to exist, is limited.

Examples of Normal P-Spaces

One example is based on this classic theorem: for any normal space X, X is paracompact if and only if the product X \times \beta X is normal. Here \beta X is the Stone-Cech compactification of the completely regular space X. Thus any normal but not paracompact space X (a non-metrizable space) is paired with \beta X, a normal P-space, to form a non-normal product.

Naturally, the next class of non-metrizable spaces to be discussed should be the paracompact spaces that are not metrizable. If there is a readily available theorem to provide a normal P-space for each non-metrizable paracompact space, then there would be a simple proof of Morita’s conjecture II. The eventual solution of conjecture II is far from simple [2]. We narrow the focus to the non-metrizable compact spaces.

Consider this well known result: for any infinite compact space X, the product \omega_1 \times X is normal if and only if the space X has countable tightness (see Theorem 1 here). Thus any compact space with uncountable tightness is paired with \omega_1, the space of all countable ordinals, to form a non-normal product. The space \omega_1, being a countably compact space, is a normal P-space. A proof that normal countably compact space is a normal P-space is given here.

We now handle the case for non-metrizable compact spaces with countable tightness. In this case, compactness is not needed. For spaces with countable tightness, consider this result: every space with countable tightness, whose products with all perfectly normal spaces are normal, must be metrizable [3] (see Corollary 7). Thus any non-metrizable space with countable tightness is paired with some perfectly normal space to form a non-normal product. Any reader interested in what these perfectly normal spaces are can consult [3]. Note that perfectly normal spaces are normal P-spaces (see here for a proof).

Examples of Non-Normal P-Spaces

Another non-normal product is X_B \times B where B \subset \mathbb{R} is a Bernstein set and X_B is the space with the real line as the underlying set such that points in B are isolated and points in \mathbb{R}-B retain the usual open sets. The set B \subset \mathbb{R} is said to be a Bernstein set if every uncountable closed subset of the real line contains a point in B and contains a point in the complement of B. Such a set can be constructed using transfinite induction as shown here. The product X_B \times B is not normal where B is considered a subspace of the real line. The proof is essentially the same proof that shows M \times \mathbb{P} is not normal (see here). The space X_B is a Lindelof space. It is not a normal P-space since its product with B, a separable metric space, is not normal. However, this example is essentially the same example as the Michael line since the same technique and proof are used. On the one hand, the X_B \times B example seems like an improvement over Michael line example since the first factor X_B is Lindelof. On the other hand, it is inferior than the Michael line example since the second factor B is not completely metrizable.

Moving away from the idea of Michael, there exist a Lindelof space and a completely metrizable (but not separable) space whose product is of weight \omega_1 and is not normal [5]. This would be a Lindelof space that is a non-normal P-space. However, this example is not as elementary as the Michael line, making it not as effective as an illustration of Morita’s three conjectures.

The next set of non-normal P-spaces requires set theory. A Michael space is a Lindelof space whose product with \mathbb{P}, the space of irrational numbers, is not normal. Michael problem is the question: is there a Michael space in ZFC? It is known that a Michael space can be constructed using continuum hypothesis [6] or using Martin’s axiom [1]. The construction using continuum hypothesis has been discussed in this blog (see here). The question of whether there exists a Michael space in ZFC is still unsolved.

The existence of a Michael space is equivalent to the existence of a Lindelof space and a separable completely metrizable space whose product is non-normal [4]. A Michael space, in the context of the discussion in this post, is a non-normal P-space.

The discussion in this post shows that the example of the Michael line and other examples of non-normal P-spaces are useful tools to illustrate Morita’s three conjectures.

Reference

  1. Alster K.,On the product of a Lindelof space and the space of irrationals under Martin’s Axiom, Proc. Amer. Math. Soc., Vol. 110, 543-547, 1990.
  2. Balogh Z.,Normality of product spaces and Morita’s conjectures, Topology Appl., Vol. 115, 333-341, 2001.
  3. Chiba K., Przymusinski T., Rudin M. E.Nonshrinking open covers and K. Morita’s duality conjectures, Topology Appl., Vol. 22, 19-32, 1986.
  4. Lawrence L. B., The influence of a small cardinal on the product of a Lindelof space and the irrationals, Proc. Amer. Math. Soc., 110, 535-542, 1990.
  5. Lawrence L. B., A ZFC Example (of Minimum Weight) of a Lindelof Space and a Completely Metrizable Space with a Nonnormal Product, Proc. Amer. Math. Soc., 124, No 2, 627-632, 1996.
  6. Michael E., Paracompactness and the Lindelof property in nite and countable cartesian products, Compositio Math., 23, 199-214, 1971.
  7. Morita K., Products of Normal Spaces with Metric Spaces, Math. Ann., Vol. 154, 365-382, 1964.
  8. Rudin M. E., A Normal Space X for which X \times I is not Normal, Fund. Math., 73, 179-186, 1971.

\text{ }

\text{ }

\text{ }

Dan Ma math

Daniel Ma mathematics

\copyright 2018 – Dan Ma

Three conjectures of K Morita

This post discusses the three conjectures that were proposed by K. Morita in 1976. These conjectures concern normality in product spaces. To start the discussion, here’s the conjectures.

Morita’s Conjecture I. Let X be a space. The product X \times Y is normal for every normal space Y if and only if X is a discrete space.

Morita’s Conjecture II. Let X be a space. The product X \times Y is normal for every normal P-space Y if and only if X is a metrizable space.

Morita’s Conjecture III. Let X be a space. The product X \times Y is normal for every normal countably paracompact space Y if and only if X is a metrizable \sigma-locally compact space.

These statements are no longer conjectures. Partial results appeared after the conjectures were proposed in 1976. The complete resolution of the conjectures came in 2001 in a paper by Zoli Balogh [5]. Though it is more appropriate to call these statements theorems, it is still convenient to call them conjectures. Just know that they are now known results rather open problems to be solved. The focus here is not on the evolution of the solutions. Instead, we discuss the relations among the three conjectures and why they are amazing results in the study of normality in product spaces.

As discussed below, in each of these conjectures, one direction is true based on prior known theorems (see Theorem 1, Theorem 2 and Theorem 4 below). The conjectures can be stated as follows.

Morita’s Conjecture I. Let X be a space. If the product X \times Y is normal for every normal space Y then X is a discrete space.

Morita’s Conjecture II. Let X be a space. If the product X \times Y is normal for every normal P-space Y then X is a metrizable space.

Morita’s Conjecture III. Let X be a space. If the product X \times Y is normal for every normal countably paracompact space Y then X is a metrizable \sigma-locally compact space.

P-spaces are defined by K. Morita [11]. He proved that a space X is a normal P-space if and only if the product X \times Y is normal for every metrizable space Y (see theorem 2 below). Normal P-spaces are also discussed here. A space X is \sigma-locally compact space if X is the union of countably many locally compact subspaces each of which is also closed subspace of X.

As we will see below, these conjectures are also called duality conjectures because they are duals of known results.

[2] is a survey of Morita’s conjecture.

Duality Conjectures

Here’s three theorems that are duals to the conjectures.

Theorem 1
Let X be a space. The product space X \times Y is normal for every discrete space Y if and only if X is normal.

Theorem 2
Let X be a space. The product space X \times Y is normal for every metrizable space Y if and only if X is a normal P-space.

Theorem 3
Let X be a space. The product space X \times Y is normal for every metrizable \sigma-locally compact space Y if and only if X is normal countably paracompact.

The key words in red are for emphasis. In each of these three theorems, if we switch the two key words in red, we would obtain the statements for the conjectures. In this sense, the conjectures are called duality conjectures since they are duals of known results.

Theorem 1 is actually not found in the literature. It is an easy theorem. Theorem 2, found in [11], is a characterization of normal P-space (discussed here). Theorem 3 is a well known result based on the following theorem by K. Morita [10].

Theorem 4
Let Y be a metrizable space. Then the product X \times Y is normal for every normal countably paracompact space X if and only if Y is a \sigma-locally compact space.

We now show that Theorem 3 can be established using Theorem 4. Theorem 4 is also Theorem 3.5 in p. 111 of [2]. A proof of Theorem 4 is found in Theorem 1.8 in p. 130 of [8].

Proof of Theorem 3
\Longleftarrow Suppose X is normal and countably paracompact. Let Y be a metrizable \sigma-locally compact space. By Theorem 4, X \times Y is normal.

\Longrightarrow This direction uses Dowker’s theorem. We give a contrapositive proof. Suppose that X is not both normal and countably paracompact. Case 1. X is not normal. Then X \times \{ y \} is not normal where \{ y \} is any one-point discrete space. Case 2. X is normal and not countably paracompact. This means that X is a Dowker space. Then X \times [0,1] is not normal. In either case, X \times Y is not normal for some compact metric space. Thus X \times Y is not normal for some \sigma-locally compact metric space. This completes the proof of Theorem 3. \square

The First and Third Conjectures

The first conjecture of Morita was proved by Atsuji [1] and Rudin [13] in 1978. The proof in [13] is a constructive proof. The key to that solution is to define a \kappa-Dowker space. Suppose X is a non-discrete space. Let \kappa be the least cardinal of a non-discrete subspace of X. Then construct a \kappa-Dowker space Y as in [13]. It follows that X \times Y is not normal. The proof that X \times Y is not normal is discussed here.

Conjecture III was confirmed by Balogh in 1998 [4]. We show here that the first and third conjectures of Morita can be confirmed by assuming the second conjecture.

Conjecture II implies Conjecture I
We give a contrapositive proof of Conjecture I. Suppose that X is not discrete. We wish to find a normal space Y such that X \times Y is not normal. Consider two cases for X. Case 1. X is not metrizable. By Conjecture II, X \times Y is not normal for some normal P-space Y. Case 2. X is metrizable. Since X is infinite and metric, X would contain an infinite compact metric space S. For example, X contains a non-trivial convergent sequence and let S be a convergence sequence plus the limit point. Let Y be a Dowker space. Then the product S \times Y is not normal. It follows that X \times Y is not normal. Thus there exists a normal space Y such that X \times Y is not normal in either case. \square

Conjecture II implies Conjecture III
Suppose that the product X \times Y is normal for every normal and countably paracompact space Y. Since any normal P-space is a normal countably paracompact space, X \times Y is normal for every normal and P-space Y. By Conjecture II, X is metrizable. By Theorem 4, X is \sigma-locally compact. \square

The Second Conjecture

The above discussion shows that a complete solution to the three conjectures hinges on the resolution of the second conjecture. A partial resolution came in 1986 [6]. In that paper, it was shown that under V = L, conjecture II is true.

The complete solution of the second conjecture is given in a paper of Balogh [5] in 2001. The path to Balogh’s proof is through a conjecture of M. E. Rudin identified as Conjecture 9.

Rudin’s Conjecture 9. There exists a normal P-space X such that some uncountable increasing open cover of X cannot be shrunk.

Conjecture 9 was part of a set of 14 conjectures stated in [14]. It is also discussed in [7]. In [6], conjecture 9 was shown to be equivalent to Morita’s second conjecture. In [5], Balogh used his technique for constructing a Dowker space of cardinality continuum to obtain a space as described in conjecture 9.

The resolution of conjecture II is considered to be one of Balogh greatest hits [3].

Abundance of Non-Normal Products

One immediate observation from Morita’s conjecture I is that existence of non-normal products is wide spread. Conjecture I indicates that every normal non-discrete space X is paired with some normal space Y such that their product is not normal. So every normal non-discrete space forms a non-normal product with some normal space. Given any normal non-discrete space (no matter how nice it is or how exotic it is), it can always be paired with another normal space (sometimes paired with itself) for a non-normal product.

Suppose we narrow the focus to spaces that are normal and non-metrizable. Then any such space X is paired with some normal P-space Y to form a non-normal product space (Morita’s conjecture II). By narrowing the focus on X to the non-metrizable spaces, we obtain more clarity on the paired space to form non-normal product, namely a normal P-space. As an example, let X be the Michael line (normal and non-metrizable). It is well known that X in this case is paired with \mathbb{P}, the space of irrational numbers with the usual Euclidean topology, to form a non-normal product (discussed here).

Another example is X being the Sorgenfrey line. It is well known that X in this case is paired with itself to form a non-normal product (discussed here). Morita’s conjectures are powerful indication that these two non-normal products are not isolated phenomena.

Another interesting observation about conjecture II is that normal P-spaces are not productive with respect to normality. More specifically, for any non-metrizable normal P-space X, conjecture II tells us that there exists another normal P-space Y such that X \times Y is not normal.

Now we narrow the focus to spaces that are metrizable but not \sigma-locally compact. For any such space X, conjecture III tells us that X is paired with a normal countably paracompact space Y to form a non-normal product. Using the Michael line example, this time let X=\mathbb{P}, the space of irrational numbers, which is a metric space that is not \sigma-locally compact. The paired normal and countably paracompact space Y is the Michael line.

Each conjecture is about existence of a normal Y that is paired with a given X to form a non-normal product. For Conjecture I, the given X is from a wide class (normal non-discrete). As a result, there is not much specific information on the paired Y, other than that it is normal. For Conjectures II and III, the given space X is from narrower classes. As a result, there is more information on the paired Y.

The concept of Dowker spaces runs through the three conjectures, especially the first conjecture. Dowker spaces and \kappa-Dowker spaces provide reliable pairing for non-normal products. In fact this is one way to prove conjecture I [13], also see here. For any normal space X with a countable non-discrete subspace, the product of X and any Dowker space is not normal (discussed here). For any normal space X such that the least cardinality of a non-discrete subspace is an uncountable cardinal \kappa, the product X \times Y is not normal where Y is a \kappa-Dowker space as constructed in [13], also discussed here.

In finding a normal pair Y for a normal space X, if we do not care about Y having a high degree of normal productiveness (e.g. normal P or normal countably paracompact), we can always let Y be a Dowker space or \kappa-Dowker space. In fact, if the starting space X is a metric space, the normal pair for a non-normal product (by definition) has to be a Dowker space. For example, if X=[0,1], then the normal space Y such that X \times Y is by definition a Dowker space. The search for a Dowker space spanned a period of 20 years. For the real line \mathbb{R}, the normal pair for a non-normal product is also a Dowker space. For “nice” spaces such as metric spaces, finding a normal space to form non-normal product is no trivial problem.

Reference

  1. Atsuji M.,On normality of the product of two spaces, General Topology and Its Relation to Modern Analysis and Algebra (Proc. Fourth Prague Topology sympos., 1976), Part B, 25–27, 1977.
  2. Atsuji M.,Normality of product spaces I, in: K. Morita, J. Nagata (Eds.), Topics in General
    Topology, North-Holland, Amsterdam, 81–116, 1989.
  3. Burke D., Gruenhage G.,Zoli, Top. Proc., Vol. 27, No 1, i-xxii, 2003.
  4. Balogh Z.,Normality of product spaces and K. Morita’s third conjecture, Topology Appl., Vol. 84, 185-198, 1998.
  5. Balogh Z.,Normality of product spaces and Morita’s conjectures, Topology Appl., Vol. 115, 333-341, 2001.
  6. Chiba K., Przymusinski T., Rudin M. E.Nonshrinking open covers and K. Morita’s duality conjectures, Topology Appl., Vol. 22, 19-32, 1986.
  7. Gruenhage G.,Mary Ellen’s Conjectures,, Special Issue honoring the memory of Mary Ellen Rudin, Topology Appl., Vol. 195, 15-25, 2015.
  8. Hoshina T.,Normality of product spaces II, in: K. Morita, J. Nagata (Eds.), Topics in General Topology, North-Holland, Amsterdam, 121–158, 1989.
  9. Morita K., On the Product of a Normal Space with a Metric Space, Proc. Japan Acad., Vol. 39, 148-150, 1963. (article information; paper)
  10. Morita K., Products of Normal Spaces with Metric Spaces II, Sci. Rep. Tokyo Kyoiku Dagaiku Sec A, 8, 87-92, 1963.
  11. Morita K., Products of Normal Spaces with Metric Spaces, Math. Ann., Vol. 154, 365-382, 1964.
  12. Morita K., Nagata J., Topics in General Topology, Elsevier Science Publishers, B. V., The Netherlands, 1989.
  13. Rudin M. E., \kappa-Dowker Spaces, Czechoslovak Mathematical Journal, 28, No.2, 324-326, 1978.
  14. Rudin M. E., Some conjectures, in: Open Problems in Topology, J. van Mill and G.M. Reed,
    eds., North Holland, 184–193, 1990.
  15. Telgárski R., A characterization of P-spaces, Proc. Japan Acad., Vol. 51, 802–807, 1975.

\text{ }

\text{ }

\text{ }

Dan Ma math

Daniel Ma mathematics

\copyright 2018 – Dan Ma

Morita’s normal P-space

In this post we discuss K. Morita’s notion of P-space, which is a useful and interesting concept in the study of normality of product spaces.

The Definition

In [1] and [2], Morita defined the notion of P-spaces. First some notations. Let \kappa be a cardinal number such that \kappa \ge 1. Conveniently, \kappa is identified by the set of all ordinals preceding \kappa. Let \Gamma be the set of all finite sequences (\alpha_1,\alpha_2,\cdots,\alpha_n) where n=1,2,\cdots and all \alpha_i < \kappa. Let X be a space. The collection \left\{A_\sigma \subset X: \sigma \in \Gamma \right\} is said to be decreasing if this condition holds: for any \sigma \in \Gamma and \delta \in \Gamma with

    \sigma =(\alpha_1,\alpha_2,\cdots,\alpha_n)

    \delta =(\beta_1,\beta_2,\cdots,\beta_n, \cdots, \beta_m)

such that n<m and such that \alpha_i=\beta_i for all i \le n, we have A_{\delta} \subset A_{\sigma}. On the other hand, the collection \left\{A_\sigma \subset X: \sigma \in \Gamma \right\} is said to be increasing if for any \sigma \in \Gamma and \delta \in \Gamma as described above, we have A_{\sigma} \subset A_{\delta}.

The space X is a P-space if for any cardinal \kappa \ge 1 and for any decreasing collection \left\{F_\sigma \subset X: \sigma \in \Gamma \right\} of closed subsets of X, there exists open set U_\sigma for each \sigma \in \Gamma with F_\sigma \subset U_\sigma such that for any countably infinite sequence (\alpha_1,\alpha_2,\cdots,\alpha_n,\cdots) where each finite subsequence \sigma_n=(\alpha_1,\alpha_2,\cdots,\alpha_n) is an element of \Gamma, if \bigcap_{n=1}^\infty F_{\sigma_n}=\varnothing, then \bigcap_{n=1}^\infty U_{\sigma_n}=\varnothing.

By switching closed sets and open sets and by switching decreasing collection and increasing collection, the following is an alternative but equivalent definition of P-spaces.

The space X is a P-space if for any cardinal \kappa \ge 1 and for any increasing collection \left\{U_\sigma \subset X: \sigma \in \Gamma \right\} of open subsets of X, there exists closed set F_\sigma for each \sigma \in \Gamma with F_\sigma \subset U_\sigma such that for any countably infinite sequence (\alpha_1,\alpha_2,\cdots,\alpha_n,\cdots) where each finite subsequence \sigma_n=(\alpha_1,\alpha_2,\cdots,\alpha_n) is an element of \Gamma, if \bigcup_{n=1}^\infty U_{\sigma_n}=X, then \bigcup_{n=1}^\infty F_{\sigma_n}=X.

Note that the definition is per cardinal number \kappa \ge 1. To bring out more precision, we say a space X is a P(\kappa)-space of it satisfies the definition for P-space for the cardinal \kappa. Of course if a space is a P(\kappa)-space for all \kappa \ge 1, then it is a P-space.

There is also a game characterization of P-spaces [4].

A Specific Case

It is instructive to examine a specific case of the definition. Let \kappa=1=\{ 0 \}. In other words, let’s look what what a P(1)-space looks like. The elements of the index set \Gamma are simply finite sequences of 0’s. The relevant information about an element of \Gamma is its length (i.e. a positive integer). Thus the closed sets F_\sigma in the definition are essentially indexed by integers. For the case of \kappa=1, the definition can be stated as follows:

For any decreasing sequence F_1 \supset F_2 \supset F_3 \cdots of closed subsets of X, there exist U_1,U_2,U_3,\cdots, open subsets of X, such that F_n \subset U_n for all n and such that if \bigcap_{n=1}^\infty F_n=\varnothing then \bigcap_{n=1}^\infty U_n=\varnothing.

The above condition implies the following condition.

For any decreasing sequence F_1 \supset F_2 \supset F_3 \cdots of closed subsets of X such that \bigcap_{n=1}^\infty F_n=\varnothing, there exist U_1,U_2,U_3,\cdots, open subsets of X, such that F_n \subset U_n for all n and such that \bigcap_{n=1}^\infty U_n=\varnothing.

The last condition is one of the conditions in Dowker’s Theorem (condition 6 in Theorem 1 in this post and condition 7 in Theorem 1 in this post). Recall that Dowker’s theorem states that a normal space X is countably paracompact if and only if the last condition holds if and only of the product X \times Y is normal for every infinite compact metric space Y. Thus if a normal space X is a P(1)-space, it is countably paracompact. More importantly P(1) space is about normality in product spaces where one factor is a class of metric spaces, namely the compact metric spaces.

Based on the above discussion, any normal space X that is a P-space is a normal countably paracompact space.

The definition for P(1)-space is identical to one combinatorial condition in Dowker’s theorem which says that any decreasing sequence of closed sets with empty intersection has an open expansion that also has empty intersection.

For P(\kappa)-space where \kappa>1, the decreasing family of closed sets are no longer indexed by the integers. Instead the decreasing closed sets are indexed by finite sequences of elements of \kappa. The index set \Gamma would be more like a tree structure. However the look and feel of P-space is like the combinatorial condition in Dowker’s theorem. The decreasing closed sets are expanded by open sets. For any “path in the tree” (an infinite sequence of elements of \kappa), if the closed sets along the path has empty intersection, then the corresponding open sets would have empty intersection.

Not surprisingly, the notion of P-spaces is about normality in product spaces where one factor is a metric space. In fact, this is precisely the characterization of P-spaces (see Theorem 1 and Theorem 2 below).

A Characterization of P-Space

Morita gave the following characterization of P-spaces among normal spaces. The following theorems are found in [2].

Theorem 1
Let X be a space. The space X is a normal P-space if and only if the product space X \times Y is normal for every metrizable space Y.

Thus the combinatorial definition involving decreasing families of closed sets being expanded by open sets is equivalent to a statement that is much easier to understand. A space that is normal and a P-space is precisely a normal space that is productively normal with every metric space. The following theorem is Theorem 1 broken out for each cardinal \kappa.

Theorem 2
Let X be a space and let \kappa \ge \omega. Then X is a normal P(\kappa)-space if and only if the product space X \times Y is normal for every metric space Y of weight \kappa.

Theorem 2 only covers the infinite cardinals \kappa starting with the countably infinite cardinal. Where are the P(n)-spaces placed where n are the positive integers? The following theorem gives the answer.

Theorem 3
Let X be a space. Then X is a normal P(2)-space if and only if the product space X \times Y is normal for every separable metric space Y.

According to Theorem 2, X is a normal P(\omega)-space if and only if the product space X \times Y is normal for every separable metric space Y. Thus suggests that any P(2)-space is a P(\omega)-space. It seems to say that P(2) is identical to P(\kappa) where \kappa is the countably infinite cardinal. The following theorem captures the idea.

Theorem 4
Let \kappa be the positive integers 2,3,4,\cdots or \kappa=\omega, the countably infinite cardinal. Let X be a space. Then X is a P(2)-space if and only if X is a P(\kappa)-space.

To give a context for Theorem 4, note that if X is a P(\kappa)-space, then X is a P(\tau)-space for any cardinal \tau less than \kappa. Thus if X is a P(3)-space, then it is a P(2)-space and also a P(1)-space. In the definition of P(\kappa)-space, the index set \Gamma is the set of all finite sequences of elements of \kappa. If the definition for P(\kappa)-space holds, it would also hold for the index set consisting of finite sequences of elements of \tau where \tau<\kappa. Thus if the definition for P(\omega)-space holds, it would hold for P(n)-space for all integers n.

Theorem 4 says that when the definition of P(2)-space holds, the definition would hold for all larger cardinals up to \omega.

In light of Theorem 1 and Dowker's theorem, we have the following corollary. If the product of a space X with every metric space is normal, then the product of X with every compact metric space is normal.

Corollary 5
Let X be a space. If X is a normal P-space, then X is a normal and countably paracompact space.

Examples of Normal P-Space

Here’s several classes of spaces that are normal P-spaces.

  • Metric spaces.
  • Compact spaces (link).
  • \sigma-compact spaces (link).
  • Paracompact locally compact spaces (link).
  • Paracompact \sigma-locally compact spaces (link).
  • Normal countably compact spaces (link).
  • Perfectly normal spaces (link).
  • \Sigma-product of real lines.

Clearly any metric space is a normal P-space since the product of any two metric spaces is a metric space. Any compact space is a normal P-space since the product of a compact space and a paracompact space is paracompact, hence normal. For each of the classes of spaces listed above, the product with any metric space is normal. See the corresponding links for proofs of the key theorems.

The \Sigma-product of real lines \Sigma_{\alpha<\tau} \mathbb{R} is a normal P-space. For any metric space Y, the product (\Sigma_{\alpha<\tau} \mathbb{R}) \times Y is a \Sigma-product of metric spaces. By a well known result, the \Sigma-product of metric spaces is normal.

Examples of Non-Normal P-Spaces

Paracompact \sigma-locally compact spaces are normal P-spaces since the product of such a space with any paracompact space is paracompact. However, the product of paracompact spaces in general is not normal. The product of Michael line (a hereditarily paracompact space) and the space of irrational numbers (a metric space) is not normal (discussed here). Thus the Michael line is not a normal P-space. More specifically the Michael line fails to be a normal P(2)-space. However, it is a normal P(1)-space (i.e. normal and countably paracompact space).

The Michael line is obtained from the usual real line topology by making the irrational points isolated. Instead of using the irrational numbers, we can obtain a similar space by making points in a Bernstein set isolated. The resulting space X is a Michael line-like space. The product of X with the starting Bernstein set (a subset of the real line with the usual topology) is not normal. Thus this is another example of a normal space that is not a P(2)-space. See here for the details of how this space is constructed.

To look for more examples, look for non-normal product X \times Y where one factor is normal and the other is a metric space.

More Examples

Based on the characterization theorem of Morita, normal P-spaces are very productively normal. Normal P-spaces are well behaved when taking product with metrizable spaces. However, they are not well behaved when taking product with non-metrizable spaces. Let’s look at several examples.

Consider the Sorgenfrey line. It is perfectly normal. Thus the product of the Sorgenfrey line with any metric space is also perfectly normal, hence normal. It is well known that the square of the Sorgenfrey line is not normal.

The space \omega_1 of all countable ordinals is a normal and countably compact space, hence a normal P-space. However, the product of \omega_1 and some compact spaces are not normal. For example, \omega_1 \times (\omega_1 +1) is not normal. Another example: \omega_1 \times I^I is not normal where I=[0,1]. The idea here is that the product of \omega_1 and any compact space with uncountable tightness is not normal (see here).

Compact spaces are normal P-spaces. As discussed in the preceding paragraph, the product of any compact space with uncountable tightness and the space \omega_1 is not normal.

Even as nice a space as the unit interval [0,1], it is not always productive. The product of [0,1] with a Dowker space is not normal (see here).

In general, normality is not preserved in the product space operation. the best we can ask for is that normal spaces be productively normal with respect to a narrow class of spaces. For normal P-spaces, that narrow class of spaces is the class of metric spaces. However, normal product is not a guarantee outside of the productive class in question.

Reference

  1. Morita K., On the Product of a Normal Space with a Metric Space, Proc. Japan Acad., Vol. 39, 148-150, 1963. (article information; paper)
  2. Morita K., Products of Normal Spaces with Metric Spaces, Math. Ann., Vol. 154, 365-382, 1964.
  3. Morita K., Nagata J., Topics in General Topology, Elsevier Science Publishers, B. V., The Netherlands, 1989.
  4. Telgárski R., A characterization of P-spaces, Proc. Japan Acad., Vol. 51, 802–807, 1975.

\text{ }

\text{ }

\text{ }

Dan Ma math

Daniel Ma mathematics

\copyright 2018 – Dan Ma

In between G-delta diagonal and submetrizable

This post discusses the property of having a G_\delta-diagonal and related diagonal properties. The focus is on the diagonal properties in between G_\delta-diagonal and submetrizability. The discussion is followed by a diagram displaying the relative strengths of these properties. Some examples and questions are discussed.

G-delta Diagonal

In any space Y, a subset A is said to be a G_\delta-set in the space Y (or A is a G_\delta-subset of Y) if A is the intersection of countably many open subsets of Y. A subset A of Y is an F_\sigma-set in Y (or A is an F_\sigma-subset of Y) if A is the union of countably closed subsets of the space Y. Of course, the set A is a G_\delta-set if and only if Y-A, the complement of A, is an F_\sigma-set.

The diagonal of the space X is the set \Delta=\{ (x,x): x \in X \}, which is a subset of the square X \times X. When the set \Delta is a G_\delta-set in the space X \times X, we say that the space X has a G_\delta-diagonal.

It is straightforward to verify that the space X is a Hausdorff space if and only if the diagonal \Delta is a closed subset of X \times X. As a result, if X is a Hausdorff space such that X \times X is perfectly normal, then the diagonal would be a closed set and thus a G_\delta-set. Such spaces, including metric spaces, would have a G_\delta-diagonal. Thus any metric space has a G_\delta-diagonal.

A space X is submetrizable if there is a metrizable topology that is weaker than the topology for X. Then the diagonal \Delta would be a G_\delta-set with respect to the weaker metrizable topology of X \times X and thus with respect to the orginal topology of X. This means that the class of spaces having G_\delta-diagonals also include the submetrizable spaces. As a result, Sorgenfrey line and Michael line have G_\delta-diagonals since the Euclidean topology are weaker than both topologies.

A space having a G_\delta-diagonal is a simple topological property. Such spaces form a wide class of spaces containing many familiar spaces. According to the authors in [2], the property of having a G_\delta-diagonal is an important ingredient of submetrizability and metrizability. For example, any compact space with a G_\delta-diagonal is metrizable (see this blog post). Any paracompact or Lindelof space with a G_\delta-diagonal is submetrizable. Spaces with G_\delta-diagonals are also interesting in their own right. It is a property that had been research extensively. It is also a current research topic; see [7].

A Closer Look

To make the discussion more interesting, let’s point out a few essential definitions and notations. Let X be a space. Let \mathcal{U} be a collection of subsets of X. Let A \subset X. The notation St(A, \mathcal{U}) refers to the set St(A, \mathcal{U})=\cup \{U \in \mathcal{U}: A \cap U \ne \varnothing \}. In other words, St(A, \mathcal{U}) is the union of all the sets in \mathcal{U} that intersect the set A. The set St(A, \mathcal{U}) is also called the star of the set A with respect to the collection \mathcal{U}.

If A=\{ x \}, we write St(x, \mathcal{U}) instead of St(\{ x \}, \mathcal{U}). Then St(x, \mathcal{U}) refers to the union of all sets in \mathcal{U} that contain the point x. The set St(x, \mathcal{U}) is then called the star of the point x with respect to the collection \mathcal{U}.

Note that the statement of X having a G_\delta-diagonal is defined by a statement about the product X \times X. It is desirable to have a translation that is a statement about the space X.

Theorem 1
Let X be a space. Then the following statements are equivalent.

  1. The space X has a G_\delta-diagonal.
  2. There exists a sequence \mathcal{U}_0,\mathcal{U}_1,\mathcal{U}_2,\cdots of open covers of X such that for each x \in X, \{ x \}=\bigcap \{ St(x, \mathcal{U}_n): n=0,1,2,\cdots \}.

The sequence of open covers in condition 2 is called a G_\delta-diagonal sequence for the space X. According to condition 2, at any given point, the stars of the point with respect to the open covers in the sequence collapse to the given point.

One advantage of a G_\delta-diagonal sequence is that it is entirely about points of the space X. Thus we can work with such sequences of open covers of X instead of the G_\delta-set \Delta in X \times X. Theorem 1 is not a word for word translation. However, the proof is quote natural.

Suppose that \Delta=\cap \{U_n: n=0,1,2,\cdots \} where each U_n is an open subset of X \times X. Then let \mathcal{U}_n=\{U \subset X: U \text{ open and } U \times U \subset U_n \}. It can be verify that \mathcal{U}_0,\mathcal{U}_1,\mathcal{U}_2,\cdots is a G_\delta-diagonal sequence for X.

Suppose that \mathcal{U}_0,\mathcal{U}_1,\mathcal{U}_2,\cdots is a G_\delta-diagonal sequence for X. For each n, let U_n=\cup \{ U \times U: U \in \mathcal{U}_n \}. It follows that \Delta=\bigcap_{n=0}^\infty U_n. \square

It is informative to compare the property of G_\delta-diagonal with the definition of Moore spaces. A development for the space X is a sequence \mathcal{D}_0,\mathcal{D}_1,\mathcal{D}_2,\cdots of open covers of X such that for each x \in X, \{ St(x, \mathcal{D}_n): n=0,1,2,\cdots \} is a local base at the point x. A space is said to be developable if it has a development. The space X is said to be a Moore space if X is a Hausdorff and regular space that has a development.

The stars of a given point with respect to the open covers of a development form a local base at the given point, and thus collapse to the given point. Thus a development is also a G_\delta-diagonal sequence. It then follows that any Moore space has a G_\delta-diagonal.

A point in a space is a G_\delta-point if the point is the intersection of countably many open sets. Then having a G_\delta-diagonal sequence implies that that every point of the space is a G_\delta-point since every point is the intersection of the stars of that point with respect to a G_\delta-diagonal sequence. In contrast, any Moore space is necessarily a first countable space since the stars of any given point with respect to the development is a countable local base at the given point. The parallel suggests that spaces with G_\delta-diagonals can be thought of as a weak form of Moore spaces (at least a weak form of developable spaces).

Regular G-delta Diagonal

We discuss other diagonal properties. The space X is said to have a regular G_\delta-diagonal if \Delta=\cap \{\overline{U_n}:n=0,1,2,\cdots \} where each U_n is an open subset of X \times X such that \Delta \subset U_n. This diagonal property also has an equivalent condition in terms of a diagonal sequence.

Theorem 2
Let X be a space. Then the following statements are equivalent.

  1. The space X has a regular G_\delta-diagonal.
  2. There exists a sequence \mathcal{U}_0,\mathcal{U}_1,\mathcal{U}_2,\cdots of open covers of X such that for every two distinct points x,y \in X, there exist open sets U and V with x \in U and y \in V and there also exists an n such that no member of \mathcal{U}_n intersects both U and V.

For convenience, we call the sequence described in Theorem 2 a regular G_\delta-diagonal sequence. It is clear that if the diagonal of a space is a regular G_\delta-diagonal, then it is a G_\delta-diagonal. It can also be verified that a regular G_\delta-diagonal sequence is also a G_\delta-diagonal sequence. To see this, let \mathcal{U}_0,\mathcal{U}_1,\mathcal{U}_2,\cdots be a regular G_\delta-diagonal sequence for X. Suppose that y \ne x and y \in \bigcap_k St(x, \mathcal{U}_k). Choose open sets U and V and an integer n guaranteed by the regular G_\delta-diagonal sequence. Since y \in St(x, \mathcal{U}_n), choose B \in \mathcal{U}_n such that x,y \in B. Then B would be an element of \mathcal{U}_n that meets both U and V, a contradiction. Then \{ x \}= \bigcap_k St(x, \mathcal{U}_k) for all x \in X.

To proof Theorem 2, suppose that X has a regular G_\delta-diagonal. Let \Delta=\bigcap_{k=0}^\infty \overline{U_k} where each U_k is open in X \times X and \Delta \subset U_k. For each k, let \mathcal{U}_k be the collection of all open subsets U of X such that U \times U \subset U_k. It can be verified that \{ \mathcal{U}_k \} is a regular G_\delta-diagonal sequence for X.

On the other hand, suppose that \{ \mathcal{U}_k \} is a regular G_\delta-diagonal sequence for X. For each k, let U_k=\cup \{U \times U: U \in \mathcal{U}_k \}. It can be verified that \Delta=\bigcap_{k=0}^\infty \overline{U_k}. \square

Rank-k Diagonals

Metric spaces and submetrizable spaces have regular G_\delta-diagonals. We discuss this fact after introducing another set of diagonal properties. First some notations. For any family \mathcal{U} of subsets of the space X and for any x \in X, define St^1(x, \mathcal{U})=St(x, \mathcal{U}). For any integer k \ge 2, let St^k(x, \mathcal{U})=St^{k-1}(St(x, \mathcal{U})). Thus St^{2}(x, \mathcal{U}) is the star of the star St(x, \mathcal{U}) with respect to \mathcal{U} and St^{3}(x, \mathcal{U}) is the star of St^{2}(x, \mathcal{U}) and so on.

Let X be a space. A sequence \mathcal{U}_0,\mathcal{U}_1,\mathcal{U}_2,\cdots of open covers of X is said to be a rank-k diagonal sequence of X if for each x \in X, we have \{ x \}=\bigcap_{j=0}^\infty St^k(x,\mathcal{U}_j). When the space X has a rank-k diagonal sequence, the space is said to have a rank-k diagonal. Clearly a rank-1 diagonal sequence is simply a G_\delta-diagonal sequence as defined in Theorem 1. Thus having a rank-1 diagonal is the same as having a G_\delta-diagonal.

It is also clear that having a higher rank diagonal implies having a lower rank diagonal. This follows from the fact that a rank k+1 diagonal sequence is also a rank k diagonal sequence.

The following lemma builds intuition of the rank-k diagonal sequence. For any two distinct points x and y of a space X, and for any integer d \ge 2, a d-link path from x to y is a set of open sets W_1,W_2,\cdots,W_d such that x \in W_1, y \in W_d and W_t \cap W_{t+1} \ne \varnothing for all t=1,2,\cdots,d-1. By default, a single open set W containing both x and y is a d-link path from x to y for any integer d \ge 1.

Lemma 3
Let X be a space. Let k be a positive integer. Let \mathcal{U}_0,\mathcal{U}_1,\mathcal{U}_2,\cdots be a sequence of open covers of X. Then the following statements are equivalent.

  1. The sequence \mathcal{U}_0,\mathcal{U}_1,\mathcal{U}_2,\cdots is a rank-k diagonal sequence for the space X.
  2. For any two distinct points x and y of X, there is an integer n such that y \notin St^k(x,\mathcal{U}_n).
  3. For any two distinct points x and y of X, there is an integer n such that there is no k-link path from x to y consisting of elements of \mathcal{U}_n.

It can be seen directly from definition that Condition 1 and Condition 2 are equivalent. For Condition 3, observe that the set St^k(x,\mathcal{U}_n) is the union of k types of open sets – open sets in \mathcal{U}_n containing x, open sets in \mathcal{U}_n that intersect the first type, open sets in \mathcal{U}_n that intersect the second type and so on down to the open sets in \mathcal{U}_n that intersect St^{k-1}(x,\mathcal{U}_n). A path is formed by taking one open set from each type.

We now show a few basic results that provide further insight on the rank-k diagonal.

Theorem 4
Let X be a space. If the space X has a rank-2 diagonal, then X is a Hausdorff space.

Theorem 5
Let X be a Moore space. Then X has a rank-2 diagonal.

Theorem 6
Let X be a space. If X has a rank-3 diagonal, then X has a regular G_\delta-diagonal.

Once Lemma 3 is understood, Theorem 4 is also easily understood. If a space X has a rank-2 diagonal sequence \{ \mathcal{U}_n \}, then for any two distinct points x and y, we can always find an n where there is no 2-link path from x to y. Then x and y can be separated by open sets in \mathcal{U}_n. Thus these diagonal ranking properties confer separation axioms. We usually start off a topology discussion by assuming a reasonable separation axiom (usually implicitly). The fact that the diagonal ranking gives a bonus makes it even more interesting. Apparently many authors agree since G_\delta-diagonal and related topics had been researched extensively over decades.

To prove Theorem 5, let \{ \mathcal{U}_n \} be a development for the space X. Let x and y be two distinct points of X. We claim that there exists some n such that y \notin St^2(x,\mathcal{U}_n). Suppose not. This means that for each n, y \in St^2(x,\mathcal{U}_n). This also means that St(x,\mathcal{U}_n) \cap St(y,\mathcal{U}_n) \ne \varnothing for each n. Choose x_n \in St(x,\mathcal{U}_n) \cap St(y,\mathcal{U}_n) for each n. Since X is a Moore space, \{ St(x,\mathcal{U}_n) \} is a local base at x. Then \{ x_n \} converges to x. Since \{ St(y,\mathcal{U}_n) \} is a local base at y, \{ x_n \} converges to y, a contradiction. Thus the claim that there exists some n such that y \notin St^2(x,\mathcal{U}_n) is true. By Lemma 3, a development for a Moore space is a rank-2 diagonal sequence.

To prove Theorem 6, let \{ \mathcal{U}_n \} be a rank-3 diagonal sequence for the space X. We show that \{ \mathcal{U}_n \} is also a regular G_\delta-diagonal sequence for X. Suppose x and y are two distinct points of X. By Lemma 3, there exists an n such that there is no 3-link path consisting of open sets in \mathcal{U}_n that goes from x to y. Choose U \in \mathcal{U}_n with x \in U. Choose V \in \mathcal{U}_n with y \in V. Then it follows that no member of \mathcal{U}_n can intersect both U and V (otherwise there would be a 3-link path from x to y). Thus \{ \mathcal{U}_n \} is also a regular G_\delta-diagonal sequence for X.

We now show that metric spaces have rank-k diagonal for all integer k \ge 1.

Theorem 7
Let X be a metrizable space. Then X has rank-k diagonal for all integers k \ge 1.

If d is a metric that generates the topology of X, and if \mathcal{U}_n is the collection of all open subsets with diameters \le 2^{-n} with respect to the metrix d then \{ \mathcal{U}_n \} is a rank-k diagonal sequence for X for any integer k \ge 1.

We instead prove Theorem 7 topologically. To this end, we use an appropriate metrization theorem. The following theorem is a good candidate.

Alexandrov-Urysohn Metrization Theorem. A space X is metrizable if and only if the space X has a development \{ \mathcal{U}_n \} such that for any U_1,U_2 \in \mathcal{U}_{n+1} with U_1 \cap U_2 \ne \varnothing, the set U_1 \cup U_2 is contained in some element of \mathcal{U}_n. See Theorem 1.5 in p. 427 of [5].

Let \{ \mathcal{U}_n \} be the development from Alexandrov-Urysohn Metrization Theorem. It is a development with a strong property. Each open cover in the development refines the preceding open cover in a special way. This refinement property allows us to show that it is a rank-k diagonal sequence for X for any integer k \ge 1.

First, we make a few observations about \{ \mathcal{U}_n \}. From the statement of the theorem, each \mathcal{U}_{n+1} is a refinement of \mathcal{U}_n. As a result of this observation, \mathcal{U}_{m} is a refinement of \mathcal{U}_n for any m>n. Furthermore, for each x \in X, \text{St}(x,\mathcal{U}_m) \subset \text{St}(x,\mathcal{U}_n) for any m>n.

Let x, y \in X with x \ne y. Based on the preceding observations, it follows that there exists some m such that \text{St}(x,\mathcal{U}_m) \cap \text{St}(y,\mathcal{U}_m)=\varnothing. We claim that there exists some integer h>m such that there are no k-link path from x to y consisting of open sets from \mathcal{U}_h. Then \{ \mathcal{U}_n \} is a rank-k diagonal sequence for X according to Lemma 3.

We show this claim is true for k=2. Observe that there cannot exist U_1, U_2 \in \mathcal{U}_{m+1} such that x \in U_1, y \in U_2 and U_1 \cap U_2 \ne \varnothing. If there exists such a pair, then U_1 \cup U_2 would be contained in \text{St}(x,\mathcal{U}_m) and \text{St}(y,\mathcal{U}_m), a contradiction. Putting it in another way, there cannot be any 2-link path U_1,U_2 from x to y such that the open sets in the path are from \mathcal{U}_{m+1}. According to Lemma 3, the sequence \{ \mathcal{U}_n \} is a rank-2 diagonal sequence for the space X.

In general for any k \ge 2, there cannot exist any k-link path U_1,\cdots,U_k from x to y such that the open sets in the path are from \mathcal{U}_{m+k-1}. The argument goes just like the one for the case for k=2. Suppose the path U_1,\cdots,U_k exists. Using the special property of \{ \mathcal{U}_n \}, the 2-link path U_1,U_2 is contained in some open set in \mathcal{U}_{m+k-2}. The path U_1,\cdots,U_k is now contained in a (k-1)-link path consisting of elements from the open cover \mathcal{U}_{m+k-2}. Continuing the refinement process, the path U_1,\cdots,U_k is contained in a 2-link path from x to y consisting of elements from \mathcal{U}_{m+1}. Like before this would lead to a contradiction. According to Lemma 3, \{ \mathcal{U}_n \} is a rank-k diagonal sequence for the space X for any integer k \ge 2.

Of course, any metric space already has a G_\delta-diagonal. We conclude that any metrizable space has a rank-k diagonal for any integer k \ge 1. \square

We have the following corollary.

Corollary 8
Let X be a submetrizable space. Then X has rank-k diagonal for all integer k \ge 1.

In a submetrizable space, the weaker metrizable topology has a rank-k diagonal sequence, which in turn is a rank-k diagonal sequence in the original topology.

Examples and Questions

The preceding discussion focuses on properties that are in between G_\delta-diagonal and submetrizability. In fact, one of the properties has infinitely many levels (rank-k diagonal for integers k \ge 1). We would like to have a diagram showing the relative strengths of these properties. Before we do so, consider one more diagonal property.

Let X be a space. The set A \subset X is said to be a zero-set in X if there is a continuous f:X \rightarrow [0,1] such that A=f^{-1}(0). In other words, a zero-set is a set that is the inverse image of zero for some continuous real-valued function defined on the space in question.

A space X has a zero-set diagonal if the diagonal \Delta=\{ (x,x): x \in X \} is a zero-set in X \times X. The space X having a zero-set diagonal implies that X has a regular G_\delta-diagonal, and thus a G_\delta-diagonal. To see this, suppose that \Delta=f^{-1}(0) where f:X \times X \rightarrow [0,1] is continuous. Then \Delta=\bigcap_{n=1}^\infty \overline{U_n} where U_n=f^{-1}([0,1/n)). Thus having a zero-set diagonal is a strong property.

We have the following diagram.

The diagram summarizes the preceding discussion. From top to bottom, the stronger properties are at the top. From left to right, the stronger properties are on the left. The diagram shows several properties in between G_\delta-diagonal at the bottom and submetrizability at the top.

Note that the statement at the very bottom is not explicitly a diagonal property. It is placed at the bottom because of the classic result that any compact space with a G_\delta-diagonal is metrizable.

In the diagram, “rank-k diagonal” means that the space has a rank-k diagonal where k \ge 1 is an integer, which in terms means that the space has a rank-k diagonal sequence as defined above. Thus rank-k diagonal is not to be confused with the rank of a diagonal. The rank of the diagonal of a given space is the largest integer k such that the space has a rank-k diagonal. For example, for a space that has a rank-2 diagonal but has no rank-3 diagonal, the rank of the diagonal is 2.

To further make sense of the diagram, let’s examine examples.

The Mrowka space is a classic example of a space with a G_\delta-diagonal that is not submetrizable (introduced here). Where is this space located in the diagram? The Mrowka space, also called Psi-space, is defined using a maximal almost disjoint family of subsets of \omega. We denote such a space by \Psi(\mathcal{A}) where \mathcal{A} is a maximal almost disjoint family of subsets of \omega. It is a pseudocompact Moore space that is not submetrizable. As a Moore space, it has a rank-2 diagonal sequence. A well known result states that any pseudocompact space with a regular G_\delta-diagonal is metrizable (see here). As a non-submetrizable space, the Mrowka space cannot have a regular G_\delta-diagonal. Thus \Psi(\mathcal{A}) is an example of a space with a rank-2 diagonal but not a rank-3 diagonal sequence.

Examples of non-submetrizable spaces with stronger diagonal properties are harder to come by. We discuss examples that are found in the literature.

Example 2.9 in [2] is a Tychonoff separable Moore space Z that has a rank-3 diagonal but not of higher diagonal rank. As a result of not having a rank-4 diagonal, Z is not submetrizable. Thus Z is an example of a space with rank-3 diagonal (hence with a regular G_\delta-diagonal) that is not submetrizable. According to a result in [6], any separable space with a zero-set diagonal is submetrizable. Then the space Z is an example of a space with a regular G_\delta-diagonal that does not have a zero-set diagonal. In fact, the authors of [2] indicated that this is the first such example.

Example 2.9 of [2] shows that having a rank-3 diagonal does not imply having a zero-set diagonal. If a space is strengthened to have a rank-4 diagonal, does it imply having a zero-set diagonal? This is essentially Problem 2.13 in [2].

On the other hand, having a rank-3 diagonal implies a rank-2 diagonal. If we weaken the hypothesis to just having a regular regular G_\delta-diagonal, does it imply having a rank-2 diagonal? This is essentially Problem 2.14 in [2].

The authors of [2] conjectured that for each n, there exists a space X_n with a rank-n diagonal but not having a rank-(n+1) diagonal. This conjecture was answered affirmatively in [8] by constructing, for each integer k \ge 4, a Tychonoff space with a rank-k diagonal but not having a rank-(k+1) diagonal. Thus even for high k, a non-submetrizable space can be found with rank-k diagonal.

One natural question is this. Is there a non-submetrizable space that has rank-k diagonal for all k \ge 1? We have not seen this question stated in the literature. But it is clearly a natural question.

Example 2.17 in [2] is a non-submetrizable Moore space that has a zero-set diagonal and has rank-3 diagonal exactly (i.e. it does not have a higher rank diagonal). This example shows that having a zero-set diagonal does not imply having a rank-4 diagonal. A natural question is then this. Does having a zero-set diagonal imply having a rank-3 diagonal? This appears to be an open question. This is hinted by Problem 2.19 in [2]. It asks, if X is a normal space with a zero-set diagonal, does X have at least a rank-2 diagonal?

The property of having a G_\delta-diagonal and related properties is a topic that had been researched extensively over the decades. It is still an active topic of research. The discussion in this post only touches on the surface. There are many other diagonal properties not covered here. To further investigate, check with the papers listed below and also consult with information available in the literature.

Reference

  1. Arhangelskii A. V., Burke D. K., Spaces with a regular G_\delta-diagonal, Topology and its Applications, Vol. 153, No. 11, 1917–1929, 2006.
  2. Arhangelskii A. V., Buzyakova R. Z., The rank of the diagonal and submetrizability, Comment. Math. Univ. Carolinae, Vol. 47, No. 4, 585-597, 2006.
  3. Buzyakova R. Z., Cardinalities of ccc-spaces with regular G_\delta-diagonals, Topology and its Applications, Vol. 153, 1696–1698, 2006.
  4. Buzyakova R. Z., Observations on spaces with zeroset or regular G_\delta-diagonals, Comment. Math. Univ. Carolinae, Vol. 46, No. 3, 469-473, 2005.
  5. Gruenhage, G., Generalized Metric Spaces, Handbook of Set-Theoretic Topology (K. Kunen and J. E. Vaughan, eds), Elsevier Science Publishers B. V., Amsterdam, 423-501, 1984.
  6. Martin H. W., Contractibility of topological spaces onto metric spaces, Pacific J. Math., Vol. 61, No. 1, 209-217, 1975.
  7. Xuan Wei-Feng, Shi Wei-Xue, On spaces with rank k-diagonals or zeroset diagonals, Topology Proceddings, Vol. 51, 245{251, 2018.
  8. Yu Zuoming, Yun Ziqiu, A note on the rank of diagonals, Topology and its Applications, Vol. 157, 1011–1014, 2010.

\text{ }

\text{ }

\text{ }

Dan Ma math

Daniel Ma mathematics

\copyright 2018 – Dan Ma

Pseudocompact spaces with regular G-delta diagonals

This post complements two results discussed in two previous blog posts concerning G_\delta-diagonal. One result is that any compact space with a G_\delta-diagonal is metrizable (see here). The other result is that the compactness in the first result can be relaxed to countably compactness. Thus any countably compact space with a G_\delta-diagonal is metrizable (see here). The countably compactness in the second result cannot be relaxed to pseudocompactness. The Mrowka space is a pseudocompact space with a G_\delta-diagonal that is not submetrizable, hence not metrizable (see here). However, if we strengthen the G_\delta-diagonal to a regular G_\delta-diagonal while keeping pseudocompactness fixed, then we have a theorem. We prove the following theorem.

Theorem 1
If the space X is pseudocompact and has a regular G_\delta-diagonal, then X is metrizable.

All spaces are assumed to be Hausdorff and completely regular. The assumption of completely regular is crucial. The proof of Theorem 1 relies on two lemmas concerning pseudocompact spaces (one proved in a previous post and one proved here). These two lemmas work only for completely regular spaces.

The proof of Theorem 1 uses a metrization theorem. The best metrization to use in this case is Moore metrization theorem (stated below). The result in Theorem 1 is found in [2].

First some basics. Let X be a space. The diagonal of the space X is the set \Delta=\{ (x,x): x \in X \}. When the diagonal \Delta, as a subset of X \times X, is a G_\delta-set, i.e. \Delta is the intersection of countably many open subsets of X \times X, the space X is said to have a G_\delta-diagonal.

The space X is said to have a regular G_\delta-diagonal if the diagonal \Delta is a regular G_\delta-set in X \times X, i.e. \Delta=\bigcap_{n=1}^\infty \overline{U_n} where each U_n is an open subset of X \times X with \Delta \subset U_n. If \Delta=\bigcap_{n=1}^\infty \overline{U_n}, then \Delta=\bigcap_{n=1}^\infty \overline{U_n}=\bigcap_{n=1}^\infty U_n. Thus if a space has a regular G_\delta-diagonal, it has a G_\delta-diagonal. We will see that there exists a space with a G_\delta-diagonal that fails to be a regular G_\delta-diagonal.

The space X is a pseudocompact space if for every continuous function f:X \rightarrow \mathbb{R}, the image f(X) is a bounded set in the real line \mathbb{R}. Pseudocompact spaces are discussed in considerable details in this previous post. We will rely on results from this previous post to prove Theorem 1.

The following lemma is used in proving Theorem 1.

Lemma 2
Let X be a pseudocompact space. Suppose that O_1,O_2,O_2,\cdots is a decreasing sequence of non-empty open subsets of X such that \bigcap_{n=1}^\infty O_n=\bigcap_{n=1}^\infty \overline{O_n}=\{ x \} for some point x \in X. Then \{ O_n \} is a local base at the point x.

Proof of Lemma 2
Let O_1,O_2,O_2,\cdots be a decreasing sequence of open subsets of X such that \bigcap_{n=1}^\infty O_n=\bigcap_{n=1}^\infty \overline{O_n}=\{ x \}. Let U be open in X with x \in U. If O_n \subset U for some n, then we are done. Suppose that O_n \not \subset U for each n.

Choose open V with x \in V \subset \overline{V} \subset U. Consider the sequence \{ O_n \cap (X-\overline{V}) \}. This is a decreasing sequence of non-empty open subsets of X. By Theorem 2 in this previous post, \bigcap \overline{O_n \cap (X-\overline{V})} \ne \varnothing. Let y be a point in this non-empty set. Note that y \in \bigcap_{n=1}^\infty \overline{O_n}. This means that y=x. Since x \in \overline{O_n \cap (X-\overline{V})} for each n, any open set containing x would contain a point not in \overline{V}. This is a contradiction since x \in V. Thus it must be the case that x \in O_n \subset U for some n. \square

The following metrization theorem is useful in proving Theorem 1.

Theorem 3 (Moore Metrization Theorem)
Let X be a space. Then X is metrizable if and only if the following condition holds.

There exists a decreasing sequence \mathcal{B}_1,\mathcal{B}_2,\mathcal{B}_3,\cdots of open covers of X such that for each x \in X, the sequence \{ St(St(x,\mathcal{B}_n),\mathcal{B}_n):n=1,2,3,\cdots \} is a local base at the point x.

For any family \mathcal{U} of subsets of X, and for any A \subset X, the notation St(A,\mathcal{U}) refers to the set \cup \{U \in \mathcal{U}: U \cap A \ne \varnothing \}. In other words, it is the union of all sets in \mathcal{U} that contain points of A. The set St(A,\mathcal{U}) is also called the star of the set A with respect to the family \mathcal{U}. If A=\{ x \}, we write St(x,\mathcal{U}) instead of St(\{ x \},\mathcal{U}). The set St(St(x,\mathcal{B}_n),\mathcal{B}_n) indicated in Theorem 3 is the star of the set St(x,\mathcal{B}_n) with respect to the open cover \mathcal{B}_n.

Theorem 3 follows from Theorem 1.4 in [1], which states that for any T_0-space X, X is metrizable if and only if there exists a sequence \mathcal{G}_1, \mathcal{G}_2, \mathcal{G}_3,\cdots of open covers of X such that for each open U \subset X and for each x \in U, there exist an open V \subset X and an integer n such that x \in V and St(V,\mathcal{G}_n) \subset U.

Proof of Theorem 1

Suppose X is pseudocompact such that its diagonal \Delta=\bigcap_{n=1}^\infty \overline{U_n} where each U_n is an open subset of X \times X with \Delta \subset U_n. We can assume that U_1 \supset U_2 \supset \cdots. For each n \ge 1, define the following:

    \mathcal{U}_n=\{ U \subset X: U \text{ open in } X \text{ and } U \times U \subset U_n \}

Note that each \mathcal{U}_n is an open cover of X. Also note that \{ \mathcal{U}_n \} is a decreasing sequence since \{ U_n \} is a decreasing sequence of open sets. We show that \{ \mathcal{U}_n \} is a sequence of open covers of X that satisfies Theorem 3. We establish this by proving the following claims.

Claim 1. For each x \in X, \bigcap_{n=1}^\infty \overline{St(x,\mathcal{U}_n)}=\{ x \}.

To prove the claim, let x \ne y. There is an integer n such that (x,y) \notin \overline{U_n}. Choose open sets U and V such that (x,y) \in U \times V and (U \times V) \cap \overline{U_n}=\varnothing. Note that (x,y) \notin U_k and (U \times V) \cap U_n=\varnothing.

We want to show that V \cap St(x,\mathcal{U}_n)=\varnothing, which implies that y \notin \overline{St(x,\mathcal{U}_n)}. Suppose V \cap St(x,\mathcal{U}_n) \ne \varnothing. This means that V \cap W \ne \varnothing for some W \in \mathcal{U}_n with x \in W. Then (U \times V) \cap (W \times W) \ne \varnothing. Note that W \times W \subset U_n. This implies that (U \times V) \cap U_n \ne \varnothing, a contradiction. Thus V \cap St(x,\mathcal{U}_n)=\varnothing. Since y \in V, y \notin \overline{St(x,\mathcal{U}_n)}. We have established that for each x \in X, \bigcap_{n=1}^\infty \overline{St(x,\mathcal{U}_n)}=\{ x \}.

Claim 2. For each x \in X, \{ St(x,\mathcal{U}_n) \} is a local base at the point x.

Note that \{ St(x,\mathcal{U}_n) \} is a decreasing sequence of open sets such that \bigcap_{n=1}^\infty \overline{St(x,\mathcal{U}_n)}=\{ x \}. By Lemma 2, \{ St(x,\mathcal{U}_n) \} is a local base at the point x.

Claim 3. For each x \in X, \bigcap_{n=1}^\infty \overline{St(St(x,\mathcal{U}_n),\mathcal{U}_n)}=\{ x \}.

Let x \ne y. There is an integer n such that (x,y) \notin \overline{U_n}. Choose open sets U and V such that (x,y) \in U \times V and (U \times V) \cap \overline{U_n}=\varnothing. It follows that (U \times V) \cap \overline{U_t}=\varnothing for all t \ge n. Furthermore, (U \times V) \cap U_t=\varnothing for all t \ge n. By Claim 2, choose integers i and j such that St(x,\mathcal{U}_i) \subset U and St(y,\mathcal{U}_j) \subset V. Choose an integer k \ge \text{max}(n,i,j). It follows that (St(x,\mathcal{U}_i) \times St(y,\mathcal{U}_j)) \cap U_k=\varnothing. Since \mathcal{U}_k \subset \mathcal{U}_i and \mathcal{U}_k \subset \mathcal{U}_j, it follows that (St(x,\mathcal{U}_k) \times St(y,\mathcal{U}_k)) \cap U_k=\varnothing.

We claim that St(y,\mathcal{U}_k) \cap St(St(x,\mathcal{U}_k), \mathcal{U}_k)=\varnothing. Suppose not. Choose w \in St(y,\mathcal{U}_k) \cap St(St(x,\mathcal{U}_k), \mathcal{U}_k). It follows that w \in B for some B \in \mathcal{U}_k such that B \cap St(x,\mathcal{U}_k) \ne \varnothing and B \cap St(y,\mathcal{U}_k) \ne \varnothing. Furthermore (St(x,\mathcal{U}_k) \times St(y,\mathcal{U}_k)) \cap (B \times B)=\varnothing. Note that B \times B \subset U_k. This means that (St(x,\mathcal{U}_k) \times St(y,\mathcal{U}_k)) \cap U_k \ne \varnothing, contradicting the fact observed in the preceding paragraph. It must be the case that St(y,\mathcal{U}_k) \cap St(St(x,\mathcal{U}_k), \mathcal{U}_k)=\varnothing.

Because there is an open set containing y, namely St(y,\mathcal{U}_k), that contains no points of St(St(x,\mathcal{U}_k), \mathcal{U}_k), y \notin \overline{St(St(x,\mathcal{U}_n),\mathcal{U}_n)}. Thus Claim 3 is established.

Claim 4. For each x \in X, \{ St(St(x,\mathcal{U}_n),\mathcal{U}_n)) \} is a local base at the point x.

Note that \{ St(St(x,\mathcal{U}_n),\mathcal{U}_n) \} is a decreasing sequence of open sets such that \bigcap_{n=1}^\infty \overline{St(St(x,\mathcal{U}_n),\mathcal{U}_n))}=\{ x \}. By Lemma 2, \{ St(St(x,\mathcal{U}_n),\mathcal{U}_n) \} is a local base at the point x.

In conclusion, the sequence \mathcal{U}_1,\mathcal{U}_2,\mathcal{U}_3,\cdots of open covers satisfies the properties in Theorem 3. Thus any pseudocompact space with a regular G_\delta-diagonal is metrizable. \square

Example

Any submetrizable space has a G_\delta-diagonal. The converse is not true. A classic example of a non-submetrizable space with a G_\delta-diagonal is the Mrowka space (discussed here). The Mrowka space is also called the psi-space since it is sometimes denoted by \Psi(\mathcal{A}) where \mathcal{A} is a maximal family of almost disjoint subsets of \omega. Actually \Psi(\mathcal{A}) would be a family of spaces since \mathcal{A} is any maximal almost disjoint family. For any maximal \mathcal{A}, \Psi(\mathcal{A}) is a pseudocompact non-submetrizable space that has a G_\delta-diagonal. This example shows that the requirement of a regular G_\delta-diagonal in Theorem 1 cannot be weakened to a G_\delta-diagonal. See here for a more detailed discussion of this example.

Reference

  1. Gruenhage, G., Generalized Metric Spaces, Handbook of Set-Theoretic Topology (K. Kunen and J. E. Vaughan, eds), Elsevier Science Publishers B. V., Amsterdam, 423-501, 1984.
  2. McArthur W. G., G_\delta-Diagonals and Metrization Theorems, Pacific Journal of Mathematics, Vol. 44, No. 2, 613-317, 1973.

\text{ }

\text{ }

\text{ }

Dan Ma math

Daniel Ma mathematics

\copyright 2018 – Dan Ma