🔢 mathematics

Automatic continuity for vector spaces with linear topology

This paper classifies all topological vector spaces with a linear topology in which every algebraic automorphism is continuous and establishes several properties of these spaces.

Original authors: Samuel Quirino, Lucas H. R. de Souza

Published 2026-03-02

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Samuel Quirino, Lucas H. R. de Souza

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you have a giant, infinite library of books (this is your Vector Space). In a normal library, the books are arranged on shelves, and there are strict rules about how you can move them around. In mathematics, these rules are called topology. They define what "close" means and what "continuous" means. Usually, if you try to rearrange the books in a weird way (an algebraic automorphism), you might break the rules of the library, making the new arrangement "discontinuous" or chaotic.

This paper asks a very specific question: Is there a special way to organize this library where any possible rearrangement of the books automatically follows the rules? In other words, is there a setup where you can't accidentally break the continuity, no matter how you shuffle the books?

The authors, Samuel and Lucas, say: "Yes, there is!"

Here is the breakdown of their discovery using simple analogies:

1. The "Co-Size" Rule (The Codimension Topology)

Usually, a library's rules are based on specific shelves. But this paper introduces a new rule called the Codimension Topology.

Imagine the library has a rule: "Any group of books that is 'small enough' compared to the whole library is considered a 'closed' group."

If you take a huge chunk of the library (say, 99% of it), the remaining tiny bit is "small."
The rule says: If a group of books is missing only a "small" number of books (mathematically, if its "missing size" or codimension is below a certain limit), then that group is treated as a solid, closed unit.

The authors prove that if you organize your library using this specific "missing size" rule, then every single way you can rearrange the books will automatically be a valid, continuous move. You can't break the rules because the rules are so flexible that they accept almost any change.

2. The "Magic Mirror" (Automatic Continuity)

In math, there's a concept called Automatic Continuity. It's like having a magic mirror. If you look at a reflection (a mathematical function), usually you have to check if the mirror is clean (continuous). But in these special libraries, the mirror is always clean.

The paper's main theorem says:

"A library has this 'magic mirror' property (where every rearrangement is continuous) if and only if it is organized using our special 'missing size' rules."

It's a perfect match. If the rules are right, continuity is automatic. If the rules are wrong, you can find a rearrangement that breaks the continuity.

3. The "Shape-Shifting" Library

One of the coolest parts of the paper is a counter-intuitive discovery.

In normal geometry, if you have two shapes that look different, they are different. But in these special vector spaces, the authors found a way to build infinite families of libraries that look completely different on the inside but have the exact same "shadow" and "echo."

The Shadow (Profinite Completion): If you squint and look at the library from far away, all these different libraries look exactly the same.
The Echo (Dual Space): If you shout into the library and listen for the echo, they all sound identical.

It's like having a thousand different houses made of different materials (wood, glass, steel), but if you look at their blueprints from a distance or listen to the wind whistle through them, they are indistinguishable. Yet, if you try to walk through them, they are totally different. This is a huge deal for mathematicians because it shows how complex these structures can be.

4. The Broken Promise (Algebras)

Finally, the authors tried to apply this logic to Algebras (libraries where the books can also be multiplied together, like numbers).

There was a famous rule (Johnson's Theorem) that said: "If you have a well-behaved library, any map between them is continuous."
The authors tested this on their special libraries and said: "Nope, it doesn't work here."

They found examples where, even with their special rules, you could still break the continuity if you tried to map one algebra to another. It's like finding a loophole in the magic mirror: the mirror works for rearranging books, but if you try to multiply the books, the magic sometimes fails.

Summary

The Problem: When does a mathematical space have the property that every rearrangement is smooth and continuous?
The Solution: When the space is organized by a "missing size" rule (Codimension Topology).
The Surprise: You can have many different spaces that look identical from the outside (same completion and dual) but are totally different inside.
The Limit: This "automatic smoothness" doesn't always hold when you start multiplying things (algebras).

In short, the paper maps out the exact conditions under which mathematical chaos is impossible, revealing a hidden order where "anything goes" is actually the most structured state of all.

1. Problem Statement and Motivation

The paper addresses the problem of automatic continuity within the specific context of topological vector spaces (TVS) equipped with a linear topology.

Context: In general topology, a space where every bijection is continuous is known to be a space with the co- $\kappa$ topology (closed sets have cardinality $<\kappa$ ). In Functional Analysis, "automatic continuity" refers to the phenomenon where algebraic homomorphisms between certain structures (like Banach algebras) are necessarily continuous. A classic result by Johnson states that homomorphisms from a Banach algebra to a semisimple Banach algebra are continuous.
The Gap: The authors investigate whether a similar classification holds for TVS with linear topology (where the neighborhood filter of 0 is generated by open vector subspaces). Specifically, they seek to classify all such spaces where every algebraic automorphism is continuous (i.e., $\text{Aut}(V) = \text{CAut}(V)$ ).
Goal: To characterize the topological structure of these spaces and explore the behavior of linear maps (homomorphisms) between them, contrasting these findings with the behavior of general topological spaces and Banach algebras.

2. Methodology and Definitions

The authors employ a combination of linear algebra, cardinal arithmetic, and topological vector space theory.

Key Definitions:
- Linear Topology: A topology on a vector space $V$ where the neighborhood filter of 0 is generated by open vector subspaces.
- $\kappa$ -Codimension Topology ( $\tau_{V,\kappa}$ ): For a cardinal $\kappa$ (where $\kappa=1$ or $\kappa \geq \aleph_0$ ), this is the topology generated by the filter of all subspaces $W < V$ such that $\text{codim}(W) < \kappa$ .
- Codimension Topology: A space $V$ has a codimension topology if it is equipped with $\tau_{V,\kappa}$ for some $\kappa$ .
- $\zeta(V)$ : Defined as $\sup\{\kappa : \exists W < V \text{ open}, \text{codim}(W) = \kappa\}$ . This invariant measures the "size" of the open subspaces in the space.
Analytical Approach:
- The authors analyze the relationship between the algebraic structure (isomorphisms) and the topological structure (continuity).
- They utilize Isomorphism Theorems to relate the dimensions of subspaces and their intersections.
- They construct specific counterexamples and proofs by contradiction regarding the existence of open subspaces with specific codimensions.
- They examine the profinite-dimensional completion and the topological dual of these spaces to understand their structural invariants.

3. Key Contributions and Results

A. Classification Theorem (Main Result)

The central contribution is Theorem 2.12, which provides a complete classification:

Theorem: Let $V$ be a topological vector space with a linear topology. Then $V$ has a codimension topology if and only if every algebraic automorphism of $V$ is continuous ( $\text{Aut}(V) = \text{CAut}(V)$ ).

Significance: This establishes a direct equivalence between the algebraic property of "automatic continuity for automorphisms" and the specific topological structure of "codimension topology." It is the linear algebraic analogue of the classification of spaces where all bijections are continuous.

B. Properties of Linear Maps

The paper extends the analysis from automorphisms to general linear maps (homomorphisms):

Proposition 2.13 & 2.14: If $V$ has the $\kappa$ -codimension topology and $V'$ has the $\kappa'$ -codimension topology, then every linear map $f: V \to V'$ is continuous if $\kappa' \leq \kappa$ .
Contrast with General Topology: In general topological spaces, the only spaces where all self-maps are continuous are the discrete and trivial spaces. However, for TVS with codimension topology, all linear self-maps are continuous (a consequence of Prop 2.14 with $\kappa' = \kappa$ ). This highlights a fundamental difference between general topology and linear topology.

C. Structural Properties

The authors derive several structural properties of spaces with codimension topology:

Hausdorff Property: For $\kappa > 1$ , these spaces are Hausdorff (Corollary 3.3).
Discreteness of Subspaces: If $W$ is a subspace with $\dim(W) < \kappa$ , then $W$ carries the discrete topology (Corollary 3.8).
Non-Linear Compactness: If $\dim(V) \geq \aleph_0$ , $V$ is not linearly compact (Proposition 3.9). This is proven by showing the existence of non-open subspaces of codimension 1, which contradicts the definition of linear compactness.
Completion: The Cauchy completion $\hat{V}$ of a space with $\aleph_0$ -codimension topology is linearly compact (Proposition 3.10).

D. Construction of Non-Isomorphic Families

Corollary 3.11 presents a striking construction:

For a vector space $V$ , one can construct a family of topologies $\{V_\kappa\}$ for $\aleph_0 \leq \kappa \leq \dim(V)^+$ .
These spaces are pairwise non-isomorphic as topological vector spaces (distinguished by their topology).
However, they all share the same profinite-dimensional completion and the same topological dual space.
Significance: This demonstrates that the profinite-dimensional completion and the dual space are insufficient invariants to distinguish between different codimension topologies on the same underlying vector space.

4. Remarks on Algebras (Failure of Johnson's Theorem)

The final section investigates whether the automatic continuity results extend to topological algebras with linear topology.

Failure of Johnson's Theorem: The authors show that Johnson's theorem (which guarantees continuity for homomorphisms into semisimple Banach algebras) fails in the context of linear topological algebras.
- Example 4.2: The identity map from a non-discrete linear topological algebra to the same algebra with the discrete topology is an algebraic epimorphism but is not continuous.
- Example 4.4: Even when restricting to pseudocompact algebras (a class often considered "close" to compact spaces), the theorem fails. A homomorphism from a pseudocompact algebra to a field (discrete) can be discontinuous if the kernel is a dense maximal ideal.

5. Significance and Conclusion

This paper makes significant contributions to the theory of topological vector spaces by:

Bridging Algebra and Topology: It rigorously classifies linear topological spaces based on the automatic continuity of their automorphisms, linking a purely algebraic condition to a specific topological definition (codimension topology).
Refining Invariants: It demonstrates that standard invariants like the dual space and profinite-dimensional completion do not uniquely determine the topology in this class, providing a rich family of counterexamples to potential uniqueness conjectures.
Limiting Automatic Continuity in Algebras: It clarifies the boundaries of automatic continuity, showing that results holding for Banach algebras do not generalize to the broader class of linear topological algebras, even under strong conditions like pseudocompactness.

The work serves as a foundational reference for understanding the interplay between linear algebraic structures and linear topologies, offering precise characterizations where automatic continuity holds and where it breaks down.