🔢 mathematics

Homotopy properties of regular mappings into real retract rational varieties

This paper investigates the homotopy properties of regular mappings from spheres into real retract rational varieties, demonstrating that the resulting homotopy classes form basepoint-independent subgroups and that all Whitehead products within these groups admit regular representatives.

Original authors: Juliusz Banecki

Published 2026-02-16

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Juliusz Banecki

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 are an architect trying to build structures using only a very specific, rigid set of blueprints. In the world of mathematics, these "blueprints" are called regular mappings. They are functions that follow strict algebraic rules (like polynomials), meaning they are smooth, predictable, and built from simple formulas.

Now, imagine you have a flexible, stretchy piece of fabric called a sphere (like a basketball). You want to wrap this sphere around a complex, multi-dimensional shape called a variety (think of it as a strange, high-dimensional landscape).

Usually, you can wrap the sphere around this landscape in many different ways. Some ways are "loopy," some are "twisted," and some are "tight." Mathematicians call these different ways homotopy classes. The big question this paper asks is: If we are forced to use only our strict algebraic blueprints, can we still create every possible way of wrapping the sphere? Or do we get stuck with only a few specific patterns?

Here is a breakdown of what the author, Juliusz Banecki, discovered, using simple analogies.

1. The Special Landscape: "Retract Rational Varieties"

The paper focuses on a special type of landscape called a retract rational variety.

The Analogy: Imagine a piece of clay that is so well-behaved that you can stretch a small, flat piece of paper over it, and then pull the paper back off without tearing it, leaving the paper perfectly flat again.
Why it matters: These shapes are "nice" enough that they behave somewhat like flat space (Euclidean space) locally, even if they look complicated globally. The author proves that if your landscape is one of these "nice" shapes, the rules of algebraic mapping become much more powerful.

2. The Main Discovery: The "Algebraic Club"

The author's biggest finding (Theorem 1.4) is about groups. In math, a "group" is a collection of things where you can combine them (like adding numbers) and get another thing in the same collection.

The Problem: Usually, if you take two "algebraic" ways of wrapping a sphere and combine them, the result might be a "messy" way that cannot be described by a strict algebraic formula.
The Solution: The author proves that for these special "retract rational" landscapes, the algebraic ways of wrapping the sphere do form a perfect club. If you take two algebraic wraps and combine them, the result is still an algebraic wrap.
The Metaphor: Imagine you have a set of LEGO bricks. Usually, if you snap two specific LEGO structures together, you might end up with a shape that doesn't fit the LEGO system. But on these special landscapes, the "LEGO system" is so robust that snapping any two valid structures together always results in another valid LEGO structure.

3. The "Whitehead Product" (The Twisty Knots)

Mathematicians love to twist spheres together to create complex knots, called Whitehead products.

The Analogy: Imagine taking a rubber band (a circle) and twisting it around a ball, then taking another rubber band and twisting it around the first one. This creates a complex knot.
The Discovery: The author shows that even these incredibly complex, twisted knots can be built using only the strict algebraic blueprints. No matter how complicated the knot is, if the landscape is "nice" (retract rational), there is a perfect algebraic formula to create it.

4. The "Path" Between Points

The paper also shows that it doesn't matter where you start on the landscape.

The Analogy: Imagine you are painting a picture of a mountain. It doesn't matter if you start painting from the peak or the base; the "algebraic rules" work the same way everywhere, as long as the mountain is connected (you can walk from one point to another without jumping).
The Result: The group of algebraic wraps is the same no matter which point on the landscape you choose as your "home base."

5. The Warning: It's Not Always Perfect

The author is careful not to overpromise. He provides two "Counterexamples" (Section 5) to show where the rules break.

Example 1: If the landscape isn't "nice" enough (not retract rational), the algebraic club falls apart. You can combine two algebraic wraps, and the result might be a "forbidden" shape that no formula can describe.
Example 2: Even on a "nice" landscape, the algebraic club might not include every possible wrap. There might be some weird, twisted knots that exist in the flexible world but simply cannot be built with the strict algebraic blueprints.
The Lesson: The "retract rational" landscapes are the "Goldilocks" zone—not too rigid, not too wild. They are the perfect size to study these algebraic properties.

Summary

In plain English, this paper says:

"If you are working with a specific, well-behaved type of geometric shape, the 'strict algebraic' ways of wrapping spheres around them are incredibly robust. They form a complete, self-contained system where you can combine them and twist them without ever leaving the system. However, if the shape is too weird, or if you try to claim that every possible wrap is algebraic, you will run into trouble."

The author essentially found the "sweet spot" in real algebraic geometry where the rigid rules of algebra and the flexible nature of topology play together perfectly.

1. Problem Statement

The paper investigates the homotopy properties of regular mappings (morphisms in real algebraic geometry) from spheres $S^n$ into real affine varieties. Specifically, it addresses the structure of the set of algebraic homotopy classes, denoted $\pi_n^{\text{alg}}(Y, y_0)$ , which consists of homotopy classes in the topological homotopy group $\pi_n(Y, y_0)$ that admit representatives which are regular mappings.

The central questions are:

Is $\pi_n^{\text{alg}}(Y, y_0)$ a subgroup of $\pi_n(Y, y_0)$ ?
Does this set depend on the choice of the basepoint $y_0$ ?
Can specific topological constructions, such as Whitehead products, be represented by regular mappings?
Under what conditions does the equality $\pi_n^{\text{alg}}(Y, y_0) = \pi_n(Y, y_0)$ hold?

The study focuses on the class of retract rational varieties, a concept introduced by Saltman, which generalizes rational and stably rational varieties.

2. Methodology

The author employs a combination of real algebraic geometry, homotopy theory, and approximation techniques.

Retract Rationality: The core assumption is that the target variety $Y$ is a nonsingular, retract rational affine variety. This means there exists a Zariski open dense subset $U \subset Y$ and a rational variety $V \subset \mathbb{R}^n$ with regular mappings $i: U \to V$ and $r: V \to U$ such that $r \circ i = \text{id}_U$ .
Approximation Theorems: The paper relies heavily on recent results (specifically Theorem 1.3, derived from [2] and [3]) stating that continuous mappings between real affine varieties can be approximated by regular mappings in the compact-open topology, provided the target is a nonsingular retract rational variety and the mapping is already regular on a compact Zariski closed subset.
Rational H-spaces: To prove the subgroup property, the author introduces the notion of a rational H-space. A pair $(Y, y_0)$ is a rational H-space if there exists a Zariski open set $A \subset Y \times Y$ and a regular multiplication map $\mu: A \to Y$ that behaves like a group multiplication up to homotopy relative to the basepoint.
Technical Lemmas:
- Lemma 2.1 & Corollary 2.2: These technical tools allow for the extension of germs of regular mappings defined near a point to neighborhoods of the product space $Y \times \{0\}$ , utilizing Lojasiewicz inequalities to control the domain of definition.
- Observation 2.3: Establishes the existence of local retractions for nonsingular retract rational varieties.
Counterexamples: The paper uses blow-ups of specific curves and surfaces to construct varieties where the expected properties fail, demonstrating the necessity of the "retract rational" hypothesis.

3. Key Contributions and Results

A. Subgroup Structure (Theorem 1.4)

The primary result establishes that for a nonsingular retract rational affine variety $Y$ , the set of algebraic homotopy classes $\pi_n^{\text{alg}}(Y, y_0)$ forms a subgroup of the topological homotopy group $\pi_n(Y, y_0)$ for all $n \ge 1$ .

Method: The author proves that $(Y, y_0)$ is a rational H-space. This allows the construction of a regular multiplication map $\mu$ . By choosing representatives of homotopy classes that are constant on specific hemispheres, the product of two algebraic classes can be represented by the regular mapping $\mu(f_1, f_2)$ . Approximation theorems ensure that inverses and products remain within the algebraic class.
Significance: This generalizes previous results known only for spheres ( $S^k$ ) to a much broader class of varieties.

B. Independence of Basepoint (Observation 1.5)

If $Y$ is a connected nonsingular retract rational variety, the groups $\pi_n^{\text{alg}}(Y, y_0)$ and $\pi_n^{\text{alg}}(Y, y_1)$ are isomorphic for any two points $y_0, y_1 \in Y$ .

Reasoning: Paths between points induce isomorphisms on topological homotopy groups. Using the approximation theorem, these isomorphisms can be shown to map algebraic classes to algebraic classes.

C. Whitehead Products (Theorem 1.6)

For any two classes $\alpha \in \pi_n(Y, y_0)$ and $\beta \in \pi_m(Y, y_0)$ , their Whitehead product $[\alpha, \beta] \in \pi_{n+m-1}(Y, y_0)$ is representable by a regular mapping.

Method: The Whitehead product is constructed via the attaching map of a cell in a torus. The author shows that the constant maps involved in the standard construction can be approximated by regular maps, and the rational H-space structure allows the composition to be realized regularly.

D. Approximation on Curves (Theorem 1.7)

If $X$ is a nonsingular affine curve (compact in Euclidean topology) and $Y$ is a nonsingular retract rational variety, any continuous map $f: X \to Y$ can be approximated by a regular map.

Corollary 1.8: Consequently, for such $Y$ , the fundamental group satisfies $\pi_1^{\text{alg}}(Y, y_0) = \pi_1(Y, y_0)$ . Every topological loop is homotopic to a regular loop.

4. Limitations and Counterexamples

The paper carefully delineates the boundaries of these results:

Necessity of Retract Rationality (Example 5.1): The author constructs a nonsingular variety $Y$ (obtained by blowing up a product of a specific curve) that is not retract rational. For this $Y$ , $\pi_1^{\text{alg}}(Y, y_0)$ is not a subgroup of $\pi_1(Y, y_0)$ . This proves the retract rational assumption in Theorem 1.4 is essential.
Strict Inequality (Example 5.2): The paper constructs a nonsingular retract rational variety $Y$ $Y$ (a blow-up of $S^3$ $S^{3}$ along a specific curve) where $\pi_2^{\text{alg}}(Y, y_0) \neq \pi_2(Y, y_0)$ $π_{2}^{alg} (Y, y_{0}) \neq = π_{2} (Y, y_{0})$ .
- Proof Strategy: Using algebraic homology and cohomology with $\mathbb{Z}/2$ coefficients, the author shows that while the topological homology is generated by certain classes, the algebraic homology is restricted by the irreducibility of the underlying curve. This discrepancy implies that not all topological homotopy classes can be represented by regular maps.

5. Significance

Characterization of Algebraic Homotopy: The paper provides a robust framework for understanding which topological features of real algebraic varieties can be captured by regular mappings. It identifies retract rational varieties as the "sweet spot" where algebraic homotopy classes form a well-behaved subgroup structure.
Generalization: It significantly extends the known results from spheres to a vast class of varieties, unifying previous scattered results regarding rational and stably rational varieties.
Delicacy of the Theory: The counterexamples highlight that while the subgroup property holds for retract rational varieties, the stronger property of equality ( $\pi_n^{\text{alg}} = \pi_n$ ) does not hold generally, even within this class. This suggests that the study of regular mappings requires nuanced distinctions between different types of rationality.
Future Directions: The results suggest that nonsingular retract rational varieties are the natural setting for studying homotopy properties in real algebraic geometry, while rationally connected varieties (a broader class) remain an open question for these specific homotopy properties.