Abstract
We study localization at a prime in homotopy type theory, using self maps of the circle. Our main result is that for a pointed, simply connected type $X$, the natural map $X \to X_{(p)}$ induces algebraic localizations on all homotopy groups. In order to prove this, we further develop the theory of reflective subuniverses. In particular, we show that for any reflective subuniverse $L$, the subuniverse of $L$-separated types is again a reflective subuniverse, which we call $L'$. Furthermore, we prove results establishing that $L'$ is almost left exact. We next focus on localization with respect to a map, giving results on preservation of coproducts and connectivity. We also study how such localizations interact with other reflective subuniverses and orthogonal factorization systems. As key steps towards proving the main theorem, we show that localization at a prime commutes with taking loop spaces for a pointed, simply connected type, and explicitly describe the localization of an Eilenberg-Mac Lane space $K(G,n)$ with $G$ abelian. We also include a partial converse to the main theorem.
Introduction
In classical algebraic topology, localization is a fundamental technique used to simplify the study of topological spaces by focusing on specific prime numbers. By "inverting" a set of primes, mathematicians can decompose complex homotopy types into simpler local components. While this theory is well-established in classical settings, its translation into Homotopy Type Theory (HoTT) presents unique challenges. HoTT operates within a constructive framework where classical tools like Whitehead’s Theorem or the Axiom of Choice are not always applicable.
This paper establishes a rigorous foundation for localization at a prime within HoTT. The authors demonstrate that for a pointed, simply connected type $X$, the type-theoretic localization $X_{(p)}$ correctly mirrors the algebraic localization of its homotopy groups. This result bridges the gap between abstract higher inductive types and the concrete algebraic structures of classical homotopy theory.
Theoretical Foundations: Reflective Subuniverses
The core of the paper’s methodology lies in the theory of reflective subuniverses. In HoTT, a reflective subuniverse $L$ acts as a localization functor. A type $A$ is considered $L$-local if it belongs to this subuniverse.
A significant contribution of this work is the development of the theory of $L$-separated types. For any reflective subuniverse $L$, a type $X$ is $L$-separated if its identity types (paths) are $L$-local. The authors prove that the collection of $L$-separated types, denoted $L'$, also forms a reflective subuniverse.
Crucially, while a general localization $L$ may not be "left exact" (i.e., it may not preserve finite limits), the authors show that $L'$ is "almost left exact." This property is essential for proving that localization commutes with certain limit-based constructions, such as taking loop spaces for simply connected types.
Localization at a Prime via Degree Maps
To localize at a prime $p$ (or more generally, a set of integers $S$), the authors utilize the degree $k$ map on the circle, $f_k: S^1 \to S^1$. By localizing a type with respect to these maps, one effectively makes the integer $k$ invertible within the type's structure.
This approach avoids the need for a general theory of homology, which is difficult to formalize generally in HoTT. Instead, it relies on the geometric properties of the circle and the use of mapping telescopes (sequential colimits) to compute the localization of loop spaces ($\Omega X$). These proofs are entirely constructive, ensuring the results hold in any $\infty$-topos without relying on the Law of Excluded Middle.
Key Results and the Main Theorem
The central result of the paper is the Main Theorem: For a pointed, simply connected type $X$ and a prime $p$, the natural map $X \to X_{(p)}$ induces algebraic localizations on all homotopy groups $\pi_n(X)$ for $n \ge 2$.
To reach this conclusion, the authors establish several intermediate results:
- Commutativity with Loop Spaces: For a pointed, simply connected type $X$, there is an equivalence $\Omega(L_{deg(S)} X) \simeq L_{deg(S)}(\Omega X)$. This allows the localization to "pass through" the loop space operator to reach the underlying homotopy groups.
- Eilenberg-Mac Lane Spaces: The authors provide an explicit characterization of the localization of $K(G, n)$ spaces, showing that $L_{deg(S)} K(G, n) \simeq K(G_{(S)}, n)$, where $G_{(S)}$ is the algebraic localization of the group $G$.
- Preservation of Connectivity: The paper establishes conditions under which localization preserves $n$-connectivity, ensuring that the "shape" of the type is maintained.
Furthermore, the authors provide a partial converse: if a map between simply connected $n$-types induces algebraic localization on all homotopy groups, then that map is itself a localization in the HoTT sense.
Discussion and Limitations
While the paper provides a robust framework for simply connected types, it also highlights areas for future exploration. In classical topology, localization is often applied to nilpotent spaces, a broader class than simply connected types. The authors acknowledge that extending these results to nilpotent types in HoTT is a natural next step, noting concurrent work in the field.
Additionally, the paper does not address "fracture theorems"—the tools used to reconstruct a type from its localizations at various primes. Finally, while the proofs are constructive, only a portion of the work was formalized in proof assistants like Coq or Agda at the time of publication, though significant progress has since been made in the HoTT Library.
Conclusion
By developing the theory of $L$-separated types and utilizing degree maps on the circle, this work provides a comprehensive treatment of localization at a prime in Homotopy Type Theory. It confirms that the synthetic approach of HoTT is consistent with classical algebraic topology, providing a solid base for further research into the local-to-global properties of types.
Summary updated