🔢 mathematics

Definability via the tilting correspondence

This paper demonstrates that the arithmetic definability of henselian valuations is preserved under the tilting correspondence, establishes that perfectoid valuations require no parameters for such definitions, and investigates the uniformity and quantifier complexity of these definitions.

Original authors: Gessica Alecci, Ihsane Hadeg, Franziska Jahnke, Margarete Ketelsen, Isabella Negrini

Published 2026-05-25

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Gessica Alecci, Ihsane Hadeg, Franziska Jahnke, Margarete Ketelsen, Isabella Negrini

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 two different worlds of numbers. One world is built on "mixed" rules (like our familiar numbers that include fractions and roots), and the other is built on "pure" rules (a simpler, positive-characteristic world). In the high-level math of this paper, these are called Perfectoid Fields.

The paper is about a magical bridge between these two worlds called the Tilting Correspondence. Think of "tilting" as a special machine that takes a complex number system, strips away its messy "mixed" features, and turns it into a simpler, cleaner version (the "tilt"). The big question the authors ask is: If we can find a specific rule to identify a special group of numbers (a "valuation") in the complex world, can we use that same rule to find the matching group in the simple world?

Here is the breakdown of their discovery using everyday analogies:

1. The Goal: Finding the "VIP Section"

In these number worlds, there is a special subset of numbers called the valuation ring. Imagine a nightclub. The whole city is the field ( $K$ ), but the VIP section inside the club is the valuation ring ( $O_v$ ).

The Problem: How do you write a rule (a formula) that tells you exactly who is in the VIP section?
The Twist: Sometimes you need a "password" (parameters) to get in. Sometimes the rule works for everyone without a password (called $\emptyset$ -definable or "parameter-free").

2. The Main Discovery: The Bridge Works

The authors prove that the "VIP status" travels perfectly across the bridge.

The Rule: If you can write a rule to find the VIP section in the complex world, you can also write a rule to find it in the simple world, and vice versa.
The "No Password" Bonus: They discovered something even cooler. If the VIP section can be found with a rule, it can always be found without a password. You never need a specific "key" or parameter; the rule works on its own.

3. Why is the VIP Section Findable? (The Three Reasons)

The paper explains that the VIP section is easy to find (definable) if the number system has one of three specific "flaws" or features. If none of these are present, the VIP section is hidden and cannot be found by a simple rule.

Think of these three reasons as three different ways to spot the VIPs:

Reason A: The "Broken Ruler" (Non-divisible Value Group)
Imagine the numbers are measured on a ruler. If the ruler has gaps (it's not perfectly divisible), you can spot the VIPs because they cluster around those gaps.
- Analogy: If your ruler only has marks for whole numbers but not halves, you can easily tell who is "whole" and who isn't.
Reason B: The "Broken Door" (Not Defectless)
Imagine a door that doesn't close perfectly, leaving a gap (a "defect"). This imperfection makes the VIPs stand out.
- Analogy: If a security gate is broken and lets some people through who shouldn't be, you can easily identify the group based on that broken gate.
Reason C: The "Strange Neighborhood" (Residue Field)
The "residue field" is like the neighborhood outside the club. If the neighborhood is weird or doesn't follow standard "tame" rules, the VIPs inside the club become obvious.
- Analogy: If the street outside is chaotic and unpredictable, the people inside the VIP lounge become very distinct by contrast.

The Verdict: If the number system is "perfectly smooth" (divisible ruler, no broken doors, and a tame neighborhood), the VIPs are invisible to simple rules. If it has any of the three flaws above, the VIPs are visible.

4. The Bridge is Symmetric

Because the "tilting" machine preserves these three features (the ruler gaps, the broken doors, and the neighborhood type), the ability to find the VIPs is preserved.

If you can find them in the complex world $\rightarrow$ You can find them in the simple world.
If you can find them in the simple world $\rightarrow$ You can find them in the complex world.

5. The "Uniformity" Trap

The authors also investigated: Can we use the exact same sentence (formula) to find the VIPs in both worlds?

The Answer: You could, but you can also easily trick the system.
The Metaphor: Imagine you have a sign that says "VIPs Only." In the complex world, the sign works. In the simple world, you could just add a tiny, invisible sticker to the sign that says "If you are in the complex world, ignore this."
The paper shows that while a universal rule exists, you can artificially create situations where the rule works in one world but fails in the other just by adding a tiny "characteristic check" (like checking if the number system is even or odd). So, while the ability to define is preserved, the exact same sentence isn't always guaranteed to work without some tweaking.

6. The Complexity of the Rules

Finally, they looked at how complicated the rules are.

Some rules are simple "Yes/No" checks (Existential).
Some are complex "For all" checks (Universal).
They found that simple "Yes/No" rules travel well across the bridge. However, complex "For all" rules sometimes get lost when crossing from the complex world to the simple world. It's like a complex instruction manual that makes sense in a high-tech lab but becomes gibberish in a simple workshop.

Summary

This paper is about a mathematical magic trick. It proves that if a special group of numbers (the valuation) is visible in a complex number system, it is always visible in its simpler "tilted" version, and vice versa. Furthermore, if it's visible, it's visible without needing any secret passwords. The visibility depends entirely on whether the number system has specific "imperfections" (like a broken ruler or a weird neighborhood). If the system is too perfect, the group remains hidden.

Technical Summary: Arithmetic Definability of Henselian Valuations via the Tilting Correspondence

Problem Statement
The paper investigates the preservation of arithmetic definability under the tilting correspondence, a fundamental equivalence in arithmetic geometry linking perfectoid fields of mixed characteristic $(0, p)$ to their tilts, which are perfectoid fields of positive characteristic $p$ . Specifically, the authors address the model-theoretic question of whether the property of a valuation ring being first-order definable (in the language of rings, $L_{ring}$ ) is preserved when transferring between a perfectoid field $(K, v)$ and its tilt $(K^\flat, v^\flat)$ . Furthermore, the work seeks to determine if such definitions can be made without parameters ( $\emptyset$ -definable) and to analyze the quantifier complexity ( $\forall$ -definability vs. $\exists$ -definability) of these definitions.

Methodology
The authors employ a combination of valuation theory, model theory, and the specific structural properties of perfectoid fields.

Structural Analysis: The paper leverages the fact that tilting preserves the value group, the residue field, and the defectless property (Fact 2.6).
Classification of Definability: The authors classify the conditions under which a henselian valuation is definable by analyzing three distinct sources of definability:
- Non-divisible value groups: Utilizing results by Hong (generalizing Koenigsmann) to show that non-divisibility allows for parameter-free definitions.
- Residue field properties: Applying the concept of $t$ -henselianity and "divisible-tame type" fields. The authors use Beth's Definability Theorem and properties of tame fields to show that if the residue field is not of divisible-tame type, the valuation is $\emptyset$ -definable.
- Defect: Building on recent work by Kuhlmann, Rzepka, Ketelsen, Ramello, and Szewczyk (KRS24), the paper examines fields with independent defect extensions.
Ultrapower Techniques: To prove non-definability results, the authors construct ultrapowers and utilize the Keisler–Shelah isomorphism theorem to demonstrate that if a valuation were definable in a specific context, it would contradict known non-definability results for tame fields with divisible value groups.
Quantifier Complexity: The analysis of quantifier complexity relies on the work of Anscombe and Fehm regarding $C$ -fields (specifically $Z$ -large fields and embedded residue properties) to distinguish between $\forall$ -definability and $\exists$ -definability.

Key Contributions and Results

Main Classification Theorem (Theorem 3.15): The paper provides a complete classification for perfectoid fields. A perfectoid valuation $v$ is definable if and only if it is $\emptyset$ -definable, which occurs if and only if at least one of the following holds:
1. $(K, v)$ is not defectless.
2. The value group $vK$ is not divisible.
3. The residue field $Kv$ is not $t$ -henselian of divisible-tame type.
  Crucially, the authors prove that for perfectoid fields, definability implies $\emptyset$ -definability; no parameters are required.
Preservation under Tilting (Corollary 3.16): Since the characterizing conditions in the Main Theorem (defectlessness, divisibility of the value group, and the $t$ -henselian nature of the residue field) are all preserved under the tilting correspondence, the paper concludes that definability is preserved. Specifically, $v$ is definable (or $\emptyset$ -definable) if and only if $v^\flat$ is definable (or $\emptyset$ -definable).
Quantifier Complexity (Section 5):
- Untilting: Both $\exists$ - $\emptyset$ -definability and $\forall$ - $\emptyset$ -definability are preserved under untilting (going from positive to mixed characteristic).
- Tilting: $\forall$ - $\emptyset$ -definability is not preserved under tilting. The authors provide an example where a valuation in mixed characteristic is $\forall$ - $\emptyset$ -definable (due to a non-divisible value group), but its tilt is not, because the residue field of the tilt is algebraically closed (and thus $Z$ -large), preventing $\forall$ - $\emptyset$ -definability.
- Open Question: Whether $\exists$ - $\emptyset$ -definability is preserved under tilting remains an open question.
Uniformity (Section 4): The paper demonstrates that while definability is preserved, uniformity of the defining formula is not. It is possible to construct formulas that define the valuation in one field but fail in the other, or conversely, to artificially construct a single formula that works for both by exploiting the difference in characteristic (using a formula that is true in characteristic 0 and false in characteristic $p$ ).
Classical Case (Section 6): For perfectoid fields where the residue field is an algebraic extension of $\mathbb{F}_p$ , the authors show that if the field is not algebraically closed, the valuation is $\emptyset$ -definable. If the residue field is not algebraically closed, the valuation is $\exists$ - $\emptyset$ -definable.

Significance
The paper establishes that the arithmetic definability of henselian valuations in the context of perfectoid fields is robust under the tilting correspondence, provided one considers the specific structural invariants (defect, value group divisibility, and residue field type) that govern definability. A significant finding is that for perfectoid fields, the need for parameters vanishes; if a valuation is definable, it is automatically $\emptyset$ -definable. This result refines the understanding of how model-theoretic properties transfer between mixed and positive characteristic, complementing existing work on decidability and elementary equivalence (e.g., by Jahnke, Kartas, and others). The paper also clarifies the limitations of this transfer regarding quantifier complexity, specifically showing that $\forall$ - $\emptyset$ -definability is not a tilting-invariant property.