⚛️ general relativity

Failure of local equi-Lipschitzness for families of Lorentz distances to Cauchy surface foliations

This paper demonstrates that the local equi-Lipschitz continuity of Lorentzian distance functions, a property known to hold for distances along complete timelike lines, generally fails for distances to the level sets of Cauchy temporal functions, while proposing conjectures linking the restoration of this property to Bartnik's splitting conjecture.

Original authors: Gregory J. Galloway, Robert J. McCann, Argam Ohanyan

Published 2026-06-23

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Gregory J. Galloway, Robert J. McCann, Argam Ohanyan

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

The Big Picture: Measuring Time in a Universe

Imagine you are living in a universe where space and time are woven together (this is called a "spacetime"). In this universe, there is a special rule: you can only travel forward in time, and there is a maximum speed (the speed of light) that limits how fast you can go.

Physicists and mathematicians love to ask: "What does this universe look like?"

Sometimes, they want to prove that the universe is actually very simple and structured, like a stack of identical pancakes (mathematically, a "product spacetime"). To prove this, they usually look for a specific feature: a perfect, infinite "time-line" that goes straight through the universe forever without bending or stopping.

The Old Trick: The "Time-Line" Method

In the past, mathematicians had a reliable trick to prove this structure.

They would find a perfect, infinite line of time (like a straight railroad track stretching forever).
They would measure the "time distance" from any point in the universe to that track.
The Magic Property: They found that if you look at a group of these time-measurements together, they behave very nicely. They are "equi-Lipschitz."

What does "equi-Lipschitz" mean?
Think of it like a group of hikers walking up a mountain.

Lipschitz means: "If you take a small step, your elevation changes by a predictable, limited amount. You can't suddenly teleport up a cliff."
Equi-Lipschitz means: "If you have a whole team of hikers (a family of measurements), they all follow the same rule. No matter which hiker you pick, or how far up the mountain they are, they all take steps of roughly the same size. None of them suddenly start sprinting or tripping over invisible rocks."

This "nice behavior" was essential. It allowed mathematicians to prove that the universe splits neatly into a stack of pancakes.

The New Problem: The "Time-Surface" Method

Recently, some scientists tried to apply this same trick to a different situation. Instead of looking for a single infinite line, they wanted to look at layers of time (like the layers of an onion or the pages of a calendar).

They imagined a "Cauchy temporal function." Think of this as a giant clock that covers the whole universe. Every time the clock ticks to a new number (say, 1:00, 2:00, 3:00), it creates a "slice" or a "surface" of the universe at that moment.

The Hypothesis:
The scientists hoped that if they measured the time distance from any point to these "time slices" (instead of the single line), the measurements would still behave nicely (be equi-Lipschitz). If they did, they could prove the universe splits into a stack of pancakes even without finding that perfect infinite line.

The Discovery: The Trick Fails

This paper says: "No, that doesn't work."

The authors (Galloway, McCann, and Ohanyan) built two specific examples to show that measuring time to these "slices" is messy. The measurements do not stay nice and predictable.

The Analogy of the Messy Hikers:
Imagine you are standing in a valley, and you want to measure the distance to a series of moving walls (the time slices) that are rushing toward you.

In the "Line" scenario, the walls were perfectly parallel and smooth. You could always predict how fast you'd hit them.
In the "Slice" scenario, the authors showed that the walls can be shaped in tricky ways. As the walls get further away (or closer), the angle at which you have to run to hit them changes wildly.
Suddenly, to hit the next "slice," you might have to sprint incredibly fast, then slow down, then sprint again. The "step size" of your measurement explodes. The "equi-Lipschitz" property breaks.

They proved that even if you choose a very "steep" and well-behaved clock (a mathematical function that is supposed to be nice), the time distances to its slices can still go haywire.

Why Does This Matter?

This failure is actually very important because it connects to a famous unsolved puzzle called Bartnik's Splitting Conjecture.

The Puzzle: If the universe is "complete" (you can travel forever without falling off the edge) and follows certain energy rules, does it have to be a simple stack of pancakes?
The Connection: The authors show that the answer to this puzzle is exactly the same as asking: "Is there any way to set up a clock in this universe such that the time distances to its slices behave nicely?"

The Conclusion:

The old idea that "time slices always behave nicely" is false. You cannot just assume it works.
However, if you can find a universe where the time slices do behave nicely, then you have proven Bartnik's conjecture (the universe is a stack of pancakes).
The paper suggests that for the universe to be a stack of pancakes, it must be possible to find a "perfect clock" that makes the time measurements behave. If you can't find such a clock, the universe might be more complex than a simple stack.

Summary in One Sentence

The authors proved that you cannot automatically assume that measuring time to "slices" of the universe behaves as smoothly as measuring time to a "line," and they showed that fixing this problem is the key to solving a major mystery about the shape of our universe.

Technical Summary: Failure of Local Equi-Lipschitzness for Families of Lorentz Distances to Cauchy Surface Foliations

Problem Statement
The classical Lorentzian splitting theorems assert that a timelike geodesically complete or globally hyperbolic spacetime $(M, g)$ containing a complete timelike line and satisfying the strong energy condition splits isometrically into a product spacetime $(\mathbb{R} \times S, dt^2 - \tilde{g})$ . A critical component in the proofs of these theorems (specifically those utilizing $p$ -d'Alembertian techniques) is the local equi-Lipschitz continuity of the families of Lorentzian distance functions $\{\ell(\cdot, \gamma(t))\}_{t>0}$ and $\{\ell(\gamma(-t), \cdot)\}_{t>0}$ in a neighborhood of the timelike line $\gamma$ .

In the absence of a pre-existing complete timelike line—such as in the context of Bartnik's cosmological splitting conjecture—researchers have considered replacing the line with a foliation by Cauchy surfaces $\Sigma_s = \{\tau = s\}$ , where $\tau$ is a Cauchy temporal function. It was previously asserted in the literature (specifically in Lemma 3.2 of [14]) that the families of Lorentz distances to and from these level sets, $\{\ell(\cdot, \Sigma_t)\}_{t>0}$ and $\{\ell(\Sigma_{-t}, \cdot)\}_{t>0}$ , are locally equi-Lipschitz continuous. This paper investigates the validity of this assertion.

Methodology
The authors employ a combination of geometric analysis and counterexample construction to test the regularity of Lorentzian distance functions relative to Cauchy surface foliations.

Theoretical Framework: The paper reviews the mechanism by which equi-Lipschitzness is established for timelike lines, relying on the "generalized timelike co-ray condition." This condition ensures that subsequential limits of maximizing curves remain timelike, preventing the blow-up of the $h$ -norm (background Riemannian norm) of initial tangent vectors of maximizing geodesics.
Counterexample Construction: To disprove the general claim for Cauchy surfaces, the authors construct specific counterexamples in Minkowski space $\mathbb{R}^{1,1}$ . They define a Cauchy temporal function $\tau(u, v) = F(u) + v$ using null coordinates, where $F$ is a smooth, steep function. By analyzing the intersection of the causal future of the origin with the level sets $\Sigma_s$ , they explicitly calculate the maximizing geodesics and their initial tangent vectors.
Analysis of Compactness and Energy Conditions: The paper examines the relationship between the equi-Lipschitz property and the existence of timelike lines. Using Lemma 3.3, the authors demonstrate that if the families of distances are locally equi-Lipschitz on a spacetime with compact Cauchy surfaces, a timelike line must exist. This is contrasted with the Ehrlich–Galloway example of a spacetime that is geodesically complete and globally hyperbolic but contains no timelike lines.

Key Contributions and Results

Refutation of General Equi-Lipschitzness: The primary result is the demonstration that the families of Lorentz distances $\{\ell(\cdot, \Sigma_t)\}_{t>0}$ ${ℓ (\cdot, Σ_{t})}_{t > 0}$ and $\{\ell(\Sigma_{-t}, \cdot)\}_{t>0}$ ${ℓ (Σ_{- t}, \cdot)}_{t > 0}$ are not locally equi-Lipschitz continuous in general, even for "steep" Cauchy temporal functions.
- Example 3.2: In 2D Minkowski space, the authors construct a steep Cauchy temporal function where the $h$ -norms of the initial tangents of maximizing geodesics from a fixed point to $\Sigma_s$ diverge to infinity as $s \to \infty$ . This directly contradicts Lemma 3.2 in [14] and invalidates the results in that paper which rely on the existence of global viscosity solutions to the Lorentzian eikonal equation derived from this assumption.
Necessity of Timelike Lines: Lemma 3.3 establishes that for a globally hyperbolic spacetime with compact Cauchy surfaces, the local equi-Lipschitzness of these distance families implies the existence of a timelike line. Consequently, in spacetimes like the Ehrlich–Galloway example (which has compact Cauchy surfaces but no timelike lines), no such Cauchy temporal function can exist that satisfies the equi-Lipschitz condition.
Equivalence to Bartnik's Conjecture: The paper formulates the existence of a Cauchy temporal function with locally equi-Lipschitz distance families as being equivalent to Bartnik's cosmological splitting conjecture.
- Corollary 3.5: Under the assumptions of global hyperbolicity, timelike geodesic completeness, compact Cauchy surfaces, and the strong energy condition, the spacetime splits isometrically if and only if such a Cauchy temporal function exists.
Proposed Conjectures:
- Conjecture 4.1: In a spacetime satisfying the conditions of Bartnik's conjecture, there exists at least one Cauchy temporal function for which the distance families are locally equi-Lipschitz.
- Conjecture 4.2: Under the same conditions, every Cauchy temporal function yields locally equi-Lipschitz distance families. The authors note that Conjecture 4.2 implies Conjecture 4.1 and is supported by Proposition 4.3, which proves the property holds for product spacetimes $(\mathbb{R} \times S, dt^2 - \tilde{g})$ with compact $S$ .

Significance
The paper clarifies a critical gap in the methodology for extending Lorentzian splitting theorems to settings without pre-existing timelike lines. By showing that the equi-Lipschitz property does not hold automatically for arbitrary Cauchy foliations, the authors demonstrate that the direct analogy to the timelike line case fails due to a lack of regularity.

The significance lies in reframing Bartnik's cosmological splitting conjecture: rather than being a purely geometric statement about curvature and completeness, it is shown to be equivalent to the existence of a specific regularity property (local equi-Lipschitzness) for a family of distance functions associated with a Cauchy temporal function. This provides a new analytical perspective for attacking the conjecture, suggesting that proving the splitting theorem may require constructing a specific "good" temporal function rather than relying on the properties of any arbitrary one. The results also correct the literature regarding the existence of global viscosity solutions to the Lorentzian eikonal equation in the general case.