⚛️ general relativity

Metric Signature as an Auxiliary Order Parameter: A Causal Viability Criterion

This paper proposes a branchwise formulation of metric gravity where the spacetime signature is encoded by an internal field, demonstrating that among various possible signature vacua, only the standard Lorentzian branch satisfies the necessary conditions for causal viability, thereby offering an anthropic explanation for the observed signature of our universe.

Original authors: Miguel Bermudez

Published 2026-09-09

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Miguel Bermudez

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

In the standard view of our universe, the fabric of space and time is woven with a specific texture. This texture, known to physicists as the metric signature, dictates the fundamental difference between moving forward in time and moving through space. It is the reason a clock ticks forward while a ruler measures a fixed distance, and it is the mathematical rule that separates the past from the future. For over a century, this distinction has been treated as a fixed starting point for the laws of gravity, a rigid backdrop against which the drama of the cosmos plays out. However, the mathematics of geometry allows for other textures. Just as a fabric can be woven in different patterns, the underlying structure of spacetime could theoretically possess different arrangements of time and space dimensions. Some of these arrangements would look like our own, while others would be entirely alien, perhaps treating all directions as space-like or mixing them in ways that defy our everyday experience of cause and effect. The question that has long lingered in the minds of theoretical physicists is whether these different textures are merely mathematical curiosities or if they could represent real, physical states of the universe that we simply do not inhabit.

A new study by Miguel Bermudez at the Université Paris Cité proposes a way to treat these different spacetime textures not as separate universes, but as different phases of a single, unified field. The researcher introduces a new internal field, a mathematical object that acts like a switch, encoding the signature of spacetime without requiring a new, independent particle to carry the signal. In this framework, the geometry of the universe is built from two parts: a standard grid that defines distances and a symmetric internal field that determines the "inertia" or the signature of those distances. When this internal field settles into a stable state, it locks the universe into a specific configuration. The study shows that this field does not need to be a dynamic actor that moves or changes on its own; instead, it functions as an auxiliary parameter, a fixed setting that defines the rules of the game for the gravity that follows.

The core discovery of this work is that while the mathematical potential allows for five distinct stable configurations, only one of them supports the kind of physics we observe. The researcher analyzed a specific model of gravity that includes a correction term to the standard theory, a common approach used to explore how gravity behaves at very high energies. By testing this model against the five possible configurations, the study found that four of them fail to describe a viable universe. Two of the configurations correspond to a purely Euclidean geometry, where time and space are indistinguishable, meaning there is no flow of time and no way for events to unfold in a sequence. Another configuration creates a split signature where the rules of cause and effect break down, making it impossible to predict the future from the present. The final configuration is a mirror image of our own universe, but it comes with a fatal flaw: it predicts that particles would behave in ways that violate the stability of matter, leading to runaway energies that would tear any structure apart.

Only one specific configuration survives this test. It is the Lorentzian signature, the one that matches our reality, where time is distinct from space and the laws of physics allow for a stable, predictable evolution of events. In this specific branch, the gravitational waves travel at the speed of light without causing instability, and the extra scalar particle predicted by the theory remains stable rather than exploding into chaos. The study demonstrates that the algebraic equations governing the internal field can assign equal value to all five configurations, yet the physics of the resulting universe is not equal. The difference lies in how the universe responds to the presence of observers. The researcher argues that if we assume the existence of observers who can record history and infer laws, we must condition our view on the fact that such observers can only exist in a branch where cause and effect are stable.

This leads to a quiet but profound conclusion: the reason we see a universe with time flowing in one direction is not necessarily because the laws of physics forced it to be that way from the beginning. Instead, it may be that the laws allow for many possibilities, but only one of those possibilities can support the complex, stable structures required for life to exist. The study does not explain how the universe might jump from one configuration to another, nor does it provide a probability for why we found ourselves in this specific branch. It simply establishes that the transition between these states would require passing through a point of mathematical breakdown where the description of space and time fails. Therefore, the universe we observe is not just a random accident, but the only viable outcome for a system that must sustain a stable, causal history. The work reframes the signature of spacetime from a fixed kinematic rule into a selection criterion, where the very existence of an observer acts as a filter, selecting the only branch of reality that can hold a story together.

Technical Summary: Metric Signature as an Auxiliary Order Parameter

Problem Statement
The paper addresses a fundamental question in gravitational theory: Can different spacetime signatures (Euclidean, Lorentzian, and split-signature) be represented as vacuum sectors of a single covariant field theory without introducing a new propagating "signature mode"? While metrics of different inertia are natural in differential geometry and appear in contexts like quantum cosmology and gravitational path integrals, it remains unclear if a unified theory can accommodate these sectors while maintaining the standard propagating degrees of freedom of metric gravity within each fixed sector.

Methodology
The author proposes a "branchwise" formulation of metric gravity using an auxiliary internal field. The core construction involves:

Field Definition: Introducing a tetrad $e^\mu_a$ and a real symmetric internal field $H_{ab}$ . The spacetime metric is defined as $g_{\mu\nu} = e^\mu_a H_{ab} e^\nu_b$ .
Action Formulation: The action is constructed as $S[e, H] = \int d^4x \sqrt{|g|} [L_g(g) - V(H)]$ , where $L_g$ is an arbitrary diffeomorphism-invariant metric Lagrangian and $V(H)$ is an algebraic potential.
Potential Choice: A specific $O(4)$ -invariant quartic potential, $V(H) = \frac{\lambda}{4} \text{Tr}(H^2 - 1)^2$ , is chosen. Its non-degenerate minima satisfy $H^2 = 1$ , yielding five distinct vacuum orbits labeled by inertia $(p, q)$ where $p+q=4$ .
Field Equation Analysis: The author derives the coupled equations of motion for the tetrad and $H$ . They utilize Sylvester's law of inertia to perform a local change of variables within connected regions where $H$ is non-degenerate, separating the variables into those encoding the metric and those encoding the internal frame.
Spectral Diagnostic: To test the viability of these branches, the author applies the framework to $R + \alpha R^2$ gravity. They analyze the spectrum of excitations (gravitons and scalarons) and the nature of the initial value problem (hyperbolicity) for each of the five signature classes.

Key Contributions and Results

Auxiliary Nature of $H$ : The paper demonstrates that in any connected region where $H$ is non-degenerate, the internal field $H$ acts strictly as an auxiliary order parameter. The components of $H$ transverse to its $O(4)$ orbit are fixed algebraically by the potential, while the tangent components correspond to gauge degrees of freedom. Consequently, the local propagating phase space of the coupled theory is identical to that of the underlying metric theory $L_g[g]$ alone; no new propagating signature modes exist.
Branchwise Separation: The theory classifies fixed-signature phases. The transition between branches (where the signature changes) necessarily passes through a degenerate point ( $\det H = 0$ ), where the standard curvature action is undefined. Thus, the theory is not a dynamical description of signature change but a classification of coexisting sectors.
Spectral and Causal Viability: Applying the framework to $R + \alpha R^2$ $R + α R^{2}$ gravity reveals that the algebraic degeneracy of the potential does not imply physical equivalence of the branches:
- Lorentzian (1, 3): Supports standard massless graviton residues and a non-tachyonic scalaron ( $m^2 > 0$ ). The initial value problem is hyperbolic.
- Lorentzian (3, 1): While hyperbolic, this branch exhibits a negative graviton residue and a tachyonic scalaron ( $m^2 < 0$ ) for fixed universal couplings.
- Euclidean (0, 4) and (4, 0): The linearized equations are elliptic, precluding real-time Cauchy evolution.
- Split (2, 2): The equations are ultrahyperbolic; generic local data on a codimension-one hypersurface do not define a well-posed unconstrained Cauchy problem.
Anthropic Selection Criterion: The paper proposes that the observed Lorentzian signature can be understood as a result of conditioning on the existence of persistent observers. Since observers require a stable causal structure and predictive evolution, the probability of observing a branch is effectively zero for sectors that are elliptic, ultrahyperbolic, or contain ghosts/tachyons. This selection mechanism operates independently of the prior probabilities assigned to the branches by the potential.

Significance and Claims
The paper claims to provide a compact, algebraic mechanism to embed all metric signatures into a single field space while preserving the standard local dynamics of metric gravity within each non-degenerate branch. It establishes that the existence of signature vacua does not necessitate additional propagating modes.

Crucially, the author modestly frames their conclusion: the theory does not provide a dynamical mechanism for transitions between signatures, nor does it define a probability measure over the branches. Instead, it offers a "causal viability criterion." The observed Lorentzian signature is interpreted not as a dynamical inevitability of the potential, but as a necessary condition for the existence of observers capable of recording history. The paper explicitly states that a full theory of dynamical transitions or prior probabilities would require additional input, such as a UV completion resolving degenerate metrics or a defined integration cycle for complex metrics.

Technical Summary: Metric Signature as an Auxiliary Order Parameter

More like this