⚛️ high-energy theory

Negative running of gravitational positivity

This paper demonstrates that while certain non-minimal interactions in effective field theories induce a negative one-loop renormalization group running of Wilson coefficients, gravitational loops generate dominant positive infrared contributions that preserve dispersive bounds, provided the number of non-minimally coupled particles respects the species bound.

Original authors: J. Fernandez, M. Ruhdorfer, J. Serra

Published 2026-03-18

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: J. Fernandez, M. Ruhdorfer, J. Serra

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 the universe as a giant, complex machine. For decades, physicists have been trying to figure out the "blueprints" for how this machine works at the smallest scales. We have a great set of instructions for how things work at medium sizes (the Standard Model), but when we try to zoom in all the way to the very center of a black hole or the moment of the Big Bang, the instructions get fuzzy. This is the realm of Quantum Gravity.

The problem? We can't build a machine big enough to test these tiny scales directly. It's like trying to figure out the rules of a game by only watching the players from a mile away.

So, physicists use a clever trick called Effective Field Theory (EFT). Think of EFT as a "rulebook" for a specific neighborhood. It doesn't need to know the secrets of the whole universe; it just needs to know how things behave in that specific neighborhood. In this rulebook, there are numbers called Wilson Coefficients. These numbers act like "knobs" or "dials" that determine how strongly particles interact.

The Big Question: Can the Dials Go Negative?

For a long time, physicists believed these "dials" had to be positive. Why? Because of a fundamental rule of the universe called Causality (cause must come before effect) and Unitarity (probabilities must add up to 100%). If a dial were negative, it would imply that particles could do impossible things, like travel back in time or create energy out of nothing.

However, this paper asks a tricky question: What happens when we turn on gravity?

The Plot Twist: Gravity is a Sneaky Neighbor

The authors of this paper investigated what happens when you have a lot of particles interacting with gravity. They found a surprising phenomenon: Negative Running.

Imagine you are watching a balloon slowly lose air. The pressure (the "coefficient") is dropping. In the world of these particles, the "pressure" of their interactions is actually decreasing as you look at lower energies (like zooming out from the machine).

Usually, this would be a disaster. A negative pressure usually means the rules of physics are broken. But here's the twist: Gravity saves the day.

The Analogy: The Tug-of-War

Think of the universe as a giant Tug-of-War game.

Team Non-Minimal (The Rebels): There are new, exotic particles interacting with gravity in weird, "non-minimal" ways. They are pulling the rope toward the negative side. They want to make the interaction coefficients negative. The paper shows that if you have enough of these particles (a lot of "species"), they can actually win this tug-of-war and drive the coefficient negative.
Team Gravity (The Referee): Gravity itself acts as a referee. In the past, we thought gravity just sat there. But this paper shows that gravity has a special move: The Graviton Pole.

When you have a "negative" coefficient caused by the Rebels, gravity steps in with a massive, positive contribution. It's like the referee blowing a whistle and adding a huge weight to the positive side of the rope.

The "Species Bound" Limit

The paper calculates exactly how many "Rebel" particles you can have before the universe breaks.

If you have a few particles, Gravity wins easily. The coefficient stays positive.
If you have too many particles (more than a specific limit related to the Planck scale, which is the energy scale of gravity), the Rebels might win, and the coefficient goes negative.
The Catch: If the coefficient goes negative, it implies the theory is inconsistent with the fundamental rules of the universe (causality).

Therefore, the paper concludes: The universe has a "species bound." You cannot have an infinite number of these exotic particles. There is a hard cap on how many types of light particles can exist in our universe, or else the laws of physics (specifically causality) would break down.

The "Smearing" Technique

How did they prove this? They used a mathematical technique called Smearing.

Imagine trying to measure the height of a mountain, but there's a thick fog (the "infrared divergence") right at the base that makes the measurement impossible. Instead of trying to measure the exact peak, they "smeared" the measurement over a wide area. This smoothed out the fog and allowed them to see the true shape of the mountain.

By "smearing" the mathematical equations over different distances, they could separate the "bad" negative effects from the "good" positive effects of gravity. They found that the positive effects of gravity (the referee) are just strong enough to cancel out the negative effects, provided you don't have too many particles.

The Takeaway

This paper is a detective story about the limits of our universe.

The Mystery: Can gravity allow particle interactions to become "negative" (which usually breaks physics)?
The Clue: Yes, if you have enough exotic particles, they can push the numbers negative.
The Solution: Gravity itself provides a counter-force.
The Verdict: The universe is consistent, but only if the number of particle types is limited. If there were too many, the "negative running" would overwhelm gravity's counter-force, and the laws of causality would collapse.

In short: The universe is a delicate balance. Gravity acts as a safety net, but it can only catch us if we don't have too many "rebel" particles trying to break the rules. This helps physicists understand the "Swampland"—the vast landscape of theories that look possible but are actually impossible because they violate these fundamental rules.

1. Problem Statement

The paper addresses a central tension in Effective Field Theory (EFT) and Quantum Gravity: the compatibility of positivity bounds (derived from unitarity, locality, and causality via dispersion relations) with renormalization group (RG) running induced by gravitational interactions.

Positivity Bounds: In the absence of gravity, dispersion relations typically require the Wilson coefficients of leading higher-derivative operators (e.g., $(\partial \phi)^4$ , $F^4$ , $C^4$ ) to be positive.
The Conflict: The authors investigate whether specific non-minimal couplings in the presence of dynamical gravity can induce a negative running (i.e., the coefficients decrease toward the infrared). If a coefficient runs negative, it might seemingly violate the positivity bounds derived from the UV completion.
Specific Question: Can non-minimal interactions (coupling scalars, photons, or gravitons to other fields) drive the leading EFT coefficients to negative values, and if so, are these theories consistent with the fundamental principles of the S-matrix?

2. Methodology

The authors employ a combination of perturbative quantum field theory techniques and S-matrix theory constraints:

One-Loop RG Evolution: They calculate the one-loop beta-functions for the leading Wilson coefficients ( $c_2$ $c_{2}$ for scalars, $\alpha_2$ $α_{2}$ for photons, $g_4$ $g_{4}$ for gravitons) in 4 dimensions.
- They utilize unitarity cuts (specifically double cuts) to extract the ultraviolet (UV) divergent parts of the amplitudes, which determine the anomalous dimensions.
- They consider theories with shift-symmetric scalars, photons, and gravitons, coupled to massless particles (vectors, fermions, scalars) via non-minimal interactions (e.g., $\phi V \tilde{V}$ , $\sigma F^2$ , $\sigma C^2$ ).
Partial-Wave Analysis: To understand the sign of the running, they perform a partial-wave decomposition of the scattering amplitudes. This allows them to isolate the contribution of the $t$ -channel graviton exchange (the source of the $1/t$ pole) versus non-minimal contact terms.
Dispersion Relations with Smearing: To test consistency with positivity, they construct non-forward dispersion relations.
- They address the Coulomb singularity ( $1/t$ pole from tree-level graviton exchange) by smearing the dispersion relations over the momentum transfer $p = \sqrt{-t}$ .
- They work in $d = 4 - 2\epsilon$ dimensions to regulate infrared (IR) divergences, treating the $1/\epsilon$ poles as physical regulators that can be traded for large impact parameters.
- They derive smeared sum rules that relate the IR coefficients to an integral over the UV spectral density, ensuring the UV contribution remains positive.

3. Key Contributions and Results

A. Discovery of Negative Running

The authors demonstrate that non-minimal three-point interactions can induce a positive beta-function (leading to a decrease of the Wilson coefficient toward the IR, i.e., negative running) for all three cases:

Scalars ( $c_2$ ): A positive $\beta$ -function arises if there are $N_V > 16$ massless vectors non-minimally coupled via $\phi V \tilde{V}$ operators.
Photons ( $\alpha_2$ ): A positive $\beta$ -function arises if there are $N_\chi > 136$ fermions (via $\bar{\chi}F\chi$ ) or $N_V > 45$ scalar-vector pairs (via $\sigma F^2$ ).
Gravitons ( $g_4$ ): A positive $\beta$ -function arises if there are scalars with $\sigma C^2$ non-minimal couplings.

Crucial Mechanism: The negative running is suppressed by the Planck scale ( $M_{Pl}$ ). It arises specifically from the interference between minimal gravitational couplings (mediated by the $t$ -channel graviton pole) and non-minimal interactions. In the decoupling limit ( $M_{Pl} \to \infty$ ), the running becomes manifestly positive (standard behavior), proving that dynamical gravity is essential for the negative running.

B. Resolution via Gravitational Dispersion Relations

The paper resolves the apparent conflict between negative running and positivity bounds:

Tree-Level Gravity: The tree-level graviton exchange contributes a negative term to the dispersion relation (due to the $1/t$ pole), which allows the Wilson coefficients to be negative by an amount proportional to $(M M_{Pl})^{-2}$ .
One-Loop Consistency: The authors show that the magnitude of the negative running derived in the RG analysis is parametrically consistent with the maximum negativity allowed by the tree-level gravitational contribution in the dispersion relation.
Species Bound: This consistency holds provided the number of non-minimally coupled species ( $N$ ) is bounded by the species scale:
$N \lesssim \left( \frac{4\pi M_{Pl}}{M} \right)^2$
If $N$ exceeds this bound, the theory would violate the positivity constraints.

C. Graviton Scattering ( $g_4$ )

For the four-graviton coefficient $g_4$ , the tree-level dispersion relation requires $g_4 > 0$ (as the graviton pole does not contribute to the spin-4 subtracted relation). However, the authors show that one-loop gravitational corrections generate positive IR contributions that restore consistency with the negative running of $g_4$ , provided the coupling strength and number of scalars satisfy:
$N_\sigma c_{\sigma C^2}^2 \lesssim \left( \frac{4\pi M_{Pl}}{M^2} \right)^2$

4. Technical Details of the Analysis

Beta-Function Calculation: The beta-functions are derived from the coefficients of the bubble integrals in the Passarino-Veltman decomposition. The authors explicitly separate contributions from:
- Pure GR (always negative running).
- Pure Non-minimal/Contact (always negative running).
- Mixed GR $\times$ Non-minimal (can be positive, driving the negative running).
Partial-Wave Decomposition: They show that the $t$ -channel double cut vanishes in the forward limit ( $t \to 0$ ) for pure non-minimal interactions (due to angular momentum selection rules), leading to negative running. However, the presence of the graviton pole ( $1/t$ ) prevents this vanishing, allowing the mixed term to dominate and flip the sign.
Smearing Functional: They construct specific polynomial smearing functionals $\psi(p)$ that ensure the UV representation of the dispersion relation is positive while regularizing the IR $1/t$ pole.

5. Significance and Implications

Swampland Program: The results provide a concrete example of how loop effects can drive EFT coefficients outside the region allowed by tree-level positivity constraints. This refines the "Swampland" criteria, suggesting that theories with many non-minimally coupled particles are only consistent if they respect the species bound.
Weak Gravity Conjecture (WGC): The findings support the WGC, as the requirement to avoid super-extremal states (which would violate positivity) is linked to the number of species and their couplings.
Gravity as a Regulator: The paper highlights that dynamical gravity is not just a background but an active regulator that modifies the RG flow and the validity of positivity bounds. The "negative running" is a unique signature of the interplay between minimal and non-minimal gravitational couplings.
Methodological Advancement: The work demonstrates the necessity of including one-loop gravitational effects when applying dispersion relations to EFTs, showing that tree-level bounds are insufficient in the presence of dynamical gravitons.

In summary, the paper proves that while non-minimal couplings can drive EFT coefficients negative, this does not violate fundamental principles of quantum gravity; rather, it imposes a strict upper bound on the number of such particles (the species bound), ensuring the theory remains within the "Landscape" of consistent quantum gravities.