Spectrum of pure $R^2$ gravity: full Hamiltonian analysis

This paper resolves controversies regarding the particle spectrum of pure $R^2$ gravity by demonstrating through a full Hamiltonian analysis that while the theory propagates three degrees of freedom globally, its linearized spectrum around Minkowski and other $R=0$ spacetimes is empty due to a constraint degeneracy that renders these backgrounds surfaces of strong coupling, although the universe can still evolve through such singularities.

The Gist

The Big Picture: A Gravity Theory with a "Ghost" Problem

Imagine gravity not just as the force that keeps your feet on the ground, but as a complex machine with moving parts. In our standard understanding (Einstein's General Relativity), this machine has specific "degrees of freedom"—think of them as independent knobs you can turn to create ripples or waves in space-time. Usually, we expect these theories to have three such knobs: two for the standard gravitational waves (like the ripples on a pond) and one extra "scalar" knob (like a breathing mode that expands and contracts space).

This paper investigates a specific, slightly weird version of gravity called Pure $R^2$ Gravity. In this theory, the rules of the game are changed so that the "energy" of the system depends on the square of the curvature of space-time, rather than just the curvature itself.

Recent studies suggested that if you look at this theory around a flat, empty universe (Minkowski space), something strange happens: all the knobs disappear. The theory seems to have zero moving parts. It's like a car engine that, when idling in a garage, has no pistons moving at all.

The authors of this paper wanted to solve the mystery: Is the engine actually broken, or is our way of looking at it the problem?

The Detective Work: The "Hamiltonian" Analysis

To get to the bottom of this, the authors didn't just look at small ripples (perturbations); they performed a "full Hamiltonian analysis."

The Analogy:
Imagine you are trying to understand a complex clock.

• The Old Way (Linear Perturbation): You gently tap the clock and listen to the sound. If the clock is in a specific state (like being frozen in a block of ice), it might not make a sound when tapped. You might conclude, "This clock has no moving gears."
• The New Way (Hamiltonian Analysis): The authors took the clock apart, counted every single gear, spring, and screw, and mapped out exactly how they are connected. They looked at the rules (constraints) that govern how the gears can move.

What They Found:

1. The Full Machine Works: When they counted the gears in the full, un-truncated theory, they confirmed there are three degrees of freedom. The engine does have moving parts. It is a healthy, functioning theory with a massive graviton and a scalar field.
2. The "Ice Block" Effect: The reason the old studies saw "zero" degrees of freedom is that they were looking at the theory in a very specific, "frozen" state (flat space or other special backgrounds like black holes). In these specific states, the rules of the game change temporarily.
   • It's like a dancer who is perfectly still. If you try to analyze their movement by only looking at the stillness, you conclude they have no ability to dance. But the ability is there; it's just hidden by the specific pose.
   • Mathematically, the "constraints" (the rules that limit movement) change their nature. Ten rules that usually stop movement become "gauge symmetries" (rules that allow for freedom), and the rules that usually allow movement become too restrictive. The result? The math says "0 degrees of freedom," but this is an illusion caused by the specific background.

The "Strong Coupling" Mystery

The paper explains that these special backgrounds (where the Ricci scalar $R=0$, like flat space or Schwarzschild black holes) are "surfaces of strong coupling."

The Analogy:
Imagine trying to walk through a field of tall, dense grass.

• Normal Ground: You can walk easily. You can take small steps (perturbations) and see where you are going.
• The Strong Coupling Surface: This is a patch of mud so thick that your small steps don't work. If you try to take a tiny step, you sink. To move, you have to make a huge, non-linear leap.

The authors show that if you try to study the theory around these special backgrounds using "small steps" (perturbation theory), you will never find the moving parts, no matter how many steps you take. The math breaks down because the "small step" assumption is invalid there. The physics becomes "non-perturbative," meaning you can't understand it by just adding up small corrections; you have to look at the whole picture at once.

The Plot Twist: Can We Cross the "Ice"?

A major question in physics is: If a theory has these "frozen" surfaces, can the universe actually evolve through them? Or are they like walls that the universe can never cross?

• The Old Belief: Singular surfaces are usually like walls (separatrices). You can approach them, but you can't cross them.
• The Paper's Discovery: The authors analyzed the "phase space" (a map of all possible states) of a cosmological universe in this theory. They found that the universe can actually cross the $R=0$ surface.

The Analogy:
Imagine a river flowing toward a waterfall (the singular surface).

• Standard physics might say the river stops at the edge.
• This paper shows that the river doesn't stop; it flows over the edge and continues on the other side. The universe can evolve from a state where $R \neq 0$ to a state where $R=0$ and then to $R \neq 0$ again.

Summary of Key Takeaways

1. The Theory is Healthy: Pure $R^2$ gravity does have three degrees of freedom (a graviton and a scalar). It is not "empty."
2. The Illusion of Emptiness: When you look at this theory around flat space or black holes using standard "small ripple" math, it looks empty. This is because the math gets confused by the specific geometry of those spaces.
3. The Limit of Small Steps: You cannot use standard perturbation theory (small steps) to study the neighborhood of these special backgrounds. The physics there is "strongly coupled," requiring a full, non-linear view.
4. Crossing the Line: The universe is not trapped on one side of these special backgrounds. It can dynamically evolve through them, passing through the "strong coupling" zone.

In short, the paper clarifies that the "empty spectrum" seen in previous studies was a mirage created by using the wrong tool (linear perturbation) in a place where that tool doesn't work. The full theory is robust, and the universe can navigate through these tricky regions.

Technical Summary

Technical Summary: Spectrum of Pure $R^2$ Gravity: Full Hamiltonian Analysis

Problem Statement
Recent works have highlighted a discrepancy regarding the particle spectrum of pure Ricci-scalar-squared ($R^2$) gravity. While the full theory is generally understood to propagate three degrees of freedom (a massive spin-2 graviton and a scalar), linearized perturbations around Minkowski spacetime were found to exhibit an empty spectrum with no propagating modes. This contradiction between the full non-linear theory and its linearized approximation around specific backgrounds raised questions about the nature of these singularities, whether they are unique to Minkowski space, and whether the absence of degrees of freedom persists at the non-linear level. Previous analyses relied on perturbative methods which may be ill-suited for strongly coupled regimes.

Methodology
The authors perform a full, non-perturbative Hamiltonian constraint analysis of pure $R^2$ gravity in the Jordan frame. The methodology proceeds as follows:

1. ADM Decomposition: The action is decomposed into Arnowitt-Deser-Misner (ADM) variables, introducing auxiliary fields to handle the higher-derivative nature of the theory.
2. Canonical Formulation: A canonical action is constructed using the Dirac-Bergmann algorithm to systematically identify primary and secondary constraints.
3. Constraint Classification: The authors classify constraints into first-class (generating gauge symmetries) and second-class (eliminating phase-space variables) to count the physical degrees of freedom ($N_{phy}$) using the formula $N_{phy} = \frac{1}{2}(N_{can} - 2N_{1st} - N_{2nd})$.
4. Perturbative Analysis: The authors analyze linear and higher-order perturbations around specific backgrounds (Minkowski, Schwarzschild, radiation-dominated FLRW) by shifting field variables and truncating the action. They track how the constraint algebra changes upon linearization.
5. Cosmological Phase-Space Analysis: To determine if singular backgrounds are dynamically accessible, the authors formulate the cosmological sector as a dynamical system in the Jordan frame, analyzing trajectories in phase space to see if they can cross the $R=0$ surface.

Key Contributions and Results

• Full Theory Degrees of Freedom: The Hamiltonian analysis confirms that the full, non-linear pure $R^2$ theory propagates exactly three degrees of freedom. This is derived from a constraint structure consisting of 24 canonical variables, 4 first-class constraints, and 10 second-class constraints.
• Empty Linearized Spectrum: The authors confirm that the linearized spectrum around Minkowski spacetime is indeed empty ($N_{phy}=0$). This occurs because the linearization process alters the nature of the constraints:
  • Ten second-class constraints of the full theory become first-class.
  • The three momentum constraints degenerate into a single longitudinal constraint.
  • This results in 12 first-class constraints and 0 second-class constraints, yielding zero propagating modes.
• Generalization to Traceless-Ricci Backgrounds: The phenomenon is not unique to Minkowski space. The authors demonstrate that all traceless-Ricci spacetimes (where the Ricci scalar $R=0$), including Schwarzschild, Kerr, and radiation-dominated cosmological spacetimes, exhibit the same empty linearized spectrum. The mechanism is the vanishing of the Poisson bracket between primary and secondary traceless constraints (specifically where the conjugate momentum $\rho \approx 0$), which triggers the change in constraint classification.
• Failure of Perturbation Theory: The authors show that expanding perturbations to arbitrary higher orders around these singular backgrounds consistently yields an empty spectrum. The constraint structure remains degenerate at every order because the perturbative expansion forces the Ricci scalar to vanish exactly at each step. This indicates that the vicinity of these backgrounds constitutes a strong coupling regime where the dynamics are non-perturbative. The absence of degrees of freedom in the perturbative expansion is a limitation of the method, not a physical property of the theory near these backgrounds.
• Accidental Gauge Symmetry: The change in constraint character corresponds to the emergence of an "accidental gauge symmetry" in the linearized theory around Ricci-flat backgrounds, constructed explicitly using the Bach tensor projector.
• Dynamical Accessibility: Contrary to the expectation that singular surfaces act as separatrices that trajectories cannot cross, the cosmological phase-space analysis reveals that the $R=0$ surface is dynamically accessible. Trajectories in the phase space of a homogeneous universe can penetrate through the singular $R=0$ surface, implying that the strong coupling feature is physically relevant for evolving cosmologies, not just static solutions.

Significance
The paper resolves the controversy regarding the spectrum of pure $R^2$ gravity by establishing that the theory consistently propagates three degrees of freedom in the full non-linear regime. The "empty spectrum" observed in linearized analyses is identified as a pathological feature of perturbation theory applied to singular backgrounds where the constraint algebra degenerates. The work clarifies that these singularities are generic to all traceless-Ricci spacetimes and that the dynamics near them are non-perturbative. Furthermore, the demonstration that the universe can dynamically cross the $R=0$ surface challenges the notion that such strong-coupling regions are dynamically isolated, suggesting that the behavior of perturbations during such transitions requires non-perturbative treatment. The authors suggest these findings may extend to general $f(R)$ theories at points where $f'(R)=0$.