Relational Cosmology: Covariant and Canonical Dressings, and the Hamiltonian of Internal Time
summary
The gist
Relational cosmology compares covariant and canonical dressings with gauge-fixing reductions for the flat FLRW minisuperspace with a homogeneous massless scalar field The analysis shows that the…
In short
The paper compares different ways to describe cosmology in flat FLRW space using covariant dressing and canonical constraint reduction. It investigates how choosing a relational frame affects the resulting equations of motion and constraints. The key finding is that a suitable relational frame must not only define coordinates but also ensure the dressed variational principle retains enough independent equations to recover the original dynamics.
Key concepts
- Covariant Dressing
- This method involves modifying the action by dressing it with frame fields. It keeps track of how a chosen reference frame changes, treating its reorientations as vector fields on field space rather than spacetime movements. This approach retains more structure compared to simpler methods.
- Canonical Constraint Reduction
- This technique simplifies the system by fixing the gauge and reducing the number of constraints. It eliminates certain structures present in covariant dressing, leading to a different set of equations and phase spaces. This method is often used for clearer Hamiltonian formulations.
- Relational Frame Admissibility
- A frame is considered suitable if it meets two criteria: first, it must define a non-degenerate coordinate along the solution space's gauge fibers. Second, the resulting dressed variational principle must still contain enough independent equations to reproduce the original constrained dynamics. This ensures the frame captures essential physics.
- Relational Hamiltonian
- This is a bulk quantity derived from the Legendre transform of a relational action. It generates evolution relative to a chosen dynamical frame, distinguishing it from gauge evolution. It arises from the structure of constraints and is not necessarily conserved in time.
Terminology used across episodes
This episode discusses
- Relational Cosmology: Covariant and Canonical Dressings, and the Hamiltonian of Internal Time · Paper Radio
- The Problem of Time in Quantum Gravity
- Relational Observables in Gravity: a Review
- Robustness of key features of loop quantum cosmology
- Time in quantum theory, the Wheeler-DeWitt equation and the Born-Oppenheimer approximation
- Singularity resolution depends on the clock
- Time in quantum cosmology
- An effective approach to the problem of time
- Effective approach to the problem of time: general features and examples
- Effective relational dynamics of a nonintegrable cosmological model
- Multiple choices of time in quantum cosmology
- What is Dynamics in Quantum Gravity?
- Clocks and dynamics in quantum models of gravity
- Internal clock formulation of quantum mechanics
- Relational quantum dynamics of the black hole interior: singularity resolution and quantum bounce
- Quantum constraints, Dirac observables and evolution: group averaging versus Schroedinger picture in LQC
- Gravity quantized
- Reduced Loop Quantization with four Klein-Gordon Scalar Fields as Reference Matter
- Chaos, Dirac observables and constraint quantization
- Conditional probabilities with Dirac observables and the problem of time in quantum gravity
- The Trinity of Relational Quantum Dynamics
The paper
Relational Cosmology: Covariant and Canonical Dressings, and the Hamiltonian of Internal Time · Read on arXiv
Francesco Sartini
Okinawa Institute of Science and Technology Graduate University
We compare covariant and canonical relational dressings with gauge-fixing reductions for the flat FLRW minisuperspace with a homogeneous massless scalar. We distinguish covariance of the dressing map from symmetries of the relational theory: fixing a clock makes its frame fields background structures, so its reorientations act as vector fields on relational field space rather than as spacetime diffeomorphisms. This loss of relational-time reparametrization covariance permits a nontrivial relational Hamiltonian, distinct from the undressed and dressed Hamiltonian constraints, which vanish on the constraint surface. We identify the conditions under which the dressed equations reproduce the original constrained dynamics. The relational generator defines a Hamiltonian flow even when relational-time translations are not symmetries of the dressed action. In each clock frame, it corresponds to the canonical complete observable and reduced Hamiltonian on the appropriate solution space, through the Legendre map and relational evaluation. The classical descriptions are compatible but retain distinct variational principles and intermediate phase spaces; their quantum equivalence remains an open question.
Transcript
Introduction to the show: ident: General Relativity and Quantum Cosmology. The day in general relativity and quantum cosmology.
Kai: We're looking at the paper: "Relational Cosmology: Covariant and Canonical Dressings, and the Hamiltonian of Internal Time".
Mira: Here's the gist of the paper. Relational cosmology compares covariant and canonical dressings with gauge-fixing reductions for the flat FLRW minisuperspace with a homogeneous massless scalar field The analysis shows that the choice of relational frame affects not only how the fields are parametrized, but also the variational principle and the constraint structure of the resulting theory A lapse condition such as N = 1, imposed at the level of the action, removes the independent lapse equation from the variational principle Suitable dynamical frames, such as a a or Φ or even πa, allow the remaining equations to recover the omitted frame equation through the off-shell Noether identity The admissibility of a relational frame therefore requires more than the non-degeneracy of the frame map: its dressed variational principle must also retain enough independent equations to reproduce the original constrained solution space Pruning redundant reasoning tokens The paper investigates how covariant relational dressing, canonical constraint reduction, and relational Hamiltonian evolution are related in flat FLRW minisuperspace with a homogeneous massless scalar field The model provides a setting in which these constructions can be compared explicitly, while retaining the essential features of a constrained gravitational system: time-reparametrization invariance, a Hamiltonian constraint, and dynamical variables that can serve as internal reference frames The authors distinguish covariance of the dressing map from symmetries of the relational theory by noting that fixing a clock makes its frame fields background structures, so its reorientations act as vector fields on relational field space rather than as spacetime diffeomorphisms This loss of relational-time reparametrization covariance permits a nontrivial relational Hamiltonian, distinct from the undressed and dressed Hamiltonian constraints, which vanish on the constraint surface The classical descriptions are compatible but retain distinct variational principles and intermediate phase spaces; their quantum equivalence remains an open question How the training signal is built The paper compares several candidate frames, including the scalar-field frame and the momentum frame, to illustrate how a frame can be admissible or not For the scalar field as a frame, the dressed variational principle retains independent dynamics when Edr a and Edr N both vanish The lapse provides a contrast but should not be regarded as a frame variable in the same sense as the scalar field because fixing N = 1 is a proper-time gauge choice rather than a relational dressing The momentum as a frame fails because πΦ is constant on shell, making the condition πΦ = u incompatible with the original solution space The relevant criterion for a relational frame is twofold: it must define a non-degenerate coordinate along the gauge fibers on the solution space of interest, and the dressed variational principle must retain enough independent equations to recover the original constrained dynamics Where the gains come from The analysis reveals distinctions between covariant dressing and canonical reduction regarding intermediate structures The covariant construction retains the dressed lapse and its associated second-class constraints, whereas canonical reduction eliminates these structures through gauge-fixing and constraint reduction The agreement of the relational generators on the constraint surface should not be interpreted as an off-shell equality The resulting relational Hamiltonian is a bulk quantity arising from the Legendre transform of the relational action, rather than a charge associated with a spatial boundary or corner The gains come from The study clarifies that generic reparametrizations of relational time are not symmetries of the dressed action, yet an integrable generator of evolution can be defined even when corresponding transformations are not symmetries of the dressed action The canonical analysis identifies the generator with the complete observable associated with the momentum conjugate to the chosen frame, on the corresponding solution branch The comparison between covariant dressing and canonical reduction yields compatible classical descriptions on the appropriate constraint surface and solution branches, but they do not produce identical intermediate variational principles or phase spaces The correspondence is also branch-dependent and requires care wherever the rescaling of the constraint or the clock condition becomes singular The classical correspondence established here does not by itself prove quantum equivalence The gains come from The paper concludes that the distinction between gauge evolution and relational evolution is reflected in the constraint structure, where time reparametrizations identify different parametrizations of the same history, whereas the relational Hamiltonian generates evolution relative to the chosen dynamical frame The extended canonical formulation shows that the primary constraint associated with the lapse momentum and the secondary dressed Hamiltonian constraint form a second-class pair The resulting relational dynamics is therefore governed by a non-vanishing Hamiltonian generator, which is not necessarily conserved in time The distinction between gauge evolution and relational evolution is also reflected in the constraint structure The final point is that the correspondence between the canonical complete observable and the covariantly dressed momentum holds only after imposing the Hamiltonian constraint The resulting generator is a bulk quantity arising from the relational Hamiltonian, rather than a charge associated with a spatial boundary or corner The gains come from
Title and authors: Kai: So we're looking at this paper, "Relational Cosmology: Covariant and Canonical Dressings, and the Hamiltonian of Internal Time," it's comparing different ways to handle time in a flat FLRW universe with a massless scalar field.
Mira: Exactly. It’s diving into how covariant dressing, canonical constraint reduction, and relational Hamiltonian evolution stack up against each other in this simple setting.
Lev: I'm curious if they are just doing some abstract math or if they are actually connecting this to what we might see on real hardware.
Kai: Well, the paper sets up a comparison between these methods and shows how the choice of frame affects not just how you parameterize your fields, but also the whole variational principle and the constraint structure of the resulting theory.
Mira: It seems like they're really digging into how fixing a specific clock can change everything—how it shifts things from being spacetime symmetries to something else entirely in terms of vector fields on relational field space.
Lev: That sounds complicated when you think about running simulations or building something, because if the frame itself isn't a symmetry, you lose some nice structure we rely on.
Kai: That's where it gets interesting. They look at a lapse condition like N equals one, imposed right in the action, and they find that this removes the independent lapse equation from the variational principle.
Mira: So they are saying that a simple choice for how we fix time early on simplifies things by getting rid of an extra equation we'd otherwise have to deal with.
Lev: That’s practical. In real error correction, reducing the number of constraints is always a win because you reduce the overhead you have to manage.
Kai: And then they show that other dynamical frames, like a frame defined by something called epsilon or even pi over alpha, let the remaining equations recover that original constraint equation through something called an off-shell Noether identity.
Mira: So it's not just about picking a clock; it’s about finding the right mathematical machinery—like those Noether identities—that lets you recover what you started with when things get complicated.
Title and authors: Lev: I wonder if this recovery is always clean, or if that off-shell identity introduces some new problems we didn't anticipate on the constraint surface.
Kai: The paper points out that for a relational frame to be useful, it has to satisfy two conditions: it needs to define a non-degenerate coordinate along the gauge fibers on the solution space of interest, and critically, the dressed variational principle must retain enough independent equations to recover the original constrained dynamics.
Mira: That second part is key. It means having enough degrees of freedom left so you don't lose the physical reality of what you started with when you dress it up.
Lev: So if we're trying to build a system, we need to make sure that whatever dressing or constraint reduction we apply doesn't accidentally eliminate the very dynamics we care about.
Kai: The paper also makes a distinction between covariant dressing and canonical reduction—the covariant construction keeps the dressed lapse and its associated second-class constraints, while canonical reduction eliminates those structures through gauge-fixing and constraint reduction.
Mira: That highlights a big difference: one method dresses the structure, the other just cuts things away with gauge fixing.
Lev: If you're aiming for real hardware implementation, which approach seems more robust when you have to deal with approximations?
Kai: The paper suggests that the agreement between the relational generators on the constraint surface shouldn't be interpreted as an off-shell equality; they are different things.
Mira: That means we can’t just assume that because two calculations match on a certain surface, they are identical everywhere, which is a lot of caution for theorists to carry with them.
Lev: So the correspondence is branch-dependent, right? That implies you have to be extremely careful where things get singular or where the rescaling of constraints becomes tricky.
Kai: It really does. The classical correspondence they establish here doesn't prove quantum equivalence on its own; that's a separate issue entirely.
Mira: So what this means for the broader picture is that we have these compatible classical descriptions, but they aren't necessarily the same intermediate phase spaces or variational principles.
Lev: That’s a lot of bookkeeping for anyone trying to map this onto a concrete computational framework where you need precise dynamics.
Title and authors: Kai: The paper concludes that the distinction between gauge evolution and relational evolution shows up in the constraint structure itself, where time reparametrizations identify different parametrizations of the same history.
Mira: And the relational Hamiltonian then generates evolution relative to that chosen dynamical frame, which isn't necessarily something conserved in time.
Lev: So we have two different kinds of evolution happening at once: one dictated by gauge choices, and one dictated by the internal clock you picked for yourself.
Kai: The final point is that the correspondence between the canonical complete observable and that covariantly dressed momentum only holds after imposing the Hamiltonian constraint.
Mira: That means you can't just look at those two things in isolation; you need to be working within the full context of those constraints to see their real relationship.
Lev: It sounds like we need a very specific setup—the full solution space—to see the dynamics through this relational lens.
Kai: So, as we wrap up on "Relational Cosmology: Covariant and Canonical Dressings, and the Hamiltonian of Internal Time," this paper really shows how different ways of defining time in cosmology lead to structurally different dynamic descriptions.
Mira: It highlights that when you move beyond simple gauge choices, the choice of frame dictates the entire structure of what we consider to be physical evolution.
Lev: For us, it means that any model we build needs to explicitly define which clock is serving as its internal reference point so we know which dynamics are actually being generated.
Kai: That's right. The paper gives us the framework to check those things systematically using those two criteria for frame admissibility.
Mira: It’s a lot of foundational work here, and it sets up a clearer path for how we might bridge the gap between different theoretical descriptions of time in gravity.
Lev: We'll keep an eye on this because if this helps constrain the structure of observables in these kinds of systems, it could guide our error correction designs.
Kai: Exactly. We’re ready to see what those next steps look like once we absorb all these details about the Hamiltonian generator and the constraint structure.
The paper's summary: Kai: So, we're looking at how this paper compares these three ways of handling time: covariant dressing, canonical constraint reduction, and relational Hamiltonian evolution in a flat universe with a massless scalar field.
Mira: It essentially shows that the way you choose to dress the theory or reduce the constraints changes not just how your math looks, but how your whole variational principle and your constraint structure actually behave.
Lev: That's interesting because it suggests that just picking a simple lapse condition, like setting N equal one in the action right away, can remove an extra equation you’d otherwise have to deal with.
Kai: Exactly. And they show that if you pick a better dynamical frame, like a frame defined by something called epsilon or pi over alpha, you can still recover that original constraint equation using something called an off-shell Noether identity.
Mira: But the crucial part is the admissibility check for any relational frame. It needs to meet two criteria: first, it has to define a non-degenerate coordinate along those gauge fibers on the solution space we care about, and second, the dressed variational principle must still have enough independent equations left over to recover your original constrained dynamics.
Lev: That second part is what really makes it hard. It means you can't just dress it up and hope everything works; you need to make sure that the dressing process didn't accidentally eliminate some of the actual physics you wanted to study.
Kai: And they draw a line between covariant dressing, which keeps those dressed lapse equations and second-class constraints, versus canonical reduction, which cuts things away using gauge fixing.
Mira: It suggests that the agreement between these relational generators on the constraint surface isn't just some kind of off-shell equality; they are actually different things.
Lev: So that means when you calculate the resulting relational Hamiltonian, it’s not just some charge associated with a boundary or a corner; it comes from a whole bulk quantity from the Legendre transform of the relational action.
Kai: It's about understanding that distinction between gauge evolution and relational evolution, where time reparametrizations identify different ways to describe the same history.
Mira: And that relational Hamiltonian then generates evolution relative to your chosen dynamical frame, which isn't necessarily something you can just say is conserved in time.
Lev: If we think about running this on real hardware, that means the primary constraint associated with the lapse momentum and a secondary dressed Hamiltonian constraint form a second-class pair.
Kai: So the final point they make is that you can only see the correspondence between your canonical complete observable and that covariantly dressed momentum after you impose that main Hamiltonian constraint.
Mira: It really boils down to this: different ways of defining time in cosmology lead to structurally different dynamic descriptions, which is a lot for us trying to build any kind of consistent model.
Lev: We need those two criteria for frame admissibility if we want to trust these relational descriptions when we start thinking about error correction overhead.
Kai: Next up, we’re going to look at how the stability of perturbations around black holes behaves when you add some spatially extended potential deformations, which is a totally different kind of complexity.
The paper's improvements: Kai: So we're looking at what this paper suggests to improve on what they already did with relational cosmology by suggesting some new ways to dress the theory and evolve time.
Mira: They focus on how you can define a better clock or frame, moving past just picking a simple background structure and instead focusing on finding frames that actually work mathematically.
Lev: I'm interested in the idea that we might be able to define an integrable generator of evolution even when the transformation isn't strictly a symmetry of the dressed action anymore.
Kai: That’s right. The canonical analysis showed that this generator can be identified with the complete observable associated with the momentum conjugate to your chosen frame on a specific solution branch.
Mira: But they have to make sure that correspondence holds only after imposing that main Hamiltonian constraint; it's not an automatic match between the two observables.
Lev: From an engineering standpoint, this means we can use a specific dynamical frame as our internal reference point, even if it’s not a symmetry of the full dressed action.
Kai: And they show that the resulting generator is actually a bulk quantity arising from the Legendre transform of that relational action, rather than some kind of charge tied to a spatial boundary or corner.
Mira: That clarifies the difference between gauge evolution and relational evolution; one is about time reparametrizations identifying different histories, and the other is about your chosen clock generating change.
Lev: So the improvement here is that you get a more robust way to define what "evolution" means in this context without relying on perfect symmetries.
Kai: The paper also refines the structure of constraints, showing that the primary constraint linked to lapse momentum and the secondary dressed Hamiltonian constraint form a second-class pair.
Mira: That means you have a non-vanishing Hamiltonian generator governing dynamics, which isn't necessarily conserved over time, which is a big point for dynamical systems.
Lev: For us building error correction hardware, knowing that the generator isn't always conserved gives us something to work with—it tells us what the expected evolution looks like relative to our internal clock.
Kai: So the improvement is less about a single new equation and more about clarifying how different theoretical tools—dressing, reduction, and canonical analysis—mesh together on the constraint surface.
Mira: It sets up a clearer picture of where you need to be careful when comparing these different descriptions of time in gravity.
Lev: And that clarity is important because it tells us which assumptions we actually need to keep when trying to map this onto something that runs on real hardware with finite fidelity.
Conclusion: Kai: So we’re wrapping up on "Relational Cosmology: Covariant and Canonical Dressings, and the Hamiltonian of Internal Time" by Kai and Mira.
Mira: Basically, this paper shows that choosing how you dress your theory or reduce your constraints changes the entire structure of what you call physical evolution in cosmology.
Lev: It’s about understanding how those different mathematical tools actually mesh together when you’re trying to build a consistent model for a universe with a scalar field.
Kai: They prove that while covariant dressing and canonical reduction look different, they can both lead to compatible classical descriptions on the right constraint surface.
Mira: But they aren't identical in terms of their intermediate variational principles or phase spaces, so you have to be careful about what you assume is true between the two methods.
Lev: For me, that’s important because it tells us exactly where the overhead lands when we try to translate these concepts into a real error correction scheme for hardware.
Kai: It highlights that defining a good dynamical frame is more than just picking an arbitrary clock; it requires satisfying those two admissibility criteria we talked about earlier.
Mira: It means any time you design a model, you have to check if your chosen frame defines a non-degenerate coordinate and if your variational principle still has enough degrees of freedom left over.
Lev: If the frame fails that second test, then the relational Hamiltonian you get might not even be something we can reliably calculate in practice.
Kai: So for those of you listening who only care about the physical setup, this paper is really about making sure our theoretical descriptions of time aren't just mathematically convenient but physically sound.
Mira: It sets a solid foundation for understanding how to compare different ways to handle time evolution in gravity models.
Lev: We still have to figure out how this maps onto those constraints we need for real hardware, which is the next big hurdle.
More episodes
- Daily Summary for 2026-11-19
- 2607.21925-From the Early Slope to Curvature: Future Prospects and challenges for Astrophysical Parameter Estimation Using the Core-Collapse Supernova High-Frequency Feature
- 2604.21755-Quantum-information diagnostics of cosmological perturbations with nontrivial sound speed in inflation
- 2605.14572-Modifications of CMB Temperature and Polarization Quadrupole Signals in Thurston Spacetimes
- 2604.06090-Posterior Predictive Checks for Gravitational-wave Populations: Limitations and Improvements
- 2606.20957-Out-of-Equilibrium Effects in Non-Radial Relativistic Stellar Perturbations: A Model-Agnostic Formulation and Mode Analysis
- 2407.02076-Gravitational Bremsstrahlung Waveform at the fourth Post-Minkowskian order and the second Post-Newtonian level
- 2512.08604-Quantum particle production and radiative properties of a new bumblebee black hole
- 2609.05666-Fermionic Casimir memory from an anisotropic gravitational pulse
- 2606.01236-Improving the resolution of double white dwarf systems with spaceborne gravitational wave observatories using a robust astrophysical prior