Consistency of PN, PM, Amplitudes, and Self-Force

Post-Newtonian (PN), post-Minkowskian (PM), scattering-amplitude, gravitational self-force (GSF), and numerical-relativity calculations describe overlapping parts of the same two-body problem, but agreement is meaningful only after their observables and expansions are put in the same frame and order. At the evidence cutoff of 2026-08-20, many demanding overlap tests succeed, including 5PM conservative scattering checks and total-flux agreement between 4.5PN and second-order self-force. The defensible conclusion is substantial, cell-by-cell consistency—not an all-orders equivalence theorem, and not a resolution of every mode, tail, spin, or strong-scattering discrepancy.

Helpful background. Map of Approximation Overlaps defines the regimes and expansion parameters. Here the task is evidentiary: identify which comparisons have actually been made, which quantities agree, and where “same physics” still hides a frame or subtraction problem.

When two approximation schemes are comparable

A comparison needs a common observable before it needs matching algebra. For bound motion that observable might be redshift, binding energy, periastron advance, or a radiative flux. For unbound motion it might be scattering angle, impulse, or radiated energy. Coordinate accelerations, intermediate potentials, and individual waveform modes can differ under gauge or Bondi-frame transformations even when invariant physics agrees.

Five axes delimit the proposition: bound versus scattering motion; conservative versus radiative dynamics; PN, PM, and self-force order; asymptotic frame and regularization prescription; and retained spin, finite-size, absorption, tail, and radiation-reaction effects. Analytic continuation between bound and unbound motion is an additional operation, not an automatic identity. A successful $O(G^5\nu)$ conservative comparison cannot be promoted to generic spinning radiation.

Assessment at 2026-08-20	Meaning
Question status	Partially resolved
Evidence status	Supported
Confidence	Moderate because the tested cells are strong but the global proposition is nonuniform
Domain	Like-for-like invariant comparisons at declared PN, PM, and self-force orders; mainly nonspinning circular or weak-deflection sectors
What could change the conclusion	A new high-order comparison, resolution of the radiative-mode scaling and frame question, or an independent strong-scattering benchmark

Entry project: compare one flux coefficient in common variables

This is a good entry for a student comfortable with series expansions, symbolic algebra, and numerical residuals who wants analytic work without first building a field solver. The manageable object is one coefficient of the circular, nonspinning energy flux. The full 5PM amplitude calculation or a new second-order self-force solution is a much larger commitment.

Read Warburton et al., arXiv:2407.00366v2, § I, Eqs. (3)–(10), and §§ II–IV. Extract the frequency and mass variables, the normalization of the flux, and the mass-ratio order being compared. Independently expand the published PN formula to the chosen self-force order, keeping logarithms symbolic. Check the quadrupole normalization and test-mass limit before comparing higher coefficients. Appendix B is the second reading: identify one example in which a coefficient crossing or limited weak-field reach defeats a simple residual-slope argument.

Deliver a short derivation and executable coefficient table that reports which powers of frequency and mass ratio are actually common to the two calculations. If the necessary numerical self-force data are available, add a residual plot with precision and fitting-window variation; otherwise label the deliverable a symbolic consistency check, with no claim to reproduce a numerical figure. A laptop suffices for this bounded algebraic task.

A candidate extension, with novelty unchecked, is a stable residual estimator near a coefficient zero crossing, tested first on synthetic data with a known remainder. A useful result would show when a slope estimate becomes non-diagnostic, rather than manufacture agreement. Stop if the two quantities use different waveform-frequency definitions, flux destinations, or asymptotic frames that you cannot translate. That is the point to seek PN/self-force supervision or narrow the comparison to the total flux. Matching coefficients copied from the same source is not an independent derivation of either method.

What the successful overlap tests establish

The accumulated lower-order evidence is extensive. Barack and Pound catalogue PN–GSF comparisons of invariant conservative and dissipative quantities and explain why the small-mass-ratio expansion can determine high-PN coefficients Barack and Pound 2019, § 8. That review predates the newest orders, but it supplies the methodological baseline.

At 5PM and first order in the symmetric mass ratio, Driesse and collaborators derive a conservative scattering result using modern worldline and amplitude methods. Its slow-velocity expansion matches independent 5PN information, and its tail-related check matches a separately known 4PM energy-loss structure Driesse et al. 2024, main result and supplemental 5PN/4PM checks. These are nontrivial tests because distinct expansions and subtraction structures meet on the same coefficients.

Radiation now supplies an equally instructive mixed result. Warburton and collaborators find agreement between the total quasicircular flux at 4.5PN and a second-order self-force calculation. Many individual spherical-harmonic modes also show the expected residual scaling, while zero crossings and self-force data extending only to about $x=0.02$ prevent a clean asymptotic test for every mode; the relative Bondi–Metzner–Sachs frame was not independently controlled Warburton et al. 2026, conclusion and Appendix B. A second 2026 study compares self-force hyperbolic radiation with PM, PN, and available numerical-relativity results over a weak-deflection domain extending to reported $v_\infty/c\simeq0.7$ Warburton 2026, §§ II–V and comparison figures. The common lesson is that precise agreement is real when the comparison cell is explicitly defined.

Results and open questions at 2026-08-20

Claim or evidence	What it supports	What it does not establish
Catalogue of PN–GSF invariants	Broad lower-order consistency and a tested comparison method	Current highest PM/2SF orders
5PM first-self-force scattering result	Conservative nonspinning scattering consistency with PN and tail information	Generic radiation, spin, or finite-size physics
4.5PN–2SF total circular flux and mode residuals	A first-principles radiative overlap, with expected scaling in many modes	Clear asymptotic agreement for every mode across the available $x$ range, or a controlled frame map
Hyperbolic radiated-energy comparison	SF–PM–PN–NR consistency at large impact parameter	Near-plunge or generic-spin scattering
Scalar-charge 4PM benchmark	High-precision numerical/analytic overlap methodology	Parameter-free validation of gravitational tidal sectors
5PM nonlocal tail match	Consistent nonlocal bookkeeping through reported 6PN overlaps	The complete local-in-time 5PM dynamics

Evidence is weighted by independence of observable and method, not by paper count. A coefficient recovered from a distinct expansion is stronger than two rearrangements of one calculation. Conversely, fitting an unknown coefficient to the benchmark being “predicted” weakens independence.

Strongest competing interpretations

One interpretation treats the existing matches as evidence that all mature approaches are converging on a single high-order two-body dynamics; remaining disagreements are bookkeeping, gauge, or resummation issues. The successful 5PM/5PN, total-flux, and hyperbolic comparisons strongly support this view within their domains.

The competing interpretation stresses nonuniformity. PN expansion can deteriorate at high velocity, finite-order PM can deteriorate near the scattering-to-plunge threshold, and a small-mass-ratio series can omit comparable-mass structure. Tail terms require careful separation of local and nonlocal pieces. Waveform modes additionally depend on an asymptotic frame. The amplitude dictionary itself involves classical limits, iteration subtractions, and canonical transformations; the SAGEX synthesis makes these choices explicit Bjerrum-Bohr et al. 2022, §§ 2–5. Thus agreement in several observables is evidence of consistency, but it cannot prove uniform convergence across the full parameter space.

Counterexamples, null results, and failed routes

The unresolved mode-by-mode asymptotic test in the 4.5PN–2SF comparison is the most important limitation. Many modes show the expected residual scaling, so the current paper does not support a claim of mode disagreement. For other modes, zero crossings in the post-Newtonian coefficients and the finite reach of the self-force data obscure the asymptotic power law. Because the relative Bondi frame was not controlled, “all modes agree in a common frame” is also stronger than the calculation establishes.

Barack and collaborators provide a useful caution in a scalar-charge model. Numerical self-force and amplitude calculations agree through 4PM, but two tidal coefficients are fixed using the self-force result Barack et al. 2023, §§ IV–VII and comparison tables. The study validates a comparison strategy and much of the analytic structure; it is not a wholly parameter-free gravitational prediction. Likewise, no cutoff search result supplied one benchmark combining generic spin, absorption, radiation reaction, and strong deflection at the highest orders.

Why frames, tails, and orders cannot be hidden

An apparent mismatch can enter before any physics is compared. A scattering angle may use different impact parameters or energy variables; a radiative multipole may live in a different Bondi frame; an amplitude potential may still contain iterations that a Hamiltonian calculation subtracts; and a tail contribution may be divided differently between local and nonlocal dynamics. The remedy is to construct the same invariant from both sides and expand only after conventions are aligned.

Dlapa and collaborators’ 5PM tail analysis reports agreement with known overlaps through 6PN, strengthening confidence in nonlocal bookkeeping Dlapa et al. 2026, abstract and overlap checks. The version of record appeared as Physical Review D 114, 024029 on 10 July 2026. The calculation remains deliberately nonlocal in time and still must be combined with the local sector, so publication closes the bibliographic question without enlarging the result’s domain.

What would change the assessment

The highest-value calculation is a common-Bondi-frame extension of the 4.5PN–2SF comparison for the modes whose asymptotic scaling is unresolved, using data farther into the weak-field regime and handling coefficient zero crossings explicitly. It should include radiative multipoles and memory pieces without fitting either side to the other. Persistent disagreement in a frame-invariant reconstructed waveform or flux would lower support; clean expected scaling would close the sharpest current radiative seam.

A second discriminant is a fresh, convergence-controlled numerical-relativity simulation of a strong but nonplunging hyperbolic encounter, compared with independently implemented PM and self-force predictions using identical asymptotic variables. Generic-spin and absorption comparisons at high order would broaden the domain, but should be assessed as new cells rather than folded into an average score.

Research checklist

• Name one invariant observable and state whether the motion is bound, scattering, or analytically continued.
• Declare PN, PM, and self-force orders separately, including powers of the mass ratio and all logarithmic or hereditary terms.
• Harmonize masses, energy, angular momentum, impact parameter, radiative time, and Bondi/BMS frame before comparing coefficients.
• State spin, tidal, absorption, radiation-reaction, and memory content; never infer an omitted sector from agreement elsewhere.
• Separate local, nonlocal, iteration, and canonical-transformation terms and show that their recombination yields the invariant.
• Publish exact symbolic coefficients and high-precision residuals; test slow-velocity, weak-field, and small-mass-ratio limits.
• Use new numerical data, not calibration data, for an independent challenge.
• Distinguish a formula comparison from a rerun. No end-to-end reproduction was performed for this assessment; this checklist is the reproduction prescription.

Source and data map

Barack and Pound provide the established PN–GSF synthesis; Bjerrum-Bohr et al. explain the amplitude/PM dictionary. Driesse et al. and Dlapa et al. provide current 5PM conservative and tail results. The two Warburton studies supply the newest PN–2SF and hyperbolic radiative comparisons. Barack et al. give a high-precision toy-model cross-check and its independence limitation.

The papers expose formulas, figures, and some data statements, but the cutoff search did not verify one immutable software environment capable of reproducing all comparisons. A valid rerun must therefore archive symbolic notebooks or code, package versions, precision settings, input data, and the asymptotic-frame transformation. The absence of that common environment is a reproduction boundary, not evidence against the analytic matches.

References

• Barack, L., et al. “Comparison of post-Minkowskian and self-force expansions: Scattering in a scalar charge toy model.” Physical Review D 108, 024025 (2023). DOI.
• Barack, L. and Pound, A. “Self-force and radiation reaction in general relativity.” Reports on Progress in Physics 82, 016904 (2019). DOI.
• Bjerrum-Bohr, N. E. J., et al. “Scattering amplitudes and the conservative Hamiltonian for binary systems.” In SAGEX Review on Scattering Amplitudes, chapter 13 (2022). Open preprint.
• Dlapa, C., et al. “Nonlocal-in-time tail effects in gravitational scattering to fifth Post-Minkowskian and tenth self-force orders.” Physical Review D 114, 024029 (2026). DOI.
• Driesse, M., et al. “Conservative Black Hole Scattering at Fifth Post-Minkowskian and First Self-Force Order.” Physical Review Letters 132, 241402 (2024). DOI.
• Warburton, N. “Gravitational radiation from hyperbolic orbits: comparison between self-force, post-Minkowskian, post-Newtonian, and numerical relativity results.” Physical Review D 113, 084059 (2026). DOI.
• Warburton, N., et al. “Comparison of 4.5PN and 2SF gravitational energy fluxes from quasicircular compact binaries.” Physical Review D 113, 084050 (2026). DOI; open preprint v2.