Pith. sign in

REVIEW 3 major objections 6 minor 52 references

PSR J0435+3233 is a hierarchical triple: its extreme apparent spin-down is an acceleration effect caused by a third star, not an intrinsic property of the pulsar.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · deepseek-v4-flash

2026-08-01 07:33 UTC pith:45TXPYTI

load-bearing objection A credible, well-argued case that PSR J0435+3233's anomalous spin-down is an acceleration effect in a hierarchical triple, but the unresolved disagreement between the frequency-derivative orbit and the timing orbit keeps it short of a clean accept. the 3 major comments →

arxiv 2607.27462 v1 pith:45TXPYTI submitted 2026-07-29 astro-ph.HE gr-qc

On the triple nature of the PSR J0435+3233 system

Paulo C. C. Freire , Colin J. Clark , Cees G. Bassa , Guillaume Voisin , Rutger van Haasteren , Lars Nieder , Benjamin W. Stappers This is my paper
classification astro-ph.HE gr-qc
keywords pulsars: individual (PSR J0435+3233)millisecond pulsarshierarchical triple systemspulsar timinggamma-ray pulsationsspin-downstrong equivalence principlewhite dwarf binaries
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper argues that PSR J0435+3233, a millisecond pulsar with a measured spin-down rate hundreds of times larger than any other known Galactic millisecond pulsar, is not an anomalously young or powerful object but the inner member of a hierarchical triple system. The apparent spin-down and the nearly identical variations of the pulsar's spin and orbital frequencies over five years are, the authors claim, the Doppler effect of the inner binary accelerating in the gravitational field of a third star on a wide, eccentric ~70-year orbit. Fitting a triple-system timing model to radio pulse arrival times and to gamma-ray photon arrival times (the latter extending the baseline to 18 years) yields a phase-coherent solution that predicts pulsations back to 2008, and identifies a ~1.2 solar-mass F-type main-sequence star as the optical counterpart. If correct, the intrinsic spin-down is at least two orders of magnitude lower than reported, in line with other millisecond pulsars, and the system becomes a promising laboratory for testing the strong equivalence principle.

Core claim

PSR J0435+3233 is a hierarchical triple: a 3.2-millisecond pulsar with a white-dwarf companion in an 8-day orbit, itself orbited by a ~1.2 solar-mass main-sequence star on a ~70-year eccentric orbit. The five measured spin-frequency derivatives and three orbital-frequency derivatives trace, almost identically, the line-of-sight acceleration of the inner binary about the common centre of mass; the reported spin-down rate therefore is not intrinsic. A timing model with two Keplerian orbits reproduces the radio residuals to 1.49 microseconds and, when weighted by gamma-ray pulsation significance, predicts pulsations across the entire gamma-ray data span (2008-2026). The optical counterpart, 11

What carries the argument

The central relation is Eq. (1): if the intrinsic spin-down and inner-orbital evolution are negligible, then the fractional spin-frequency derivative equals the fractional orbital-frequency derivative, and both equal minus the line-of-sight acceleration of the inner binary in the outer star's gravitational field divided by the speed of light. Because all five spin and all three orbital frequency derivatives are negative and evolve together, they are interpreted as Doppler shifts of a single accelerating frame, not as timing noise. This allows the five Keplerian parameters of the outer orbit (projected semi-major axis, period, eccentricity, argument of periastron, true anomaly) to be estimate

Load-bearing premise

The interpretation rests on Eq. (1): the measured spin and orbital frequency derivatives are dominated by the line-of-sight acceleration from the outer companion, with negligible intrinsic spin-down and intrinsic orbital-period evolution, so that if the pulsar exhibits strong timing noise mimicking this correlated acceleration the derived outer orbit could be spurious.

What would settle it

Measure the radial velocity of the optical counterpart, a star 11 milliarcseconds from the pulsar, over the next decade: the model predicts a smooth ~4 km/s decline until 2036 followed by an ~8 km/s rise to 2050; a different pattern (e.g., constant radial velocity) would falsify the triple interpretation. Alternatively, continued pulsar timing before the predicted 2036 periastron passage should show the steep, non-linear change in orbital range; failure to see that would rule out the outer-orbit solution.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • The intrinsic spin-down of PSR J0435+3233 is at least two orders of magnitude smaller than the observed value, making it a normal millisecond pulsar and removing the need for a new formation channel.
  • The gamma-ray efficiency is no longer anomalously low; the modest gamma-ray flux is consistent with typical millisecond pulsars.
  • The non-detection of continuous gravitational waves from the pulsar no longer constrains the fraction of spin-down power radiated as gravitational waves.
  • The large projected semi-major axis of the inner orbit (8 seconds versus 1.2 seconds for PSR J0337+1715) could allow tighter tests of the strong equivalence principle, potentially surpassing current limits.
  • The model makes a testable prediction: the outer star's radial velocity should decline by ~4 km/s until the 2036 periastron passage, then rise by ~8 km/s to 2050.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the triple interpretation holds, the same orbital motion should appear as non-linear terms in the pulsar's apparent position and proper motion over the coming decades; multi-epoch astrometry of both the pulsar and the optical companion should reveal a common orbital signature independent of timing.
  • The current two-Keplerian model absorbs orbital perturbations into parameters like the inner orbit's period derivative and semi-major-axis derivatives; a full three-body integration will be needed to separate tidal and relativistic effects from purely kinematic ones, especially near the 2036 periastron.
  • High-resolution spectroscopy of the optical counterpart could measure the predicted radial-velocity trend, giving a theory-independent mass ratio and, combined with the pulsar-timing inclination, a direct dynamical distance and the system's three-dimensional velocity.
  • If the apparent acceleration were instead timing noise mimicking the signature, the model would likely break down once the outer orbit is traversed beyond the current 25% coverage; continued timing through the 2036 periastron passage is the decisive test.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. The paper argues that PSR J0435+3233 is a hierarchical triple system: a 3.2 ms pulsar with a white-dwarf companion in an 8-day orbit, itself orbited by a ~1.2 M_sun main-sequence star on a wide (~70 yr), eccentric (e~0.6) orbit. The anomalously large observed spin-down rate is attributed to line-of-sight acceleration of the inner binary by the outer companion, not to intrinsic pulsar spin-down. The authors first fit the five spin and three orbital frequency derivatives published by Wu et al. (2026) with an outer Keplerian orbit, then fit a hierarchical triple timing model to the FAST ToAs, obtaining a reduced chi-square of ~1.0. Using Fermi-LAT data, they report gamma-ray pulsations back to 2008 and a refined solution. They identify a Gaia/2MASS star 11 mas away as the optical counterpart, derive the outer orbit inclination and distance, and discuss future tests of the strong equivalence principle.

Significance. If correct, the paper resolves a major anomaly in the MSP population and establishes the second known hierarchical triple with a pulsar in the Galactic disk, with the potential for competitive strong-equivalence-principle tests. The radio-only timing solution already improves the residual rms over Wu et al. (2026) and detects gamma-ray pulsations back to 2014, which is independent, predictive evidence. The optical counterpart has a tiny chance-alignment probability, and the paper provides falsifiable radial-velocity predictions. These strengths justify serious consideration. However, the central acceleration assumption is not internally verified because the outer orbit derived from frequency derivatives disagrees with the timing fit, and the full-baseline gamma-ray detection is obtained after a posterior selection, so the statistical evidence is weaker than the abstract implies.

major comments (3)
  1. [§3.1 vs §3.2, Table 1] The outer orbit fitted to the W26 Taylor coefficients (x_O≈1900 s, P_B,O≈58 yr, e≈0.55) is not consistent with the radio timing fit (Table 1: x_O=2460(30) s, P_B,O=26040(320) d≈71.3 yr, e=0.600(3)). Since Eq. (1) and the §3.1 fit are the only quantitative bridge from the published derivatives to the triple hypothesis, this discrepancy is load-bearing. The statement in §5 that 'the cause of this is unclear' is insufficient. The authors should quantify the covariance between the Taylor coefficients and the timing-model parameters, fit the W26 derivatives with the Table 1 model, and show explicitly whether the empirical ˙P_B, ˙x, ¨x and intrinsic f_dot can account for the difference. Otherwise the acceleration assumption is not internally verified.
  2. [§3.3, Table 1 right] The full-baseline gamma-ray detection (H=171.9) is obtained by scanning all radio posterior samples and re-weighting by exp(0.398405H), so it is a selected maximum rather than a blind prediction. This introduces a trials factor that must be quantified, for example via a posterior predictive p-value or an out-of-sample split. The independent evidence is the radio-only model's detection from MJD 57500 (Table 1 left, H=112.5), and the paper should separate that predictive check from the parameter-refinement step. The abstract's wording 'detection of gamma-ray pulsations back to the beginning of the Fermi-LAT data' should be qualified to reflect the selection procedure.
  3. [§3.2, Table 1] The timing model requires empirical ˙P_B=−140(90)×10^-12, ˙x=362(5)×10^-15 and ¨x=2.00(8)×10^-21 with no physical model, and the joint radio/gamma solution gives f_dot=30(10)×10^-15 Hz/s, an apparent spin-up at roughly 3σ. This suggests that the 'two non-interacting Keplerian orbits plus polynomial terms' model is absorbing unmodeled effects. The authors should demonstrate that these terms are not mimicking the outer-orbit acceleration, e.g. by comparing fits with and without them and by fitting the Voisin et al. Eq. (B.23) perturbation model directly instead of only estimating its amplitude. Without this, the outer-orbit parameters and their uncertainties may be biased.
minor comments (6)
  1. [Eq. (1)] Please define f_B explicitly and state that the equality holds only if intrinsic spin and orbital derivatives are negligible; the text says this, but the equation as written could be misread as exact.
  2. [Fig. 1] The axis labels contain placeholder squares ('10□6', '10□14 s□1') and the units are garbled. Please fix the LaTeX/rendering issue and state unambiguously that the plotted quantities are fractional variations relative to the reference epoch.
  3. [Table 1] The spin-frequency derivative f_dot is positive in both columns, while the derived 'P_dot' is listed as negative. Since pulsar timing papers usually define P_dot as positive for spin-down, the sign convention should be stated explicitly to avoid confusion.
  4. [§3.2] The text 'not yet been merged into the main repository' is missing a verb ('has not yet been merged'). Also, for reproducibility, please provide the exact commit or version of the PINT triple-model branch used in the analysis.
  5. [§5, Fig. 7] The term 'range' is non-standard for the projected Rømer delay of the outer orbit. Please rename it (e.g. 'projected Rømer delay') and make the sign convention in Fig. 7 explicit.
  6. [§4] The DM-based distance estimates (1.2–1.5 kpc) and the Gaia photo-astrometric distances (2.0–2.2 kpc) overlap only marginally. The text says they are 'consistent within uncertainties', but no uncertainties are given for the DM-model distances. Please provide the full uncertainty ranges and assess the degree of consistency quantitatively.

Circularity Check

0 steps flagged

No significant circularity: the triple model is tested against independent radio timing, gamma-ray, and optical astrometric data.

full rationale

The paper's derivation chain is not circular. The central assumption, Eq. (1), that the observed spin and orbital frequency derivatives are dominated by line-of-sight acceleration from a third body, is an explicitly stated physical hypothesis, not a restatement of the conclusion. It is testable: the near-identical evolution of f and f_B is an observed property of the W26 fit, and the subsequent radio timing solution is a phase-coherent fit to the raw ToAs, not a re-output of the derivatives. The fact that the derivative-based outer orbit (P_B,O ≈ 58 yr) differs from the timing orbit (P_B,O ≈ 72 yr, acknowledged in Section 5) demonstrates that the timing analysis is not merely recovering its own input. The gamma-ray evidence is partly co-fitted: the 2008 phase-connected detection is obtained after re-weighting radio posterior samples by the H-statistic, so that final solution is a joint radio/gamma fit rather than a pure prediction. However, the radio-only solution already detects gamma pulsations back to 2014, providing independent external confirmation before any gamma re-weighting; the gamma photons themselves are independent data, and the re-weighting is a transparent parameter-refinement step, not a reduction of the prediction to its input. The informative prior on the intrinsic spin-down derivative is based on the known Galactic MSP population, not derived from the triple model, and the paper openly reports that a uniform-prior fit produced an unphysical large positive derivative. The optical counterpart analysis uses Gaia astrometry and photometry that were not used in the timing fit; the small chance-alignment probability and the consistency of the projected orbit with the star's position and proper motion constitute an independent astrometric check. Self-citations (e.g., Voisin et al. 2025) are used for background estimates and future prospects, not as load-bearing justification for the triple claim. No quoted step reduces to its own inputs by construction.

Axiom & Free-Parameter Ledger

5 free parameters · 5 axioms · 0 invented entities

The paper introduces no new physical entities. The free parameters listed are fitted timing parameters or assumed mass values that the central claims (triple nature, low intrinsic spin-down, optical counterpart association) depend on. The most significant ad hoc elements are the empirical inner-orbit derivatives and the gamma-ray re-weighting scheme.

free parameters (5)
  • Intrinsic spin frequency derivative f_dot = 2.8e-14 Hz/s (joint radio/gamma, with 1-sigma uncertainty 1.0e-14)
    Fitted with a Gaussian prior centered at zero with width 5e-14 Hz/s; the claim that intrinsic P_dot is two orders of magnitude lower than observed depends on this fit and on the prior choice.
  • Inner orbital period derivative P_dot_B = 36e-12 (joint radio/gamma, uncertainty 24e-12)
    Empirical term added to the timing model to achieve an acceptable fit; not predicted by the triple model and attributed to unmodeled tidal/time-dilation effects.
  • Inner semi-major axis derivative x_dot = 365e-15 (joint radio/gamma, uncertainty 5e-15)
    Empirical term added to the fit; absorbs effects not modeled by the two-Keplerian approximation.
  • Inner semi-major axis second derivative x_ddot = 1.90e-21 (joint radio/gamma, uncertainty 0.09e-21)
    Additional empirical term needed for a good fit to the radio ToAs.
  • Mass ratio R_O = M_O/M_b = 0.6 (assumed)
    Used for sky-projection of the outer orbit (Fig. 6) and for the radial-velocity prediction; not yet directly measured.
axioms (5)
  • domain assumption Equation (1): f_dot/f ≃ f_dot_B/f_B ≃ -a_l/c, i.e., observed frequency derivatives are dominated by the line-of-sight acceleration from a third body
    Load-bearing assumption that intrinsic spin-down and intrinsic orbital variations are negligible compared to Doppler acceleration; used to derive the outer orbit from W26's frequency derivatives.
  • standard math The Joshi & Rasio (1997) model relating high-order spin frequency derivatives to Keplerian orbital parameters
    Standard astrophysical method, implemented in Bassa et al. (2016); used for the initial outer orbit estimate from frequency derivatives.
  • domain assumption Two non-interacting Keplerian orbits (BT1P-like) are sufficient for the timing model
    The paper states that orbital perturbations, tidal effects, and time dilation are not fully modeled, but the fit is good; this approximation is used throughout the timing analysis.
  • ad hoc to paper Gamma-ray weighting W ∝ exp(0.398405 H) is a valid proxy for the pulsation likelihood
    Used to re-weight posterior samples from the radio timing analysis; it is described as a proxy, not a rigorous joint likelihood, and its use without a trials correction affects the final parameter uncertainties.
  • ad hoc to paper Assumed masses M_p=1.4 M⊙, M_i=0.6 M⊙, M_b=2.0 M⊙, M_O=1.2 M⊙ for derived inclination and distance
    The outer orbit's inclination (i=149°) and distance (d=2.2 kpc) are derived using these fiducial masses; the optical companion's mass is constrained to 1.0–1.7 M⊙ from photometry, and 1.2 M⊙ is a representative value chosen to match the timing solution.

pith-pipeline@v1.3.0-daily-deepseek · 17143 in / 13988 out tokens · 142306 ms · 2026-08-01T07:33:09.739261+00:00 · methodology

0 comments
read the original abstract

Context. The recent pulsar timing ephemeris of PSR J0435+3233 indicates that this millisecond pulsar (MSP) has a spin-down rate that is much higher than observed in other MSPs and challenges our understanding of the formation and evolution of MSPs. Aims. We propose that this system is a hierarchical triple, and that the high spin-down rate is caused by varying acceleration due to a tertiary in a wide orbit. Methods. We use pulsar timing methods with radio and gamma-ray observations of PSR J0435+3233 to determine the system properties. Results. We find that a hierarchical triple timing model describes the timing observations of PSR J0435+3233 and that this results in the detection of gamma-ray pulsations back to the beginning of the Fermi Large Area Telescope (LAT) data in 2008. The intrinsic spin-down rate remains uncertain as it correlates with the parameters of the outer orbit, but large spin-down rates are excluded and the intrinsic rate is at least two orders of magnitude lower than the observed rate, in line with other Galactic MSPs. We identify a star located 11 mas from the pulsar position as the optical counterpart to the tertiary companion. From the 1.5-2.5 kpc distance and colours, we infer that the tertiary is a 1.2 solar mass F-type main-sequence star. Along with the pulsar binary, it orbits the common centre of mass with an eccentric (e ~ 0.6), wide (~ 70 yr) orbit that is likely seen at a low orbital inclination. Conclusions. We conclude that PSR J0435+3233 is a hierarchical triple system. We discuss the motivation and prospects for the continued study of this system. Spectral measurements of the outer star in addition to continued astrometric measurements will yield mass ratio and inclination estimates, while continued pulsar timing may yield a tighter constraint on violations of the Strong Equivalence Principle than are currently obtained from PSR J0337+1715.

Figures

Figures reproduced from arXiv: 2607.27462 by Benjamin W. Stappers, Cees G. Bassa, Colin J. Clark, Guillaume Voisin, Lars Nieder, Paulo C. C. Freire, Rutger van Haasteren.

Figure 1
Figure 1. Figure 1: Evolution of the spin and orbital frequencies (top) and [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Corner plot showing the Keplerian parameters for the outer orbit and the spin frequency derivative obtained from our timing [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Timing residuals obtained from the FAST barycentric [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Detection of γ-ray pulsations for PSR J0435+3233. The spin phases of the γ-ray photons, shown as points in the lower panel, were calculated using the timing solution in the second column of [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Distance estimates of PSR J0435+3233 based on the pul￾sar dispersion measure and Galactic electron density models, and of the outer companion 2MASS J04353375+3233080 based on Gaia parallax, Galactic stellar density models and/or broadband optical and near-infrared photometry. and would not be detectable with Gaia. The 11 mas separation to 2MASS J04353375+3233080 with G = 16.66 would also greatly hinder det… view at source ↗
Figure 7
Figure 7. Figure 7: Evolution of the range (radial light-travel-time from [PITH_FULL_IMAGE:figures/full_fig_p008_7.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

52 extracted references · 7 linked inside Pith

  1. [1]

    2020, ApJS, 247, 33

    Abdollahi, S., Acero, F., Ackermann, M., et al. 2020, ApJS, 247, 33

    2020

  2. [2]

    2022, ApJS, 260, 53

    Abdollahi, S., Acero, F., Baldini, L., et al. 2022, ApJS, 260, 53

    2022

  3. [3]

    2019, A&A, 628, A94

    Anders, F., Khalatyan, A., Chiappini, C., et al. 2019, A&A, 628, A94

    2019

  4. [4]

    Anders, F., Khalatyan, A., Queiroz, A. B. A., et al. 2022, A&A, 658, A91

    2022

  5. [5]

    M., Gusinskaia, N

    Archibald, A. M., Gusinskaia, N. V ., Hessels, J. W. T., et al. 2018, Nature, 559, 73 7 This is a dimensionless parameter that determines the strength of the coupling between a scalar field and the spacetime geometry in JFBD gravity (Brans & Dicke 1961). As its approaches infinity, the theory converges asymptotically to general relativity. 8 Wex, private ...

    2018

  6. [6]

    A., & Thorsett, S

    Arzoumanian, Z., Joshi, K., Rasio, F. A., & Thorsett, S. E. 1996, in Astronomi- cal Society of the Pacific Conference Series, V ol. 105, IAU Colloquium 160: Pulsars: Problems and Progress, ed. S. Johnston, M. A. Walker, & M. Bailes, 525–530

    1996

  7. [7]

    2013, in Proceedings of the 4th Fermi

    Atwood, W., Albert, A., Baldini, L., et al. 2013, in Proceedings of the 4th Fermi

    2013

  8. [8]

    Symposium, Monterey, California, 2012, ed. T. J. Brandt, N. Omodei, & C. Wilson-Hodge, eConf C121028, 8, arXiv:1303.3514

    Pith/arXiv arXiv 2012

  9. [9]

    B., Abdo, A

    Atwood, W. B., Abdo, A. A., Ackermann, M., et al. 2009, ApJ, 697, 1071

    2009

  10. [10]

    Bailer-Jones, C. A. L., Rybizki, J., Fouesneau, M., Demleitner, M., & Andrae, R. 2021, AJ, 161, 147

    2021

  11. [11]

    Bailer-Jones, C. A. L., Rybizki, J., Fouesneau, M., Mantelet, G., & Andrae, R. 2018, AJ, 156, 58

    2018

  12. [12]

    H., Lott, B., & The Fermi-LAT collaboration

    Ballet, J., Bruel, P., Burnett, T. H., Lott, B., & The Fermi-LAT collaboration. 2023, arXiv e-prints, arXiv:2307.12546

    Pith/arXiv arXiv 2023

  13. [13]

    G., Janssen, G

    Bassa, C. G., Janssen, G. H., Stappers, B. W., et al. 2016, MNRAS, 460, 2207

    2016

  14. [14]

    1995, PASP, 107, 1047

    Bergeron, P., Wesemael, F., & Beauchamp, A. 1995, PASP, 107, 1047

    1995

  15. [15]

    2003, Nature, 425, 374

    Bertotti, B., Iess, L., & Tortora, P. 2003, Nature, 425, 374

    2003

  16. [16]

    2025, A&A, 698, A239

    Blanchard, C., Guillemot, L., V oisin, G., Cognard, I., & Theureau, G. 2025, A&A, 698, A239

    2025

  17. [17]

    2018, JAX: composable transforma- tions of Python+NumPy programs

    Bradbury, J., Frostig, R., Hawkins, P., et al. 2018, JAX: composable transforma- tions of Python+NumPy programs

    2018

  18. [18]

    & Dicke, R

    Brans, C. & Dicke, R. H. 1961, Physical Review, 124, 925

    1961

  19. [19]

    H., Digel, S

    Bruel, P., Burnett, T. H., Digel, S. W., et al. 2018, arXiv e-prints, arXiv:1810.11394

    Pith/arXiv arXiv 2018

  20. [20]

    2024, arXiv e-prints, arXiv:2402.10797

    Cabezas, A., Corenflos, A., Lao, J., et al. 2024, arXiv e-prints, arXiv:2402.10797

    Pith/arXiv arXiv 2024

  21. [21]

    Cordes, J. M. & Lazio, T. J. W. 2002, arXiv e-prints, astro

    2002

  22. [22]

    & Schäfer, G

    Damour, T. & Schäfer, G. 1991, Phys. Rev. Lett., 66, 2549 de Jager, O. C., Raubenheimer, B. C., & Swanepoel, J. W. H. 1989, A&A, 221, 180

    1991

  23. [23]

    Dutta, A., Freire, P. C. C., Gautam, T., et al. 2025, A&A, 697, A166

    2025

  24. [24]

    W., Lang, D., & Goodman, J

    Foreman-Mackey, D., Hogg, D. W., Lang, D., & Goodman, J. 2013, PASP, 125, 306

    2013

  25. [25]

    2022, A&A, 662, A125

    Fouesneau, M., Andrae, R., Dharmawardena, T., et al. 2022, A&A, 662, A125

    2022

  26. [26]

    Freire, P. C. C. & Wex, N. 2024, Living Reviews in Relativity, 27, 5 Gaia Collaboration, Vallenari, A., Brown, A. G. A., et al. 2023, A&A, 674, A1

    2024

  27. [27]

    Hoffman, M. D. & Gelman, A. 2011, arXiv e-prints, arXiv:1111.4246

    Pith/arXiv arXiv 2011

  28. [28]

    Joshi, K. J. & Rasio, F. A. 1997, ApJ, 479, 948

    1997

  29. [29]

    Kopeikin, S. M. 1996, ApJ, 467, L93

    1996

  30. [30]

    2018, IEEE Microwave Magazine, 19, 112

    Li, D., Wang, P., Qian, L., et al. 2018, IEEE Microwave Magazine, 19, 112

    2018

  31. [31]

    Lidov, M. L. 1962, Planet. Space Sci., 9, 719

    1962

  32. [32]

    2021, ApJ, 911, 45

    Luo, J., Ransom, S., Demorest, P., et al. 2021, ApJ, 911, 45

    2021

  33. [33]

    N., Hobbs, G

    Manchester, R. N., Hobbs, G. B., Teoh, A., & Hobbs, M. 2005, AJ, 129, 1993

    2005

  34. [34]

    2026, arXiv e-prints, arXiv:2607.18219

    McGloughlin, B., Ming, J., Alessandra Papa, M., et al. 2026, arXiv e-prints, arXiv:2607.18219

    Pith/arXiv arXiv 2026

  35. [35]

    2015, Tempo: Pulsar timing data analysis, Astrophysics Source Code Library, record ascl:1509.002

    Nice, D., Demorest, P., Stairs, I., et al. 2015, Tempo: Pulsar timing data analysis, Astrophysics Source Code Library, record ascl:1509.002

    2015

  36. [36]

    J., et al

    Nieder, L., Kerr, M., Clark, C. J., et al. 2022, ApJ, 931, L3

    2022

  37. [37]

    1968, Physical Review, 170, 1186

    Nordtvedt, K. 1968, Physical Review, 170, 1186

    1968

  38. [38]

    Ocker, S. K. & Cordes, J. M. 2026, ApJ, 1002, 3

    2026

  39. [39]

    Perera, B. B. P., Stappers, B. W., Lyne, A. G., et al. 2017, MNRAS, 468, 2114

    2017

  40. [40]

    C., Flynn, C., & Deller, A

    Price, D. C., Flynn, C., & Deller, A. 2021, PASA, 38, e038

    2021

  41. [41]

    M., Stairs, I

    Ransom, S. M., Stairs, I. H., Archibald, A. M., et al. 2014, Nature, 505, 520

    2014

  42. [42]

    Shannon, R. M. & Cordes, J. M. 2010, ApJ, 725, 1607

    2010

  43. [43]

    Shklovskii, I. S. 1970, Soviet Ast., 13, 562

    1970

  44. [44]

    F., Cutri, R

    Skrutskie, M. F., Cutri, R. M., Stiening, R., et al. 2006, AJ, 131, 1163

    2006

  45. [45]

    A., Abdollahi, S., Ajello, M., et al

    Smith, D. A., Abdollahi, S., Ajello, M., et al. 2023, ApJ, 958, 191

    2023

  46. [46]

    S., Urquhart, R., et al

    Strader, J., Ray, P. S., Urquhart, R., et al. 2025, ApJ, 980, 124

    2025

  47. [47]

    L., Archibald, A

    Susobhanan, A., Kaplan, D. L., Archibald, A. M., et al. 2024, ApJ, 971, 150 V oisin, G., Cognard, I., Freire, P. C. C., et al. 2020, A&A, 638, A24 V oisin, G., Cognard, I., Saillenfest, M., et al. 2025, A&A, 693, A143

    2024

  48. [48]

    Will, C. M. 1993, Theory and Experiment in Gravitational Physics

    1993

  49. [49]

    Will, C. M. 2018, Theory and Experiment in Gravitational Physics

    2018

  50. [50]

    2026, Nature Astronomy

    Wu, Q., Wang, N., Yuan, J., et al. 2026, Nature Astronomy

    2026

  51. [51]

    M., Manchester, R

    Yao, J. M., Manchester, R. N., & Wang, N. 2017, ApJ, 835, 29

    2017

  52. [52]

    2026, arXiv e-prints, arXiv:2607.16119 Article number, page 10 P

    Zhang, M., Pei, S., & Zhang, P. 2026, arXiv e-prints, arXiv:2607.16119 Article number, page 10 P. C. C. Freire et al.: On the triple nature of the PSR J0435+3233 system Appendix A: Full timing results Fig. 5 in the main text shows the posterior distribution for the outer-orbital parameters and the pulsar’s spin-frequency derivative. Fig. A.1 here shows th...

    Pith/arXiv arXiv 2026