We explore whether information-theoretic constraints—Bekenstein bounds, holographic entropy limits, Shannon channel capacity, and computational capacity arguments—provide novel explanatory traction on observed correlations in black hole–galaxy co-evolution, AGN feedback, and cosmic structure. Rather than replacing ΛCDM or general relativity, we ask whether information-theoretic quantities yield scaling relations or predictions that complement existing models.
We develop a framework in which black holes are treated as information-bounded systems operating at or near fundamental processing limits. Applying Landauer's principle to AGN feedback, we derive an information-reorganization equilibrium condition (Proposition 3.1) that yields MBH ∝ σ5, matching the Silk & Rees (1998) energy-driven scaling and consistent with the observed M–σ exponent (α ≈ 4–5). The derivation introduces a Landauer temperature factor absent from momentum-driven models, generating a novel prediction: residual scatter in the M–σ relation should correlate with bulge virial temperature (Prediction 4).
We examine Planck-scale information saturation as a singularity resolution mechanism, discuss the Bullet Cluster constraint on information-theoretic reinterpretations of dark matter, and formulate five falsifiable predictions testable with existing or near-term data.
In 1965, Penrose proved that gravitational collapse inevitably produces spacetime singularities—regions of infinite curvature where geodesics terminate (Penrose 1965). Hawking extended this to cosmological settings (Hawking & Penrose 1970). These theorems are mathematically rigorous. They are also, by broad consensus, not descriptions of physical reality: infinite density and the termination of time violate quantum mechanics (unitary evolution), thermodynamics (entropy bounds), and information theory (conservation of information).
The standard response is to invoke quantum gravity. But there is a complementary interpretation: the infinities arise because General Relativity lacks the degrees of freedom to represent what is physically happening at those scales. Shannon's channel capacity theorem (Shannon 1948) establishes that any channel with finite bandwidth distorts signals whose information content exceeds its capacity. We propose—as an interpretive framework—that GR's singularities are instances of this phenomenon: formalism overflow.
Separately, Hawking's discovery that black holes emit thermal radiation (Hawking 1975) created the information paradox. Recent progress on the Page curve via the island formula (Almheiri et al. 2019, Penington 2020) has shifted consensus toward information conservation, but the mechanism remains debated. Both problems point toward the question: what is the information-theoretic structure of black holes and spacetime?
This framework builds on: Wheeler's "It from bit" (1990); Bekenstein's entropy bounds (1973, 1981); the holographic principle ('t Hooft 1993, Susskind 1995, Bousso 2002); Landauer's erasure principle (1961); Lloyd's computational limits (2000, 2002); Jacobson's thermodynamic derivation of Einstein's equation (1995); Verlinde's entropic gravity (2011); the ER=EPR program (Maldacena & Susskind 2013, Van Raamsdonk 2010); Blandford & Znajek's jet mechanism (1977); and the M–σ models of Silk & Rees (1998) and King (2003).
This paper is a speculative framework, not a complete theory. We distinguish three tiers of claims:
| Type | Manifestation | Example |
|---|---|---|
| Divergences | Mathematical infinities | GR singularities, UV divergences |
| Importations | Free parameters | Cosmological constant |
| Postulations | Unexplained axioms | Born rule, measurement postulate |
| Nominations | Named placeholders | Dark matter, dark energy |
For Sagittarius A* (M ≈ 4 × 106 M☉), the Schwarzschild radius is rs ≈ 1.2 × 1010 m, giving horizon area A ≈ 1.8 × 1021 m². With Planck area ℓP² ≈ 2.6 × 10−70 m²:
The maximum operation rate (Margolus & Levitin 1998):
This is an upper bound, not a claim about what a black hole "computes."
An adjacent analysis proposed deriving the Schwarzschild metric by minimizing an "information distortion functional" 𝒟[f] = ∫√(−g) R d⁴x. This is not an independent derivation. The functional is the Einstein-Hilbert action. Extremizing it is exactly the variational procedure that produces the Einstein field equations. Relabeling it as "information distortion" does not constitute derivation from information-theoretic first principles.
Jacobson (1995) achieved something closer to the aspiration: deriving the Einstein equation from thermodynamic reasoning applied to local Rindler horizons. This remains the template for future work.
If information density saturates at the Planck scale, the Schwarzschild metric is regularized:
Properties: (1) recovers Schwarzschild for r ≫ ℓP; (2) f(0) = 1 − rs/ℓP (finite); (3) curvature at origin R(0) = rs/ℓP³ (finite). For Sgr A*: R(0) ≈ 2.9 × 10114 m−2. Enormous but finite.
The M–σ relation (Ferrarese & Merritt 2000, Gebhardt et al. 2000, Kormendy & Ho 2013):
This implies MBH ∝ σα with α = 4.24 ± 0.44. The scatter is ~0.3 dex.
Energy-driven outflow (Silk & Rees 1998): Energy balance between feedback and bulge binding energy yields MBH ∝ σ5.
Momentum-driven outflow (King 2003): Radiation pressure momentum balance yields MBH ∝ σ4.
Both parameterize the feedback coupling efficiency without deriving it.
By Landauer's principle (1961), erasing one bit in a thermal environment at temperature T requires minimum energy kT ln 2. The maximum rate at which AGN feedback reorganizes bulge degrees of freedom:
where ε ~ 0.01–0.1 is the coupling fraction (Fabian 2012).
where μ ≈ 0.6 is the mean molecular weight.
Substituting Lemma 3.3 into 3.1:
The bulge generates entropy through N-body gravitational scattering (Binney & Tremaine 2008). With relaxation time trelax = N*·tcross/(8 ln N*) and virial relation Rb ~ GMb/σ²:
The black hole grows until its information reorganization rate matches the bulge's dynamical entropy production rate.
Setting Γreorg = Γgen and solving for MBH:
Result: MBH ∝ σ5, matching Silk & Rees (1998) and consistent with the observed exponent α = 4.24 ± 0.44.
The Landauer factor kTvir introduces temperature dependence absent from King (2003). For two bulges with identical σ but different virial temperatures:
Prediction: Residual scatter in M–σ correlates positively with bulge virial temperature. Hotter bulges → larger BH masses. This is the opposite sign from Silk & Rees (cooling-function) predictions.
| Feature | King (2003) | Silk & Rees (1998) | This work |
|---|---|---|---|
| Mechanism | Momentum balance | Energy balance | Information reorganization |
| Exponent | σ4 | σ5 | σ5 |
| T dependence | None explicit | Inverse (cooling) | Direct (Landauer) |
| Scatter prediction | None specific | Cooling-dependent | Tvir correlation (+) |
| Novel testable prediction | No | No | Yes (Prop 3.2) |
The MeerKAT Galactic Center survey (Heywood et al. 2019, 2022; Yusef-Zadeh et al. 2022) reveals hundreds of non-thermal radio filaments within ~150 pc of Sgr A*: lengths up to 150 pc, aspect ratios >10:1, remarkably straight, often parallel, highly polarized (20–40%). Standard MHD models account for much of the morphology. The information-theoretic observation is that the filaments' regularity—straightness, parallelism, quasi-uniform spacing—is reminiscent of waveguide architectures. This becomes scientific only if it predicts observable signatures that differ from MHD (see Prediction 1).
AGN jets (Blandford & Znajek 1977, Blandford et al. 2019) show collimation over Mpc scales, quasi-regular knots, and stability over millions of years. A relativistic jet has a calculable Shannon capacity: Cjet ≤ Ljet/(kTjet ln 2). For a powerful FR II jet: Cjet ~ 1050 bits/s. Whether this constrains the jet's structure or variability is the open question.
The cosmic web's topology matches optimal network models with preferential attachment in hyperbolic geometry (Krioukov et al. 2012, Boguñá et al. 2021): scale-free degree distribution P(k) ∝ k−2.2, small-world clustering, filamentary connectivity. This mathematical fact is independent of physical interpretation. The question is whether the similarity is deep or coincidental.
An earlier version of this framework proposed reinterpreting dark matter as "information infrastructure." We abandoned this interpretation because the evidence rules out the naive version.
The Bullet Cluster (1E 0657-56; Clowe et al. 2006) provides direct proof that dark matter decouples from baryonic matter during mergers. Weak lensing maps show mass tracking the collisionless components, offset from X-ray-emitting gas stripped by ram pressure. This rules out infrastructure anchored to baryonic nodes.
The long-term aspiration remains an information-theoretic characterization of dark matter. Any such characterization must satisfy three non-negotiable conditions: (1) collisionless dynamics reproducing Bullet Cluster decoupling; (2) NFW density profiles without ad hoc tuning; (3) gravitational lensing signatures indistinguishable from particulate dark matter. An information field theory satisfying all three is not obviously impossible but is not currently formulated.
Claim: The two-point polarization correlation function shows excess at angular scales of filament spacing (~4–8 arcmin), beyond the smooth power-law from MHD turbulence.
Falsification: No excess at any preferred scale. Data: MeerKAT SARAO archive (DDT-001). Timeline: 12 months.
Claim: Knot spacing scales with the square of the Lorentz factor: Δz ∝ γ². The information prediction (coherence length scaling) gives slope 2 in log-log space. Recollimation models predict different scaling.
Falsification: Spacing uncorrelated with γ, or slope ≠ 2. Data: MOJAVE VLBI survey. Timeline: 24 months.
Claim: Time lag τ = D/c (light-travel) rather than D/cs (sound-crossing). These differ by ~10³. For D ~ 10 kpc: τinfo ~ 3 × 104 yr vs τthermal ~ 3 × 107 yr.
Falsification: Lags consistent with thermal timescales. Data: Multi-wavelength surveys (LOFAR, Herschel, GALEX). Timeline: 48 months.
Claim (Proposition 3.2): Positive correlation between M–σ residuals and bulge TX. King predicts none; Silk & Rees predicts opposite sign.
Falsification: No correlation (ρ ≈ 0) or negative correlation. Data: Kormendy & Ho (2013) + Babyk et al. (2018) X-ray temperatures. Timeline: 6 months.
Claim: Cosmic web Betti numbers match information-optimal network models better than pure N-body predictions. Speculative and underspecified. Data: DESI, Euclid.
Verlinde (2011) proposed gravity as entropic force, with extensions to cosmic acceleration. The approach faces criticisms (Kobakhidze 2011, Visser 2011). Information-theoretic explanations of dark energy remain highly speculative.
Timeline constraint: Cosmic acceleration onset (~5 Gyr ago, z ~ 0.7) does not coincide with peak SMBH formation (z ~ 2–3, ~10–11 Gyr ago). JWST has identified massive SMBHs at z > 6. Any information-theoretic account must address this.
GW signals from compact binary mergers encode masses, spins, and orbital parameters. If black holes are information-processing systems near fundamental limits, mergers might produce waveform features beyond GR predictions. No specific prediction can currently be formulated. Flagged as long-term research direction only.
The primary contribution is Proposition 3.1: a derivation of M–σ scaling from Landauer-limited information reorganization, with a novel testable prediction (Proposition 3.2). Secondary contributions: the formalism overflow taxonomy (Definition 2.1), the Planck-saturated metric (Proposition 2.1), and five testable predictions.
It does not derive GR from information theory (Section 2.4). It does not explain dark matter (Section 4.4). It does not resolve the cosmological constant problem. The Schwarzschild "derivation" is a relabeling.
The most novel conceptual contribution may be Definition 2.1: classifying singularities, free parameters, unexplained axioms, and named placeholders as different manifestations of a theory exceeding its representational capacity. Even if the astrophysical predictions fail, this tool may have independent value in philosophy of physics.
Almheiri, A. et al. (2019). JHEP 2019, 63.
Babyk, Iu. V. et al. (2018). ApJ 857, 32.
Bardeen, J. M. (1968). Proc. GR5, Tbilisi, 174.
Bekenstein, J. D. (1973). Phys. Rev. D 7, 2333.
Bekenstein, J. D. (1981). Phys. Rev. D 23, 287.
Binney, J. & Tremaine, S. (2008). Galactic Dynamics, 2nd ed. Princeton.
Blandford, R. D. & Znajek, R. L. (1977). MNRAS 179, 433.
Blandford, R. et al. (2019). ARA&A 57, 467.
Boguñá, M. et al. (2021). Nature Reviews Physics 3, 114.
Bond, J. R. et al. (1996). Nature 380, 603.
Bousso, R. (2002). Rev. Mod. Phys. 74, 825.
Clowe, D. et al. (2006). ApJ 648, L109.
Curiel, E. (2019). Nature Astronomy 3, 27.
Earman, J. (1995). Bangs, Crunches, Whimpers, and Shrieks. Oxford.
Fabian, A. C. (2012). ARA&A 50, 455.
Ferrarese, L. & Merritt, D. (2000). ApJ 539, L9.
Gebhardt, K. et al. (2000). ApJ 539, L13.
Hawking, S. W. (1975). Commun. Math. Phys. 43, 199.
Hawking, S. W. & Penrose, R. (1970). Proc. R. Soc. A 314, 529.
Hayward, S. A. (2006). Phys. Rev. Lett. 96, 031103.
Heywood, I. et al. (2019). Nature 573, 235.
Heywood, I. et al. (2022). ApJ 925, 165.
Jacobson, T. (1995). Phys. Rev. Lett. 75, 1260.
King, A. (2003). ApJ 596, L27.
Kobakhidze, A. (2011). Phys. Rev. D 83, 021502.
Kormendy, J. & Ho, L. C. (2013). ARA&A 51, 511.
Krioukov, D. et al. (2012). Scientific Reports 2, 793.
Landauer, R. (1961). IBM J. Res. Dev. 5, 183.
Lloyd, S. (2000). Nature 406, 1047.
Lloyd, S. (2002). Phys. Rev. Lett. 88, 237901.
Maldacena, J. & Susskind, L. (2013). Fortschr. Phys. 61, 781.
Margolus, N. & Levitin, L. B. (1998). Physica D 120, 188.
Modesto, L. (2010). Int. J. Theor. Phys. 49, 1649.
Navarro, J. F. et al. (1996). ApJ 462, 563.
Penington, G. (2020). JHEP 2020, 2.
Penrose, R. (1965). Phys. Rev. Lett. 14, 57.
Rubin, V. C. & Ford, W. K. (1970). ApJ 159, 379.
Shannon, C. E. (1948). Bell Syst. Tech. J. 27, 379.
Silk, J. & Rees, M. J. (1998). A&A 331, L1.
Springel, V. et al. (2005). Nature 435, 629.
Susskind, L. (1995). J. Math. Phys. 36, 6377.
't Hooft, G. (1993). arXiv:gr-qc/9310026.
Van Raamsdonk, M. (2010). Gen. Rel. Grav. 42, 2323.
Verlinde, E. (2011). JHEP 2011, 29.
Visser, M. (2011). JHEP 2011, 140.
von Neumann, J. (1955). Math. Found. Quantum Mech. Princeton.
Wheeler, J. A. (1990). In Complexity, Entropy, and the Physics of Information.
Yusef-Zadeh, F. et al. (2022). ApJ 925, L18.