Indefinite survival of life in an ever-expanding universe: the unresolved assumptions of Dyson's 'Time without end' scenario

Dyson's 1979 analysis 'Time without end: Physics and biology in an open universe' argued that life could survive indefinitely in an open, ever-cooling universe, conditional on assumptions the paper itself left open: the Biological Scaling Hypothesis (subjective time scales with operating temperature), the stability of matter against proton decay together with the minimum meaningful black-hole mass, the realizability of radiators and receivers operating near the theoretical dipole-power and sum-rule bounds, the tunneling lifetimes for macroscopic collapse including lower-action non-symmetric modes, and the feasibility of low-temperature life architectures. Later work transformed but did not settle the question: Krauss and Starkman (2000) argued that a positive cosmological constant imposes a de Sitter horizon-temperature floor and, more generally, that finite physical systems cannot process information indefinitely; Barrow and Hervik (2003) and Freese and Kinney (2003) exhibited ever-expanding backgrounds (anisotropic, or quintessence-driven with $w > -1$) in which indefinite information processing appears possible, leaving the dispute unresolved and coupled to open empirical questions: the dark-energy equation of state and proton stability (Super-Kamiokande bounds on the partial lifetime exceed $2.4\times10^{34}$ yr with no decay observed). The problem asks for adjudication of the overarching question, can life as information processing persist forever in an ever-expanding universe, by resolution of its load-bearing physical and biological assumptions, with every surviving conditionality made explicit.

Background

In 1979 Freeman J. Dyson published "Time without end: Physics and biology in an open universe", asking whether life could survive indefinitely in a universe that expands forever without re-collapsing. The paper analyzed the open Friedmann-Robertson-Walker (FRW) cosmology — the standard homogeneous, isotropic solution of general relativity with zero cosmological constant $\Lambda$ (the constant vacuum-energy density, which acts as a repulsive stress and accelerates expansion) and sub-critical matter density, which expands forever at a decelerating rate — in which the temperature of the cosmic microwave background (CMB), the relic radiation bath, falls toward zero as the universe grows. Physical eschatology, the quantitative study of the far future of the universe and its contents, takes this paper as a foundational text.

Dyson's argument has three load-bearing pieces. First, the Biological Scaling Hypothesis: if a living creature is copied quantum state by quantum state so that its Hamiltonian becomes $H_c=\lambda U H U^{-1}$ (with $U$ unitary and $\lambda>0$) and its environment's temperature is rescaled from $T$ to $\lambda T$, then the copy is alive and subjectively identical, with all vital functions slowed by the factor $\lambda$. Under this hypothesis the relevant clock is "subjective time" $u(t)=f\int_0^t T(t')dt'$ rather than proper time, so an ever-cooler civilization can in principle accumulate unbounded subjective time. Second, energy and heat-disposal limits: a creature of complexity $Q$ — roughly the entropy produced per moment of subjective awareness; a human dissipating 200 W at 300 K gives $Q\sim10^{23}$ — has a minimum metabolic rate scaling as $T^2$, while the power available to dispose of its waste heat is bounded by the limit on electromagnetic dipole radiation from its $N_e$ electrons; every other channel (gravitational quadrupole radiation, weak-interaction neutrino emission) falls faster with temperature. Third, hibernation: alternating active and dormant eras, a civilization whose operating temperature falls as $t^{-p}$ achieves infinite total subjective time with finite total energy consumption whenever $1/3<p\le 1/2$ and the background cools at least as fast.

The argument is explicitly conditional on the far-future physics of matter. If protons decay, ordinary matter disintegrates on the decay timescale. If they do not, cold degenerate matter is slowly processed by pycnonuclear fusion (fusion driven by the quantum zero-point motion of nuclei rather than heat, which continues even at zero temperature) toward iron on timescales of order $10^{1500}$ yr, and the resulting "iron stars" can in principle collapse into black holes by quantum tunneling; black holes of mass $M$ then evaporate through Hawking radiation — thermal emission at temperature $1/(8\pi G M)$ — over timescales growing as $M^3$. Which futures remain open for material embodiments of life therefore depends on the proton lifetime, on the minimum meaningful black-hole mass $M_B$, and on the tunneling lifetimes; Dyson's estimates assumed the most symmetric (spherical) collapse paths, and lower-action non-symmetric modes could shorten them. Dyson also flagged that no hardware has been designed to approach the dipole power bound or the corresponding receiver "sum-rule" limits (bounds on the total power a receiver can absorb, obtained by integrating its frequency-dependent absorption cross-section over all frequencies, in analogy with the spectral sum rules of atomic physics) at ultra-low temperature, and that the biological side — functional equivalents of sensors, actuators, and memory (including analog memories of growing precision) at arbitrarily low temperature — remained entirely speculative.

Dyson wrote before cosmic acceleration was discovered. Type-Ia supernova surveys in 1998–1999 indicated that the expansion is accelerating, driven by "dark energy" consistent with a positive $\Lambda$. Krauss and Starkman then observed that a $\Lambda$-dominated universe approaches de Sitter space (the maximally symmetric vacuum solution of general relativity with positive $\Lambda$), whose horizon carries a thermal Gibbons-Hawking temperature $T_{dS}=(\Lambda/12\pi^2)^{1/2}$, of order $10^{-30}$ K for the value inferred from observation — a floor beneath which nothing can cool. Dark energy, however, need not be a cosmological constant: in quintessence models its equation-of-state parameter $w$ (pressure over energy density) exceeds $-1$ and generally evolves in time, and the horizon temperature then falls with the Hubble rate rather than saturating. Krauss and Starkman further argued from the thermodynamics of computation: computation can in principle be made thermodynamically reversible (Bennett/Landauer) if performed arbitrarily slowly, but erasing and resetting finite memory is dissipative, and no finite system possesses an infinite tower of ever-lower energy states, so a finite physical system cannot perform infinitely many irreversible computations. Whether indefinite information processing is possible has thereby become a model-dependent question at the interface of cosmology, the thermodynamics of computation, particle physics, and speculative biology.

Problem Statement

The problem is Dyson's conditional question, with the precision later work has added: in an ever-expanding universe, can life — understood as indefinitely continued information processing — persist forever, and which of the load-bearing assumptions does that verdict rest on?

The question resolves into three linked parts, which jointly support this single verdict:

1. Settle the cosmological core. For each specified ever-expanding background — open or critical-density FRW with $\Lambda=0$; $\Lambda$-dominated (de Sitter) space; accelerating expansion driven by dark energy with $w>-1$ (constant or time-varying, e.g. quintessence potentials $V\propto\phi^n$); and anisotropic ever-expanding models — determine whether Dyson's condition, infinite integrated subjective time $u(t\to\infty)\to\infty$ with finite total energy consumed under the Biological Scaling Hypothesis, can be met. This requires adjudicating the unresolved dispute between Krauss and Starkman's general no-go arguments (the de Sitter temperature floor; alarm-clock complexity limits; loss of thermal contact with cooling backgrounds; breakdown of the dipole-radiation estimate once the thermal wavelength exceeds the system size; and the ground-state/infinite-tower argument that no finite system can perform infinitely many irreversible computations) and the counterarguments of Freese and Kinney (in backgrounds with growing horizons the infrared cutoff — the longest-wavelength, lowest-energy scale available to a finite system, set by the horizon size — falls, so the effective ground state keeps receding and relaxation takes infinite subjective time) and of Barrow and Hervik (anisotropic shear supplies persistent temperature gradients exploitable as free energy).

2. Resolve the load-bearing physical assumptions that Dyson explicitly left open: (i) whether protons are absolutely stable or decay, and on what timescale — the experimental lower bound on the partial lifetime in the $p\to e^+\pi^0$ mode currently exceeds $2.4\times10^{34}$ yr with no decay observed, leaving both branches live; (ii) the minimum meaningful black-hole mass $M_B$ (candidate values discussed in the source range from zero through the Planck mass and the quantum-classical crossover mass $M_Q$ to the Chandrasekhar mass — the maximum mass, about 1.4 solar masses, that electron-degeneracy pressure can support against gravitational collapse — with qualitatively different consequences for the fate of material objects); (iii) whether radiators and receivers operating near the theoretical dipole-power and sum-rule bounds are realizable at arbitrarily low temperature, including the macroscopic quantum effects (discrete excitation spectra once the thermal wavelength exceeds the system size) that Krauss and Starkman argued suppress cooling below Dyson's estimate; (iv) the microscopic tunneling pathways and lifetimes for macroscopic objects (iron stars and degenerate remnants) to fuse, collapse, or nucleate black holes, including non-symmetric collapse modes of lower action than the symmetric estimates.

3. Resolve the biological assumptions: the validity of the Biological Scaling Hypothesis itself, and whether any feasible architecture for low-temperature life exists — functional equivalents of sensors, actuators, and memory, including analog memories of growing precision — or whether surviving physical information processing must depart from the hypothesis's $T^2$ metabolic scaling (for example by reversible or quantum computation, which Freese and Kinney showed changes time constants but not, by itself, the finite-versus-infinite verdict).

Each of the three parts is load-bearing for the same adjudication: an answer that resolves only some of them (the pattern of the existing literature) must state precisely which surviving assumptions its conclusion remains conditional on, and does not complete the problem. The verdict depends on the cosmological model and its parameter values: the resolution must map out which expanding universes permit indefinite information processing and why. The verdict must cover the four background classes of part 1: open or critical-density FRW with $\Lambda=0$; $\Lambda$-dominated (de Sitter) space; accelerating expansion with $w>-1$; and anisotropic ever-expanding models.

For the sub-questions whose underlying fact is itself empirically undetermined — proton stability, and the feasibility of a low-temperature life architecture — resolution consists in carrying the live branches explicitly and stating which part of the verdict is conditional on each, not in settling the undetermined fact itself.

References

1. Freeman J. Dyson, 1979, 'Time without end: Physics and biology in an open universe', Reviews of Modern Physics 51, 447-460. DOI: 10.1103/RevModPhys.51.447, https://doi.org/10.1103/RevModPhys.51.447
2. Lawrence M. Krauss and Glenn D. Starkman, 2000, 'Life, The Universe, and Nothing: Life and Death in an Ever-Expanding Universe', The Astrophysical Journal 531, 22-30. DOI: 10.1086/308434, https://arxiv.org/abs/astro-ph/9902189
3. Fred C. Adams and Gregory Laughlin, 1997, 'A dying universe: the long-term fate and evolution of astrophysical objects', Reviews of Modern Physics 69, 337-372. DOI: 10.1103/RevModPhys.69.337, https://arxiv.org/abs/astro-ph/9701131
4. John D. Barrow and Sigbjørn Hervik, 2003, 'Indefinite information processing in ever-expanding universes', Physics Letters B 566, 1-7. DOI: 10.1016/S0370-2693(03)00775-5, https://arxiv.org/abs/gr-qc/0302076
5. Katherine Freese and William H. Kinney, 2003, 'The ultimate fate of life in an accelerating universe', Physics Letters B 558, 1-8. DOI: 10.1016/S0370-2693(03)00239-9, https://arxiv.org/abs/astro-ph/0205279
6. Super-Kamiokande Collaboration (A. Takenaka et al.), 2020, 'Search for proton decay via p→e+π0 and p→μ+π0 with an enlarged fiducial volume in Super-Kamiokande I-IV', Physical Review D 102, 112011. DOI: 10.1103/PhysRevD.102.112011, https://arxiv.org/abs/2010.16098
7. Steven Frautschi, 1982, 'Entropy in an Expanding Universe', Science 217, 593-599. DOI: 10.1126/science.217.4560.593, https://doi.org/10.1126/science.217.4560.593
8. Milan M. Ćirković, 2004, 'Forecast for the Next Eon: Applied Cosmology and the Long-Term Fate of Intelligent Beings', Foundations of Physics 34, 239-261. DOI: 10.1023/B:FOOP.0000019583.67831.60, https://doi.org/10.1023/B:FOOP.0000019583.67831.60
9. Dan Hooper, 2018, 'Life versus dark energy: How an advanced civilization could resist the accelerating expansion of the universe', Physics of the Dark Universe 22, 74-79. DOI: 10.1016/j.dark.2018.09.005, https://doi.org/10.1016/j.dark.2018.09.005
10. Anders Sandberg, Stuart Armstrong, and Milan M. Ćirković, 2017, 'That is not dead which can eternal lie: the aestivation hypothesis for resolving Fermi's paradox', arXiv:1705.03394, https://arxiv.org/abs/1705.03394
11. M. E. Caplan, 2020, 'Black dwarf supernova in the far future', Monthly Notices of the Royal Astronomical Society 497, 4357-4362. DOI: 10.1093/mnras/staa2262, https://arxiv.org/abs/2008.02296