Skip to content
CosmosExplorer
Chapter 03 · The Nature of Space and Time/Emergent Spacetime & Gravity

Causal Set Theory

1987 · Luca Bombelli, Joohan Lee, David Meyer, Rafael Sorkin
Frontier

Spacetime is fundamentally discrete: a partially ordered set of events where causal relations are primary and continuum geometry emerges by counting.

Skip 3D content

In one sentence

Spacetime is fundamentally a discrete set of events ordered by causality, with continuum geometry emerging as a large-scale approximation when the discrete set is dense, analogous to how a fluid emerges from molecules.

The claim

Causal Set Theory (Bombelli, Lee, Meyer & Sorkin 1987) posits that spacetime is at root a discrete set of elementary 'events,' with the only fundamental structure being a partial order specifying which event can causally influence which. There is no underlying continuum manifold; all geometric quantities (distances, volumes, dimensions) are recovered by counting and ordering at large scales when the discrete set is sufficiently dense.

When the discrete causal set is 'sprinkled' densely into a target Lorentzian manifold (events sampled randomly with constant density), the order structure of the causal set approximates the causal structure of the target manifold. Volumes are recovered by counting events. Lengths come from longest chains. The full geometry of a manifold is encoded in 'order plus number,' to use Sorkin's slogan.

The program is small but active in 2024-2026. The major technical challenge is showing that generic random causal sets approximate 4D Lorentzian manifolds with realistic dynamics; this hasn't been convincingly demonstrated outside specific construction schemes (classical sequential growth, transition-amplitude models). Phenomenological work explores 'everpresent Λ' models in which the cosmological constant fluctuates around its observed small value, and stochastic Lorentz-invariant fluctuations from discreteness that might be observable in high-precision cosmic-ray or gravitational-wave timing experiments.

The family stance

Spacetime is not fundamental. It emerges from a deeper structure: entanglement patterns, thermodynamic relations on horizons, or discrete causal ordering. None of these has been confirmed; each makes some testable predictions but most operate at conceptual or structural levels.

Predictions

  • 'Everpresent Λ' models: a cosmological constant that fluctuates around its observed small value, leaving specific imprints on large-scale structure and the cosmic expansion history
  • Discreteness at the Planck scale: stochastic, Lorentz-invariant fluctuations in particle propagation, potentially observable in cosmic-ray spectra or gravitational-wave timing
  • Black-hole entropy as a count of discrete elements crossing the horizon, agreeing with Bekenstein-Hawking only in specific limits

Evidence

  • The order + number = geometry program has demonstrated continuum approximation for specific classes of causal sets sprinkled in Minkowski and de Sitter spacetimes
  • Lorentz invariance can be preserved statistically at the ensemble level even with a discrete underlying structure, a non-trivial result given that most discretizations break Lorentz invariance manifestly
  • 'Everpresent Λ' gives a candidate dynamical explanation for the small observed cosmological constant without fine-tuning

Counterpoints

  • A central technical challenge is showing that generic random causal sets (not specially constructed ones) admit realistic continuum limits with 4D Lorentzian manifold structure; this has not been convincingly demonstrated
  • The dynamics of causal sets (how they grow, how quantum amplitudes are computed) is less developed than alternatives like Loop Quantum Gravity or string theory; multiple dynamics proposals exist with none universally accepted
  • Phenomenological predictions are typically very small or model-dependent, making them hard to distinguish from other quantum-gravity approaches or from standard GR
  • The program is small relative to LQG or holographic approaches; fewer researchers means slower convergence on key technical questions
0votes
Go deeperTechnical detail with proper terminology

Causal set basics: a set C with a partial order 'precedes' satisfying transitivity (x precedes y, y precedes z implies x precedes z), irreflexivity (not x precedes x), and local finiteness (the cardinality of intermediate elements between any two related events is finite). The order encodes causality; local finiteness encodes discreteness.

Sprinkling: sample events randomly in a target Lorentzian manifold with Poisson density ρ. The discrete causal set inherits the order from the manifold's causal structure. The continuum is recovered as ρ -> infinity. A key result is that sprinklings can preserve Lorentz invariance statistically, in contrast with regular lattices which break it.

Dynamics: classical sequential growth models (Sorkin and collaborators) specify how a causal set grows event-by-event with specific transition probabilities; alternative proposals use quantum amplitudes summed over causal sets in a path-integral fashion. Neither has been fully connected to a continuum quantum field theory.

Phenomenology: stochastic Lorentz-invariant Planck-scale fluctuations might produce energy-dependent dispersion in cosmic-ray and gamma-ray-burst propagation, in principle observable at high enough energies, but typically at levels comparable to other Planck-scale quantum-gravity predictions and hard to distinguish from them.

References

  1. Established
    Bombelli, Lee, Meyer & Sorkin (1987). Space-Time as a Causal Set. Phys. Rev. Lett. 59, 521
  2. Established
    Surya (2019). The causal set approach to quantum gravity. Living Rev. Rel. 22, 5

Last reviewed May 17, 2026

Spotted an error? Have a source to add?

Prefer email?

You can also send a prefilled email with the variant URL already filled in.

Related theories