There is no sample of a black hole. Nothing has been removed from one and placed beneath an instrument. What reaches us has already been filtered through the exterior: light from heated matter, the motion of nearby stars, a shadow against luminous plasma, or the gravitational wave left when two horizons settle into one.

To ask what lies inside is therefore to ask a theory to continue beyond observation. General relativity gives a remarkably exact account of causal structure. Quantum theory insists that the account cannot be final. Between them sits an object whose exterior is mathematically spare and whose microscopic constitution remains unknown.

A boundary made of causality

Einstein’s field equations do not describe gravity as a substance. They relate the curvature of spacetime to the distribution of energy and momentum. Soon after Einstein completed the equations in 1915, Karl Schwarzschild published the first exact vacuum solution in 1916: the geometry outside a spherical, nonrotating mass, determined by a single parameter.1

In Schwarzschild coordinates, coefficients become singular at r = 2GM/c² and at r = 0. The two failures are physically different. The first disappears when the spacetime is written in coordinates that cross the horizon smoothly. The second is accompanied by divergent curvature and cannot be removed by relabeling coordinates. The horizon is not a material shell or a wall of infinite density. It is a null boundary separating events that can communicate with distant infinity from those that cannot.

A sufficiently large black hole need not announce the crossing locally. An observer in free fall can pass the horizon in finite proper time without encountering a sudden surface. What changes is the orientation of every allowable future. Inside, even outward-directed light moves toward smaller radius. Escape is not prevented by an unusually strong force acting through space; it is absent from the future-directed geometry.

Light cones outside and inside a black-hole event horizon, with every future direction inside tilted toward smaller radius
Figure 1Outside the horizon, a future-directed light ray may reach larger radius. Inside, both edges of the future light cone lead toward smaller radius. The diagram is schematic and uses Schwarzschild geometry.

The singularity is more difficult. The singularity theorems establish, under broad assumptions, that gravitational collapse produces geodesic incompleteness: some paths cannot be continued indefinitely.2 They do not provide a microscopic photograph of an infinitesimal object. A singularity is best understood as a verdict on the classical description. General relativity has reached a regime in which it no longer supplies its own continuation.

When geometry acquires entropy

Black holes became thermodynamic objects before anyone knew what their microscopic states might be. Bekenstein associated entropy with horizon area. Hawking’s quantum-field calculation assigned a temperature and showed that an isolated black hole can radiate and lose mass.3,4 In conventional units, the Bekenstein-Hawking entropy is

This is an enormous entropy. Entropy ordinarily counts, in a coarse-grained way, how many microscopic configurations correspond to the same macroscopic description. A gas may have a pressure and temperature while concealing the positions and momenta of its molecules. A stationary black hole in Einstein-Maxwell theory is externally characterized by mass, angular momentum, and charge, yet its entropy suggests an immense number of underlying states. The question “what is it made of?” becomes sharper: what degrees of freedom are being counted?

One sparse macroscopic black-hole description corresponding to many possible microscopic states
Figure 2A few exterior parameters can correspond to an enormous number of possible microstates. The diagram expresses the statistical question posed by the Bekenstein-Hawking entropy; it does not select a microscopic theory.

Hawking radiation intensifies the problem. The familiar picture of particle-antiparticle pairs dividing at the horizon is only a mnemonic and can be misleading when treated literally. The calculation concerns quantum fields on a curved background and produces an approximately thermal spectrum. If evaporation ends in radiation containing no recoverable record of the initial state, pure quantum states appear to evolve into mixed ones, conflicting with unitary evolution.

The information paradox is therefore not the ordinary fact that fallen objects are inaccessible. Inaccessibility is compatible with quantum mechanics. The conflict concerns whether the complete final quantum state can still encode the correlations present before the black hole formed.

Candidate microphysics

String theory supplies controlled examples in which black-hole entropy can be reproduced by counting quantum microstates.5 The fuzzball and microstate-geometry programs push this idea further: in certain settings, individual states are horizonless configurations, while the familiar black-hole geometry appears after many states are coarse-grained together. These results are substantial, but they have not established that a generic astrophysical Kerr black hole is literally a fuzzball. The distance from highly symmetric constructions to objects formed by collapsing stars remains part of the problem.

Loop quantum gravity begins elsewhere, quantizing geometric quantities themselves. In canonical formulations, area and volume operators possess discrete spectra, with spin networks supplying a basis for quantum states of geometry.6 Loop-inspired interior models can replace the classical singular region with a quantum transition or bounce. Such models show how singularity resolution might occur. They do not yet constitute an experimentally established account of a black-hole interior.

Holography changes the question again. In anti-de Sitter settings, gravitational physics in a bulk spacetime is equivalent to a nongravitational quantum theory on its boundary. Calculations involving quantum extremal surfaces and replica wormholes reproduce the expected Page curve for evaporating black holes in idealized models, supporting information-preserving evolution.7,8 They clarify how semiclassical gravity may recover unitarity without furnishing a single agreed inventory of what an astrophysical interior contains.

What observation has earned

Gravitational-wave ringdowns, stellar orbits, accretion flows, and horizon-scale imaging now test the exterior geometry with increasing precision. Event Horizon Telescope measurements of Sagittarius A* find a ring size consistent with the expected Kerr shadow, while analyses of the gravitational-wave event GW250114 test the frequencies with which a merged compact object settles.9,10 These observations constrain alternatives and deviations from general relativity. They do not resolve Planck-scale microstructure or directly observe a singularity.

Four schematic observational channels: stellar orbits, accretion light, a horizon-scale ring, and gravitational-wave ringdown
Figure 3Each channel reaches us from the exterior and tests a different part of the macroscopic geometry. The image and ringdown panels are schematic, informed by the Event Horizon Telescope and LIGO-Virgo-KAGRA; they are not reproductions of measured data.

Inside the horizon, disciplined language matters. The horizon marks a limit of causal access, not the location of a known substance. The singularity marks a failure of the classical theory, not a discovered particle. Entropy strongly implies microscopic structure, while several frameworks show how portions of that structure might be represented. No experiment has chosen among them.

A black hole is therefore not “made of” geometry in quite the way a table is made of atoms, nor can its contents responsibly be named by selecting the most elegant conjecture. We know the equations governing the exterior, the thermodynamic quantity that must be explained, and the precise conflicts that arise when quantum fields meet a horizon. The unknown is no longer shapeless. It has boundaries drawn by what the successful theories cannot simultaneously preserve.

References

  1. Karl Schwarzschild, “On the Gravitational Field of a Mass Point According to Einstein’s Theory”, 1916. ↩
  2. Roger Penrose, “Gravitational Collapse and Space-Time Singularities”, Physical Review Letters 14, 1965. ↩
  3. Jacob D. Bekenstein, “Black Holes and Entropy”, Physical Review D 7, 1973. ↩
  4. Stephen W. Hawking, “Particle Creation by Black Holes”, Communications in Mathematical Physics 43, 1975. ↩
  5. Andrew Strominger and Cumrun Vafa, “Microscopic Origin of the Bekenstein-Hawking Entropy”, Physics Letters B 379, 1996. ↩
  6. Carlo Rovelli and Lee Smolin, “Discreteness of Area and Volume in Quantum Gravity”, Nuclear Physics B 442, 1995. ↩
  7. Ahmed Almheiri et al., “The Page Curve of Hawking Radiation from Semiclassical Geometry”, 2019. ↩
  8. Geoffrey Penington et al., “Replica Wormholes and the Black Hole Interior”, 2019. ↩
  9. Event Horizon Telescope Collaboration, “First Sagittarius A* Event Horizon Telescope Results VI: Testing the Black Hole Metric”, Astrophysical Journal Letters 930, 2022. ↩
  10. LIGO-Virgo-KAGRA Collaboration, “Testing General Relativity and the Nature of the Remnant with GW250114”, 2026. ↩