What Is Quantum Gravity — The Theory That Could Unite Physics?

Physics has two extraordinary, deeply tested theories. General relativity, Einstein's theory of gravity, describes gravity as the curvature of spacetime caused by mass and energy. It correctly predicts the precession of Mercury's orbit, the bending of starlight by the Sun, black holes, gravitational waves and the expansion of the universe. Quantum mechanics, and its extension quantum field theory, describes the subatomic world. It underlies chemistry, semiconductors, lasers and nuclear physics.

Each theory works superbly in its own domain. The trouble starts when you need both at once. If you try to treat gravity with the same quantum methods that work for the other forces, the calculations produce infinities that cannot be removed, and the theory stops making predictions at high energies.

Finding a theory of quantum gravity, a single framework that respects both quantum mechanics and general relativity, is widely regarded as the deepest open problem in fundamental physics. This article separates what is established, what is a well-defined mathematical result, and what remains a speculative research programme. That distinction matters, because popular accounts often present candidate theories as if they were settled.

Key takeaways

• Established: general relativity and the quantum field theories of the Standard Model have each passed every experimental test so far. Neither has been seen to fail.
• Established: treating gravity as an ordinary quantum field theory gives infinities that cannot be absorbed in the usual way. A 1986 calculation by Goroff and Sagnotti showed this for pure gravity at two loops.
• The Planck length, about 1.616 × 10⁻³⁵ metres (CODATA 2022), marks the scale where quantum effects of gravity are expected to become strong. It is far beyond any experiment.
• Speculative: string theory, loop quantum gravity, asymptotic safety and causal dynamical triangulations are serious research programmes with important mathematical results, but none has been confirmed by experiment.
• Testable soon, perhaps: tabletop experiments proposed in 2017 aim to test whether gravity can entangle two masses, which would show that gravity has quantum properties.

What is known, and what is not

Before looking at the candidate theories, it helps to sort the claims by how secure they are.

Claim	Status	Basis
General relativity describes gravity accurately from labs to galaxy clusters	Established	Solar system tests, binary pulsars, LIGO/Virgo gravitational waves
Gravitational waves travel at the speed of light to about 1 part in 10¹⁵	Established (measured)	GW170817 and GRB 170817A, 2017
Quantum field theory describes the other three forces	Established	Standard Model tests, including the 2012 Higgs discovery
Gravity quantised as an ordinary field theory is non-renormalisable	Established (calculation)	't Hooft and Veltman 1974; Goroff and Sagnotti 1986
Black holes emit thermal (Hawking) radiation	Strong theoretical result, not observed	Hawking 1974–75
Spacetime is made of strings, loops or discrete events	Speculative	No experimental evidence yet
Gravitons exist	Expected, never detected	Follows if gravity is quantum

Keeping this table in mind avoids two common errors: thinking that quantum gravity has been "solved" by one of the candidate theories, and thinking that the problem is pure philosophy with no hard results.

Why the two theories conflict

General relativity is a classical theory of geometry

In general relativity, gravity is not a force in the usual sense. Mass and energy curve spacetime, and objects move along the straightest possible paths in that curved geometry. The Einstein field equations link the curvature of spacetime to the matter and energy within it. Given a starting state, they predict the future exactly. There is nothing probabilistic in the theory. For a gentler introduction, see our explainer on Einstein's relativity.

Quantum theory is probabilistic

Quantum mechanics describes systems with wavefunctions that give probabilities for measurement outcomes. Quantities such as energy often come in discrete steps, and some pairs of quantities, such as position and momentum, cannot both be sharply defined at once. Quantum field theory goes further: every particle is an excitation of a field that fills space, and forces are carried by particles such as the photon. Our guide to quantum mechanics covers the basics.

The clash in one sentence

In quantum theory, the things in spacetime are uncertain and can be in superpositions. In general relativity, spacetime's geometry depends on where those things are. So if a mass is in a superposition of two places, what is the geometry of spacetime? Classical general relativity has no answer, because it assumes a single, definite geometry.

The renormalisation problem

The Standard Model treats electromagnetism, the weak force and the strong force as quantum field theories. When physicists calculate quantum corrections in these theories, they meet infinities, but they can absorb them into a finite number of measured constants, such as the electron's charge and mass. This procedure is called renormalisation, and theories where it works are renormalisable.

Gravity resists this. Its coupling constant, Newton's constant, has units that make quantum corrections grow with energy. In 1974 Gerard 't Hooft and Martinus Veltman showed that gravity coupled to matter produces infinities at one loop that cannot be absorbed. For pure gravity, the decisive calculation came from Marc Goroff and Augusto Sagnotti, published as "The Ultraviolet Behavior of Einstein Gravity" in Nuclear Physics B 266 (1986). They found a divergence at two loops that requires a new type of counterterm built from the curvature of spacetime. To cancel infinities at each new order, you would need new constants without limit, so the theory cannot predict anything at high energy.

This does not mean quantum gravity is impossible to calculate with. At low energies, general relativity can be treated as an effective field theory, a theory valid below some energy scale, and quantum corrections to Newton's law can be computed reliably. The breakdown happens near the Planck energy. The problem is not that we cannot do any quantum gravity calculations; it is that we do not know the theory that takes over at very high energies.

The Planck scale: where the problem becomes acute

Combining Newton's constant, the speed of light and Planck's constant gives a natural length, the Planck length. The CODATA 2022 value published by NIST is 1.616255 × 10⁻³⁵ metres. The corresponding Planck time is about 5.4 × 10⁻⁴⁴ seconds and the Planck energy about 1.2 × 10¹⁹ GeV (both derived from the same constants).

Quantity	Approximate value	Comparison
Planck length	1.6 × 10⁻³⁵ m	About 10²⁰ times smaller than a proton
Planck time	5.4 × 10⁻⁴⁴ s	Time for light to cross one Planck length
Planck energy	1.2 × 10¹⁹ GeV	About 10¹⁵ times the LHC's collision energy
Planck mass	about 2.2 × 10⁻⁸ kg	Roughly the mass of a flea egg

Source for the Planck length: NIST CODATA 2022. Other values derived from the same constants.

In most situations the conflict never arises. Planets are so large that quantum uncertainty in their position is negligible; atoms are so light that their gravity is negligible. But there are places where both matter:

• Inside black holes. General relativity predicts a singularity, a point of infinite density, at the centre. Long before that, curvature reaches the Planck scale and the classical theory can no longer be trusted. See black holes explained.
• The very early universe. Running the expansion of the universe backwards leads to a moment of infinite density. Before about one Planck time, classical physics fails. Our explainer on the Big Bang describes what is known about the moments that follow.
• Black hole evaporation. Hawking's calculation describes radiation from black holes using quantum fields on curved spacetime. Its endpoint raises the information paradox, discussed below.

The Large Hadron Collider collides protons at 13.6 TeV, or about 1.4 × 10⁴ GeV. The Planck energy is about a million billion times higher. No accelerator on Earth, or plausibly ever, will reach it directly. That is why tests of quantum gravity have to be indirect.

Clues from black holes

Some of the strongest hints about quantum gravity come from black holes, even though no one has observed the effects involved.

In the early 1970s, Jacob Bekenstein argued that black holes must have entropy proportional to the area of their event horizon. Stephen Hawking then showed, in 1974 and 1975, that quantum fields near a horizon cause a black hole to radiate like a hot body, with a temperature inversely proportional to its mass. The resulting Bekenstein-Hawking formula says a black hole's entropy is one quarter of its horizon area in Planck units. For background, see Hawking radiation explained and our profile of Stephen Hawking.

Entropy, in ordinary physics, counts microscopic states, as explained in our article on entropy and the arrow of time. So what microscopic states does a black hole have? General relativity says a black hole is described by just mass, charge and spin. Any theory of quantum gravity should explain where the hidden states are. This is one of the few concrete targets the field has.

Then there is the information paradox. If a black hole radiates away completely and the radiation is exactly thermal, the information about what fell in seems to be lost. Standard quantum mechanics forbids this, because quantum evolution is reversible. Either the radiation carries the information out in subtle correlations, or quantum mechanics must change, or something else happens at the end. This remains debated, although many theorists now think the information does come out.

Approach 1: String theory

The idea. String theory replaces point particles with tiny one-dimensional strings. Different vibration patterns of a string look like different particles. One pattern has exactly the properties of a graviton, the expected quantum of gravity. So gravity is not added to string theory; it is required by it. We cover the basics in string theory explained.

Established mathematical results.

• Perturbative string theory gives finite results order by order for graviton scattering in the cases calculated, which is the problem that defeated ordinary quantised gravity.
• In 1996, Andrew Strominger and Cumrun Vafa used string theory to count the microscopic states of a class of five-dimensional extremal black holes and reproduced the Bekenstein-Hawking entropy, S = A/4, exactly (Physics Letters B 379, 1996). This was the first time a microscopic count matched the formula.
• In November 1997, Juan Maldacena proposed the AdS/CFT correspondence, a conjectured equivalence between string theory in a particular curved spacetime (anti-de Sitter space) and an ordinary quantum field theory without gravity living on its boundary (Advances in Theoretical and Mathematical Physics 2, 1998). It is one of the most-cited papers in physics and has become a tool for studying strongly interacting systems.

Open problems.

• Extra dimensions. Consistent superstring theories need ten spacetime dimensions; M-theory has eleven. The six or seven extra dimensions must be curled up too small to see. None has been observed.
• The landscape. There appear to be a vast number of ways to curl up the extra dimensions, often quoted as around 10⁵⁰⁰, an estimate from the early 2000s. Each gives different low-energy physics, so string theory does not obviously predict the particles and constants we see.
• Our universe. AdS/CFT works best for spacetimes with a negative cosmological constant. Our universe appears to have a small positive one, and constructing such solutions reliably in string theory is still disputed.
• No experimental confirmation. No stringy effects have been seen, and searches at the LHC for some low-scale versions, such as large extra dimensions, have found nothing.

Status: a mathematically rich framework with strong internal results, especially about black holes, but not confirmed as a description of our world.

Approach 2: Loop quantum gravity

The idea. Loop quantum gravity (LQG) tries to quantise general relativity directly, without new particles or extra dimensions. Its starting point was a reformulation of general relativity by Abhay Ashtekar in 1986, which made the equations look more like those of the gauge theories of the Standard Model. The central claim is that space itself is granular: the quantum states of geometry are described by networks, called spin networks, whose links carry quanta of area.

Established mathematical results.

• In a paper submitted in November 1994 and published in Nuclear Physics B 442 (1995), Carlo Rovelli and Lee Smolin showed that in this framework the operators for area and volume have discrete spectra. Space, in LQG, comes in smallest units near the Planck scale.
• LQG reproduces the Bekenstein-Hawking entropy of black holes up to a parameter (the Barbero-Immirzi parameter) that is fixed by matching the formula.
• Applied to cosmology, a simplified version called loop quantum cosmology produces models in which the Big Bang singularity is replaced by a "bounce" from an earlier contracting phase, notably in work by Abhay Ashtekar, Tomasz Pawlowski and Parampreet Singh in 2006.

Open problems.

• The classical limit. It has proven very hard to show that LQG reproduces smooth spacetime and ordinary general relativity at large scales.
• Dynamics. Defining how quantum geometry evolves in time is not settled; several versions exist.
• Matter and unification. LQG is a theory of quantum spacetime. It does not explain why the Standard Model has the particles it has.
• Predictions. Some early versions predicted that the speed of light might depend slightly on energy. Observations of distant gamma-ray bursts have put tight limits on such effects, which rules out the simplest versions of that idea in any approach.

Status: a serious programme with clear results about quantum geometry, but its connection to the smooth spacetime we observe is not yet established.

Approach 3: Asymptotic safety

The idea. Maybe gravity does not need new ingredients at all. In 1979 Steven Weinberg suggested that gravity could be "asymptotically safe": as energy increases, its couplings might approach a fixed value instead of growing without limit. If so, the theory would be predictive even at very high energies, despite failing the usual renormalisation test. This is similar in spirit to asymptotic freedom in the strong force, where the coupling becomes weak at high energy.

Results. Since the late 1990s, approximate calculations using a method called the functional renormalisation group have repeatedly found evidence for such a fixed point. As Astrid Eichhorn's 2018 review "An asymptotically safe guide to quantum gravity and matter" explains, the programme has moved toward asking what the fixed point implies for matter, including possible constraints on Standard Model parameters.

Open problems. The calculations use truncations, approximations that keep only some terms. Whether the fixed point survives in the full theory is not proven. There are also questions about whether the approach is consistent with basic principles such as unitarity.

Other approaches

Causal dynamical triangulations (CDT). Spacetime is approximated by gluing together tiny four-dimensional building blocks, with a rule that keeps cause and effect well defined. Computers then sum over all such geometries. In a 2004 paper in Physical Review Letters (93, 131301), Jan Ambjørn, Jerzy Jurkiewicz and Renate Loll reported that a large-scale four-dimensional universe emerges from this sum, which earlier approaches without the causality rule had failed to produce.

Causal set theory. Spacetime is a discrete set of events with only one structure: which events can influence which. Smooth geometry is meant to emerge on large scales, much as a fluid emerges from molecules.

Emergent or entropic gravity. Some proposals treat gravity as not fundamental at all, but a large-scale effect of something deeper, such as thermodynamics or quantum entanglement. Erik Verlinde's 2010 paper "On the Origin of Gravity and the Laws of Newton" (published in JHEP in 2011) argued that gravity could be an entropic force. Ideas from AdS/CFT, in which spacetime geometry is linked to entanglement in the boundary theory, have given this line of thinking a more precise form, though it remains speculative.

Approach	Core idea	Gravity is...	Extra dimensions?	Key result	Main open problem
String theory	Particles are vibrating strings	One string vibration (graviton)	Yes (10 or 11)	Black hole entropy count (1996); AdS/CFT (1997)	Landscape; no experimental test
Loop quantum gravity	Space is quantised into spin networks	Quantum geometry	No	Discrete area and volume (1995)	Recovering smooth spacetime
Asymptotic safety	Couplings reach a fixed point	An ordinary quantum field	No	Fixed point found in approximations	Proof beyond truncations
Causal dynamical triangulations	Sum over causal building blocks	Emergent geometry	No	4D universe emerges (2004)	Link to continuum physics
Causal sets	Spacetime is discrete events	Order plus counting	No	Lorentz-invariant discreteness	Dynamics
Emergent gravity	Gravity from thermodynamics or entanglement	Not fundamental	Depends	Entropic derivations	Precise, testable formulation

Can quantum gravity be tested?

For decades the answer seemed to be "not in practice." That has started to change, not because anyone can reach the Planck energy, but because experiments have found cleverer questions.

Is gravity quantum at all? In 2017, two groups independently proposed tabletop experiments. One, by Sougato Bose and colleagues, titled "A Spin Entanglement Witness for Quantum Gravity" (Physical Review Letters 119, 240401), imagines two tiny masses, each placed in a superposition of two positions. If the masses become entangled purely through their gravitational attraction, then gravity must be able to carry quantum information, which a classical field cannot do. Chiara Marletto and Vlatko Vedral published a related proposal at the same time. These experiments have not yet been performed; they require keeping even very small masses in superposition, which is far beyond current technology, but research groups are working toward it.

Constraints from astronomy. On 17 August 2017, LIGO and Virgo detected gravitational waves from merging neutron stars (GW170817), and the Fermi and INTEGRAL satellites saw a gamma-ray burst about 1.7 seconds later. Since both signals travelled about 130 million light-years, the LIGO, Virgo, Fermi and INTEGRAL teams showed in The Astrophysical Journal Letters 848 (2017) that the speed of gravity differs from the speed of light by less than about one part in 10¹⁵. This rules out many modified gravity theories and constrains some quantum gravity ideas. See our account of the LIGO gravitational wave detection.

Cosmology. If the early universe went through a phase of rapid expansion called inflation, quantum fluctuations of spacetime itself would have left a faint pattern, called B-mode polarisation, in the cosmic microwave background. Detecting it would be evidence that spacetime fluctuations were quantised. Searches so far have only set upper limits.

Black hole observations. Gravitational-wave "ringdown" signals and images from the Event Horizon Telescope test whether black holes match general relativity's predictions near the horizon. So far, they do, within the uncertainties.

Common misconceptions

• "Gravitational waves are gravitons." No. LIGO detects classical ripples in spacetime. Detecting a single graviton is thought to be practically impossible with any foreseeable detector, because gravity interacts so weakly.
• "String theory has been disproved." No. It has not been confirmed, and some low-energy versions have been constrained, but its core claims are at energies no experiment can reach.
• "Quantum gravity means we cannot calculate anything about gravity and quantum mechanics together." No. Quantum fields on curved spacetime, which gives Hawking radiation, and low-energy effective field theory of gravity are both well-developed and reliable within their limits.
• "The Planck length is the smallest possible length." Not established. It is the scale where we expect new physics. Some approaches, such as LQG, predict a minimum area; others do not.
• "A theory of everything is the same as quantum gravity." Not quite. A quantum theory of gravity could exist without unifying all the forces. LQG and asymptotic safety, for example, do not aim to explain the Standard Model.

What quantum gravity must explain

Any successful theory should:

1. Reduce to general relativity at large distances and weak fields, consistent with every gravitational test so far.
2. Reduce to quantum field theory in flat spacetime, consistent with the Standard Model.
3. Explain black hole entropy by identifying the microscopic states behind the Bekenstein-Hawking formula.
4. Resolve the information paradox, saying what happens to information that falls into a black hole.
5. Describe the Big Bang and black hole singularities without infinities.
6. Make testable predictions that differ from general relativity somewhere we can look.

The first two are constraints every candidate tries to meet. String theory and LQG both claim progress on the third. The last is where every programme still falls short.

Frequently asked questions

Q: What is quantum gravity in simple terms? A: It is the missing theory that would describe gravity using the rules of quantum mechanics. General relativity explains gravity as curved spacetime, but it is a classical theory. Quantum gravity would explain what happens when spacetime itself behaves quantum mechanically, as near black hole centres or at the Big Bang.

Q: Why can't gravity be quantised like the other forces? A: When gravity is treated as an ordinary quantum field, calculations produce infinities that cannot be removed with a finite number of measured constants. Goroff and Sagnotti showed this for pure gravity at two loops in 1986. The theory still works at low energies but fails near the Planck energy.

Q: Is string theory the answer to quantum gravity? A: It is the most developed candidate and contains a graviton automatically. It has produced notable results, such as counting the states of certain black holes in 1996. But it requires extra dimensions, allows an enormous number of possible universes, and has no experimental confirmation.

Q: How is loop quantum gravity different from string theory? A: Loop quantum gravity quantises spacetime geometry directly, in four dimensions, without strings or extra dimensions. It predicts that area and volume come in discrete units near the Planck scale. Unlike string theory, it does not try to unify all forces.

Q: Have gravitons been detected? A: No. Gravitons are the expected quantum particles of gravity, but they interact so weakly that detecting even one is considered practically impossible. LIGO's gravitational waves are classical waves, not individual gravitons.

Q: Could quantum gravity ever be tested? A: Possibly, indirectly. Proposed tabletop experiments would test whether gravity can entangle two masses, and cosmology and black hole observations can constrain theories. The 2017 neutron-star merger already showed gravity travels at the speed of light to about one part in 10¹⁵.

The bottom line

Quantum gravity is the theory physics is missing: a description of gravity consistent with quantum mechanics. What is established is that general relativity and quantum field theory each work, that naive quantised gravity breaks down at high energies, and that black holes appear to have an entropy that needs explaining. What is not established is which candidate, if any, is right. String theory, loop quantum gravity, asymptotic safety and causal dynamical triangulations each have genuine mathematical achievements and serious gaps. The most promising near-term progress may come not from accelerators but from precise tabletop and astronomical experiments that ask a simpler question first: is gravity quantum at all?

Sources

• NIST, CODATA 2022 value of the Planck length.
• M. H. Goroff and A. Sagnotti, "The Ultraviolet Behavior of Einstein Gravity," Nuclear Physics B 266 (1986) 709–736; see the Goroff Caltech thesis record.
• C. Rovelli and L. Smolin, "Discreteness of area and volume in quantum gravity", Nuclear Physics B 442 (1995).
• A. Strominger and C. Vafa, "Microscopic Origin of the Bekenstein-Hawking Entropy", Physics Letters B 379 (1996).
• J. Maldacena, "The Large N Limit of Superconformal Field Theories and Supergravity", Adv. Theor. Math. Phys. 2 (1998).
• J. Ambjørn, J. Jurkiewicz and R. Loll, "Emergence of a 4D World from Causal Quantum Gravity", Physical Review Letters 93 (2004).
• E. Verlinde, "On the Origin of Gravity and the Laws of Newton", JHEP (2011).
• S. Bose et al., "A Spin Entanglement Witness for Quantum Gravity", Physical Review Letters 119 (2017).
• LIGO, Virgo, Fermi GBM and INTEGRAL, "Gravitational Waves and Gamma-rays from a Binary Neutron Star Merger: GW170817 and GRB 170817A", ApJL 848 (2017).
• A. Eichhorn, "An asymptotically safe guide to quantum gravity and matter" (2018).

Related Reading

• What Is the Speed of Dark — And Other Things That Move Faster Than Light
• What Is Spacetime? The Fabric of the Universe Explained
• What Is the Higgs Boson — and Why It Matters
• What Is the Largest Known Prime Number — And Why Do Mathematicians Keep Searching?
• What Is Entropy — and Why the Universe Keeps Running Down