The nature of dark energy is one of the central unresolved questions in contemporary physics and cosmology. The term refers to whatever is responsible for the observed accelerating expansion of the universe - a phenomenon confirmed by independent lines of evidence but whose physical cause remains deeply contested. Competing positions range from interpreting dark energy as a true physical substance pervading all of space, to treating it as a mathematical artifact of general relativity, to rejecting the acceleration inference altogether in favor of modified gravitational theories or revised assumptions about cosmic structure. The dispute is empirical, theoretical, and interpretive simultaneously: the competing positions differ not only on what dark energy is, but on whether the concept is well-founded at all.

The acceleration of the universe's expansion was inferred in 1998 from observations of Type Ia supernovae by two independent teams - the Supernova Cosmology Project and the High-Z Supernova Search Team - whose results were mutually consistent and unexpected.¹⁾²⁾ The standard cosmological model (ΛCDM) incorporates this acceleration through Einstein's cosmological constant (Λ) - a term he originally introduced and later abandoned - which acts as a repulsive energy density associated with empty space. Under ΛCDM, dark energy accounts for approximately 68% of the total mass-energy content of the universe.³⁾ The question of what dark energy physically is, or whether the label picks out anything real, remains open.

The dominant position in professional cosmology holds that the accelerating expansion is real and best described by Einstein's cosmological constant. On this view, Λ represents the energy density of the vacuum - empty space carries an intrinsic energy that drives expansion. The ΛCDM model incorporating this term fits a wide range of independent observations: Type Ia supernova distances, the cosmic microwave background (CMB) power spectrum, baryon acoustic oscillations (BAO), and large-scale structure surveys. Proponents argue that the convergence of these independent datasets on consistent cosmological parameters is strong evidence that the acceleration is a genuine feature of the universe, not an artifact of any single measurement technique.

The principal theoretical difficulty with this position is the cosmological constant problem: quantum field theory predicts a vacuum energy density many orders of magnitude larger than the value of Λ inferred from observation - estimates of the discrepancy range from roughly 60 to 120 orders of magnitude, depending on the energy scale assumed.⁴⁾ Advocates of the cosmological constant position treat this as an outstanding problem for quantum field theory and the theory of gravity's interface with quantum mechanics, not as a reason to doubt the observational inference. Some invoke anthropic reasoning or the landscape of string theory to explain why Λ has the value it does rather than the much larger theoretically expected value.⁵⁾

See Dark Energy - Cosmological Constant Viewpoint.

A second class of positions holds that the accelerating expansion is real but that it is not driven by a true cosmological constant - that is, by a fixed, unchanging vacuum energy. Instead, these accounts propose that dark energy is a dynamical field whose energy density and equation-of-state parameter (w) vary over time. The simplest and most studied such proposal is quintessence: a scalar field slowly rolling down a potential energy curve, analogous in some respects to the inflaton field proposed for cosmic inflation.⁶⁾ Other proposals include k-essence, phantom energy (in which w < -1, potentially leading to a “Big Rip” future), and tracker fields that naturally approach their observed values from a wide range of initial conditions.

Advocates of dynamical dark energy argue that it is more theoretically natural than a pure cosmological constant, that it may resolve or sidestep the fine-tuning problem associated with Λ, and that future observations measuring the time-evolution of w could distinguish between these models and a static Λ. Current observational constraints are consistent with w = -1 (the cosmological constant value) but do not exclude values close to -1.⁷⁾ Results from the Dark Energy Spectroscopic Instrument (DESI) released in 2024 found mild but notable tension with a static cosmological constant, with the best-fit equation-of-state parameter deviating from w = -1 at modest statistical significance - a finding that has renewed interest in dynamical dark energy models and is subject to ongoing analysis.⁸⁾ Critics note that quintessence models introduce new fine-tuning problems of their own and that no compelling theoretical motivation for any specific scalar field potential has been established.

See Dark Energy - Quintessence Viewpoint.

A third position holds that the apparent acceleration of expansion does not require any new substance or field - that it can instead be explained by modifying the gravitational law at cosmological scales. On this view, general relativity in its standard form is not the correct theory of gravity at large scales, and the inferred dark energy is an artifact of applying an incorrect gravitational framework to the data.

Several modified gravity frameworks have been proposed with cosmological implications. f(R) gravity theories replace the Ricci scalar R in the Einstein-Hilbert action with a function of R, producing additional degrees of freedom that can drive accelerated expansion.⁹⁾ DGP gravity (Dvali-Gabadadze-Porrati) introduces extra spatial dimensions to modify gravity at large distances.¹⁰⁾ Galileon and scalar-tensor theories represent further variants. Proponents argue that these frameworks provide economical explanations for the acceleration without the conceptual burden of an undetected substance, and that they make distinctive predictions - for instance, about the growth rate of large-scale structure - that differ measurably from ΛCDM.

Critics argue that modified gravity theories face stringent constraints. The 2017 detection of gravitational waves from a neutron star merger (GW170817) combined with simultaneous gamma-ray observations effectively ruled out a broad class of scalar-tensor theories by demonstrating that gravitational waves travel at the speed of light to high precision.¹¹⁾ Many of the most natural f(R) and Galileon theories predict a different speed for gravitational waves and are therefore excluded by this result. Surviving modified gravity proposals must account for this constraint while still fitting other cosmological data.

See Modified Gravity - Cosmological Scales Viewpoint.

A minority position within theoretical cosmology holds that the apparent acceleration of expansion is not a fundamental feature of the universe but rather an artifact of applying the idealized FLRW metric - which assumes perfect homogeneity and isotropy - to a universe that is in fact highly inhomogeneous on small scales. On this account, the averaging of a lumpy, structured universe introduces corrections to the apparent expansion rate that mimic acceleration, without any dark energy or modified gravity required. This class of models is referred to as backreaction or inhomogeneous cosmology.¹²⁾

Proponents argue that the standard approach of fitting a smooth FLRW background to observational data obscures potentially significant effects of structure, and that a proper averaging procedure - Buchert's scalar averaging formalism is the most developed - can in principle reproduce the apparent acceleration without exotic new physics. Critics respond that numerical simulations of cosmic structure formation suggest backreaction effects are far too small to account for the observed acceleration, and that the CMB and BAO evidence for dark energy does not rest solely on supernova observations in a way that homogeneity assumptions could compromise.¹³⁾ The debate remains technically unresolved.

See Dark Energy - Backreaction Viewpoint.

Some researchers have questioned the inferential chain from supernova observations to cosmic acceleration more directly - not by proposing alternative cosmological models but by challenging the quality or interpretation of the underlying data. Challenges have included potential systematic errors in Type Ia supernova standardization, selection effects in high-redshift supernova samples, and questions about whether supernova intrinsic brightness is as uniform as assumed.¹⁴⁾ Some analyses have argued the statistical evidence for acceleration from supernovae alone, without combining multiple datasets, is weaker than typically presented.

Defenders of the standard interpretation note that the acceleration is now supported by multiple independent lines of evidence - CMB, BAO, weak gravitational lensing, and galaxy cluster counts - and that the case does not rest on supernovae alone. They argue that criticisms focused on supernova systematics do not affect these independent confirmations, and that the overall statistical case is robust.

Researchers across these positions generally agree that:

• The original 1998 supernova observations represented a genuine and reproducible finding, whatever their correct interpretation.
• The cosmological constant problem - the discrepancy between theoretically predicted and observed vacuum energy density - is a serious unresolved problem in fundamental physics, regardless of one's view on dark energy.
• The Hubble tension - the persistent discrepancy between the value of H0 inferred from the CMB and early-universe physics and the value measured by local distance-ladder methods - is an unresolved problem that bears on ΛCDM and may have implications for dark energy models, though its connection to dark energy specifically remains debated.
• Future surveys (including the Vera C. Rubin Observatory's LSST, the Euclid satellite, and the Nancy Grace Roman Space Telescope) will provide substantially improved constraints on the dark energy equation of state and on alternative models.
• The ΛCDM model, whatever its conceptual difficulties, is currently the most predictively successful framework for cosmological observations.

• Dark Energy
• General Relativity
• Standard Cosmological Model (ΛCDM)
• Cosmology - Physics Consensus
• Dark Matter - Debate
• Quantum Gravity - Debate
• General Relativity - Debate
• Dark Energy - Cosmological Constant Viewpoint
• Dark Energy - Quintessence Viewpoint
• Modified Gravity - Cosmological Scales Viewpoint
• Dark Energy - Backreaction Viewpoint
• Cosmology - History