The universe is accelerating, and the standard explanation is a positive cosmological constant, i.e., a de Sitter vacuum. String theory, as a candidate for quantum gravity, should be able to describe such a vacuum, but after decades of effort, no reliable construction has emerged. Alongside this work, other studies either try to build dark energy through symmetry, look for quantum corrections to black holes, or explore dark bubbles. This paper fits into the broader search for constraints on the cosmological constant, but takes a more radical step: not just difficulty, but a fundamental prohibition on de Sitter in quantum gravity.
The authors formulate the swampland criterion: the gradient of the scalar potential V must satisfy the inequality |∇V| ≥ c·V with a constant c of order one. They then test it on concrete examples from string theory. For M-theory, they consider compactifications of 11-dimensional supergravity on arbitrary manifolds without singularities; the potential depends on the volume modulus, and its slope is estimated through curvature and flux contributions. For the heterotic string O(16)×O(16), they analyze the effective potential in ten and lower dimensions. In type II supergravity, they generalize no-go theorems for compactifications with D-branes, O-planes, and fluxes. They also derive constraints from the strong and null energy conditions.
In all constructions considered, the ratio |∇V|/V turns out to be no less than a number of order one. For M-theory in four dimensions, they obtain a lower bound of 6/√14 ≈ 1.6. The heterotic string gives a value of 5/√2 ≈ 3.5 in ten dimensions. In type II supergravity, for various dimensions of D-branes and O-planes, values range from 0.8 to 3.7, all exceeding √2, which rules out accelerated expansion. Experimental data on dark energy give c ≲ 0.6, only slightly below the theoretical estimates. The authors call this closeness of numbers 'tantalizingly close.'
This work turns scattered no-go theorems into a unified principle: de Sitter vacua fall into the swamp. Unlike neighboring studies that seek specific mechanisms for dark energy or quantum effects in black holes, here a universal criterion is proposed, applicable to any low-energy theory emerging from quantum gravity. This changes the view of the cosmological constant: instead of fine-tuning parameters, we get a natural explanation of its smallness through the slope of the potential. If the criterion holds, dark energy must be quintessence, not a vacuum.
The hypothesis opens several directions. The value of the constant c needs to be pinned down: currently it is bounded above by values of order one, but below only the experimental estimate of 0.6 is known. The criterion must be tested on a wider class of string constructions, including non-geometric compactifications and unstable vacua. A separate question is whether the inequality can be derived from more fundamental principles, such as unitarity or the absence of global symmetries. If confirmed, the criterion would become a powerful tool for selecting realistic models of inflation and dark energy.
The swampland criterion affects cosmology, string phenomenology, and quantum gravity. It constrains models of inflation and dark energy, ruling out overly flat potentials. For string theory, this means that the search for de Sitter vacua in the spirit of KKLT may be futile, and efforts should be redirected toward quintessence or other dynamical scenarios.
The immediate task is to test the criterion against known counterexamples, such as models with broken supersymmetry or nontrivial topology. It is also important to see whether the assumptions about compactification can be relaxed without losing the constraint.
This work is directly related to the cosmological constant problem: why is vacuum energy so small but not zero? It also touches on the question of the string landscape: if de Sitter is forbidden, the landscape shrinks dramatically, and the anthropic principle loses its footing. Finally, the swampland conjecture links cosmology with constraints on effective field theories, which may shed light on the nature of dark energy.
The authors call the coincidence between the experimental bound c ≲ 0.6 and the theoretical estimates c ~ 1 'tantalizingly close.' If the value were slightly larger, accelerated expansion would be impossible within quintessence, and we would have to look for entirely different explanations.
The idea of a ban on stable vacuum echoes Greg Egan's novel 'Schild's Ladder,' where physicists discover that the vacuum is unstable and could decay. The paper deals with a milder but related restriction: de Sitter vacuum cannot be a resting point, but only a slowly rolling state.