Magnetic Weak Gravity Conjecture

• Magnetic Weak Gravity Conjecture is an ultraviolet cutoff formulation for U(1) gauge theories, stating that weak gauge coupling forces the EFT to break down at Λ ≲ g Mₚ.
• It derives constraints from monopole mass estimates and black hole formation criteria, emphasizing that excessively heavy monopoles imply gravitational collapse.
• The conjecture extends to higher-form fields, axionic frameworks, and string theory, illustrating varied UV completion mechanisms in a quantum gravitational context.

The magnetic Weak-Gravity Conjecture is the ultraviolet-cutoff formulation of the Weak Gravity Conjecture for gauge sectors coupled to gravity. In its standard abelian form, it states that a $U(1)$ effective field theory with gauge coupling $g$ cannot remain valid above a scale $\Lambda \lesssim g M_{\rm Pl}$, up to order-one factors, because otherwise the magnetic monopole implied by the gauge sector would become so massive that it would itself lie inside its Schwarzschild radius. In the modern literature, this statement is typically presented as complementary to the electric WGC, whose canonical formulation requires a superextremal charged state satisfying $\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext}$. The magnetic statement constrains not the existence of a light electric particle, but the validity range of the effective field theory itself (Rudelius, 2024).

1. Standard abelian statement and its place within the WGC

For a $U(1)$ gauge force, the electric WGC is formulated in terms of extremal black holes. In natural units $G=c=4\pi\varepsilon_0=1$, charged black holes are classified as subextremal, extremal, or superextremal according to

$Q<M,\qquad Q=M,\qquad Q>M.$

The electric conjecture requires at least one superextremal or extremal particle,

$$\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext},$$

so that charged black holes can decay. In simple units, $\left.\frac{Q}{M}\right|_{\rm ext}$ can be set to $1$, making the condition roughly $g$0. With massless scalars, the extremality bound can instead depend on moduli through an order-one function $g$1 (Rudelius, 2024).

The magnetic WGC is usually not presented in the same theorem-style form. Its standard content is instead the cutoff estimate

$g$2

again up to order-one uncertainty. The heuristic quantities entering this statement are

$g$3

The claim is that a consistent quantum-gravitational EFT cannot keep a weakly coupled $g$4 sector valid to arbitrarily high energies. Weak gauge coupling and ultraviolet validity are therefore not independent parameters.

This distinction is central to the subject. The electric WGC constrains the charged spectrum. The magnetic WGC constrains the maximal regime of EFT validity. In that sense, the magnetic statement is the ultraviolet-cutoff version of the broader principle that gravity must be the weakest force.

2. Monopoles, black holes, and generalized symmetry breaking

The usual monopole argument proceeds by combining the monopole mass estimate with black-hole collapse. A $g$5 gauge theory admits magnetic monopoles; in a weakly coupled theory their mass scales as $g$6, and their physical size is set by the UV scale through $g$7. If $g$8 is taken too high at fixed small $g$9, the monopole becomes so massive that it lies داخل its Schwarzschild radius. Requiring that the monopole not already be a black hole gives the parametric bound $\Lambda \lesssim g M_{\rm Pl}$0 (Rudelius, 2024).

A structurally different derivation comes from generalized global symmetry breaking. In four-dimensional Maxwell theory without monopoles there is an exact magnetic one-form symmetry $\Lambda \lesssim g M_{\rm Pl}$1, with magnetic current

$\Lambda \lesssim g M_{\rm Pl}$2

The claim advanced in recent work is that approximate generalized symmetries should be badly broken at or below the quantum-gravity scale. In a Higgsed $\Lambda \lesssim g M_{\rm Pl}$3 UV completion,

$\Lambda \lesssim g M_{\rm Pl}$4

The low-energy abelian EFT cutoff is $\Lambda \lesssim g M_{\rm Pl}$5, and imposing bad breaking below $\Lambda \lesssim g M_{\rm Pl}$6 yields

$\Lambda \lesssim g M_{\rm Pl}$7

Equivalently,

$\Lambda \lesssim g M_{\rm Pl}$8

This reframes the magnetic WGC as a consequence of the impossibility of keeping the magnetic one-form symmetry only weakly broken all the way to the Planck scale (Cordova et al., 2022).

This symmetry-based perspective is significant because it isolates what the magnetic statement is really diagnosing: the failure of a low-energy $\Lambda \lesssim g M_{\rm Pl}$9 description to remain autonomous in quantum gravity. The monopole is then not merely a soliton used in a heuristic estimate, but the dynamical agent that breaks the magnetic one-form symmetry.

3. EFT status, Higgsing loopholes, and composite monopoles

A major qualification to the naive cutoff formula comes from Higgsed theories. In a model with two $\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext}$0 gauge fields $\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext}$1 and $\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext}$2, both with coupling $\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext}$3, plus a scalar of charge $\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext}$4 with $\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext}$5, Higgsing leaves a massless combination

$\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext}$6

with effective low-energy charge quantum

$\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext}$7

Naively applying the magnetic WGC to the infrared theory suggests a cutoff of order $\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext}$8. The explicit construction shows that the infrared EFT can badly violate this usual magnetic bound even though the UV theory obeys the WGC (Saraswat, 2016).

The reason is magnetic confinement in the Higgs phase. A monopole of the surviving $\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext}$9 is not a simple pointlike object whose size directly measures the scale of new local degrees of freedom. Instead, it can be built from $U(1)$0 monopoles of one gauge factor and one anti-monopole of the other, connected by Nielsen–Olesen flux tubes. The flux tubes have radius $U(1)$1 and tension

$U(1)$2

The crucial failure in the standard monopole derivation is the assumption that monopole size $U(1)$3 sets the EFT cutoff through $U(1)$4. In the Higgsed example, $U(1)$5 is an emergent scale controlled by long-range forces and flux tubes, whereas the actual new local states can appear at a much higher scale. On this basis, the usual magnetic WGC should not be treated as a universal veto on low-energy EFTs (Saraswat, 2016).

The same paper proposes a much weaker bound motivated by entropy considerations,

$U(1)$6

and argues that Higgsed theories produced from a UV theory satisfying the multifield WGC can saturate but not violate it. This shifts the interpretation of the magnetic WGC away from a strict bottom-up EFT axiom and toward a statement about UV completion in quantum gravity.

A later analysis of the same Saraswat model studied the composite magnetic monopole itself as a dynamical object. In the large-$U(1)$7 regime, the composite monopole has size

$U(1)$8

and supports rotational and oscillatory excitations whose energies become parametrically light in the same limits interpreted as approaching a boundary of moduli space. The proposed conclusion is that the expected tower associated with the magnetic-WGC–distance-conjecture connection can be localized on the monopole rather than propagating as a bulk four-dimensional tower (Pathak et al., 20 Jun 2025).

4. Higher-form, axionic, domain-wall, and topological extensions

For axions, the magnetic WGC takes a different form because the axion is dual in four dimensions to a two-form gauge field, and the magnetically charged object is a string. A conservative formulation is therefore that the minimally charged cosmic string coupled to the dual two-form should exist as a valid effective-field-theory object. A naive continuation of the generic magnetic-WGC cutoff formula, together with $U(1)$9, suggests that super-Planckian axion decay constants are problematic. In the static Cohen–Kaplan geometry, the exterior of an axionic string becomes difficult to match to a sensible UV core when $G=c=4\pi\varepsilon_0=1$0 is too large, while Gregory’s time-dependent solutions and topological inflation offer a more conservative loophole. Multi-axion constructions with

$G=c=4\pi\varepsilon_0=1$1

produce composite strings built from several fundamental strings and domain walls, but their tension is generically of order $G=c=4\pi\varepsilon_0=1$2, so they do not provide a clean evasion of the magnetic tension problem (Hebecker et al., 2017).

For domain walls, the subject is usually phrased in higher-form language. In four dimensions, a $G=c=4\pi\varepsilon_0=1$3-form gauge potential has electrically charged membranes or domain walls. The magnetic side is obtained by analytic continuation of the higher-form magnetic-WGC scaling and yields

$G=c=4\pi\varepsilon_0=1$4

In axion monodromy with

$G=c=4\pi\varepsilon_0=1$5

the effective coupling is identified as

$G=c=4\pi\varepsilon_0=1$6

so the cutoff becomes

$G=c=4\pi\varepsilon_0=1$7

up to order-one factors. In this setting the electric side can be weak, because the domain walls produced by the cosine wiggles can be made arbitrarily light by lowering $G=c=4\pi\varepsilon_0=1$8; the magnetic side remains nontrivial because it constrains the EFT cutoff independently of the wiggle amplitude (Hebecker et al., 2015).

Dimensional reduction leads to a further generalization. In a five-dimensional Einstein–Maxwell theory compactified on $G=c=4\pi\varepsilon_0=1$9, the four-dimensional magnetic graviphoton charge $Q<M,\qquad Q=M,\qquad Q>M.$0 uplifts to a NUT-like topological charge in five dimensions. The resulting four-dimensional dyonic black holes impose stronger inequalities than the purely five-dimensional electric and magnetic black objects, including

$Q<M,\qquad Q=M,\qquad Q>M.$1

and

$Q<M,\qquad Q=M,\qquad Q>M.$2

The proposed implication is that the logic of the WGC should apply not only to ordinary Maxwell charges but also to topological charges associated with the global structure of spacetime (Cremonini et al., 2020).

5. String-theoretic, lattice, and species-scale realizations

String theory furnishes realizations of magnetic-WGC behavior that are naturally phrased in terms of brane forces, brane tensions, and charge lattices rather than an explicit field-theoretic cutoff. In type I with Scherk–Schwarz supersymmetry breaking, the wrapped D1-brane self-energy gives a negative one-loop correction to the tension,

$Q<M,\qquad Q=M,\qquad Q>M.$3

The RR charge is not comparably reduced at this order, so the large-distance force becomes repulsive when

$Q<M,\qquad Q=M,\qquad Q>M.$4

In magnetized type II systems, or equivalently D6-branes at angles, SUSY can be broken while the open-string spectrum remains tachyon-free if the magnetic parameters $Q<M,\qquad Q=M,\qquad Q>M.$5 satisfy the triangle inequalities

$Q<M,\qquad Q=M,\qquad Q>M.$6

In precisely this tachyon-free regime, the long-distance force is repulsive. The identified pattern is

$Q<M,\qquad Q=M,\qquad Q>M.$7

with the underlying mechanism being flux-induced charges and loop-induced tension renormalization (Bonnefoy et al., 2020).

In nonabelian settings, the magnetic story is refined into a lattice statement. Reece and Rudelius verified in heterotic toroidal orbifold compactifications that violations of the Lattice Weak Gravity Conjecture are correlated with fractionally charged confined monopoles. In the examples studied, there exists a discrete subgroup

$Q<M,\qquad Q=M,\qquad Q>M.$8

such that superextremal particles exist at every site in the charge lattice of

$Q<M,\qquad Q=M,\qquad Q>M.$9

while confined monopoles exist at all sites in the corresponding magnetic charge lattice. The conjectural relation is

$$\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext},$$0

This suggests that LWGC violation is controlled by the global form of the gauge group and, in particular, by the center $$\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext},$$1 (Reece et al., 4 Mar 2026).

A related species-scale perspective appears in four-dimensional F-theory weak-coupling limits. There the magnetic WGC scale is identified as

$$\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext},$$2

and its physical meaning depends on its comparison with the KK species scale. Geometrically, weak-coupling limits fall into three classes labeled by $$\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext},$$3. Only the $$\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext},$$4 emergent heterotic-string limit yields a four-dimensional tower of non-BPS EFT-string excitations satisfying the asymptotic Tower WGC. The $$\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext},$$5 and $$\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext},$$6 cases instead describe decompactification to higher dimensions or a defect gauge theory decoupling from the gravitational bulk, rather than a conventional four-dimensional magnetic-WGC tower (2208.00009).

6. Applications, refined variants, and unresolved scope

One prominent application is to de Sitter critical points in $$\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext},$$7 gauged supergravity with vector multiplets. The magnetic input is the cutoff

$$\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext},$$8

In the models studied, the same gauging that generates the scalar potential also charges the gravitini, and the Hubble scale obeys

$$\frac{q}{m} \ge \left.\frac{Q}{M}\right|_{\rm ext},$$9

A controlled de Sitter EFT would require $\left.\frac{Q}{M}\right|_{\rm ext}$0, hence $\left.\frac{Q}{M}\right|_{\rm ext}$1, but charge quantization obstructs arbitrarily small gravitino charge. The proposed consequence is a Dine–Seiberg-type obstruction, with a degenerate gravitino mass matrix,

$\left.\frac{Q}{M}\right|_{\rm ext}$2

serving as a swampland criterion for the class of de Sitter points considered (Cribiori et al., 2020).

A different line of development studies magnetic and dyonic black holes directly. In four-dimensional Einstein–Maxwell–dilaton theory with four-derivative corrections, the magnetic WGC is implemented by requiring the corrected extremal black hole to become superextremal, $\left.\frac{Q}{M}\right|_{\rm ext}$3. For pure magnetic charge, the relevant condition is

$\left.\frac{Q}{M}\right|_{\rm ext}$4

where

$\left.\frac{Q}{M}\right|_{\rm ext}$5

In the dyonic $\left.\frac{Q}{M}\right|_{\rm ext}$6 and $\left.\frac{Q}{M}\right|_{\rm ext}$7 families the inequalities become $\left.\frac{Q}{M}\right|_{\rm ext}$8-dependent linear combinations of Wilson coefficients. The interpretation is that ultraviolet consistency constrains higher-derivative couplings so that extremal magnetic or dyonic black holes can decay (Loges et al., 2019).

At the same time, several neighboring literatures do not establish the magnetic WGC itself. Work on force-based versus charge-to-mass versions of the WGC argues for tower and sublattice strengthenings, and proposes the Maximal $\left.\frac{Q}{M}\right|_{\rm ext}$9 Conjecture as a unifying criterion, but it does not derive the standard magnetic cutoff formula from monopole physics (Heidenreich et al., 2019). Holographic arguments based on near-horizon thermalization similarly reformulate the electric WGC for electrically charged probes in Reissner–Nordström backgrounds without producing a magnetic inequality (Urbano, 2018). Likewise, asymptotic-safety analyses modify the electric extremality condition of charged black holes through the running of $1$0 and $1$1, but leave the magnetic WGC essentially open because they do not derive a cutoff bound of the form $1$2 (Ghosh, 27 Apr 2025).

The present status is therefore differentiated rather than uniform. In its standard abelian form, the magnetic WGC remains the statement that weak gauge coupling forces a low EFT cutoff through monopole and black-hole consistency. That statement is structurally supported by one-form symmetry breaking, by higher-form analogues, by string-theoretic force calculations, and by several compactification arguments. But its interpretation as a universal bottom-up EFT constraint is challenged by Higgsed examples, by composite magnetic objects, and by cases in which the relevant UV completion is organized around confinement, species scales, or extended defects rather than a pointlike monopole. This suggests that the magnetic WGC is best regarded not as a single universally theoremized inequality, but as a family of tightly related ultraviolet-consistency statements whose precise implementation depends on the sector—abelian, higher-form, axionic, topological, or lattice-valued—in which magnetic charge is realized.