In differential geometry, holonomy describes the transformation induced on the fibers of a vector bundle or the tangent space of a manifold by parallel transport along closed loops, capturing the global geometric structure through local connection properties.^[1][2] For a connection $\nabla$ on a vector bundle $E$ over a manifold $M$, the holonomy group $\mathrm{Hol}_p(\nabla)$ at a point $p \in M$ is the Lie subgroup of $\mathrm{GL}(E_p)$ generated by the parallel transport maps $P^\nabla_\gamma: E_p \to E_p$ along all piecewise smooth loops $\gamma$ based at $p$, with the restricted holonomy $\mathrm{Hol}^0_p(\nabla)$ considering only contractible loops.^[2][3] The holonomy representation refers to the action of the holonomy group on the fiber $E_p$ (or on the tangent space $T_p M$ in the case of the Levi-Civita connection), considered up to isomorphism class, as changing the base point yields conjugate groups. The holonomy principle states that questions about covariantly constant (parallel) objects, such as vector fields or differential forms, reduce to algebraic questions about invariants under this representation.^[4] The concept originated in the early 20th century, with Élie Cartan developing it in the context of Levi-Civita connections on Riemannian manifolds to study spaces of constant curvature and generalized spaces, building on earlier ideas from classical mechanics like Heinrich Hertz's distinction between holonomic and non-holonomic constraints in 1895.^[1] Holonomy is intrinsically linked to curvature: the Ambrose–Singer theorem states that the Lie algebra of the holonomy group is generated by the curvature endomorphisms $\Omega(X,Y)$ evaluated on vector fields $X, Y$, where $\Omega(X,Y) = d\omega(X,Y) + [\omega(X), \omega(Y)]$ and $\omega$ is the connection form, implying that flat connections (vanishing curvature) yield trivial restricted holonomy group $\mathrm{Hol}^0_p = \{\mathrm{Id}\}$, while the full holonomy group $\mathrm{Hol}_p$ may be nontrivial. However, the full holonomy group may be nontrivial in the presence of non-contractible loops. A classic example is the non-trivial real line bundle over the circle (Möbius bundle), which admits a flat connection (locally trivial with zero connection form) but has holonomy -1 around the generating loop of S^1, as parallel transport flips the sign of the fiber elements.^[5][3][6] A manifold's holonomy group determines key geometric features, such as the existence of parallel vector fields or differential forms; for instance, irreducible holonomy implies no non-trivial parallel subbundles, while decomposable holonomy allows splitting the manifold into factors with irreducible holonomy via de Rham's theorem.^[1] Special holonomy groups—subgroups of the full orthogonal group $\mathrm{SO}(n)$ preserving additional structures—classify Ricci-flat manifolds of interest in physics and geometry, including Kähler manifolds with holonomy $\mathrm{U}(m)$, Calabi–Yau manifolds with $\mathrm{SU}(m)$, hyperkähler manifolds with $\mathrm{Sp}(m)$, and exceptional cases like $G_2$ for 7-dimensional manifolds or $\mathrm{Spin}(7)$ for 8-dimensional ones, as classified by Berger in 1955.^[2] These groups not only encode integrability conditions for metrics and connections but also underpin applications in string theory and supersymmetry, where reduced holonomy ensures the existence of covariantly constant spinors.^[1]

Fundamental Definitions

Holonomy in Vector Bundles

In a vector bundle $E \to M$ over a smooth manifold $M$, equipped with a linear connection $\nabla$, parallel transport along a piecewise smooth curve $\gamma: [0,1] \to M$ with $\gamma(0) = \gamma(1) = p$ defines an automorphism $\mathrm{Hol}_\gamma: E_p \to E_p$ of the fiber over the base point $p$, obtained as the linear isomorphism induced by lifting $\gamma$ to a parallel section of $E$ along the curve.^[7] This holonomy map $\mathrm{Hol}_\gamma$ measures the failure of parallel transport to be path-independent, arising from the curvature of $\nabla$.^[8] The explicit construction of holonomy proceeds via the associated frame bundle or local trivializations. In a local trivialization of $E$ over an open set $U \subset M$, the connection $\nabla$ is represented by a $\mathfrak{gl}(r,\mathbb{R})$-valued 1-form $A$ (the connection form), and the parallel transport $\tau_\gamma$ along $\gamma$ is given by the path-ordered exponential

$$\tau_\gamma(v) = \mathcal{P} \exp\left( -\int_\gamma A \right) v,$$

where $\mathcal{P}$ denotes the ordering along the path, ensuring non-commuting matrix exponentials are handled correctly; for a closed loop, $\mathrm{Hol}_\gamma = \tau_\gamma$.^[8] For flat connections (curvature zero), this simplifies without higher-order terms, and holonomy can also be constructed via horizontal lifts in the principal frame bundle associated to $E$, where loops in $M$ lift to paths in the total space preserving the fiber structure.^[7] A particularly simple and instructive case arises for abelian connections on complex line bundles (rank-1 complex vector bundles). Let $L$ be a complex line bundle over $M$ equipped with a connection $A$. Suppose that $\gamma = \partial D$ is an oriented closed curve that is the boundary of an oriented surface $D$ (i.e., $\gamma$ is a homological boundary). Then the holonomy (parallel transport) around $\gamma$ is given by

$$\mathrm{Hol}_\gamma = e^{-\int_D F_A} \cdot \mathrm{Id},$$

where $F_A = dA$ is the curvature 2-form of the connection $A$ and $e^z$ denotes the complex exponential map. Proof. By subdividing if necessary, we may assume without loss of generality that the image of $\gamma$ is contained in a coordinate chart over which $L$ is trivial. Write $A = A_i \, dx^i$ for the connection form in this trivialization. Given $s_0 \in L_{\gamma(0)}$, we solve the parallel transport ODE for a section of the form

$$s = e^\varphi s_0, \quad \varphi(0) = 0,$$

where $\varphi$ is a complex-valued function along $\gamma$. The ODE reads

$$\frac{Ds}{dt} = 0 = \frac{ds}{dt} + \frac{d \gamma^i}{dt} A_i \cdot s = \left( \frac{d\varphi}{dt} + \frac{d \gamma^i}{dt} A_i \right) s,$$

which implies

$$\frac{d \varphi}{dt} = - \frac{d \gamma^i}{dt} A_i.$$

Integrating gives

$$\varphi(1) = \varphi(1) - \varphi(0) = -\int_\gamma A.$$

Applying Stokes's theorem yields

$$\varphi(1) = -\int_D dA = -\int_D F_A.$$

Therefore,

$$s(\gamma(1)) = e^{\varphi(1)} s_0 = e^{-\int_D F_A} s_0,$$

so the parallel transport is multiplication by $e^{-\int_D F_A}$, which is the desired formula. This result shows that, in the abelian setting of complex line bundles, the holonomy around any loop bounding a surface is determined exactly by the integrated curvature (flux) over that surface. In particular, if the connection is flat ($F_A = 0$), then $\mathrm{Hol}_\gamma = \mathrm{Id}$ for any such bounding loop. A representative example occurs for the trivial bundle $E = M \times \mathbb{R}^n \to M$ equipped with the Euclidean (flat) connection $\nabla = d$, where sections are identified with $\mathbb{R}^n$-valued functions and parallel transport reduces to the constant map, yielding $\mathrm{Hol}_\gamma = \mathrm{Id}$ for any loop $\gamma$; this identity lies in the orthogonal group $O(n)$ with respect to the standard inner product on $\mathbb{R}^n$.^[9] In contrast, flat connections on nontrivial bundles can have nontrivial full holonomy despite trivial restricted holonomy. For instance, consider the Möbius line bundle over $S^1$, with transition function -1 on half the overlap; a flat connection (zero local forms) yields parallel transport around the circle as multiplication by -1, resulting in holonomy group {1, -1}.^[6] This occurs because the generator loop of $\pi_1(S^1) \cong \mathbb{Z}$ is not homologous to zero, so the above theorem does not apply; the holonomy remains nontrivial even though the curvature vanishes. Key properties include that $\mathrm{Hol}_\gamma$ depends only on the homotopy class of $\gamma$ relative to its endpoints, with homotopic loops inducing the same automorphism; if $M$ is simply connected, all closed loops are homotopic to a point, so holonomy is determined by the restricted holonomy group from contractible loops.^[7] Moreover, if parallel transport along every closed loop is the identity (as for flat connections on simply connected bases), the holonomy group is trivial.^[7]

Holonomy in Principal Bundles

In principal bundles, holonomy generalizes the concept from vector bundles by incorporating the action of a Lie group $G$, providing the natural framework for connections in gauge theories. Consider a principal $G$-bundle $\pi: P \to M$ over a smooth manifold $M$, equipped with a connection $\omega$, which is a Lie algebra-valued $\mathfrak{g}$-valued 1-form on $P$ satisfying the equivariance condition $R_g^* \omega = \mathrm{Ad}(g^{-1}) \omega$ for $g \in G$ and the normalization $\omega(\xi_P) = \xi$ for fundamental vector fields $\xi_P$ generated by $\xi \in \mathfrak{g}$.^[10] For a piecewise smooth curve $\gamma: [0,1] \to M$ with $\gamma(0) = \gamma(1) = p \in M$, the holonomy $\mathrm{Hol}_\gamma \in G$ at a point $u_0 \in P_p = \pi^{-1}(p)$ is defined as the unique group element such that the horizontal lift $\hat{\gamma}: [0,1] \to P$ of $\gamma$, starting at $\hat{\gamma}(0) = u_0$ and satisfying $\pi \circ \hat{\gamma} = \gamma$ with $\omega(\hat{\gamma}'(t)) = 0$ for all $t$, ends at $\hat{\gamma}(1) = u_0 \cdot \mathrm{Hol}_\gamma$, where $\cdot$ denotes the right $G$-action on $P$.^[10] This parallel transport along $\hat{\gamma}$ measures the failure of the connection to be integrable, capturing the geometric obstruction to global trivialization.^[11] The connection form $\omega$ plays a central role in determining horizontal subspaces, defined as the kernel of $\omega$ at each point in $P$, which are complementary to the vertical subspaces tangent to the $G$-orbits. Along the horizontal lift $\hat{\gamma}$, the condition $\omega(\hat{\gamma}'(t)) = 0$ ensures that the transport is purely horizontal, avoiding vertical (infinitesimal gauge) directions. The holonomy element $\mathrm{Hol}_\gamma$ arises as the solution to the parallel transport differential equation: if $U(t) \in G$ represents the time-dependent group element such that $u(t) = u_0 \cdot U(t)$ along $\hat{\gamma}$, then $U$ satisfies the ODE $\frac{dU}{dt} = -U(t) \cdot \omega(\hat{\gamma}'(t))$ with initial condition $U(0) = e$, the identity.^[10] This equation integrates the connection along the path, yielding $\mathrm{Hol}_\gamma = U(1)$. For abelian structure groups, the solution simplifies without ordering issues, but in general, it requires careful path dependence.^[11] The explicit form of the holonomy is given by the path-ordered exponential $\mathrm{Hol}_\gamma = \mathcal{P} \exp\left( -\int_\gamma \omega \right)$, where $\mathcal{P}$ denotes the ordering along $\gamma$ to account for non-commutativity in non-abelian Lie algebras $\mathfrak{g}$. This formula encapsulates the cumulative effect of the connection over the loop, with the negative sign arising from the right-action convention.^[10] In the limit of small loops, it relates to the curvature 2-form $d\omega + \frac{1}{2}[\omega, \omega]$, though the full holonomy encodes global path information. Key properties of holonomy include its multiplicative nature under loop concatenation: $\mathrm{Hol}_{\gamma_1 \cdot \gamma_2} = \mathrm{Hol}_{\gamma_2} \cdot \mathrm{Hol}_{\gamma_1}$, making it a representation of the fundamental groupoid. The holonomy group at $p \in M$ is the subgroup $H_p = \{ \mathrm{Hol}_\gamma \mid \gamma \text{ loop based at } p \} \subseteq G$, a closed Lie subgroup generated by all such elements. The restricted holonomy subgroup consists of those arising from contractible loops, often a connected normal subgroup of $H_p$.^[10] These groups determine the local symmetry preserved by the connection and facilitate structure group reductions. Holonomy in vector bundles arises naturally as the associated bundle construction from principal $G$-bundles, where the representation on the fiber induces linear holonomy maps.^[11] A prominent application occurs in Yang-Mills theory on principal bundles with compact structure groups like $\mathrm{SU}(2)$, where the holonomy of connections satisfying the Yang-Mills equations (self-dual instantons) classifies solutions near singularities via limit holonomy conditions. Specifically, for singular Sobolev connections on 4-manifolds, the asymptotic holonomy around codimension-two singular sets determines removability of singularities and the integer invariants labeling instanton moduli, linking to topological invariants like the second Chern class.

Holonomy Groups and Bundles

In differential geometry, the holonomy group of a connection $\nabla$ on a principal $G$-bundle $P \to M$ is defined pointwise: for a point $p \in M$, the holonomy group $H_p$ is the subgroup of $G$ generated by the parallel transport maps along all piecewise smooth loops based at $p$.^[7] The restricted holonomy group $\mathrm{Hol}^0_p$ (sometimes denoted $\mathrm{Hol}^\circ_p$) is the subgroup generated by parallel transports along contractible loops based at $p$.^[6] The full holonomy group $\mathrm{Hol}(M, \nabla)$ is then the union $\bigcup_{p \in M} H_p \subseteq G$, which forms a Lie subgroup of $G$ closed under conjugation and acts on the fibers of the bundle.^[7] The holonomy bundle associated to a point $u_0 \in P$ is the subbundle $P(u_0) \subseteq P$ generated by the orbits under parallel transport from $u_0$, equivalently viewed as the pullback of $P$ over the loop space of $M$ via the holonomy map.^[7] This bundle inherits the connection $\nabla$ restricted from $P$, with structure group reduced to the holonomy group $H = \mathrm{Hol}(u_0)$ at $u_0$.^[7] A key reduction theorem states that if $H \subseteq G$ is a Lie subgroup closed under conjugation by elements of $G$, then the original bundle $P$ admits a reduction to a principal $H$-bundle preserving the connection and its curvature, determining the integrability of the horizontal distribution defined by $\nabla$.^[12] This reduction captures how the holonomy encodes the global twisting of the bundle that prevents trivialization. For flat connections, where the curvature vanishes identically, the restricted holonomy group is trivial ($\mathrm{Hol}^0_p = \{e\}$) while the full holonomy group $H_p$ is a discrete subgroup of $G$ that can be nontrivial, reflecting the representation of the fundamental group $\pi_1(M,p)$ on the fiber. A standard example is the flat connection on the real line bundle over $S^1$ known as the Möbius bundle, with transition function changing sign over half the circle (e.g., $\sigma_{01} = 1$ for $0 < \theta < \pi$ and $-1$ for $\pi < \theta < 2\pi$) and locally trivial connection forms $A_0 = A_1 = 0$; parallel transport preserves the absolute value but yields holonomy $-1$ around the non-contractible loop generating $\pi_1(S^1)$.^[6] The parallel transport depends only on the homotopy class of loops, leading to constructions of covering spaces over $M$ whose deck transformations correspond to the holonomy representation.^[7] In such cases, the holonomy bundle often simplifies to a product structure, facilitating explicit geometric realizations like those in representation theory.^[13] Properties of the holonomy group include a dimension for its Lie algebra that equals the rank of the curvature tensor evaluated over the holonomy bundle, linking algebraic size directly to geometric obstruction.^[7] Additionally, when the holonomy group is amenable—such as finite extensions of solvable groups—it implies solvability conditions on the bundle's topology, aiding in cohomology computations for flat bundles.^[13] This algebraic analogue parallels monodromy in covering space theory, where representations of the fundamental group encode similar branching phenomena.^[7] The action of the holonomy group $H_p$ on the fiber over $p$ defines the holonomy representation, a representation of $H_p$ on the fiber of the bundle. For vector bundles, this is the direct action on the vector fiber; for principal bundles, it is induced through the representation defining any associated vector bundle. The holonomy representation is considered up to isomorphism, as different choices of base point or frame yield conjugate representations, preserving the isomorphism class. The holonomy principle states that a section of the bundle (or associated tensor field) is covariantly constant (parallel) if and only if its value at any point is invariant under the action of the holonomy group via this representation. This principle transforms geometric questions concerning covariantly constant objects into algebraic questions about invariants under the holonomy representation.^[4]

Related Concepts

Monodromy

In complex analysis, monodromy refers to the transformation induced on the values of a multi-valued holomorphic function when it is analytically continued around closed loops on a Riemann surface. Specifically, for a multi-valued function $f$ defined on a Riemann surface $S$, the monodromy associated with a loop $\gamma$ in the base space is the permutation or linear map on the fiber over a point that results from following the analytic continuation of $f$ along $\gamma$.^[14] The monodromy group arises as the image of the homomorphism from the fundamental group $\pi_1$ of the base space to the automorphism group $\mathrm{Aut}(F)$ of the fiber $F$, capturing the global topological structure of the continuations. In Picard–Lefschetz theory, this group acts on the homology of the fibers, where the monodromy around a critical value is described by a Dehn twist along the vanishing cycle, providing a precise description of how cycles transform under variation of the function parameter.^[15] A classic example is the complex logarithm function $\log z$ on the punctured complex plane $\mathbb{C}^*$, where analytic continuation around a loop encircling the origin once adds $2\pi i$ to the value, generating a monodromy group isomorphic to $\mathbb{Z}$. Monodromy exhibits distinct properties depending on the nature of singularities in the defining differential equation. At regular singular points, the monodromy is Fuchsian, meaning it can be represented by a quasi-unipotent matrix, reflecting the polynomial growth of solutions near the singularity. In contrast, at irregular singularities, the monodromy is wild, involving more complex exponential growth and non-unipotent transformations that cannot be diagonalized over the algebraic closure.^[16][17] The monodromy theorem in complex analysis guarantees that analytic continuation along homotopic paths yields the same result, ensuring the local triviality of the associated covering spaces over simply connected domains. This topological relation underscores monodromy as the discrete analogue to holonomy groups in smooth geometry.

Local and Infinitesimal Holonomy

In the context of a connection on a vector bundle or principal bundle over a manifold, local holonomy refers to the transformations induced by parallel transport along loops that are contractible within small neighborhoods of a base point $p \in M$. For such loops, the holonomy map $ \mathrm{Hol}\gamma: E_p \to E_p $ (or the corresponding group element in the structure group) can be approximated using the curvature of the connection, as the infinitesimal behavior is governed by the local geometry. Specifically, for a small contractible loop $\gamma$ bounding a surface $S$, the holonomy is given approximately by $ \mathrm{Hol}\gamma \approx \exp\left( \int_S R \right) $, where $R$ denotes the curvature 2-form, reflecting how curvature accumulates over the enclosed area.^[18] The infinitesimal holonomy algebra $\mathfrak{h}_p$ at a point $p$ is the Lie subalgebra of the structure Lie algebra generated by the values of the curvature operator $R(X,Y)$ for all tangent vectors $X, Y \in T_p M$. This algebra captures the first-order deformations of parallel transport near $p$, with $\mathfrak{h}_p$ consisting of endomorphisms that span the image of the curvature tensor acting on the fiber. In flat connections, where $R = 0$, the infinitesimal holonomy algebra vanishes, $\mathfrak{h}_p = \{0\}$, implying that the bundle is locally trivializable and parallel transport is path-independent in a neighborhood of $p$.^[7][19] A more precise expansion for the holonomy along a small loop $\gamma$ arises from the path-ordered exponential of the connection form $\omega$, yielding $ \mathrm{Hol}_\gamma \approx \exp\left( \int_S R \right) $, where the curvature integral provides the leading non-trivial contribution for contractible paths via Stokes' theorem applied to a spanning surface $S$. This highlights the role of the connection $\omega$ and the curvature $R$ in the approximation.^[18][7] In the special case of abelian connections, such as those on complex line bundles where the structure group is commutative, path-ordering is trivial, and the holonomy around any closed curve bounding an oriented surface is exactly the exponential of minus the integrated curvature over that surface. Theorem. Let $L$ be a complex line bundle with a connection $A$. Suppose that $\gamma = \partial D$ is an oriented closed curve that is the boundary of an oriented surface $D$. Then

$$P^{A,\gamma} = e^{-\int_D F_A} \cdot \mathbbm{1}.$$

Here $e^z$ is the complex exponential map. Proof. By subdividing if necessary, assume without loss of generality that the image of $\gamma$ is contained in a coordinate chart over which $L$ is trivial. Write $A = A_i \, dx^i$ for the connection form. Given $s_0 \in E_{\gamma(0)}$, solve the parallel transport ODE for a section of the form

$$s = e^{\varphi} s_0, \quad \varphi(0) = 0.$$

The ODE reads

$$\frac{Ds}{dt} = 0 = \frac{ds}{dt} + \frac{d \gamma^i}{dt} A_i \cdot s = \left( \frac{d\varphi}{dt} + \frac{d \gamma^i}{dt} A_i \right) s,$$

so

$$\frac{d \varphi}{dt} = - \frac{d \gamma^i}{dt} A_i.$$

Integrating yields

$$\varphi(1) = -\int_\gamma A.$$

By Stokes' theorem,

$$\int_\gamma A = \int_D dA = \int_D F_A,$$

since $F_A = dA$ in the abelian case. Thus,

$$\varphi(1) = -\int_D F_A,$$

and the parallel transport operator is multiplication by $e^{\varphi(1)} = e^{-\int_D F_A}$, giving the desired result. The algebra $\mathfrak{h}_p$ spans the space of curvature operators at $p$, meaning every element arises from combinations of $R(X,Y)$, and under assumptions of manifold completeness, the Lie algebra of the full holonomy group coincides with $\mathfrak{h}_p$, as established by global extensions like the Ambrose–Singer theorem.^[19][7]

Core Theorems

Ambrose–Singer Theorem

The Ambrose–Singer theorem, established by Warren Ambrose and Isadore M. Singer in their 1953 paper, characterizes the restricted holonomy group of a linear connection in terms of the curvature form, resolving key questions about the representation of the holonomy algebra for general connections.^[20] This result extends earlier work by Élie Cartan on spaces of constant curvature and provides a foundational link between global holonomy and local curvature invariants.^[20] For a smooth manifold $M$ equipped with an affine connection $\nabla$, let $\mathrm{Hol}^0(M, \nabla)$ denote the restricted holonomy group at a base point $p \in M$, acting on the