In this paper we show that weak Hopf (co)quasigroups can be characterized by a Galois-type condition. Taking into account that this notion generalizes the ones of Hopf (co)quasigroup and weak Hopf algebra, we obtain as a consequence the first fundamental theorem for Hopf (co)quasigroups and a characterization of weak Hopf algebras in terms of bijectivity of a Galois-type morphism (also called fusion morphism).Keywords. Hopf algebra, weak Hopf algebra, Hopf (co)quasigroup, weak Hopf (co)quasigroup, Galois extension.MSC 2010: 18D10, 16T05, 17A30, 20N05.algebraic structures remain valid under this unified approach (in particular, the Fundamental Theorem of Hopf Modules associated to a weak Hopf quasigroup [4]), and it is very natural to ask for other wellknown properties related with Hopf algebras. In particular, Nakajima [10] gave a characterization of ordinary Hopf algebras in terms of bijectivity of right or left Galois maps (also called fusion morphisms in [14]). This result was extended by Schauenburg [13] to weak Hopf algebras, and by Brzeziński [7] to Hopf (co)quasigroups. The main purpose of this work is to give a similar characterization in the weak Hopf (co)quasigroup setting. More precisely, we state that a weak Hopf (co)quasigroup satisfies a right and left Galois-type condition, and these Galois morphisms must have almost right and left (co)linear inverses, and conversely. As a consequence we get the characterization of Hopf (co)quasigroups given by Brzeziński [7] (called the First Fundamental Theorem for Hopf (co)quasigroups), and the one obtained by Schauenburg [13] for weak Hopf algebras.
A characterization of weak Hopf quasigroupsThroughout this paper C denotes a strict monoidal category with tensor product ⊗ and unit object K. For each object M in C, we denote the identity morphism by id M : M → M and, for simplicity of notation, given objects M , N , P in C and a morphism f :From now on we also assume that C admits equalizers and coequalizers. Then every idempotent morphism splits, i.e., for every morphismAlso we assume that C is braided, that is: for all M and N objects in C, there is a natural isomorphism c M,N : M ⊗ N → N ⊗ M , called the braiding, satisfying the Hexagon Axiom (see [8] for generalities). If the braiding satisfies c N,M • c M,N = id M⊗N , the category C will be called symmetric.By a unital magma in C we understand a triple A =, the unital magma will be called a monoid in C. Given two unital magmas (monoids) A = (A, η A , µ A ) and B = (B, η B , µ By duality, a counital comagma in C is a triple D = (D, ε D , δ D ) where D is an object in C and ε D :the counital comagma will be called a comonoid. If D = (D, ε D , δ D ) and E = (E, ε E , δ E ) are counital comagmas (comonoids),Moreover, if D is a comagma and A a magma, given two morphisms f, g : D → A we will denote by f * g its convolution product in C, that is f *We shall denote by C A the category of right A-modules. In an analogous way we can define the category of left A-modules and we denote it by A C.Let D be a comonoid. The...