One cannot hear orbifold isotropy type (Q2431763)

One of the most successful generalization of the notion of a smooth manifold is the one of smooth orbifold, which is locally modelled on an orbit space of \(\mathbb R^n\) under the action of a finite group of diffeomorphisms. A Riemannian orbifold is locally modelled on an orbit of a Riemannian manifold under the action of a finite group of isometries. This class of singular spaces has wide applications, in particular in the study of three manifolds and string theory. The main advantage that orbifolds enjoy, with respect to other ``wilder'' singular spaces is that global analysis on them is relatively tracable. For example, on a Riemannian orbifold there is a natural Laplace operator acting on functions. A natural question then arises: what information about the singular set of a Riemannian orbifold, that is a set of nontrivial isotropy, is encoded in its Laplace spectrum? In the paper under review, the authors, motivated by the above question, address and settle in the negative the famous question of \textit{Mark Kac}, ``Can one hear the shape of a drum?'' [Am. Math. Mon. 73, No. 4, Part II, 1--23 (1966; Zbl 0139.05603)], by constructing arbitrarly large families of mutually isospectral orbifolds with different isotropy types and by showing that singular strata of two isospectral orbifolds are not necessary homeomorphic. To prove their results the authors rely on and improve some ideas and techniques introduced by \textit{T. Sunada} [Ann. Math. (2) 121, 169--186 (1985; Zbl 0585.58047)] and \textit{P. Bérard} [Astérisque 177--178, 127--154 (1989; Zbl 0703.53035); Math. Ann. 292, No. 3, 547--560 (1992; Zbl 0735.58008)], in particular by adapting them to the case of orbifolds.