We aim at studying the macroscopic constitutive models for fractured rocks deduced from an analysis of the microscopic-scale mechanisms of sliding along discontinuities. We will use homogenization techniques in the framework of large-deformations. The objective of this research project is threefold. First, starting from the work of Hill on elastoplastic heterogeneous material, we will study the homogenization process taking into account large deformations and surfaces of discontinuity. Special attention will be paid to the rotation part. Secondly, we shall analyze the rigid-plastic mode obtained for multi-fractured media using a multislip approach similar to that used for monocrystal plasticity. Different friction laws will be examined in order to understand their differences at the macroscopic level. Finally, different distributions of fractures will be studied to make precise the type of anisotropy of the resulting elastoplastic constitutive macroscopic law. Periodic and stochastic distributions will be specially emphasized. The mathematical techniques used will be adaptations of those from the analysis of crystal plasticity and polycrystalline materials. As a result of the proposed work, we will gain a better understanding of the origins of large-scale rock plasticity, which is important in the understanding of the brittle behavior of rocks. Application of the expected results to the dynamics of earthquakes as well as to plate tectonics are envisaged.