One of the primary objectives of this work is to better understand the frictional behavior of joints under shear loads, including the creation of damage zones. Discontinuities have an important influence on the deformational behavior of rock systems. The choice of a general criterion to determine the shear strength of rough rock joints is a general problem that has been investigated for many years. Numerous shear models have been proposed in the last decades to relate shear-strength to measurable joint parameters, but their limitations have to be recognized. The problem is how to measure and then to express the roughness with a number (e.g. JRC) or a mathematical expression in order to introduce the morphology of the joint into a shear strength criterion. In the frame of this work it has been pointed out that the geometry of roughness influences the size and distribution of contact areas during shearing. In order to locate and estimate the contact area during the shearing, it was argued that only the zones of the surface faced to the shear direction, and steeper than a threshold inclination are involved in the shearing. An empirical relation between the potential contact area and the minimal apparent dip inclination of the surface is proposed. The close agreement between this empirical description of the potential contact area, and experimental points permits to predict the real contact area involved in the phenomena. A new constitutive law, relating stress and displacements, is proposed to model the shear resistance of joints under constant normal load conditions. It is based on the empirical surface description, and on the results from more than fifty constant-normal-load direct-shear tests performed on both replicas of tensile joints, and induced tensile fractures for seven rock types. It is shown that this constitutive model is able to describe experimental shear tests realized in laboratory. Moreover, the parameters required in the model can be easily obtained through standard laboratory tests. The proposed model was also used to estimate the JRC value. The expression obtained to evaluate the joint roughness coefficient is capable of predicting the JRC. It was successfully compared with JRC values obtained by back analysis of shear tests. In the current research no attention was paid to investigate the influence of the scale on the shearing. The results have been validated only in the range of the samples tested in laboratory. Further studies are needed to explore the applicability of the proposed model in field conditions.