In the statistical analysis of spatial point patterns, stationarity is often assumed to mean that the spatial point process has constant intensity and uniform correlation depending only on the lag vector between pairs of points (Møller and Toftaker 2012). In other words, it is assumed that the correlation between the elements of a spatial distribution is a function of the Euclidean distance between them. This framework has been vastly used in Spatial Analysis to describe settlement processes, taking into account a homogenous and undifferentiated surface that is easy to generalise. These assumptions fail when we consider the historical and economical dynamics that took place in space. This failure arises from the fact that the possibility of movement constraints have hardly been taken into account. Social action is supposed to have been performed in a simple and homogeneous Euclidean surface where only straight line distances are considered.