We propose new methods to detect paradigmatic fields through simple statistics over a scientific content database. We propose an asymmetric paradigmatic proximity metric between terms which provide insight into hierarchical structure of scientific activity and test our methods on a case study with a database made of several millions of resources. We also propose overlapping categorization to describe paradigmatic fields as sets of terms that may have several different usages. Terms can also be dynamically clustered providing a high-level description of the evolution of the paradigmatic fields.