We report results of an experimental study with Dutch agrammatic aphasics that investigated their ability to interpret pronominal elements in transitive clauses and Exceptional Case Marking constructions (ECM). Using the obtained experimental results as a tool, we distinguish between three competing linguistic theories that aim at determining principles responsible for pronoun distribution: Government and Binding (Chomsky, 1981), Reflexivity (Reinhart & Reuland, 1993), and Primitives of Binding (Reuland, 2001). The obtained results are inconsistent with Government and Binding, but consistent with the latter two theories. We further show that the Primitives of Binding framework most naturally explains our results. Our proposal is based on the different performance demonstrated by patients in transitive clauses and ECM constructions, in combination with the often-made claim in the literature regarding the limitation of processing resources in this population (e.g., Haarmann and Kolk, 1991, Haarmann and Kolk, 1994 and Zurif et al., 1993, among others).