This article is based on a large corpus of data collected as part of the European Science Foundation Project on Second Language Acquisition by Adult Immigrants (Perdue, 1984; 1993). In this study, some empirical observations are discussed from a follow-up on the ESF project. This follow-up study dealt with the conceptual domain of people (Broeder, 1991; 1992). Mostly, the encoding of reference to people is done through nouns and pronouns. Nominal reference involves a set of proper names and more or less complex noun phrases (e.g. Ahmet, my brother, my neighbour's sister). Pronominal reference is based on an exhaustive list of frequently used and predominantly monosyllabic words (e.g., I, you, he, etc.). Learners necessarily have to use a restricted set of nouns and pronouns as efficiently as possible in daily interactions with native speakers. This study examines the acquistion of pronouns. The relevant questions are: how do people start our encoding pronominal reference, how does their repertoire of pronouns develop and why do they make the choices they make?