The issue of subject pronominal expression (SPE) in language contact situations, as illustrated in Nosotros somos muy buenos estudiantes [‘We are very good students’] vs. Ø somos muy buenos estudiantes [‘(we) are very good students’], has been the focus of decades of research in both variationist and generative studies (Bentivoglio, 1987; Cameron, 1992, 1993, 1995, 1996; Chomsky, 1981, 1986; Huang, 1984; Orozco & Guy, 2008; Travis, 2005a, 2005b; Otheguy & Zentella, 2007, 2012; Rizzi, 1982; Morales, 1980; Silva-Corvalán, 1982, 1994, 1997; inter alios). While generative studies have shown that null subject languages (NSLs) can either be licensed by rich verbal paradigms or by discursive mechanisms, variationist studies have shown that there are predictors that condition SPE in NSLs. Furthermore, they have argued that the rates of SPE reflect uniformity across different varieties of Spanish (Orozco, 2015). In spite of this significant body of evidence in monolingual varieties of Spanish, studies involving bilingual groups and indigenous languages are relatively sparse, as most studies have studied Spanish in contact with other Indo-European languages. This dissertation investigates the SPE in the Spanish of a group of highly proficient bilingual speakers of Uwa and Spanish in a language contact situation. The research reported in this dissertation also studies the nature of cross-linguistic influence between the two languages of the bilinguals. Specifically, the idiosyncratic morphosyntactic traces in SPE resulting from the contact between Uwa and Spanish and whether those traces evidence patterns of variation, and, if so, how they can be accounted for. Spanish and Uwa are both NSLs and thus both allow for referential null subject pronouns (SPs). However, each language has specific syntactic and discursive predictors responsible for null subjects. For instance, Uwa relies heavily on discursive clues whereas Spanish is a sentence-oriented language. This means that while in Uwa the subject reference is understood from context and discursive clues governed by chains of topics, the rich verbal paradigm of Spanish is responsible for licensing null subjects. The fact that Spanish and Uwa are both NLs but still have dissimilar typological status provides a rich testing ground for the mechanisms involved in SPE as well as for investigating the effects of bilingualism. This dissertation aims at enhancing our linguistic knowledge in relation to the principles and mechanisms involved in SPE in bilingual communities and at providing a better understanding of the nature of cross-linguistic influence in highly proficient bilinguals.