Language, identity, and investment in the English language of a group of Mexican women living in Southeastern Pennsylvania

Open Access
Ross, Brenda M.
Graduate Program:
Applied Linguistics
Doctor of Philosophy
Document Type:
Date of Defense:
December 10, 2010
Committee Members:
  • Dr Susan Strauss And Sr Suresh Canagarajah, Dissertation Advisor
  • Susan G Strauss, Committee Chair
  • Athelstan Canagarajah, Committee Chair
  • Karen E Johnson, Committee Member
  • Esther Susana Prins, Committee Member
  • ethnic identity
  • discourse analysis
  • investment
  • identity
  • language
  • gender
  • agency
This dissertation has two goals. First, it demonstrates how the participants, 14 Mexican women living in Southeastern Pennsylvania, use contextualization cues (Gumperz, 1982, 1989, 1992a, 1992b, 1997; Schiffrin, 1996) and subjectivity (Banfield, 1982; Benveniste, 1971) in narratives about their own lives to express different aspects of their identities. Secondly, it discusses these identities in relation to how they affect the participants’ investment (Norton, 1997, 2000; Norton-Pierce, 1995) in the English language. This study combines discourse analysis with ethnographically rich data collection methods. It focuses on structural aspects of the discourse of the participants that closely parallel those analyzed by De Fina in her (2004) study of “Identity in narrative.” These are: the use of terms of reference to ethnic or nationality groups, the use of reported speech, and expressions of agency. However, in this dissertation, these structures are studied in relation to three different predominant identities that emerged in the participants’ discourse: ethnic, gendered, and literate identities, respectively. Analysis revealed that the participants’ use of references to ethnic or nationality groups indexes their identification mainly with their own family, and with the people with whom they associate socially, rather than to their ethnic group as a whole. A certain degree of identification with people who are able to understand them and a simultaneous distancing from monolingual English speakers who are unable or unwilling to communicate with them was also found. Within the analysis of the participants’ gendered identity, it was found that they exhibited a marked preference for citing themselves rather than their spouses. This challenges traditional views of Hispanic women’s gendered roles. The use of reported speech also appears to have the function of highlighting important sections of narrative, and of enhancing the affect in these sections. The analysis of the participants’ literate identity reveals that they present themselves as not having had full control over their past learning experiences, and that they feel they have little or no control over their future education. The participants exhibited a tendency to downplay their agency in events related to the learning or studying of language, literacy, and other skills. They also frequently presented themselves in non-agentive positions within these events. It becomes apparent that of these three predominant identities, the participants’ ethnic and literate identities have a clearly more detrimental effect on their investment in the English language. However, it was also found that obtaining a rewarding job in which they have to use English has highly motivated some of the participants to further their studies of the language. In conclusion, a better understanding of the identities of the participants has yielded a clearer picture of why they have not become invested in the English language, in spite of the overt declaration of their intent.