A panel discussion presented at the Second Language Acquisition Graduate Student Symposium on 17 April 2010 at the University of Wisconsin-Madison.

Background

The symposium, co-hosted by the University of Wisconsin-Madison (UW-Madison) and the University of Iowa, focused on identity and was entitled ‘Who are we? SLA in a global society’. Keynote speaker Bonny Norton (University of British Columbia) presented on ‘Identity, investment and digital language learning’, and graduate students explored identity in connection with issues such as race, multilingualism, and immersion contexts. With this growth in identity research comes an increased need to discuss methodologies; the symposium ended with such a panel discussion, emphasizing the need for identity research to be rigorous, ethical, and relational. Chaired by Norton, the panel featured Monica Chavez, Peter De Costa, Chiemi Hanzawa, Margaret Hawkins, Kristine Muñoz, and Carol Severino.

Themes

Rigorous research must be grounded in theory and have robust evidentiary support. Hawkins (UW-Madison) emphasized that ‘identity in and of itself is not a theory’, arguing that researchers cannot examine the discursive construction of identity simply by asking questions and compiling answers. Instead, choosing specific theories to guide research questions both determines research methodology and focuses researchers’ attention on particular aspects of identity. Norton noted the importance of differentiating between ‘scientific’ and ‘rigorous’ research, highlighting that defending claims through careful use of evidence can gain respect for qualitative research. Using data triangulation to achieve ‘thick description’ (Geertz Reference Geertz1973) was viewed as key to realizing this goal. In her research on identity in L2 writing, Severino (University of Iowa) examines multiple drafts of multiple genres types over multiple papers, combining paper analysis with student and instructor interviews. Muñoz (University of Iowa), in her research on personal relationships and persuasion, analyzes naturally-occurring conversations between partners as well as interview data to investigate dissonances between what individuals say and do. De Costa (UW-Madison), in his recent examination of immigrant student identity in a Singapore secondary school classroom, used recorded classroom interactions, field notes, artifacts, and interviews to illustrate the impact of ideology on focal learners' language development. De Costa also noted the need for longitudinal data collection to examine participant positioning over time, which allows for the ‘thickening of identities’ (Wortham Reference Wortham2006). Hawkins noted that as researchers interpret this triangulated data, they must be aware that participant stories are not stories of identity, but are partial pictures; researchers must analyze, then, not only what they hear, but also what they do not or cannot hear.

Regarding ethical dimensions of research, Norton, who positions herself as ‘an educator at heart’, stated that she begins research from participant perspectives through asking about challenges they have experienced. By working collaboratively and responsibly so as not to disrupt participants’ lives, researchers can be respectful of participants and their lived experiences. However, not all research contexts allow researchers or participants freedom to make research relevant to participants. Chavez (UW-Madison) observed that although identity is situated, the power differentials present in certain research contexts may actually demonstrate the limits of discursive identity construction. Drawing on her research in German foreign language classrooms, where teaching assistants and students are to speak only German, she noted that because the language may not be ‘real’ to them, language learners may be deprived of certain resources and identities as they cope with being positioned by others according to specific identity categories. Another ethical issue confronting identity researchers is the need to collaborate with community members and organizations while advancing social justice on behalf of research participants. Severino noted that to avoid exploitation of participants’ stories and experiences, researchers need to move beyond the role of ‘researcher’ and advocate for participants in their local communities. Hawkins elaborated on the challenges in partnering with communities, noting that in her work linking women and children from US community centers to African schools, projects can become viewed as ‘US-’ or ‘university-based’.

Panel members also highlighted the need for relational research, which takes into account the power relations between the researcher and the researched, adding another dimension to data analysis and representation. According to Hawkins, power relations between researcher and researched cannot be dissolved and must be accounted for. Although researchers may approach identities as multiple, dynamic, and locally constructed, participants may feel they carry a particular identity with them. Norton noted that within a perspective viewing identities as fluid, certain identities can be seen as more salient in certain contexts, and De Costa added that some identities, such as religious identity, may be more ‘fixed’ than others. One way to address discrepancies between researcher and participant representations, as Muñoz pointed out, is to acknowledge that both perspectives are ‘right’; the job of the researcher is therefore not to make them coalesce ‘perfectly’, but to make them coherent. De Costa, however, observed that ontologically, if data is analysed from a postmodern perspective that recognizes the existence of multiple realities, the issue of ‘discrepancies’ becomes less problematic; consequently, coherence becomes less significant in qualitative research (Leung, Harris & Rampton Reference Leung, Harris, Rampton, Norton and Toohey2004). De Costa also highlighted that while researchers should be aware of their influence on the research context, they should refrain from overemphasizing their role in the research process (Canagarajah Reference Canagarajah2005).

In closing, Muñoz added that research needs to be not only rigorous, ethical, and relational, but must also be rhetorical, in that researchers must consider audience expectations and align their research agenda with reader expectations.