1 Introduction
Intimate discourse, interaction between couples, families and close friends in private, non-professional settings, lies at the heart of our everyday linguistic experience (McCarthy 1998; Clancy 2016). It deserves more of the research gaze than it currently commands; therefore, this chapter explores how Irish intimates index their identity through their use of personal pronouns. Specifically, the analysis focuses on the use of the items he, she and we in the Limerick Corpus of Intimate Talk (LINT) a 600,000-word sub-corpus of the Limerick Corpus of Irish English (LCIE; see Clancy 2016). In its most elemental form, an index is a linguistic item that depends on either the co-textual or situational context for interpretation, for example the first or second person pronouns I and you (Silverstein 1976). Pronouns are, according to Wales (1996: xii), a ââsyntacticâ phenomenon [that] cannot actually be satisfactorily explained syntactically.â They are, instead, highly complex pieces of language, inevitably bound up in notions of speaker identity, relationships and power (Pennycook 1994; Wales 1996) and are, therefore, inherently pragmatic in nature. A speakerâs choice of which pronoun to use indexes their identity and positions them in respect to others. Schiffrin (2006) maintains that key to the creation of an identity for the âselfâ is the creation of an âotherâ in relation to whom the âselfâ is defined. Personal pronouns, which, by and large, enable the creation of both a linguistic âselfâ in the form of a speaker/writer and âotherâ in the form of a hearer/reader (or indeed a third party), are fundamental to this identity creation process.
Bucholtz and Hall (2005) maintain that indexicality provides the connection between linguistic form and social meaning. In this respect, indexical items function on a number of discourse domains or levels. On the one hand, their discourse meaning needs to be understood in order to determine who is talking or, more pertinently for this chapter, who is being talked about. This enables conversational participants to adopt identities on a micro level, such as speaker-addressee, or meso level, such as protagonist in a narrative (see Greatbatch and Dingwall 1998; Ochs and Taylor 1995; Bucholtz and Hall 2005; Moore and Podesva, 2009). On the other hand, once these identities have been indexed, they then need to be considered in light of âthe social worlds of actions, turns, and relationships in which the reference is produced and interpretedâ (Schiffrin 2006: 130). In this way, âspeakers make relevant their social identities through the discourse identities they take upâ (Koester 2006: 6). For example, Kendall (2008) illustrates how, at the family dinner table, parents can have up to fifteen different meso identities such as head chef, social secretary, moral guardian, etc. She then explores these at a macro level by linking them with gendered parental identities within the family (see also Ochs and Taylor 1995). The mother assumes, almost exclusively, an identity of ânurturing disciplinarianâ. In contrast, the father, primarily through his use of humour, constructs an identity of ârebellious comedianâ. It is through our micro- and meso-level identities that our wider social identities emerge and are constructed (Bucholtz and Hall 2005; Koester 2006; Schiffrin 2006; Handford 2014). Thus, in a discourse and social sense, indexical items are used to anchor identity in interaction (Bucholtz and Hall 2005).
The chapter employs a corpus pragmatic methodology where tools such as word frequency lists are used to extrapolate pronomial patterns surrounding he, she and we that can be said to characterise intimate identity. In order to do this, WordSmith Tools Version 7.0 (Scott 2017) is used to determine the raw and normalised frequency of occurrence of these pronouns, to generate the most frequent word clusters associated with them and, finally, to facilitate detailed examination of concordance lines that contain these items. This highly iterative approach demonstrates that the markedly frequent use of the pronouns he and she, and the two-word clusters is s/he and did s/he, is a way of speaking that is recognisably intimate. It is argued that the use of these pronouns constitutes the indexing of the linguistic style of intimates and, therefore, their identity. The analysis also focuses on the pronoun we and examines how although it is used less frequently in intimate discourse than in other context-types, it is an important linguistic tool in the creation of intimate identities at a variety of levels given its flexibility and multifunctionality with respect to the creation of in-groups and out-groups.
2 Corpus linguistic studies of indexicality and identity
Before dealing specifically with corpus studies that have explored the link between indexicality and identity, it is worth commenting on the ubiquitous presence of personal pronouns on spoken and written corpus word frequency lists. RĂźhlemann (2007: 66â69) posits four reasons for the frequency of I and you in casual conversation; (i) I is prone to repetition (I is repeated at a frequency of about 200 times per million words in conversation (see Biber et al. 1999: 334)); (ii) I and you have a high frequency of collocation especially with cognitive verbs, for example, I think and you know; (iii) speakers in conversation show a clear tendency to prefer a direct mode than an indirect mode and (iv) conversation is co-constructed, with speakers taking turns and each new turn requires the reconstruction of the new speakerâs deictic system. Regarding the distribution of these pronouns, Biber et al. (ibid: 333), using the Longman Corpus of Written and Spoken English (LWSE), have shown that the pronouns I and you are far more common in casual conversation than in other registers such as academic prose. The pronomial forms he/him and she/her are most common in fiction, closely followed by conversation; with the masculine pronoun more common than its corresponding feminine equivalent in both fiction and conversation. In the British National Corpus (BNC), however, RĂźhlemann (2007) has shown that the lemmas HE (he, him, âim and âe) and SHE (she, her and âer) are more frequent in the demographically sampled spoken component of the BNC than in the written component as conversational participants âshow a strong tendency to relate what happened to them and/or othersâ (p.71). He notes that the presence of narrative in conversation means that speakers need to be able to distinguish betwee...