Grammatical Flexibility as a Resource in Explicating Referents

Abstract

This article examines one aspect of interplay between grammar and social interaction: how speakers of different languages explicate referents that had been referred to tacitly, i.e., without using an explicit referential expression. The focus is on situations when speakers go on to explicate the referent in the transition space, after bringing the turn constructional unit to a possible completion. Depending on the grammatical affordances of the language, rendering a tacit reference explicit may either expose or mask this operation. Focusing on the latter, we show that the grammars of Russian and Brazilian Portuguese (and, to a lesser extent, English) enable speakers of these languages to explicate referents by extending a possibly complete turn constructional unit with a grammatically fitted increment and, thereby, embed this remedial operation into the progressive construction of the turn without engaging repair machinery. We discuss how tacit referring and flexible word order can enable speakers to carry out this repair-like operation in a covert or embedded fashion, and we examine some interactional functions of this referent-explicating operation.

Notes

¹See Ono and Thompson (1997) for a critical discussion of the term “zero anaphora.” “Zero reference” is a broader term, as, unlike “zero anaphora,” it does not invoke a previous mention of the referent. In some languages, initial references (to persons) may be accomplished tacitly, or with zero reference (e.g., Levinson, 2007). In languages considered here—English, Brazilian Portuguese, and Russian—tacit referring is mostly done in locally subsequent positions (one exception may be the first person reference “I,” which can often be elided). Depending on the researcher's theoretical orientation and analytic goals, similar phenomena may be referred to as pro, subject/object ellipsis, unexpressed, omitted or null subject/object, etc. (e.g., Holmberg, 2005; McShane, 2002).

²We are adopting Schegloff's (1996b, 2000) definition of increments as grammatically fitted extensions of the host turn constructional unit. Ford et al. (2002) refer to such increments as “extensions.” On the other hand, what Ford et al. call “free constituents”—another category of “increments” in their scheme—are not considered “increments” here.

³In linguistics, this is known as “obligatory transitivity”—the requirement that the direct object of the verb be overtly expressed.

⁴A discussion of conditions under which this may be possible is beyond the scope of this article. For linguistic accounts of object omission in Russian see, for example, Gordishevsky & Avrutin (2003) and McShane (2002); for Brazilian Portuguese, see Cyrino (2004)and Cyrino & Matos (2002)

⁵There is some theoretical debate about the status of Russian and Brazilian Portuguese as “true” pro-drop languages (e.g., Gordishevsky & Avrutin, 2003; Holmberg, 2005). There is no question, however, that these languages allow omission of grammatical subjects in some environments. A discussion of these environments is, however, beyond the scope of this article.

⁶A comparison with languages like Japanese or Korean suggests that the prevalence of tacit referring is only partially due to a language's morphological richness (Fox, Hayashi, & Jasperson, 1996; Holmberg, 2005).

⁷Linguists studying Russian (e.g., Turk, 2000; Yokoyama, 1986) and Brazilian Portuguese (e.g., Silva, 2001) argue that word order in these languages relates to the information structure of discourse (specifically, the distinction between “old” versus “new” information). To our knowledge, no interactional studies on word order in either Russian or Brazilian Portuguese have been conducted. A discussion of what constitutes an “underlying” word order in these languages or of the interactional functions of word order is beyond the scope of this article.

⁸For instance, Japanese and Korean freely allow for zero reference, but—as strictly verb-final languages—clearly mark the end of a TCU as such (with sentence-final particles). So, when referring is accomplished tacitly and a TCU is

brought to a possible completion, any explication of referents in the transition space is exposed. These “postpredicate” extensions are understood to be operating on the prior TCU, but they do so disjointedly and are marked (see Couper-Kuhlen & Ono, 2007; Kim, 2007).

⁹During line 2, Lena's face can only be seen in profile on the video. From what can be seen, Lena simply holds her gaze on Mira and displays no signs of nonunderstanding (such as poking or tilting her head-cf. Seo & Koshik, 2010).

¹⁰It should be noted that transition space repair on an indexical reference may result in a grammatical construction called “right dislocation” (Quirk, Greenbaum, Leech, & Svartvik, 1985), as in the following excerpt (line 3): Vick (CDII) 01 SHA: ‘T's free’ee sh'd be'appy. 02 MIC: W[asn' h a p p y.] 03 NAN:   [He- he's there ]evry night Vick 04     (0.4) 05 MIC: Mm hm? 06     (1.2) 07 SHA:  ^○ Yeah ^○  The relationship between transition space repair and right dislocation and what it might mean for “grammar for conversation” deserve further investigation (see, for example, Geluykens, 1994).

¹¹The problem with the initial assessment term might be that the expression é ( … ) f:::ogo” is similar to the vulgar, but more common, é foda (meaning “it's hard”). The new version (at line 9) avoids this association.

¹²In Russian, the verb “will live” (budut zhit) is marked for the third person plural or “they.”