Processing grammatical structuresMorphosyntactic complexity and efficiency in varieties of english around the world with special reference to pronoun omission

Resumo

At the beginning of the twenty-first century, several metrics that attempted to measure language complexity in an empirical manner were postulated (cf., for instance, Miestamo et al. 2008). However, there is one important distinction, put forward by Dahl (2004, 43–44), that has been largely ignored in the specialized literature: structural versus system complexity. System complexity pertains to the grammatical rules that produce the structures used by speakers: the more rules or constraints mediating between meanings and their formal expression, the more complex the grammar (Nichols 2009, 112; Davydova 2011, 103). Structural complexity, on the other hand, focuses on the outputs of those rules: generally, the fewer forms in a structure, the simpler it is (Hawkins 2004, 31–61). A notion related to that of language complexity is communicative efficiency. Communicative acts are efficient if the intended message is transmitted by the speaker to the hearer “in rapid time and with the most minimal processing effort that can achieve this communicative goal” (Hawkins 2014, 34). An efficient communication may thus require more or less structural complexity, depending on the message that the speaker wants to convey. The main aim of this dissertation is to estimate the contribution of one grammatical feature, namely pronoun omission, to the complexity and efficiency of the grammars of several varieties of English. In addition, the influence of language contact, a well-known source of grammatical simplification (cf., for instance, Kusters 2003, 2008) on the degree of complexity and efficiency of the varieties is also assessed. Pronoun omission, that is, the presence of a gap in a structure that could have been filled by a personal pronoun, minimizes the structural complexity of utterances by decreasing the number of forms they contain. Nevertheless, the alternation between overtly expressed and deleted pronouns is not random, that is, it depends on several grammatical constraints. The existence of an additional referential expression and additional rules to regulate its use in turn entails an increase in system complexity. On the other hand, omission results in a more efficient transmission of information from the speaker to the hearer, as the resulting structures contain fewer forms that must be articulated and processed but that can still be successfully decoded by the hearer if the antecedents of omitted pronouns can be easily retrieved from the context. Two case studies are presented that approach these issues from methodologically different but complementary perspectives. The first one (Chapter 4) is a cross-varietal study of the attestation and pervasiveness of pronoun deletion features in 76 varieties of English. The second case study (Chapter 5) complements the first with an investigation of fine-grained distributional patterns of use of omitted referential subject pronouns in three varieties, namely British English, Indian English, and Singapore English. The results of these two case studies allow us to conclude that (i) pronoun omission indeed simplifies structures at no expense to their communicative efficiency, since the antecedents of omitted pronouns are almost always highly accessible; (ii) pronoun omission features are favoured in varieties of English with a history (or a present) of contact with other languages, that is, contact has a simplifying effect on grammars; and (iii) the hypothesized increase in system complexity as a result of the availability of omitted pronouns in the grammar does not take place in varieties that are learned and used as second languages by the majority of their speakers. Therefore, this dissertation contributes new and interesting findings to the literature at the crossroads of research on World Englishes, language complexity and communicative efficiency, and pronoun omission.