Rhetorical questions in Italian between theory and acquisition

Abstract

This dissertation investigates rhetorical questions (RhQs) in Italian, their linguistic marking, and their acquisition in a monolingual and a bilingual setting. RhQs are utterances with the syntactic and semantic properties of a question, but they are not used to ask for information: rather, the speaker utters a RhQ to indicate that the answer is obvious. The answer to the question is presupposed by the speaker, who uses the RhQ to “make a point”. The rhetoricity of a question needs to be signalled to the addressee through contextual and/or linguistic cues, which convey aspects of the speaker's emotive or epistemic attitude or point to the shared knowledge between the speaker and the addressee. Thus, producing and understanding RhQs requires a complex interaction of linguistic phenomena, pragmatic reasoning, and conversational competence. Such complex and multi-faceted phenomena at the interface between language and general cognition are predicted to be acquired in late childhood or early adolescence, and to be particularly challenging in bilingual acquisition. Comparing monolingual and bilingual children on the acquisition of the same phenomenon allows for a comparison of two different modes of acquisition. The target bilingual population of this work are Italian-German bilingual children, acquiring Italian as (one of) the language(s) of the family (i.e., the heritage language) and German as the language of the environment (i.e., the majority language). A heritage language is acquired under distinct conditions, as the input the child receives is usually quantitatively and qualitatively different from the input that a monolingual receives, and the heritage language may undergo the influence of the majority language; degrees of language experience, proficiency and dominance vary widely across different speakers. In Chapter 2, I investigate which lexical and morphosyntactic cues (henceforth, lexical-syntactic cues) mark RhQs in colloquial Italian, focusing on questions that presuppose a negative answer (such as Who eats broccoli for breakfast?! = No one eats broccoli for breakfast!). Through an elicitation experiment, I show that this type of RhQs can be marked by a variety of cues, which are all optional and contribute to the rhetorical interpretation of a question by various means. Importantly, most cues do not force the rhetorical interpretation; rather, they are ambiguous with other types of questions that express the speaker's negative attitude, surprise, or inability to find an answer, or even with canonical information-seeking questions (ISQs). The discrepancy between the type of utterance (interrogative) and the discourse function of RhQs (more similar to an assertion) poses them as an interesting topic for acquisition, both in monolingual and bilingual settings. Through a series of tasks, I investigate the acquisition of optional pragmatic marking in RhQs by 6-to-9-year-old children, as well as the similarities and differences between monolingual and bilingual acquisition. The multiplicity, optionality, and ambiguity of pragmatic cues, as well as their being a phenomenon at the interface between language, pragmatics and general cognition, predicts this phenomenon to develop later in childhood, potentially throughout the primary school period. In Chapter 3, I investigate through a Comprehension task whether monolingual and bilingual children are able to distinguish between RhQs and ISQs without a context, solely based on prosodic and lexical-syntactic cues. The results indicate that children were able to correctly interpret questions as rhetorical or non-canonical when provided with a combination of lexical-syntactic and prosodic cues, while the ability to do so from prosody alone is greatly variable among children. Bilingual and monolingual children behave similarly in this task. In Chapter 4, I address the acquisition of lexical-syntactic cues in production. I compare 6-to-9-year-old Italian monolingual children and Italian monolingual adults on a Production task, observing the patterns of optional modification of RhQs and ISQs in each group. In this task, children modified RhQs using the same types of cues as adults (with the exception of formulaic embedding under want, a strong rhetorical cue). This result indicates that some children have acquired some cues, but it does not imply that all children did. A developmental pattern is found in the rate of modification, as younger children (6- and 8-year-olds in particular) use on average fewer cues than adults, while 9-year-olds display the same pattern and frequency as adults. The findings suggest that the acquisition of optional pragmatic marking in RhQs is still ongoing in this age range, confirming the complexity and multidimensionality of the phenomenon. In Chapter 5, bilingual children are tested on the same Production task and their results are put in relation with those of monolingual children. Bilingual children exploit the same types of cues as monolinguals to mark RhQs throughout the investigated age range. Nevertheless, they do so at lower rates and do not present the same developmental pattern shown by monolinguals, such that, as a group, 9-year-olds are at the same level as 6-year-olds. Additionally, bilingual children show some instances of transfer from the ML, in particular in the occasional translation of German discourse particle schon ‘already’, a strong rhetorical cue, into the Italian equivalent già. I discuss a number of possible reasons for such divergences, including different levels of language experience and of proficiency in the HL, as proven by the effect of background measures, in particular vocabulary diversity (obtained from a narrative task) and dominance (calculated from language experience). The results of this study indicate that children can only acquire and exploit the fine patterns of pragmatic expression through linguistic means if they are sufficiently exposed to their HL, Italian, in quantitatively and qualitatively rich contexts. In Chapter 6, I bring together and discuss the findings of the various experiments; while the explicit goal of the thesis is to examine the patterns of linguistic marking in RhQs, I draw some inferences on children’s acquisition of rhetoricity itself, suggesting that in the investigated age range children are acquiring not only the overt linguistic marking, but also RhQs as a pragmatic category. Finally, in Chapter 7, I conclude and suggest several opportunities for future research. Taken together, the findings of the present dissertation shed light on a yet unexplored topic in monolingual and bilingual language acquisition, contributing to the wider issue of how the complex relations between communication and language interact with different modes of acquisition, and paving the way for many further research directions on the acquisition of RhQs.