Does prosody constrain code-switching?: Stress, tone and speech rate in Papiamento-Dutch spontaneous speech

Research has shown that code-switching (CS) is morpho-syntactically constrained (e.g. Poplack, 1980; Myers-Scotton, 1993; MacSwan, 1999; Lipski, 2019). The fact that phonology and syntax interface in bilingual performance (Bullock, 2009) has been largely neglected in CS research. It is likely that the interface between prosody and morphosyntax, and not merely morphosyntax alone, may play a role in constraining CS. However, the phonetic and phonological reflexes of CS remain relatively unexplored. This thesis aims to improve our understanding of prosodic constraints on CS by examining the speech from a Papiamento- Dutch conversation corpus (Gullberg, Indefrey & Muysken, 2004; 2009). This language pair is eminently suitable for a prosodic analysis because Papiamento has a tonal system with two level tones that interacts with lexical stress, and Dutch a different lexical system: with stress, without tone. I examined whether stress constrains CS in the nominal domain (Akinremi, 2016),...Show moreResearch has shown that code-switching (CS) is morpho-syntactically constrained (e.g. Poplack, 1980; Myers-Scotton, 1993; MacSwan, 1999; Lipski, 2019). The fact that phonology and syntax interface in bilingual performance (Bullock, 2009) has been largely neglected in CS research. It is likely that the interface between prosody and morphosyntax, and not merely morphosyntax alone, may play a role in constraining CS. However, the phonetic and phonological reflexes of CS remain relatively unexplored. This thesis aims to improve our understanding of prosodic constraints on CS by examining the speech from a Papiamento- Dutch conversation corpus (Gullberg, Indefrey & Muysken, 2004; 2009). This language pair is eminently suitable for a prosodic analysis because Papiamento has a tonal system with two level tones that interacts with lexical stress, and Dutch a different lexical system: with stress, without tone. I examined whether stress constrains CS in the nominal domain (Akinremi, 2016), and whether Papiamento tone constrains Dutch insertions (Zheng, 1997; Tuc, 2003). Furthermore, I examined whether speech rate in bilingual vs. unilingual utterances differ to add to the research on speech planning in CS (Johns & Steuck, 2021). My findings are that the stress of switched nouns does not constrain CS, but the stress of adjacent words might; Dutch insertions occur mostly in a context where the prosodic systems of Papiamento and Dutch coincide; and speech rate in bilingual vs. unilingual utterances does not significantly differ. I conclude that congruency in prosody facilitates CS and that CS does not inhibit speech planning. Taken together, my findings are compatible with the view that CS may be an opportunistic strategy that bilinguals use to aid speech planning as prosody in both languages openly contributes to production (Beatty-Martinez, Navarro-Torres & Dussias, 2020).Show less

Leiden University Student Repository

Documents

In Collections

Does prosody constrain code-switching? Stress, tone and speech rate in Papiamento-Dutch spontaneous speech