Arróniz, S. & Amaral, P. On prosody and constructions: A phonological description of the Spanish reduplicative-topic construction. Hispanic Linguistics Symposium. Wake Forest, NC. October 2021.

Abstract

In recent years, constructions have been back at the center of linguistic theory, remarking that they are not epiphenomenal elements and they encode semantic, pragmatic, and syntactic information (Fillmore, Kay & O’Connor, 1988; Kay & Fillmore, 1999; Goldberg, 1995; Hilferty & Valenzuela, 2001; Hilferty, 2003). Not much is known about the role of phonology in the characterization of specific constructions (but see Elvira-García, 2015; Elvira-García et al., 2017). The present study aims to address this gap by examining the phonological features of the construction in (1), called the Spanish reduplicative-topic construction (Valenzuela et al., 2005). The approach we adopt is a Laboratory Phonology (LabPhon) approach and aims to shed light on the role of pitch accents and rhythm in the production of this construction, in order to contribute to a better understanding of the interface between prosody and syntax in these units. This construction is characterized by a productive topicalization pattern in which the topic phrase at the beginning of the clause is reduplicated (cf. Barcelona, 1986; Narbona, 2000). The construction is analyzed as containing three Periods (A, B, and C, see [1]). In their study, Valenzuela et al. (2005) propose that the reduplicative-topic construction exhibits a set of idiosyncratic features that are not strictly predictable from its constituents. Accordingly, they provide a description of the semantic, syntactic (polarity, agreement), and phonological factors that shape this construction. Nevertheless, prosodic factors such as rhythm, intonation, or constituent weight, albeit informally described, are not examined from a LabPhon approach. The present study addresses these aspects by adopting the Autosegmental-Metrical (AM) system for the analysis of intonation.

(1) [A] Comer, comer, [B] no come, [C] pero bebe como un cosaco.
to-eat to-eat no eats but drinks like a cossack
‘He doesn’t eat very much, but he drinks like a fish.’

An oral elicitation instrument was used to collect production data from four speakers of peninsular Spanish, with 3 tasks: (i) a context-free oral production task, in which subjects read a list of different examples of reduplicative-topic constructions; (ii) a contextualized oral production task, in which participants read a narrative containing several instances of the reduplicative-topic construction; and (iii) an oral Discourse Completion Task, introducing a context that elicited a response using a reduplicative-topic construction. The productions were analyzed acoustically with Praat and labeled according to the AM system. Results from our pilot study show that, while there is some variation in the production of patterns, a similar contour is observed in the production of Period A, L*+H (Figure 1). In addition, unlike what is claimed by Valenzuela et al. (2005), we have found different behaviors among speakers regarding rhythmic patterns. While some speakers showed extended pauses between Periods B and C, others had a continuity in the sentence with no pauses between the periods. An additional finding pertains to vowel lengthening (Figure 2). Several speakers showed a lengthening of the last vowel in Period A (/ko.ˈmeːɾ/), or in the lexical item that initiates Period C (/ˈpeː.ɾo/). This aspect is not mentioned in the Valenzuela et al. (2005) study. We are currently testing a higher number of speakers in order to further investigate these findings. This study provides the first analysis of the prosodic properties of the Spanish reduplicative-topic construction using a LabPhon approach. While important properties of this construction had been identified by Valenzuela et al. (2005), an empirically-based analysis of this type can unveil properties of the construction. This approach can provide in-depth insight into the multiple levels of information encoded by constructions.n recent years, constructions have been back at the center of linguistic theory, remarking that they are not epiphenomenal elements and they encode semantic, pragmatic, and syntactic information (Fillmore, Kay & O’Connor, 1988; Kay & Fillmore, 1999; Goldberg, 1995; Hilferty & Valenzuela, 2001; Hilferty, 2003). Not much is known about the role of phonology in the characterization of specific constructions (but see Elvira-García, 2015; Elvira-García et al., 2017). The present study aims to address this gap by examining the phonological features of the construction in (1), called the Spanish reduplicative-topic construction (Valenzuela et al., 2005). The approach we adopt is a Laboratory Phonology (LabPhon) approach and aims to shed light on the role of pitch accents and rhythm in the production of this construction, in order to contribute to a better understanding of the interface between prosody and syntax in these units.

This construction is characterized by a productive topicalization pattern in which the topic phrase at the beginning of the clause is reduplicated (cf. Barcelona, 1986; Narbona, 2000). The construction is analyzed as containing three Periods (A, B, and C, see [1]). In their study, Valenzuela et al. (2005) propose that the reduplicative-topic construction exhibits a set of idiosyncratic features that are not strictly predictable from its constituents. Accordingly, they provide a description of the semantic, syntactic (polarity, agreement), and phonological factors that shape this construction. Nevertheless, prosodic factors such as rhythm, intonation, or constituent weight, albeit informally described, are not examined from a LabPhon approach. The present study addresses these aspects by adopting the Autosegmental-Metrical (AM) system for the analysis of intonation.

An oral elicitation instrument was used to collect production data from four speakers of peninsular Spanish, with 3 tasks: (i) a context-free oral production task, in which subjects read a list of different examples of reduplicative-topic constructions; (ii) a contextualized oral production task, in which participants read a narrative containing several instances of the reduplicative-topic construction; and (iii) an oral Discourse Completion Task, introducing a context that elicited a response using a reduplicative-topic construction. The productions were analyzed acoustically with Praat and labeled according to the AM system. Results from our pilot study show that, while there is some variation in the production of patterns, a similar contour is observed in the production of Period A, L*+H (Figure 1). In addition, unlike what is claimed by Valenzuela et al. (2005), we have found different behaviors among speakers regarding rhythmic patterns. While some speakers showed extended pauses between Periods B and C, others had a continuity in the sentence with no pauses between the periods. An additional finding pertains to vowel lengthening (Figure 2). Several speakers showed a lengthening of the last vowel in Period A (/ko.ˈmeːɾ/), or in the lexical item that initiates Period C (/ˈpeː.ɾo/). This aspect is not mentioned in the Valenzuela et al. (2005) study. We are currently testing a higher number of speakers in order to further investigate these findings. This study provides the first analysis of the prosodic properties of the Spanish reduplicative-topic construction using a LabPhon approach. While important properties of this construction had been identified by Valenzuela et al. (2005), an empirically-based analysis of this type can unveil properties of the construction. This approach can provide in-depth insight into the multiple levels of information encoded by constructions.