Different cognitive systems struggling for word order
Marina Nespor
University of Milano-Bicocca
(work in collaboration with PhD student Alan Langus)
I will present arguments indicating that the grammatical diversity observed among the world’s languages emerges from the struggle between individual cognitive systems trying to impose their preferred structure on human language.
Evidence from language change, grammaticalization, stability of order, parsing advantages, and theoretical arguments, indicates a syntactic preference for SVO. There reason for the prominence of SOV languages is not as clear.
I will present experiments aimed at establishing the cognitive bases of the two most common word orders in the world’s languages: SOV (Subject–Object–Verb) and SVO. In two gesture-production experiments and one gesture comprehension experiment, I will show that SOV emerges as the preferred constituent configuration in participants whose native languages (Italian and Turkish) have different word orders. I will propose that improvised communication does not rely on the computational system of grammar.
The results of a fourth experiment, where participants comprehended strings of prosodically flat words in their native language, shows that the computational system of grammar prefers the orthogonal Verb–Object orders.
Further preferences of the computational system of grammar will be examined in relation to language acquisition.
Cues to early language discrimination: the rhythm class hypothesis revisited
Laura Bosch
Universitat de Barcelona
A first step into language acquisition involves getting some knowledge about the general characteristics of the sound pattern of the native language. While very young infants can rely on rhythm information to discriminate between sets of utterances that come from different languages, slightly older infants’ abilities to differentiate their native language from another, rhythmically close language, are not completely well understood. While some authors have suggested that increasing experience with language gives infants access to more subtle rhythm cues that might be used to solve the discrimination task, the availability of alternative (or complementary) cues cannot be neglected. Languages differ in their segmental inventories and infants’ ability to track statistical information in the speech signal is present very early on. Thus, cues derived from distributional information of segments in fluent speech are good candidates that can facilitate language discrimination in the absence of clear rhythm differences.
In this talk, a review of infant studies focused on language (and dialect) discrimination abilities will be offered. Results from recent work revealing an early failure to discriminate languages that belong to different rhythm categories will also be discussed. The role of variables related to the procedures and material used to test infants’ language discrimination abilities will also be considered in interpreting the experimental evidence.
The DEPE project: general goals and research topics
DEPE Project Research Team
Universidade de Lisboa
Largely focusing on European Portuguese (EP) but taking a cross-linguistic approach, the DEPE project has as general goals to contribute to the understanding of (i) what is the pattern of emergence and development of different aspects of prosodic knowledge, (ii) what is the contribution of prosody for the emergence and development of other aspects of grammar, such as word segmentation, the development of the lexicon, the emergence of multiword sentences and word order setting, and (iii) what in language acquisition is biologically determined and/or universal and what is dependent of the ambient language, and in the latter case what is the role of frequency vs. grammar in the acquisition of phonology. These goals are pursued by research in four key areas: (1) the emergence and development of prosodic domains at the word and phrase levels; (2) the development of the intonational system; (3) the pace of temporal patterns in early child speech; (4) the processing of prosody in early child and adult grammars. Three other languages are also investigated for some of the topics, namely Catalan, Spanish, and Dutch, following similar research questions and methodologies, and allowing for cross-linguistic generalizations.
Intonational focus-marking in different word orders in children
Aoju Chen
Max Planck Institute for Psycholinguistics
Many languages employ word order and intonation in tandem to mark focus, though they can differ in the extent to which each device is used. The use of intonation in focus marking in child speech has received some attention but past work is mostly restricted to one word order, namely, the SVO order. Taking a cross-linguistic approach, the current study aimed to develop our understanding of the intonational realisation of focus in different word orders in children acquiring languages that differ in the degree of the use of word order in focus marking. To this end, we examined how four- to five-year-olds children used intonation to encode broad focus (B), contrastive focus (CF), and narrow focus (NF) in SVO and OVS word orders in German and Dutch. BF, CF and NF sentences in SVO and OVS word orders were elicited from 20 German and 14 Dutch 4- to 5-year-olds via a controlled but interactive answer-reconstruction task. We analysed f0-maximum, f0-minimum, f0-range, and duration of the target nouns.
Initial results showed that in both languages, children distinguished focus types phonetically only when the subject was focused in NF and CF. However, German and Dutch children differed from each other in the exact uses of the cues. More specifically, in the SVO word order, German children made a more intensive use of f0-related cues than Dutch children to distinguish CF from BF and NF from BF. However, Dutch children varied duration to distinguish CF from BF whereas German children didn’t. In the OVS word order, German children varied f0-range to distinguish CF from BF and duration to distinguish NF from BF, whereas Dutch children only varied f0-maximum to distinguish CF from BF. We suggest that the difference in the OVS word order may reflect the more common use of OVS (in marking contrastive focus on the object) in German than in Dutch. But the difference in the SVO word order may point to a general difference in how pitch and duration are used in the canonical word order in German and Dutch.
Early intonational development in Catalan: results from perception and production
Pilar Prieto
ICREA- UPF
In this talk I will review complementary results from two recent research papers dealing with early intonational development in Catalan. Taken together, the results seem to suggest a very early development of intonational grammar both in the production and in the perception realms.
The first production study (Esteve & Prieto 2010) had the goal of investigating the prosody of early vocalizations of Catalan infants. Audiovisual recordings of three Catalan-babbling infants born in monolingual families were made. From ages 0;6 to 1;0, these infants were recorded weekly for 30 minutes per session, at their homes and while playing normally with their parents. Vocalization analysis consisted of the independent marking of two prosodic features (pitch range and duration) and the codification of the communicative status of each vocalization. If they were judged to be communicative, vocalizations were further classified according to their specific pragmatic function. A total of 2,222 vocalizations were analyzed. Results show that the prosodic patterns that infants produce are different depending on the communicative status of the vocalization: vocalizations are shorter and with a higher pitch range when they are communicative than when they are not, irrespective of age and child. Further analyses show that infants display different prosodic patterns depending on the specific pragmatic functions they seem to be expressing: vocalizations expressing requests and protests have a wider pitch range and longer duration than vocalizations expressing response or statement. Thus, the results of this study confirm the hypothesis that prelinguistic infants are able to control their prosodic patterns in the second half of their first year to indicate not only communicativeness but also certain pragmatic functions.
The second pilot perception study (Artès 2010 MA thesis) had the goal of exploring whether Catalan children at 14-18 months can identify phonological contrasts based on stress (iambic vs trochaic patterns) by taking into account the intonational variation present in the input language. If children have learnt that intonational variation does not affect lexical meaning they will be able to rightly associate several intonation patterns with their corresponding contrastive stress patterns (wS and Sw, for example). We first ran an adults’ experiment to assess the difficulties posed by some intonations in combination with two stress conditions (iamb and trochee). This experiment showed us that statement, vocative and yes-no question intonations were the most easily discriminated, whereas some others, namely gentle request and insistent request intonations, were the most difficult ones to distinguish. In order to test infants’ abilities, we chose the three easier intonations and one of the complex ones (the insistent request) for the child experimental task. We used a visual habituation procedure, one of the most common techniques in infant behavioural studies, implementing a switch procedure (Werker et al., 2002) with some modifications involving a visual choice design (Yoshida et al., 2009). The five children tested were able to rightly associate a particular contour with each stress condition, thus having learned the correlation between stress and intonation. Moreover, the infants in our study did not have difficulty with the most easily discriminable intonations, while the insistent request was very often misinterpreted, especially in the trochee condition. Though the sample used was not so large as to offer us a reliable conclusion, we can tentatively conclude that 14-19 month olds already have a (generally) stable intonational grammar. Importantly, these results are in agreement with those of recent production studies (Prieto & Vanrell 2007 for Catalan, Frota & Vigário 2008 for European Portuguese).
|
