EARLY INTONATIONAL DEVELOPMENT IN CATALAN

Pilar PrietoICREA-UAB

Introduction

• Our main interest was to investigate early intonational development in Catalan children between the ages of 1;0 and 2;6, focusing on the capacity to use appropriate intonational forms for specific pragmatic meanings.

Introduction

Specific aims

Methodology

Results

Conclusions

Introduction• Very few studies on early intonational development (Snow 2004, 2005, 2006; Oller 1998, D’Odorico 2004, Frota & Vigário 1995, among others).

• A general assumption is that children’s first intonation patterns deviate drastically from the target system.

Introduction

Specific aims

Methodology

Results

Conclusions

• A very productive line of research has been on the frequency of appearance of rising and falling contours.

Introduction• Some studies on prosodic development seem to indicate that the acquisition of intonation is paced by the child’s development of grammar.

•As Snow (2006) points out, “the milestone event in children’s acquisition of expressive syntax is the appearance of two-word combinations at about 18 months, which coincides exactly with the dramatic growth in intonation that was observed in this and other studies.”

Introduction

Specific aims

Methodology

Results

Conclusions

Specific Aims• The aim of the study is threefold:

– To describe the development of different intonation patterns and pragmatic meanings at different stages of development;– To assess whether the first intonation contours produced by the children reflect the language-specific prosodic properties of the input language;–To assess whether the mastery of certain intonation patterns correlate with grammatical development and the start of the two-word period.

Introduction

Specific aims

Methodology

Results

Conclusions

Participants• We used the Serra-Solé corpus on Catalan available in CHILDES: contains transcribed speech of four Catalan children (Gisel·la, Guillem, Laura, and Pep).

•The children and both parents used Central Catalan almost exclusively in their family context, with slightly different degrees of contact with Spanish.

Introduction

Specific aims

Methodology

Results

Conclusions

Materials, 1• Each child was video-taped on a monthly basis from the start of the use of 25 words up until four years of age.

• Data was collected following a naturalistic design, that is, spontaneous situations were recorded at home.

Introduction

Specific aims

Methodology

Results

Conclusions

Materials, 2• Table 1 presents a summary of the data used for this study.

•The data analysis ranges from the beginning of the 25-word vocabulary stage up until 26 months.•A total of 6090 utterances were analyzed.

Introduction

Specific aims

Methodology

Results

Conclusions

Phonetic and prosodic

transcription, 1• After digitizing the original videotapes, the data were segmented and phonetically transcribed with PHON. • In this first stage, all utterances uttered by the children were analyzed, including speech-like utterances like vocalizations, cries, or whisperings. •The acoustic part of each utterance was exported for acoustic and prosodic analysis.

Introduction

Specific aims

Methodology

Results

Conclusions

Phon Interface

Introduction

Specific aims

Methodology

Results

Conclusions

Perceptual and prosodic analysis, 1• After exporting the sound files, each meaningful utterance was manually annotated for the following fields:

– orthographic transcription; – prosodic transcription in the Catalan version of ToBI, CatToBI (Prieto et al. 2007); – pragmatic meaning; – annotation of the main differences between the adult and the children’s intonation patterns.

Introduction

Specific aims

Methodology

Results

Conclusions

Cat-ToBI Squematic representation

Prototypical sentence types and intonational

meaning

L* L-L% Neutral statements

H+L*/ ¡H+L* L-L% Yes-no questions

(L)+H* L-L% Wh-questions

L* H-H% Neutral yes-no questions, confirmation seeking-

questions

L+H* H-H% Echo questions with surprise

H* H-H% Yes-no questions with invitation or offer meaning

L* L-H% Obvious statement

L+H* L-L% Imperatives and emphasis statements

L* H-L% Statements with convictions and obviety

meaning

L+H* !H-!H% Vocatives

Introduction

Specific aims

Methodology

Results

Conclusions

Cat-ToBI Squematic representation

Prototypical sentence types and intonational

meaning

L+H* H-L% Vocative with insistent interest

L+H* L-H% Yes-no echo questions

L+H* LH-L% Insistent request

L+H* L-!H% Categorical-obvious statements

L* L-!H% Disapproval statements

L+H* H-!H% Hesitation statements

Introduction

Specific aims

Methodology

Results

Conclusions

Perceptual and prosodic analysis, 2

Figure 1: Waveform display, spectrogram, f0 contour, and prosodic labelling of the utterance [] hola ‘hello’ produced

by Guillem at 1;4.26.

Introduction

Specific aims

Methodology

Results

Conclusions

Methodological issues: MLU

• In this study, we calculated both using the “mlu” and “wlen” commands in CLAN.

Introduction

Specific aims

Methodology

Results

Conclusions

Figure 2 shows the MLUw for each of the sessions, for each child.

Methodological issues: MLU

Introduction

Specific aims

Methodology

Results

Conclusions

• The graph shows that:– while Pep and Guillem reach an MLUw level of 1.5 between 1;5 and 1;8,

Pep Guillem

• The graph shows that:– while Pep and Guillem reach an MLUw level of 1.5 between 1;5 and 1;8,

Pep Guillem

Methodological issues: MLU

Introduction

Specific aims

Methodology

Results

Conclusions

• The graph shows that:– while Pep and Guillem reach an MLUw level of 1.5 between 1;5 and 1;8, –Laura and Gisel·la do not reach this level until they are 2;1 or 2;2.

Gisel·la Laura

Thus…

•The dual distribution of the data makes it possible to test the claim that there is a sound correlation between grammatical and intonational development (Snow 2000, 2006, among others).

Introduction

Specific aims

Methodology

Results

Conclusions

Results•Summary of the percentage of occurrence of sentence-types for three of the children.

Introduction

Specific aims

Methodology

Results

Conclusions

Nuclear accent configurations:

statements

Introduction

Specific aims

Methodology

Results

Conclusions

L+H* LL%

H* LL%

Introduction

Specific aims

Methodology

Results

Conclusions

Nuclear accent configurations:

questionsL+H* HH%

L* HH%

Introduction

Specific aims

Methodology

Results

Conclusions

Nuclear accent configurations:

commandsL+H* LL%

H* LL%

Guillem and Pep: 1-word stage

•Guillem and Pep start producing two-word combinations at around 1;6. • The intonational analysis reveals that both begin to use a handful of intonational contours at about 13-15 months of age.• Importantly, the children master the tune-text alignment patterns in these contours.

Introduction

Specific aims

Methodology

Results

Conclusions

Video file (left) and waveform display, spectrogram, f0 contour, and prosodic labelling (right) of the utterance [] pilota ‘ball’ produced by Pep at 1;2.3.

Statements: L+H* LL%

Introduction

Specific aims

Methodology

Results

Conclusions

Pitch accent: L+H* LL%

Video file (left) and waveform display, spectrogram, f0 contour, and prosodic labelling of [,] Laia, Laia ‘proper name’ produced by Pep at 1;2.28.

Focal accents: L+H* LL%

Introduction

Specific aims

Methodology

Results

Conclusions

Pitch accent: L+H* LL%

Video file (left) and waveform display, spectrogram, f0 contour, and prosodic labelling of the utterance [] hola ‘hello’ produced by Guillem at 1;4.26.

Calling contours: L+H* LH%

Introduction

Specific aims

Methodology

Results

Conclusions

Pitch accent: L+H* LH%

Video file (left) and waveform display, spectrogram, f0 contour, and prosodic labelling of [] aquesta? ‘this one?’ uttered by Pep at 1;5.22.

.

Interrogative contours: L*HH%

Introduction

Specific aims

Methodology

Results

Conclusions

Pitch accent: L* HH%

Guillem and Pep: 2-word stage

•In this period, the two children start producing a variety of complex tunes to express:

– requests– complaint, discontent– insistence

• The production of these contours demonstrate that relatively early both children have an outstanding control over the complex alignment of edge tunes.

Introduction

Specific aims

Methodology

Results

Conclusions

Video file (left) and waveform display, spectrogram, f0 contour, and prosodic labelling of [, ] home, una cullera! ‘man, a spoon!’ uttered by Pep at 1;8.0.

.

Complaint contours: L*HL%

Introduction

Specific aims

Methodology

Results

Conclusions

Pitch accent: L* HL%

Video file (left) and waveform display, spectrogram, f0 contour, and prosodic labelling of the utterance [] mira.. ‘please take a look’ uttered by Guillem at 1;11.13.

Insistence contours: H* LHL%

Introduction

Specific aims

Methodology

Results

Conclusions

Pitch accent: H* LHL%

Gisel·la and Laura: 1-word stage

•Gisel·la and Laura show a great increase in the use of intonation well before they start using two-word combinations. •By this time both produce statements and a variety of exclamative, imperative and interrogative intonation contours correctly, and also a variety of tunes to express requests, discontent or insistence. • Importantly, the children master the tune-text alignment patterns in these contours.

Introduction

Specific aims

Methodology

Results

Conclusions

Video file (left) and waveform display, spectrogram, f0 contour, and prosodic labelling of the sequence [ :] té? ‘do you want it?’ uttered by Laura at 1;7.10.

Interrogative contours: L*HH%

Introduction

Specific aims

Methodology

Results

Conclusions

Pitch accent: L* HH%

Video file (left) and waveform display, spectrogram, f0 contour, and prosodic labelling of the utterance [,] aigua, pilota ‘water and a ball’ uttered by Gisel·la at 1;10.07.

Complaint contours: H+L* HL

%

Introduction

Specific aims

Methodology

Results

Conclusions

Pitch accent: H+L* HL%

Video file (left) and waveform display, spectrogram, f0 contour, and prosodic labelling of the utterance [] aquest. ‘this one’ uttered by Gisel·la at 1;8.24.

Insistent requests: H*LHL%

Introduction

Specific aims

Methodology

Results

Conclusions

Pitch accent: H* LHL%

Gisel·la and Laura, conclusions

•In conclusion, Laura and Gisel·la’s examples of intonational development between 1;7 and 1;11 show a good phonetic and phonological command of a variety of pitch accents and boundary tones, all produced during the one-word stage. • No substantial increase in intonational grammar is attested when they started producing two-word combinations.

Introduction

Specific aims

Methodology

Results

Conclusions

Conclusions, 1•Our results demonstrate that, contrary to what has been claimed, children’s emerging intonation is largely independent of grammatical development.

• Despite the fact that the start of the two-word period was so different across the two groups of children, we did not find a substantial difference in the production of nuclear pitch accents and boundary tones.

Introduction

Specific aims

Methodology

Results

Conclusions

Conclusions, 2•Children have an important knowledge of intonational grammar before they produce two-word combinations.

– Specifically, they control the phonetic production and intonational meaning of a variety of phonologically distinct pitch accents and boundary tones by about 1;9, regardless of their level of grammatical development. –They have an early mastery of tone-text alignment of pitch accents and boundary tones, which are finely produced from the beginning. –Yet they do show some phonetic implementation discrepancies with the adult language with respect to the scaling of the L% boundary tone.

Introduction

Specific aims

Methodology

Results

Conclusions

Acknowledgments:

•Joint work with Maria del Mar Vanrell and Ana Estrella.

•We are very grateful to Aoju Chen, Paula Fikkert and Laura D’Odorico for very helpful comments on the first version

of the paper.•Sponsoring: BFF-00478 from the Spanish

Ministry of Education.

Summary

Introduction

Specific aims

Methodology

Results

Conclusions

Summary

Introduction

Specific aims

Methodology

Results

Conclusions