CHILDES French Champaud Corpus

Christian Champaud Laboratoire de Psychologie CNRS kail@ext.jussieu.fr website

Browsable transcripts

Download transcripts

Citation information

Champaud, C. (1994). The development of verb forms in French children at around two years of age: some comparisons with Romance and non-Romance languages. Paper presented at the Fisrt Lisbon Meeting on Child Language, Lisbon, Portugal.

In accordance with TalkBank rules, any use of data from this corpus must be accompanied by the above reference.

Project Description

This subdirectory contains data that were videotaped, transcribed, and coded in CHAT by Christian Champaud, Chargé de Recherches at the C.N.R.S. (UA 316). The participant of the study is Grégoire. The work was done in collaboration with Grégoire’s family with particular assistance from his mother Dominique. Thanks are due to Catherine Marlot, Danièle Boussin, and Françoise Roland for transcription, rechecking, and typing. This work is still in progress and new data are still being added to the corpus.

Grégoire was the third child of Dominique (mother, born on 3-AUG-46) and Michel (father, born on 5-APR-48, whose usual name is Kôfy). Grégoire was born on 28-APR-86. His elder brother, Adrien, was born on 21-APR-80. The other brother, Victor, was born on 16-JUN-83. Grégoire’s parents lived in Paris, France, and the only spoken language in the family was French. Both parents had college degrees. Dominique was a professor of French, who left her job in order to take care of her family. She had experience in developmental psycholinguistics (investigations conducted with Laurence Lentin) and in linguistics. Michel was a professor of German at the University of Paris III. The socioeconomic status of Grégoire’s family can be characterized as upper-middle-class.

In French, when a word begins with a vowel, this leads in some cases to the disappearance of the final vowel of the preceding word: “l’ami” and not “le ami.” The vowel -e is elided, and in the spelling, the two words are linked together by an apostrophe. This is the case for a determiner preceding a noun, or for a clitic pronoun preceding a verb, and so forth. It is important to add a space after the apostrophe in these cases. It allows users to make searches for some determiners or articles, for pronouns, and for the words that follow, as whole words. It allows also to include them in frequency counts, or to take them into account for computing the MLU in words. This decision implies that, in French CHAT transcriptions, the strings c’, d’, j’, l’, m’, n’, qu’, s’, t’, and y’ must be obligatorily followed by a space. Presently, this has only been done with some transcriptions.

In hyphenated words like abat-jour, the hyphen (-) must be replaced by a plus (+): abat+jour, in order to avoid confusions with suffixes. In French, the hyphen symbol is sometimes used between words, like in est-ce que; in these cases, the hyphen symbol must be omitted and replaced by a space to yield est ce que to ensure that all words are included in frequency counts.

In order not to create an unreadable transcript and in order to avoid overinterpretation often inherent in error analysis, a decision was made to rely on the %pho tier for the definitive form of words. It is then possible to revise interpretations on the main line without returning to the videotape.

The corpus is composed of 10 files recorded roughly monthly from age 1;9.18 to 2;5.27. An additional directory called GREGX contains observations and notes of the parents or investigator for which no audio recording is available and no double-checking can be done. These files begin when Grégoire is 1;9 and run for about 28 months, but the later ones are not yet transcribed.