CHILDES Afrikaans Stellenbosch Corpus

Ondene Van Dulm
Department of General Linguistics
University of Stellenbosch


Cecile le Roux
Organization and Behavior
Carnegie Mellon University in Qatar


Participants: 2
Type of Study: naturalistic
Location: South Africa
Media type: audio no longer available
DOI: doi:10.21415/T5801X

Browsable transcripts

Download transcripts

Citation information

No citation yet.

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

Project Description

This project was headed by Cecile le Roux, formerly of the Department of General Linguistics of the University of Stellenbosch. Ondene van Dulm collected, transcribed, and coded the data. The aim of the project was the creation of a detailed computerized database on the acquisition of South African Languages. Data were collected for the following :

However, only the data from Chanel and Jean are included in CHILDES.

The data consist of the spontaneous utterances of the children, elicited in environments familiar to the children, at daycare, school or at home. The data collectors included assistants and lecturers of the General Linguistics Department of Stellenbosch University. These people included Ondene van Dulm, Simone Conradie, and Debra Aarons. The utterances were elicited in a free play situation with the younger children, and in conversation with the older children. Video recordings were made of the two toddlers, and audio recordings of the rest of the subjects. Transcriptions were done by the various data collectors, and every transcription was checked for reliability by another member of the team. Only the utterances relevant to the study were transcribed - please see warnings. To date, only the data of the two toddlers, acquiring Afrikaans as L1 has been coded. The remaining data will be coded in 2000.

Chanel Leuvenink. (Permission to use real name.) Born 29/10/96. Female. Only child. Afrikaans-speaking, both parents Afrikaans-speaking. The 32 files for Chanel represent sessions that begin on 30/4/98 at the age of 1;6 and continue every other week to 14/10/99 at the age of 2;11.7. Chanel was recorded by video camera, either at her daycare centre or at home. At the daycare centre she usually sat in a quiet room alone with the investigator, Ondene van Dulm, and looked at books and played with toys. At home, her mother Ria was included in the recordings, and Chanel played freely in the sitting-room, study and her own bedroom. She had a great variety of toys and books. At times, other people were also present for the recordings, including her father, Kimo, members of her extended family, and the daughter of the investigator, Enya.

Jean Kriel. (Permission to use real name.) Born 14/10/96. Male. Younger of two. Older brother 13 years of age. Afrikaans-speaking, both parents and brother Afrikaans-speaking. The 32 file for Jean begin on 30/4/98 at the age of 1;6.15 and continue every other week to 5/10/99 at the age of 2;11;21. Jean was recorded by video camera, at home, at first by the recorder, Ondene van Dulm, and later by his mother. The video camera was simply switched on to record at various times during his daily activities, including dinner time, bath time, play time, story time etc. At times, Jean's older brother Lexie and/or father Lex was also present.

Chanel used the following onomatopoeic forms: moo = koei = cow, woef-woef or woef or woefie = hond = dog, peep-peep = toet = hoot, and ooo = bird song. Chanel used the following child-based forms: doedoe = slaap = sleep, nana = lekker(s) = sweet(s) or nice, babie = baba = baby, Ponia = Tolia = Pretoria, pokkipaai = poppiepaia = porcupine, bollie / gebollie = pooh / poohed. hello = telefoon = telephone, eina = seer = sore, siesa = sies = ugh, oraait = all right, nana = piesang = banana

Jean used the following onomatopoeic forms: woef = hond = dog, aargh = brul = roar, uughh = sleg = bad/yuk. Jean used the following child-based forms: mama = kamera = camera, ba = boom = tree, pang = slang = snake, ku = kous = sock, mys = roomys = ice-cream, Charla = Charlie = Charlie, pillie = pilletjie = pill-dim, pikillie = pilletjie = pill-dim, gaga = vuil (doek) = soiled, Jenna = Jemina = Jemina, ies = iets = something, eina = seerplek = sore, eina = seer (adj) = sore, bottie = bottel = bottle, babie = baba(tjie) = baby, vakie = vakie(s) = tired, trekkie = toutjie = rope-dim, bzz = by = bee, opskim = opklim = climb up, and buks = hit.


The investigators would like to be informed if the data is to be used or published in any form. Users can let us know via e-mail to We would appreciate being sent copies of any articles etc in which the data is mentioned or used.

As the researchers were interested only in various aspects of syntactic development as from the stage of two-word utterances, only such utterances were transcribed and coded. Recording only began when the subject was consistently using two- or more word utterances, and single-word utterances were not transcribed or coded. Utterances of the other participants, which surrounded single-word utterances of the subject, were also excluded from the transcription. Only the contextually relevant utterances of other participants, were transcribed. Also note that cases of suspected ellipsis were not filled in, as the investigators did not feel justified in making assumptions about what the child intended to say.