Grimm Corpus

Angela Grimm
Department of German Linguistics
University of Osnabrück


Participants: 4
Type of Study: naturalistic
Location: Potsdam, Germany
Media type: not available
DOI: doi:10.21415/T5M89T

Browsable transcripts

Phon data

CHAT data

Link to media folder

Citation information

Grimm, Angela. 2006. Intonational Patterns and Word Structure in Early Child German. In David Bamman, Tatiana Magnitskaia & Colleen Zaller (eds.), Proceedings of the 30th Annual Boston University Conference on Language Development, 237–248.

Grimm, Angela. 2007. The Development of Early Prosodic Word Structure in Child German: Simplex Words and Compounds. Universität Potsdam Ph.D. Dissertation.

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

Name Age Range Sessions Sex

All four children are firstborns without siblings, being raised in monolingual German middle-class environments. Their parents either have a university degree or some other kind of professional training. During the recording time, the children were day-cared at home by their mothers. Apart from occasional colds, none of the children suffer from health problems or hearing impairments or have deficits in cognitive and motor development. Also, there were no complications during pregnancy and childbirth.

Regular recordings started as soon as the parents identified the first meaningful words in the speech of their children. Sandra, Eleonora, and Nele produced their first words in September 2002; Wiglaf in October 2002. The recordings took place over a period of 9 - 12 months at the children’s homes in presence of a parent (mostly the mothers) and the investigator.

During the earliest phase when the number of word productions increased very slowly, bi-weekly speech samples were taken for 45 - 60 minutes. As soon as the parents reported a rapid increase of the productive vocabulary (‘vocabulary explosion’), weekly recordings took place for 30 - 40 minutes.23 The mean recording period is 10 months and 6 days; the mean number of recordings containing meaningful speech is 25 per child.

The speech samples were taken as audio recordings using a SONY TCD-D8 DAT-recorder and a SONY ECM-MS957 microphone. The microphone was placed in front of the child and adopted to the child’s position if she or he moved through the room. In order to keep background noises as minimal as possible, the caretakers were asked to remove crackling or other noisy toys before starting the session.

The children were observed during natural interaction with their caretaker(s) and the author, while looking at picture books or playing with toys. As far as possible, the data represent spontaneous productions. The setting was controlled to some degree because, in order to elicit tri- and quadrisyllabic words, five plastic animals were introduced into the spontaneous interaction: Papagei ‘parrot’, Krokodil ‘crocodile’, Elefant ‘elephant’, Kamel ‘camel’, and Giraffe ‘giraffe’. Occasionally, the attention of the child was shifted to other objects to elicit longer object names. If the child was not willing to speak, she or he was asked to repeat particular target words from time to time but imitations did not follow a fixed criterion.

In natural interaction, parents usually pick up children’s word productions and extend (‘Yes, this is a X.’) or expand them (‘Yes, a green X.’). Such parental behaviours helped the identification of a word and for separating it from babbling sequences.