CHILDES Finnish Kirjavainen-MPI Corpus

Minna Kirjavainen English Language and Linguistics University of the West of England minna.kirjavainen-morgan@uwe.ac.uk

Browsable transcripts

Download transcripts

Link to media folder

Citation information

Kirjavainen, M., Kidd, E. & Lieven, E. (2017). How do language-specific characteristics affect the acquisition of different relative clause types? Evidence from Finnish. Journal of Child Language, 44(1), 120–157. doi:10.1017/S0305000915000768

Additional references and publications

Kirjavainen, M. & Lieven, E. (2011). The acquisition of relative clauses in Finnish; the effect of input. In E. Kidd (Ed.) Acquisition of Relative Clauses: Processing, Typology, and Function. pp. 107-139 (Ch.5) John Benjamins.

Kirjavainen, M., Nikolaev, A. & Kidd, E. (2012). The effect of frequency and neighbourhood density on the development of past tense morphology in Finnish children. Cognitive Linguistics 23(2), 273-315. https://doi.org/10.1515/cog-2012-0009

Surakka, M. & Kirjavainen, M. (2025). To what extent does children’s temporal expressions actually refer to time? An investigation into temporal and discursive functions of adverbs. Journal of Child Language, 52(3), 498-531. 2025;52(3):498-531. doi:10.1017/S0305000924000114

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 corpus consists of care-giver – child interactions of one monolingual Finnish-speaking female, ‘Piia’.

Piia is a monolingual first-born child born in Kotka, South-East Finland. The family and the wider family are all native speakers of Finnish and the regional dialect where they live. SES-wise (e.g., income, local area where they live) the family is middle class. At the time of the recordings, both parents had tertiary education, but neither had been to university.

Piia was audio recorded by her parents for approx. three hours per week between the ages of 1;7.21 and 4;0.13 (278 recordings, approx. 260 hours of data in total) in 2006-2008. The interaction during the recordings was completely spontaneous, consisting of everyday family interactions (e.g., meal-times, bath time, arts and crafts, joint book reading, play time). Both parents were regularly present during the recordings. However, there is significantly more data for the mother than for the father, because the father worked full-time while the mother didn’t work for the duration of the recording period. Relatives (aunt, grandparents) and friends were occasionally present during the recordings. Piia’s younger sister was born when Piia was 2;4 after which the younger sister was often present during the recordings, but rarely took part in the interactions (due to her early developmental stage at the time).

The transcription was done manually. This was carried out largely by native Finnish speakers studying Finnish language at the University of Jyväskylä. A small number of the recordings were transcribed by other native Finnish speakers. Each utterance has been transcribed as it was spoken in dialect on the main-line. The corresponding standard Finnish form of the utterance has been transcribed on the %flo-line.

Kirjavainen, as a native speaker of the corpus family’s dialect, checked the transcription quality of each recording by listening to approx. 10-15% of each recording and correcting the transcription lines where necessary. If there were systematic transcription errors in a given recording, she would check the whole recording. In addition, she listed all speech transcribed as inaudible (i.e. xxx) and amended where possible.

Acknowledgements

This work was funded by the Max Planck Institute for Evolutionary Anthropology, Leipzig grant awarded to Minna Kirjavainen via the Max Planck Child Study Centre, the University of Manchester.

I am beyond grateful to the family for their time, effort and willingness for incorporating the frequent recordings into their family life for a approx. 2 ½ year period without any financial compensation.

I would like to thank the team of Finnish-language students and other transcribers, who took part in the lengthy process of manually transcribing the main-line and %flo-line line for the approx. 260h of data, Lea Nieminen (University of Jyväskylä) for her help with creating the transcription protocol and training the transcribers at Jyväskylä, Jeannine Goh and the other Max Planck Child Study Centre staff (University of Manchester) who helped with the practicalities related to the recordings, data storage and in resolving technical issues and Elena Lieven for her support throughout the project.

I would also like to thank Brian MacWhinney for running the CHECK function and for the creation of the %mor and %gra lines.