Request for Advice on Using CHAT and CLAN to code codeswitching on a phrasal level

whenyouonlyhaveone at whenyouonlyhaveone at
Thu Feb 7 23:42:55 EST 2019


My name is Wong Jin Yi and I am a research assistant based in Singapore. I 
am under a linguistics lab run by Suzy Styles and Luca Onnis of Nanyang 
Technological University. 

I am currently trying to prepare some transcripts for uploading into 
CHILDES. The transcripts are being coded via CHAT. However, I have realised 
that the transcription convention of CHAT doesn't allow me to code 
codeswitching on any level above the individual word. 

This convention is not going to work for Singaporean Colloquial English, 
because Singaporean Colloquial English (henceforth known as SCE) is a 
creole which makes use of english as its lexifier language and has much of 
its grammatical and syntactic structures derived from Chinese 
Languages/Malay languages. As a result of this, using CHAT's existing 
coding convention for codeswitching would not be able to code for SCE, 
because the words being used are clearly English in origin, and one would 
only be able to grasp the difference if the entire phrase is being marked. 

For example, here is a sample of a phrase in SCE and how it would be marked 
under the current conventions. 

*MOT: and no i've got eight, and now i've got seven only!

In this example, SCE sentence structure is apparent in the unusual way in 
which only is used at the end of the clause instead of the beginning, as it 
would be in standard English. By CHAT convention, one would mark the word 
"only" with @s:sce, however, that would lose the essence of what sce is 
entirely, notwithstanding the fact that the word "only" is undoubtedly an 
English lexical form. 

*MOT: sit down.

*MOT: aurelia sit down, <mama read to you at s:sce> [x 1] come.

The above fragment also gives an example of SCE speech patterns, in this 
case the non-usage of the future marking article <will>, which would 
usually be used in standard English. Again, in this case, it is unclear how 
it would be possible to mark it as codeswitching according to CHAT 
conventions, since all the words used are undoubtedly standard English 

Currently, going with CHAT convention, the best way I have found to note 
these instances is as follows:

*MOT: aurelia sit down, <mama read to you at s:sce> [x 1] come.

This is the best way I have found that gets past the Check function i.e. 
doesn’t give me an error message when I run Check. However, I’m not sure 
regarding how well this format allows me to run a variety of commands in 
CLAN using this transcript. Would anyone who has encountered similar 
problems previously be able to advise me on this? I have tried ELAN, but 
that is also not ideal, considering that i will have to manually code the 
timings on each utterance before conversion. 

As a result of the above problem, I would like to ask, if it is possible, 
to include a means in CHAT that can more clearly show codeswitching at a 
phrasal level?This would be very helpful in our understanding of how 
language mixing occurs in Singapore. The solving of this problem would also 
be likely to contribute to the using of CHAT to code for creoles, 
especially those that use a known lexifier. I believe this would definitely 
help in research of creole languages. 

In addition, would anyone be able to recommend a means to add Singapore 
Colloquial English as a codable language in the CLAN metadata? This would 
also be greatly helpful in our efforts to code our transcripts. 

Thank you very much, and I hope to hear from you all soon! :) 


Wong Jin Yi


You received this message because you are subscribed to the Google Groups "chibolts" group.
To unsubscribe from this group and stop receiving emails from it, send an email to chibolts+unsubscribe at
To post to this group, send email to chibolts at
To view this discussion on the web visit
For more options, visit
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <>

More information about the Chibolts mailing list