- Reading time: 3 minutes
- Price: Free download
- Published: 4th October 2026
- Word count: 629 words
- File format: Text
Personal statement example
The voice note lasted forty-three seconds, and my cousin switched between isiZulu and English eleven times. I know because I transcribed it, with her permission, as the first item in what became my honours project. She was telling her sister about a job interview, and the English arrived in predictable places: 'shortlisted', 'panel', 'competency-based'. What interested me more were the switches that had nothing to do with vocabulary gaps. She moved into English to quote the interviewer, then back into isiZulu to comment on him, and the change of language did the work that a change of tone of voice might do in a monolingual retelling.
My project analysed eighty-six voice notes contributed by twelve friends and relatives aged eighteen to thirty. I chose voice notes because they sit between speech and writing: they are spoken, but people can rerecord them, and several participants told me they did. I coded each switch by position and apparent function, using Carol Myers-Scotton's Matrix Language Frame model to describe the grammatical structure and drawing on Gumperz's idea of conversational code-switching for the functional categories. The structural coding went smoothly; the functional coding did not, because many switches plausibly served two purposes at once. I ended up recording a primary and a secondary function and reporting where my categories overlapped, rather than forcing a single label. My supervisor's comment that the overlap was a finding in itself is the reason I want to continue.
At postgraduate level I would like to look at a question the project raised but could not answer: whether the conventions of code-switching in recorded messages differ from those in live conversation, given that speakers can plan and revise. Answering it properly would need a larger and more systematically sampled corpus, a clearer ethics framework for collecting private messages, and stronger quantitative skills than I currently have. I took one statistics course as an elective and can run basic analyses in R, but I know that is a starting point rather than a toolkit.
Since graduating I have worked full-time in an insurance call centre, handling claims enquiries in isiZulu and English. I did not take the job for research purposes, but it has sharpened my ear. Our scripts are written in English, and I have noticed how often colleagues translate the formal phrasing into isiZulu loosely for callers while keeping certain English terms, such as 'excess' and 'claim number', exactly as written, perhaps because those terms carry institutional weight. I keep notes on this privately and do not record calls, but it has given me a sense of how language choice works under time pressure and in front of an audience of supervisors and quality checkers. The job has also taught me to work steadily through repetitive tasks, which I expect will be useful when coding a large dataset.
On Saturdays I coach an under-twelve netball team at our local community hall. This has nothing to do with linguistics, although I have noticed that the girls invent nicknames for set plays in a mixture of languages. Mostly it has made me better at explaining things clearly, organising other people, and turning up every week whatever the weather.
I am applying for a coursework and dissertation master's because I want structured training alongside an independent project. I need a firmer grounding in phonetics, which I studied only at introductory level, in sociolinguistic research methods, and in the theory of language contact in multilingual societies, particularly in southern Africa, where much of the classic literature was built on other contexts. I would bring a working dataset, careful transcription habits, and experience of collecting language data from people who trust me. I hope to leave able to design a study that answers my question properly.