In the first post in this series, we looked at structuring contributions in the speaking test, giving full, developed answers. In the second post, we are going to look at the language we use in the speaking exam.
In all levels of Cambridge exam, from YLE Starters up to Proficiency, there is, logically, a specific mark for pronunciation. When we talk about this aspect of language, there is a tendency to focus on accent, and specifically whether the candidate is capable of reproducing a particular native speaker accent. However, the examiner is not measuring the non-native candidate against a native-speaker norm. The emphasis is instead on reproducing the individual sounds, intonation and stress patterns of English in a way which does not impede comprehension. While higher levels of exam require the candidate to be ‘intelligible’, lower levels, such as KET or PET allow for a fairly intrusive L1 accent which may make comprehension more difficult at times.
The above video, from the Cambridge English TV channel on You Tube, focuses on word stress, and how a change in stress may mark a change in meaning, and so impede understanding if not reproduced accurately. This word stress can be realised in any accent, native or non-native. Similarly, sentence stress is not dependent on accent. English is traditionally a stress-timed language, as opposed to a syllable-timed language like, for example, Spanish. This means that a successful candidate should be able to place the stress on the correct syllables within an utterance, and at higher levels (CAE or CPE particularly) the candidate should be able to use stress to suplement the meaning of the utterance.
Another important aspect which can be reproduced accurately whatever the accent of the candidate is intonation. A successful candidate should be able to use rising and falling tones within the utterance in order to indicate the internal structure of the utterance, usually rising at the end of each element of a list, for example, or at breaks in an utterance usually represented graphically by a comma, then falling at the end of an utterance, represented graphically by a full stop.This can actually have more effect on understanding at times than accuracy in individual sounds. Several years ago, I examined a PET candidate who reproduced individual sounds acurately, but whose intonation was so wrong that he was almost impossible to understand.
Clearly examiners must also focus on the accurate reproduction of individual sounds. However, different accents imply different versions of individual sounds. Here too, the important thing is to be understood with relative ease, avoiding as far as possible L1 intrusions. It doesn’t matter if the student pronounces ‘Tomato’ as in British English or in American English, but if they say ‘city’ as ‘thity’ (a typical Spanish error, since in Spain, the ‘ci’ and ‘ce’ are pronounced ‘thi’ and ‘the’), that impedes understanding, and so is marked down.