A shtrange shift
Arecent question from a reader of this column concerns their observation that ‘‘some people can’t say ‘straight’ without inserting an h (‘shtraight’)’’. This is something that linguists have been noticing too, and not just in New Zealand.
The phenomenon involves the pronunciation of the /s/ sound using a tongue position that is further back in the mouth than expected. (Note that I am using / / here to indicate pronunciation.) Linguists call this /s/-retraction. To get a sense of what is involved, produce a /s/ sound and move your tongue to where it would be for the ‘‘sh’’ sound. You will probably find that, as well as a change in position, this also involves a different grooving of the tongue.
This /s/-retraction happens in certain contexts more than in others – it is particularly frequent in the sequences /str/ and /stj/, where /j/ is the sound sometimes called ‘‘yod’’. This is the sound at the beginning of the word yellow and which, in some English varieties, also occurs after /t/ in words like Stewart .In these varieties, /s/-retraction can result in ‘‘Shtewart’’. Former prime minister John Key is a standout exponent of /s/-retraction, with pronunciations such as ‘‘Aushtralia’’ and ‘‘shtudent’’.
As the correspondent’s spelling ‘‘shtraight’’ implies, this retracted /s/ is similar to the sound often written with ‘‘sh’’, as in shop. There is of course no h inserted in the pronunciation, and the spelling ‘‘sh’’ represents a single speech sound. ‘‘Sh’’ is not the only spelling for this sound – think of words ending in ‘‘-tion’’, such as vaccination. Spelling reformers have suggested that English spelling is so daft that fish could be written ‘‘ghoti’’.
Although the origins of this example are unclear, it is often attributed to Irish playwright George Bernard Shaw. This spelling is based on the use of ‘‘ti’’ as in nation for the final ‘‘sh’’ sound, ‘‘gh’’ for /f/, as in enough, and ‘‘o’’ for the vowel, as in the first syllable of women. (I will comment in another column on the widespread NZ pronunciation of women as ‘‘woman’’, but the example still works for such speakers, as the vowel is still the one they would have in fish.)
It is not clear why /s/-retraction occurs, but the fact that it is found in some sound contexts (notably /str/ and /stj/) more than others suggests it may be related to changes in other parts of those sound sequences, such as how /tr/ and /tj/ sequences are pronounced in train and tune, for instance. The /s/-retraction is not unique to New Zealand – an international team of researchers has tracked it in accents of English in Scotland, the United States and Canada.
Nor is it new – recent VUW MA graduate Reuben Sanderson investigated a historical database of NZ English and found evidence of /s/-retraction in recordings of speakers born from the 1950s onwards.
He also noted that it shows up in some other contexts – in /st/ sequences with no following /r/ or /j/. In all of these contexts, the shift from /s/ to ‘‘sh’’ is possibly tolerated because it does not lead to ambiguity. English has no words with ‘‘shtr’’ that would be confused with words with ‘‘str’’; ‘‘shtreet’’ is a variant pronunciation of street, not a different word.
There are historical parallels in a closely related language, German, where most dialects (apart from some northern varieties) shifted a long time ago from initial /sp/, /st/, /sl/, /sm/, /sn/, /sw/ to ‘‘shp’’, ‘‘sht’’, etc. For instance, sprechen (to speak) has a ‘‘shpr’’ onset. So perhaps English /s/-retraction is not so shtrange after all.
Contact us
Got a language query? Email opinion@ stuff.co.nz. Not all queries will be answered.