Syllabification and Allophony

Satisfactory Essays
Syllabification and Allophony


In his complete revision of Daniel Jones's English Pronouncing

Dictionary (1977), Gimson did away with the hyphen symbol which Jones

had considered 'necessary ... [as] a means of showing "syllable

separation" ... in all circumstances where the absence of suitable

marking might lead to ambiguity in the interpetation of a phonetically

transcribed word' (Jones 1963: xxvi). Thus in the twelfth edition, his

last, Jones transcribed toe-strap as /ˈtəʊ-stræp/ but toast-rack as

/ˈtəʊst-ræk/. In Gimson's revision these words appear simply as

/ˈtəʊstræp, ˈtəʊstræk/. While recognising that 'the situation of the

syllable division (juncture) has implications for the duration and

quality of the sounds involved', Gimson justifies his decision on the

grounds that 'such divisions and their implications for pronunciation

are generally evident from the orthography and from the meaningful

segmentation (morpheme boundaries) of the word' (Jones 1977: xiv).

I think this decision was unfortunate. Not only did it mean the

removal of information which is undoubtedly part of the specification

of a word's pronunciation, but it also made it impossible to show

competing pronunciations that differ only in syllabification. Jones

was able to indicate that the word teaspoon is commonly pronounced in

RP as if it were monomorphemic: he gave /ˈtiËspuËn/ as the first

pronunciation, /ˈtiË-spuËn/ only as a less common variant. Under

Gimson's revision this information disappeared from the dictionary.

Another of Jones's examples is shellfish /ˈʃel-fiʃ/, where the /l/ 'is

treated as if it were final and is consequently pronounced long',

while the absence of any hyphen in selfish /ˈselfiʃ/ implies that the

/l/ 'is not so treated, but is short' on account of the presence of

the following /f/ (1963: xxvii). In Gimson's revision these two words

are transcribed as if they were perfect rhymes (which they are not):

/ˈʃelfɪʃ, ˈselfɪʃ/.

Both shellfish and selfish contain internal morpheme boundaries. Yet,

as Jones implicitly recognised, it is not necessary to give any

explicit notational recognition to the boundary between self- and -ish

in selfish.
Get Access