Wiktor Jassem's short article on rhythm (‘Indication of speech rhythm in the transcription of educated Southern English’) was published in 1949 in Le Maître Phonétique, the ancestor of today's Journal of the International Phonetic Association. The author was a young man aged 27 at the time, who was working on a longer treatment of the intonation of conversational English which was to be published three years later (Jassem Reference Jassem1952), just 60 years ago this year. Despite his youth, and despite the fact that he was not a native speaker of English, Jassem did not hesitate to take on giants, and to suggest in this paper that his treatment of rhythm is superior to that of both Henry Sweet and Harold Palmer, as well, he adds in the 1952 book, as that of his mentor Daniel Jones.
Jassem illustrates here what he considered to be a crucial aspect of English pronunciation which in later work he calls the anacrusis, consisting of any unaccented syllables which are not attached to a preceding Narrow Rhythm Unit and which are generally pronounced more quickly than other syllables. Although in the example:
summer dresses /ˈsʌmə ˈdresiz/ vs. some addresses /ˈsʌm əˈdresiz/
the space he introduces as a notational device corresponds to the word boundary, this is not always the case. An example proposed by Scott (Reference Scott1940),
Take Greater London vs. Take Grey to London
could be transcribed, using Jassem's proposal, as
/ˈteɪk ˈgreɪtə ˈlʌndn/ vs. /ˈteɪk ˈgreɪ təˈlʌndn/
where the spaces are no longer identical to those of the orthographic version.
Jassem's notation can even make a distinction which is not made in normal orthography. The sentence
He bought her chocolates.
can be interpreted in two ways, depending on whether her is taken to be an indirect object (i.e. He bought chocolates for her) or whether it is a possessive determiner (i.e. He bought the chocolates she was selling). Jassem's model predicts that in the first of these, her will be in the same Narrow Rhythm Unit as the verb bought, but that in the second it will be part of the anacrusis, so that the two interpretations would be transcribed respectively:
Jassem's notational device, which he uses in the last paragraph of his article, was unfortunately not generally adopted by English phoneticians, although it was used in O'Connor's (1976) Better English Pronunciation, although without reference to Jassem's work. Jassem's distinction between a rhythm unit and an anacrusis makes empirical predictions about English speech which later research (Jassem, Hill & Witten Reference Jassem, Hill, Witten, Gibbon and Richter1984, Hirst & Bouzon Reference Hirst and Bouzon2005) has shown to be largely justified and which are still the subject of ongoing research.
Wiktor Jassem celebrated his 90th birthday on 11th June 2011, and the republication here of one of his first articles is intended as a tribute to this great phonetician, alongside the publication of a collection of articles in his honour (Gibbon, Hirst & Campbell 2010), which includes a complete bibliography and an updated version of the tribute to him written ten years ago by Jack Windsor Lewis (Windsor Lewis Reference Windsor Lewis2003).
Indication of speech rhythm in the transcription of educated Southern English
A number of languages, Southern English among them, have been transcribed phonemically and such a transcription, also called ‘broad transcription’, has the twofold advantage of both showing the actual phonological structure of the language represented and being convenient for the purpose of practical teaching. It seems however that, at least in the case of educated Southern English, the current ‘broad transcription’ is, in one respect, inconsistent, namely, in the way it employs spacing. In some cases a space is indicative of phonetic or phonological differences. Thus some addresses ˈsʌm əˈdresiz and summer dresses ˈsʌmə ˈdresiz are different phonetically (ʌ and m being longer and ə being shorter in the former collocation than in the latter) and phonologically (the so-called juncture occurring in the first example between the phonemes m and ə, and in the other between the phonemes ə and d(r)).
Comparing, however, the natural pronunciation of such pairs as can form and conform, offence and a fence, in tense and intense, we find that the current transcription: kən foːm and kənfoːm etc. appears to suggest differences that do not exist. Can form and conform are phonetically identical and should also be considered phonologically identical.
Now, the spelled form: an old English grammar means: ‘an old handbook of English grammar’ and: an Old English grammar means ‘a grammar of Old English’. In a transcription which employs stress-marks the difference in pronunciation can be shown thus: (1) ən ˈould ˈiŋgliʃ ˈgramə, (2) ən ˈould iŋgliʃ ˈgramə. The lack of rhythmical stress on iŋgliʃ causes all the syllables of the sequence ˈould iŋgliʃ to be shorter than in ˈould ˈiŋgliʃ. Similarly, the unstressed syllables are you cause considerable shortening of the stressed syllable where in Where are you going? as compared with But where? It might appear that any unstressed syllables shorten a preceding stressed one, whether in the same ‘word’ or not. But this is not so. aː is shorter in to harden steel than in hard and very heavy. In fact one or more unstressed syllables may or may not influence the length of a preceding stressed syllable. They do if they belong to the same rhythmical unit; they do not if they belong to the following rhythmical unit. In the sentence I ˈheard a most pe ˈ culiar ˈsound (with high pitch on heard) there are three unstressed syllables after heard and yet that syllable ˈhəːd is noticeably longer there than in the sentence I heard him sing, where it's only followed by one unstressed syllable. The rhythmical juncture occurs after heard in the former and after him in the latter case. The rhythmical junctures are (1) ai ʲhəːd | ə moustpiˈkjuːliə | ˈsaund, (2) aiai ʲhəːdim | siŋ. Such a transcription would be both scientifically strict and practical in the teaching of English to foreigners whose own language may be governed by different rules of speech rhythm. The learner has to realise that: (1) unstressed syllables preceding rhythmical stress within a unit (if there are any) must be pronounced as short as possible, (2) the amount of time taken for the utterance of each of those parts of rhythm units that come between stress and juncture is approximately the same throughout a passage pronounced at a certain speed, (3) there is a strong tendency for that amount of time to be equally divided between any number of syllables that may occur: between stress and juncture, so that the length of such syllables tends to be inversely proportional to their number.
It should be noticed that a transcription like the one I suggested is different from both that at one time advocated by Henry Sweet, who, in some of his primers, linked together anything that occurs between two stresses and that at a later date proposed by Harold Palmer, who was once in favour of linking all words together that occur within a breath-group.
A transcription using spacing for purposes of indicating rhythm and juncture may weaken the association between transcript and historical spelling but it's very likely to increase a foreign learner's capacity of recognition of actual speech units and consequently increase his ability of imitating the peculiar rhythm of English speech.
W. Jassem
