SCAN0086 (2)

260 Introduction to the Pronunciation of English

seems m

it seems impossible

wants, have in

he wants to have it now

forty in

it was in nineteen forty-nine

(cf. fifty in

it was in nineteen fifty-nine

where the stressed /i/ of fifty may not have the same qualitative prominence as the stressed /o:/ of forty)

said, go in

they all said they would go

(where the vowels of both cases of they are łikely to be weakened)

understand, say in *

does he always understand what you say?

(For variations of word accentual patterns in connected speech, see §11.03.)

10.04 Quantity and Quality

(1) Accented words.—Vowels and continuant consonants in ac-cented syllables which form the hub of a rhythmic group are shortened according to the number of unaccented syllables (especially following) in the group. Thus, the /ai/ of /taid/ (tide) shows progressive shortening in such rhythmic groups as tidy, tidily, she tidied it, etc.; the /a/ and /m/ of /kAm/ (come) are similarly shortened in comfort, comfortablc, come for me, circumference. Or again, a comparable phonemic seąuencc will have slight variations of sound length according to the division into rhythmic groups: cf. aboard a liner /a'bo:d 3'laino/ and a border-liner /a'bo:d3 .laino/—/o:/ being slightly shorter in the second case; minor ojficial /'mainar a'fi.fl/ and minę are ojficial /'mam oroTijl/—/ai/ being longer in the second case. Such variation of rhythmic grouping, involving changes of ąuantity, constitutes a reality for the speaker, bul it is doubtful whether slight modifications of this kind are markedly significant to a listener, sińce the choice of meaning for such similar phoneme seąuences is normally determined by the context, such cues as are provided by rhythmic variation of ąuantity being redundant.

(2) Unaccented words.—A morę marked effect is that which characterizes the ąuantity and ąuality of unaccented words. Content words (monosyllables and polysyllables) generally have in connected speech the qualitative pattern of their isolate form and therefore retain some measure of qualitative prominence even when no pitch prominence is associated with them and when they are relatively unstressed, e.g.:—

why, try, jump in _e    o    «

why don’t you try and jump?

nuisance in    .

what a nuisance

thirty-nine in    "'j

thirty-nine and eleven

But many form words have two or morę qualitative and quantitative patterns according to whether they are unaccented (as is usual) or accented (in special situations or when said in isolation). As compared with the accented realizations of these words (the ‘strong’ forms), the unaccented (‘weak’ form) varieties of these words show reductions of the length of sounds, obscuration of vowels towards /o, i, u/, and the elision of vowels and consonants. The following list of examples pre-sents the most common of these words, first in their unaccented (normal) weak forms and secondly in their less usual¹ accented strong form:—

	Unaccented	Accented
a	N	/«/
am	/m, om/	/aem/
an	/n, an/	/aen/
and	/and, nd, on, n/	/aend/

^lrrhe following 42 items occur in the first 200 most common words in connected speech: the, you, I, to, and, a, that, we, of, have, is, are, for, at, he, but, there, do, as, be, them, will, me, was, can, him, had, your, been, from, my, or, she, by, some, her, his, us, an, am, has, shall. It is significant that of these the following 19 have over 90% unaccented occurrences with a weak form: at, of, the, to, as, and, or, a, his, an, but, been, for, her, we, be, shall, was, them. Other items of rarer occurrence, such as than, only exceptionally occur with the strong form.