Advances in Sociophonetics

120 Rosalind A. M. Temple

0

I went back

 w   m  b a k

Time (s) 0.4695

Figure 8. Spectrogram showing extract from and then I went back (63); female speaker.

(63) and then I went [wɛ ̰m̰] back to work again

Singleton alveolars assimilate fairly frequently to following bilabials and velars,

as in (64)–(66), although this seems to be limited to certain individual speakers

and there are plentiful examples in preceding sections of non-assimilated tokens.

This shows, nevertheless, that regressive assimilation of alveolars is a feature of

this variety of English.

(64) the next morning they were all brought back [bɹɔːp ̚baʰkʰ] again

(65) they were really like sad people [sab ̚pipl ̩] straight up

(66) and my leg could move [kʰʊb ̚muːv]

The absence of assimilation in (t,d) tokens is in fact not so surprising when the

phonetic details of the data are considered. There are clearly non-assimilated alve-

olar stop articulations illustrated in §2.1, but there are many tokens where it was

impossible to determine the place of articulation of the (t,d) consonant because of

the absence of formant transitions into and out of the closure, due to the presence

of the preceding and following consonants. An example with preceding /l/ is given

in (67) and Figure 9 which is acoustically and auditorily ambiguous.

(67) we’ve been told by [tʰɔːld ̚baː]/[tʰɔːlb ̚baː] that many people

Glottally reinforced tokens are equally difficult to identify even in singleton con-

sonants, as illustrated by (3) above, which is reproduced here for convenience: