Advances in Sociophonetics

Chapter 7. Ejectives in English and German 195

on the presence of two glottal stops either side of [ə], e.g. [pʊʔə] put a but [pʊɾəʔ]
put it (see also Trudgill 1974; Lodge 1984). In an attempt to investigate this phe-
nomenon instrumentally using electroglottography (EGG), the same speaker was
recorded producing a list of short sentences. Although not of direct relevance at
the time of the experiment, the words back or work had been placed at the end
of several of the sentences. Despite the lexical and syntactic simplicity of the sen-
tences (e.g. She’ll look at it at work), the speaker exhibited a number of dysfluencies
(pauses, false starts, truncations) when producing many of the sentences. This was
most likely due to a considerable discrepancy between her naturally occurring
patterns and the standard-like pronunciation the speaker considered appropriate
to the formality of the recording situation. However, of direct interest to the dis-
cussion here was the realisation of the final plosives in ‘back’ and ‘work’. The plo-
sives were always released, but in approximately one third of cases the plosive was
realised as an ejective. Figure 2a shows an example of an ejective release, Figure 2b
a pulmonically fuelled release. The oscillogram shows the EGG trace, which I will
return to below. Approaching these patterns from a sociophonetic point of view,
it is possible to treat the ejectives as being one part of the standard-like patterns
being produced by this speaker in the context of a formal recording situation, per-
haps again being a correlate of articulatory place enhancement, as Ogden (2009)
has suggested. But even here, it is possible to propose that, although the sentences
may represent structural identity from the analyst’s point of view, there is no guar-
antee that the speaker is treating them this way. So, even if our speaker is produc-
ing a sequence of sentences in a studio setting, it is not legitimate to state that
observable differences in the realisation of final plosives in a series of sentences

5000

k’

0

4000

3000

2000

1000

0

0.9 1 1.1 1.2 1.3

5000

0

4000

3000

2000

1000

0

1.3 1.4 1.5 1.6 1.7

Figure 2. EGG and spectrograms of (a) ejective vs. (b) pulmonically released plosives
from sentence-final tokens of the word back from a Suffolk English speaker.