Minds and Computers : An Introduction to the Philosophy of Artificial Intelligence

Figure 16.2
Pronunciation chart for English obstruent phonemes.

---

The other axis of classification is the place of articulation, which
describes the primary articulatory apparatus involved in their pro-
duction.
Bilabial phonemes are those which involve the use of both lips in
their production. Labiodental phonemes involve placing the upper
teeth against the bottom lip. Interdental phonemes involve placing
the tongue between the teeth.
The alveolar ridge is the hard raised ridge that is just behind the
upper teeth. Alveolar phonemes are those whose production involves
placing the tongue against or near the alveolar ridge.
Continuing behind the alveolar ridge along the upper jaw is the
hard palate. Alveopalatal phonemes are produced by placing the
tongue behind the alveolar ridge and against, or close to, the hard
palate.
If you run a finger along the roof of your mouth towards the back
of your throat to the point where the hard palate ends, you’ll feel the
squishy soft palate. The soft palate is also known as the velum. Velar
phonemes are those which are produced by placing the tongue near,
or against, the velum.
The glottis is the space between the vocal chords. The glottal frica-
tive /h/ is produced by restricting the passage of air through the
glottis, but without vibrating the vocal cords.
Speakers of some dialects of English – such as the London cockney
dialect – produce a glottal stop rather than the alveolar stop in certain
phonemic contexts (such as in ‘bottle’ or ‘sorted’). This is not a dis-
tinct English phoneme but an allophonic variant of the phoneme /t/.
We will learn more about allophones shortly, but first let’s taxonomise
the sonorant (vowel) phonemes.
16.2 SONORANT PHONEMES
All sonorant phonemes are ipso facto voiced. They also all have the
same manner of articulation – they are open sounds. In other words,
the passage of air through the articulatory apparatus is not impeded
but resonates freely in the oral cavity.
Sonorant phonemes are taxonomised along two axes which
describe the position of the tongue in the mouth during their produc-
tion. One axis corresponds with the height of the tongue and the other
corresponds with the part of the tongue that is raised or lowered.
Unlike the obstruent phonemes which admit of clear discontinu-
ities between phonemes given the distinct articulatory apparatus
involved in their production, the sonorant phonemes are placed on a
 
167

---

continuum, since that the possible positions of the tongue in the
mouth are continuous, not discrete. As such, vowel phonemes fall
within a two-dimensional space of possible vowel sounds.
Consequently, we identify cardinal vowels within this vowel space
as shown in Figure 16.3. The vowel space is continuous from the
highest, most fronted sonorant phoneme, through to the lowest and
least fronted sonorant phoneme.
The cardinal vowels represented in Figure 16.3 are all monoph-
thongs. This means that their production involves a single continuous
tongue position. As well as the monophthong phonemes, however,
there are also diphthong phonemes. Diphthongs are phonemes whose
production begins at one of the cardinal vowel positions but moves,
during the production of the sound, towards the tongue position of
another of the cardinal vowels.
The pronunciation chart provided in Figure 16.4 represents the
vowel sounds in the dialect of Australian English that I speak.
Note, however, that vowel pronunciation will vary between dialects
of English, particularly with respect to the production of diph-
thongs, but also with respect to some monophthongs. Consequently,
some of the triples in Figure 16.4 which all rhyme in my dialect of
English may not rhyme in yours, depending on where you learned to
speak English. Also, speakers of dialects of English other than
Australian – particularly American dialects – are likely to produce
monophthongs in place of some of the diphthongs.
168
  