Andrew Radford - Linguistics An Introduction [Second Edition]

Exercises

1.

Below is a sample of words from the first stages of development of

Amahl Smith. Assuming that the child’s underlying representations

are identical to the adult surface representations, what neutralisation

processes (processes which ensure that Amahl does not make a dis-

tinction which is made in the adult system) affect Amahl’s speech at

this time? (The transcriptions have been simplified slightly.)

word

adult pronunciation

child’s pronunciation

apple

/apl̩/

/ɛbu/

bath

/bɑːθ/

/bɑːt/

brush

/bɹʌʃ/

/bʌt/

bus

/bʌs/

/bʌt/

caravan

/karavan/

/gawəwan/

church

/ʧəːʧ/

/dəːt/

dark

/dɑːk/

/gɑːk/

feet

/fiːt/

/wiːt/

finger

/fɪŋgə/

/wɪŋə/

flower

/flaʊə/

/wawə/

John

/ʤɒn/

/dɒn/

knife

/naɪf/

/maɪp/

leg

/lɛg/

/gɛk/

light

/laɪt/

/daɪt/

nipple

/nɪpl̩/

/mɪbu/

other

/ʌðə/

/ʌdə/

sing

/sɪŋ/

/gɪŋ/

Child phonology

107

snake

/sneɪk/

/ŋeːk/

sock

/sɒk/

/gɒk/

stop

/stɒp/

/bɒp/

table

/teɪbl̩/

/beːbu/

taxi

/taksɪ/

/gɛkiː/

uncle

/ʌŋkl̩/

/ʌgu/

write

/ɹaɪt

/daɪt/

yes

/jɛs/

/dɛt/

zoo

/zuː/

/duː/

2.

Neil Smith, Amahl’s father, uses the following data to argue that his

son’s phonological processes must apply in a strictly defined order.

State the processes in as general a form as possible. Then show why,

when so stated, they must apply in a set order (assume that the child’s

underlying forms are identical to the adult surface forms):

bottle

/bɒkəl/

colour

/kʌlə/

gentle

/dɛŋkəl/

gollywogs

/gɒlɪwɒgd/

kennel

/kɛŋəl/

kettle

/kɛkəl/

metal

/mɛkəl/

muzzle

/mʌdəl/

nice

/naɪt/

nose

/noːd/

nozzle

/nɒdəl/

pedal

/pɛgəl/

pencil

/pɛntəl/

pickle

/pɪkəl/

puddle

/pʌgəl/

sew

/təu/

shoe

/tuː/

tassel

/tatəl/

television

/tɛlɪwɪdən/

whistle

/wɪtəl/

zoo

/duː/

3.

Marlys Macken has argued that Amahl Smith has actually misstored

the pronunciation of a word such as puddle and represented it not with

the adult pronunciation but as /pʌgəl/. If this were the case, how would

it affect your conclusions in exercise 2?

4.

Here are two sets of words from an early and a later stage of Amahl

Smith’s development. Formulate two syllable templates, one for each

of the two sets of data. Comment on the differences between the

templates. How do the templates account for the child’s data?

word

adult pronunciation

early stage

later stage

ant

/ant/

ɛt

ant

black

/blak/

pak

blak

break

/bɹeɪk/

peːk

bɹeɪk

child

/ʧaɪld/

taɪl

taɪld

clean

/kliːn/

kiːn

kliːn

count

/kaʊnt/

kaʊt

kaʊnt

drink

/dɹɪŋk/

kɪk

dɹɪŋk

friend

/fɹɛnd/

wɛn

fɹɛnd

hand

/hand/

ɛn

and

hold

/hould/

uːd

uːld

jump

/ʤʌmp/

tʌp

dʌmp

lunch

/lʌnʃ/

lʌt

lʌnt

mend

/mɛnd/

mɛn

mɛnd

108

sounds

monkey

/mʌŋkiː/

mʌgiː

mʌŋkiː

pencil

/pɛnsɪl/

pɛtəl

pɛntəl

Smith

/smɪθ/

mɪt

mɪt

snake

/sneɪk/

ŋeːk

neɪk

spider

/spaɪdə/

paɪdə

paɪdə

spring

/spɹɪŋ/

pɪŋ

plɪŋ

stamp

/stamp/

tap

thamp

stroke

/stɹouk/

koːk

tɹoːk

swing

/swɪŋ/

wɪŋ

wɪŋ

think

/θɪŋk/

kɪk

thɪŋk

5.

Below are two sets of words from different stages in Amahl Smith’s

development.

(a) Describe the syllable template for the child at each stage.

(b) The words at Stage A show two phonological processes that affect

consonants, one of which affects only final consonants at Stage A

in particular circumstances. Describe these processes in words.

(c) What crucial difference between Stage A and Stage B might

account for the change in the pronunciation of the final consonants

between the two stages?

Stage A

Stage B

quick

kɪp

kwɪk

queen

kiːm

kwiːn

squeeze

kiːb

kwiːz

quite

kaɪp

kwaɪt

twice

daɪp

twaɪs

win

wɪn

wɪn

sweet

wiːt

swiːt

spoon

puːn

spuːn

7

Processing sounds

There are two aspects to the real-time processing of language in which we all indulge on a day-to-day basis. One is hearing what others say to us, or in the case of written language and sign languages, seeing what others are saying to us. This is the problem of speech perception, and a fundamental part of it for spoken languages is the recognition of speech sounds. The other is producing language ourselves, speech production. For spoken varieties of language, this includes the problem of control of the muscles of the vocal tract (lungs, throat, tongue, lips) responsible for making the sounds. For sign languages, it is the problem of control of movements of the hands and face. In psychology, the organisation of movement is referred to as motor control.

Speech perception

Suppose you are singing a note on a certain pitch. If you wish to sing

a different note, one option you have is to shift to the new note gradually and continuously (you can also jump straight to it, but this option doesn’t concern us here). This indicates that the pitch of the human voice, determined by the rate at which the vocal cords vibrate, admits of any number of gradations. Now contrast this with someone playing two notes on a piano. A piano has a finite number of discrete notes, and as a consequence it isn’t possible to play a note between C and C#; it is, however, perfectly feasible to sing such a note.

What are speech sounds like? Do they gradually shade into one another like the notes we sing, or are they discrete like the notes of a piano? If we recall our descriptions of the way speech sounds are produced in section 2, we should be immediately attracted by the former possibility. Take place of articulation and the difference between, say, a dental and an alveolar sound. The former requires contact between the tip of the tongue and the upper teeth, whereas the latter requires contact between the tip of the tongue and the alveolar ridge. But the space between the bottom of the upper teeth and the back of the alveolar ridge is a continuous space and the tip of the tongue can make contact at any of the infinite number of points in this space. This suggests that the shift from [t̪] to [t] or from

[s̪] to [s] will be gradual and continuous rather than discrete. Or consider vowel sounds and the front/back and high/low axes, which are fundamental in categoris-ing these sounds. Given any two points on either of these axes, there will always 109

110

sounds

be another point between them, suggesting that the shift from a high to a mid to a low vowel or from a front to a central to a back vowel will again be gradual and continuous.