Second Language Learning and Language Teaching

allophones
phoneme
initial
medial
final 
cluster (CC) etc.
misc
1. /
p/
pin
❒
supper
❒
map
❒
spit
❒
2. /
b/
bin
❒
suburb
❒
rub
❒
bleed
❒
3. /
t/
tip
❒
bitter
❒
pet
❒
sting
❒
4. /
d/
doll
❒
rudder
❒
fed
❒
drain
❒
5. /
k/
cash
❒
tucker
❒
luck
❒
create
❒
6. /
/
goat
❒
rugger
❒
mug
❒
glade
❒
7. /
tʃ/
chew
❒
Richard
❒
rich
❒
8. /
d /
joke
❒
lodger
❒
fudge
❒
9. /
f/
fast
❒
differ
❒
off
❒
flame
❒
10. /
v/
view
❒
river
❒
of
❒
11. /
θ/
thigh
❒
rethink
❒
bath
❒
three
❒
12. /
ð/
then
❒
rather
❒
bathe
❒
13. /
s/
soon
❒
lesson
❒
mess
❒
strain
❒
14. /
z/
zoom
❒
razor
❒
was
❒
sizzle
❒
15. /
ʃ/
show
❒
usher
❒
fish
❒
shrine
❒
16. /
/
genre
❒
measure
❒
rouge
❒
17. /
h/
who
❒
18. /
l/
lip
❒
pillar
❒
hill
❒
plain
❒
19. /
r/
read
❒
direct
❒
far (0)
❒
there is
❒
20. /
m/
mix
❒
summer
❒
aim
❒
dims
❒
21. /
n/
nod
❒
dinner
❒
sin
❒
likes
❒
22. /
ŋ/
banger
❒
sang
❒
finger
❒
23. /
j/
yes
❒
reunite
❒
student
❒
24. /
w/
wet
❒
dissuade
❒
saw it
❒
What does this test tell you about (a) the person’s first language (b) the person’s
first writing system?

---

4.1 Phonemes and second language acquisition
Phonemes and second language acquisition 69
●
What do you think are the crucial sounds in your first language?
●
How do you think you learnt them?
Focusing questions
phonemes: the sounds of a language that are systematically distinguished from
each other, for example, /
s/ from /t/ in ‘same’ and ‘tame’
allophones: different forms of the phoneme in particular contexts, for example,
the aspirate /
p/ (with a puff of air) in ‘pill’ versus the unaspirated /p/ (with-
out a puff of air) in ‘lip’
distinctive feature: the minimal difference that may distinguish phonemes,
such as voice and aspiration in ‘din’ and ‘tin’
voice onset time (VOT): the moment when voicing starts during the produc-
tion of a consonant
Keywords
Each language uses a certain number of sounds called phonemes that distinguish
words and morphemes from one other. The spoken word ‘sin’ is different from the
word ‘tin’ because one has the phoneme /
s/, the other the phoneme /t/; ‘sin’ dif-
fers from ‘son’ in that one has the phoneme /
i/, the other the phoneme //. And
so on for all the words of the language – ‘bin’, ‘kin’, ‘din’, ‘gin’, ‘soon’, ‘sawn’,
‘seen’, ... Phonemes signal the difference between words and meanings: the spo-
ken distance between ‘I adore you’ and ‘I abhor you’ is a single phoneme, /
d/ ver-
sus /
b/.
A phoneme is a sound which is conventionally used to distinguish meanings in
a particular language. Any language only uses a small proportion of all the sounds
available as phonemes; English does not have the /x/ phoneme heard in German
words like ‘Buch’, or the click sounds used in South African languages; Japanese
does not have two phonemes for the /
l/ in ‘lip’ and the /r/ in ‘rip’; nor does French
recognize a distinction between short /
i/ in ‘bin’ and long /i/ in ‘been’. Human
languages have between 11 and 141 phonemes, English being about average with
44 or so (depending on accent).
As well as phonemes, there are allophones – variant pronunciations for a
phoneme in different situations. For instance, in English the phoneme /
l/ has
three main allophones. At the beginning of a word such as ‘leaf’, it is a so-called
‘clear’ [
l], sounding more like a front high vowel. At the end of a word such as
‘feel’, it can be pronounced as a ‘dark’ [
l], sounding lower and more like a back low
vowel. For many British speakers it is nowadays pronounced as /
w/, that is, ‘tell’ is
pronounced /
tew/. It is not going to affect the meaning if you pronounce ‘leaf’
with the wrong dark /
l/ but it will certainly convey a particular foreign accent.
The problem for second language acquisition is that each language has its own
set of phonemes and allophones. Two phonemes in one language may correspond
to two allophones of the same phoneme in another language, or may not exist at

---

all: the two Polish phonemes that distinguish ‘prosie’ (pig) from ‘prosze’ (please)
sound like allophones of /
ʃ/ (ship) to an English ear, while the two English
phonemes /
θ/ ‘thigh’ and /ð/ ‘thy’ seem to be allophones of one phoneme to a
Spanish speaker.
When the phonemes of spoken language connect one-to-one to the letters of
alphabetic written language, the writing system is called transparent, as in Finnish
or Italian. The English writing system is far from transparent because there are
many more sounds than letters to go round: 44 phonemes will not go into 26 let-
ters. So pairs of written letters go with single sounds, like ‘th’ for /
θ/ in ‘three’ or
‘ea’ for /
i/ in ‘bean’; or single letters go with two sounds, like ‘x’ for /ks/ ‘six’; or
letters have multiple pronunciations, like the 
a in ‘pat’ //, ‘atomic’ /ɘ/, ‘ska’
/
a/ and ‘swan’ /ɒ/. And of course letters are used very differently in the spelling
of, say, English, Polish and Arabic.
In the early days of the direct method, such phonetic scripts were often used
directly for language teaching, and they are still common at advanced levels where
people are often taught ‘ear-training’ by transcribing spoken language. Most EFL
coursebooks use a phonetic script as a resource to be consulted from time to time
rather than as the main vehicle for teaching; charts of the phonetic alphabet for
English can be seen pinned up in many classrooms. The elementary coursebook
New Headway Beginners (Soars and Soars, 2002) has a chart of the symbols for
English at the end of the book and uses them in the vocabulary lists, but only a
handful of exercises in the book actually use them. Joanne Kenworthy’s The
Pronunciation of English: A Workbook (2000), intended more for teachers than stu-
dents, uses phonetic symbols to train the listener to locate and discuss phonemes
in authentic English speech.
Over the years the concept of the phoneme has proved useful in organizing mate-
rials for teaching pronunciation, even when it has been largely superseded in much
phonological research. Pronunciation textbooks like Ship or Sheep? (Baker, 1981)
present the student with pairs of words: ‘car’ /
ka/ versus ‘cow’ /ka / or ‘bra’ /bra/
versus ‘brow’ /
bra /. This technique originated from the ‘minimal pairs’ technique
used by linguists to establish the phonemes of a language from scratch; you present
the native speaker with a series of likely or unlikely pairs of words and ask them
whether they are different. This allows you, in principle, to build up the whole
phoneme inventory – in practice, it is very hard to do, as I discovered when I naively
tried to demonstrate it in a lecture with a native speaker of a language I did not
know (Russian).
In typical pronunciation materials the student learns how to distinguish one
phoneme from another by hearing and repeating sentences with a high concentra-
tion of particular phonemes, such as ‘I’ve found a mouse in the house’ or ‘This is the
cleanest house in town’, or traditional tongue-twisters such as ‘He ran from the
Indies to the Andes in his undies’. Like the teaching of structural grammar, this
activity emphasizes practice rather than communication and sees pronunciation as
a set of habits for producing sounds. The habit of producing the sound /
n/ is believed
to be acquired by repeating it over and over again and by being corrected when it is
said wrongly. Learning to pronounce a second language means building up new
pronunciation habits and overcoming the bias of the first language. Only by saying
‘car’ /
ka/ and ‘cow’ /ka / many times is the contrast between /a/ and /a / acquired.
In other areas of language teaching, such as grammar, people would scorn making
students simply repeat sentences. Nevertheless it remains a popular technique for
pronunciation teaching.
Acquiring and teaching pronunciation