Day 5:

PHONOLOGY
1

LING/ANTH 106
DISCUSSION

DONALD GONG
GONG@KU.EDU
OFFICE HOURS: (BLAKE 419)
TUESDAY/THURSDAY 12:30-
1:30PM OR BY APPOINTMENT
 Agenda
2

 Review terminology.

 Part 1: Intro to Phonology

 Part 2: More Phonology

 Group Phonology Problems

 Individual Phonology Problems

 HW 5: Phonology
 Linguistics Terms!
3

 Phonetics
 Phon-etics

 Phonology
 Phon-ology

 Morphology
 Morph-ology

 Syntax
Part 1: Intro to Phonology
4
 What is Phonology?

 Phonetics: the physical properties of speech sounds

 How they’re produced

 How they’re represented in IPA: a few dozen features
are needed to describe every sound in every human
language.
 How we perceive sounds and how our perceptual
abilities change
 How we can characterize differences across language
varieties in terms of pronunciation

5
 What is Phonology?

 Phonology: how speech sounds are organized in

different languages.

Phonology is the linguistic knowledge a speaker has

of the sound patterns that are possible in a language:
 what combination of sounds is possible (blik) versus
the ones that aren’t (*nguk).

6
 What is Phonology?

We are going to be telling you a lot about rules and

processes in English that you already implicitly
know.

Keep this in mind when we look at other languages:

When you solve a phonology problem, you are really

discovering the unconscious knowledge that a
speaker of that language has!

7
 Questions that phonologists ask

 What is the organization of sounds in a given

language?

 Which sounds are predictable (based on the sound

environment), and which sounds are unpredictable?

 What is the phonetic context (environment) which

allows us to predict the appearance of one sound
versus another?

 Which sounds affect the meaning of words?

8
 An Example of Predictability: Plural in English

 The general rule for pluralization in English is: add

–s to the end of the word.

 We know that this rule is inadequate for irregulars

(child-children) but how well does it capture even
the regulars?

 Q:What did you buy your mother for Mother’s Day?

A: two [dags], two [kats], and two [waʧɪs]

 Two of these should sound weird.

9
An Example of Predictability: Plural in English

 How do we pluralize regular nouns in English?

A B C
cabs [kæb__] cats [kæt__] rashes [ræʃ___]
dogs [dag__] backs [bæk__] raises [rez___]
coins [kɔɪn__] cuffs [kʌf__] ledges [lɛʤ__]
scraps [skræp__] patches [pæʧ__]
loads [lod__]

 If the plural -s is realized in 3 different ways, then how do we

know which one to use?

 Do we memorize every noun, or is there some kind of

predictability governing pluralization?
10
 Wug Testing (Jean Berko Gleason 1958)

 If we rely on memorization, how would we

handle unknown nouns?

 This is a wug:

 Now there are two of them.

Now there are two _______.
11
 Wug Testing

Let’s try some other nonsense words.

How would you pluralize the following?

 “Two days ago I had only one shola, but yesterday I

bought another one. Now I have two ___________.”

 “My pet flot just had babies, so now I have a bunch of

____________.”

 “I started my collection with just one huz, but now I have

shelves of _________.”

 “This is my sister’s klup, and I have one, too. Together

we have two .”
12
 Wug Testing

Singular (in IPA) Plural (in IPA)

wug [wʌg] [wʌgz]
shola [ʃolə] [ʃoləz]
flot [flɑt] [flɑts]
huz [hʌz] [hʌzɪz]
klup [klʌp] [klʌps]

o You know how to pluralize novel words—although the plural –s can

be realized in three different ways, you have clear intuitions about
when to use each realization.

o Phonologists investigate this type of knowledge—speakers’

knowledge about sound patterns and distribution within their own
language. This knowledge is often unconscious.
13
 Back to the Plurals

 The behavior of the plural –s is predictable. We can

predict which of the 3 forms will appear in any given
phonetic environment.

 Let’s look at our list of plurals again to see if we can

identify the relevant environment.
A B C
[kæbz] [kæts] [ræʃɪz]
[dagz] [bæks] [rezɪz]
[kɔɪnz] [kʌfs] [lɛʤɪz]
[lodz] [skræps] [pæʧɪz]

14
 Back to the Plurals

 FIRST: list the actual sounds that [z], [s], and [ɪz] appear after.
[-z] : [b], [g], [n], [d]
[-s]: [t], [k], [f], [p]
[-ɪz]: [ʃ], [z], [ʧ][ʤ]

A B C
[kæbz] [kæts] [ræʃɪz]
[dagz] [bæks] [rezɪz]
[kɔɪnz] [kʌfs] [lɛʤɪz]
[lodz] [skræps] [pæʧɪz]

15
 Back to the Plurals

 NEXT: determine which feature(s) each group

of sounds has in common. This allows us to
make some generalizations regarding the plural
–s. Groups (a) and (b) are fairly easy.
A B
[kæbz] [kæts]
[dagz] [bæks]
[kɔɪnz] [kʌfs]
[lodz] [skræps]

+ Voice -Voice

16
 Back to the Plurals

 To categorize the sounds in group (c), we need to go back to

phonetics. The sounds [z], [ʃ], [ʧ], and [ʤ] belong to a special class of
sounds called sibilants (also includes [s] and [ʒ]). Sibilants are
sounds that have a buzzing, hissy quality.

 [-ɪz] appears after

sibilants

A B C
[kæbz] [kæts] [ræʃɪz]
[dagz] [bæks] [rezɪz]
[kɔɪnz] [kʌfs] [lɛʤɪz]
[lodz] [skræps] [pæʧɪz]

+ Voice -Voice Sibilant

Back to the Plurals

sibilants
 A Side Note on Plurals

 With which words is a child more likely to make

a mistake in forming the plural?

 cat or foot
 monkey or sheep

 What is different about the words that children

make errors with? What does this tell us about
the knowledge they have?
 Children know the rule: they just over-extend it.
 Another Example of Predictability

Up until now when we have compared

languages, we have only considered whether or
not a given language has a particular sound or
contrast in its inventory:

 English does not have clicks while Khoisan does.

 Japanese does not distinguish between [l] and

[r] but English does.

20
 Another Example of Predictability

In phonology we explore how different sounds

behave in a given language.

Two languages may have the same sounds but

they might have a very different status in the two
languages.

21
 Another Example of Predictability

An example: [s] - [ʃ]

 Both English and Japanese have the sounds [s]

and [ʃ] in their phonetic inventories

 English:[slæʃ] slash
 Japanese: [ʃimasu] do

22
 Another Example of Predictability

Question:
 In English, can we predict where we will get [s]
and where we will get [ʃ]?

 To figure this out, we can look at a minimal pair

like
seep [sip]
sheep [ʃip]

 This shows us that given the ending of a word [__-ip],

there is no way to predict (looking at the word in
isolation) which of the two sounds the word started
with. Both sounds are equally plausible.

23
 Another Example Of Predictability

Not so in Japanese!

 In Japanese, we can predict where will get [s] and

where we will get [ʃ]!

 The surrounding sounds (or sound environment) tell

us when we can have [s], and when we can have [ʃ].
 if there is an [i], it will be preceded by a [ʃ]

 [s] cannot appear before [i]

 therefore, anytime we see an [i] we know that the

sound before it must be an [ʃ] and cannot be an [s]
24
 Another Example Of Predictability

 English and Japanese are alike in that they both

contain [s] and [ʃ] in their inventories.

 But they differ in that in Japanese we can predict

the occurrence of one sound versus the other:

 Japanese: anytime we see an [i] we know that the

sound before it must be an [ʃ] and cannot be an [s].

 In English we cannot predict which sound will

appear.

25
 Another Example Of Predictability

 Another difference: if an English speaker

learning Japanese used [s] instead of [ʃ] before
an [i], it would not change the meaning of the
word.

Rather a native Japanese speaker would just

think that particular learner has an accent.

 Now imagine the case in English…..

26
If you substitute [ʃ] for [s] in English when you tell a
story about your vacation and how a lifeguard
__aved you at the beach, you are going to get some
really strange looks…

27
 Another Example Of Predictability

In the English case, the meaning of the word

changes!

Are you……..[ʃɔr] or are you [sɔr]?

[ʃ] and [s] are very different in English: they contrast

meaning.

28
 Summary: [s] and [ʃ]

 English and Japanese both have [s] and [ʃ] in their phonetic
inventories, but the sounds pattern differently in each
language.

 The two sounds do not behave predictably in English, but

we can predict the occurrence of [ʃ] in Japanese.

 In Japanese, [ʃ] always occurs preceding an [i], and [s]

never does.

 This tells us that [s] and [ʃ] are allophones of the same
phoneme in Japanese—they are like two different versions
of the same sound to the speaker’s mind.
29
 So What Is an Allophone?

 Allophones are the sounds in a given language that are

considered to be the “same sound” by native speakers.

 Some allophones in English

pit spit

 What is your intuition about these—are they “the same”

sound?

 We generally perceive them to be “the same.”

In the brain out the mouth

[p]
phoneme /p/ allophones
[ph]
30
 So What Is an Allophone?

 Do you remember when [p] is aspirated in

English?

 Voiceless oral stops are aspirated when they

appear in initial position (pit vs. spit).

31
 Allophones

 Allophones are the sounds that actually get pronounced.

 Each allophone must differ from all other allophones with

respect to at least one feature: [p] is different from [ph]
because [ph] is aspirated and [p] is not.
 Allophones are represented by putting an IPA symbols
between brackets; i.e., [p], [ph]

 Allophones are different realizations (pronunciations) of

the same phoneme.
 When there are multiple allophones for one phoneme, we
can often predict which sound will be produced based on
the environment ([ph] in initial position and [p] elsewhere).
32
 Allophones vs. Phonemes

 Phonemes are the mental representations of a sound that

every speaker of a language creates.
 Phonemes cannot be pronounced or heard; they exist only in
the minds of speakers.
 One phoneme may have multiple allophones associated with it.

 When multiple allophones belong to the same phoneme, these

allophones are interpreted as “the same sound” by speakers.
 Phonemes are represented by putting an IPA symbol between
slashes: /p/

33
 Minimal Pairs

 How do we determine whether two sounds are

allophones of the same phoneme or allophones of
separate phonemes?

 The first step is to look for a minimal pair.

 Two words form a minimal pair when they vary by

only a single sound. These are minimal pairs in English:
[sip] - [ʃip] [sɪk] - [θɪk]
[but] - [sut] [khæp] - [khæb]
[splɪt] - [splɪʃ] [but] - [bot]
 We can use Minimal Pairs to solve
Phonology Problems

 Minimal pairs help us determine the distribution

of allophones and phonemes in other languages.

 If we find a minimal pair, we know that the

sounds in question are allophones of different
phonemes.

 Why? Because the speaker’s mind MUST tell them

apart to be able to access the contrasting
meanings.
 Minimal Pairs: Be Careful!

 If two words differ by more than one sound then

they are not minimal pairs—even if they sound
very similar!

 The following pairs of English words are not

minimal pairs:

[blæst] - [pæst] [mʌg] - [gʌm] [hæm] - [spæm]

 We can use Minimal Pairs to solve
Phonology Problems

 Consider the following data from Hindi.

[kap] cup [phəl] fruit
[kaph] phlegm [pəl] moment

 Are [p] and [ph]:

A. allophones of the same phoneme
B. allophones of different phonemes

 You’ve just discovered something about the unconscious

linguistic knowledge of a Hindi speaker!
 Notice how English and Hindi differ with respect to aspiration.
 ASPIRATION IN HINDI

In Hindi, aspiration is used to contrast meaning.

We know this because we can find minimal pairs.
 Part 2
39

CONTRASTIVE VS. COMPLEMENTARY

DISTRIBUTION

DISTINCTIVE VS. NON-DISTINCTIVE

FEATURES

WRITING RULES

WRITING RULES FOR NATURAL CLASSES

REVIEW: MINIMAL PAIRS

If we find a minimal pair, it means that speakers of that

language do not consider the sounds in question to be “the
same.”

A minimal pair means that a speaker hears the two sounds

as different and uses that difference to distinguish between
two different words (pet and bet).

Remember: if we find a minimal pair, we know that the

sounds in question are allophones of different phonemes;
contrasting sounds in the speaker’s mind.
 Contrastive Distribution

 One basic property of phonemes is that they are

contrastive.

 We know that there are two phonemes /m/ and /n/ in

English, because they are the only basis for contrast in
the pairs mat and gnat, or mine and nine.

 If we substitute /m/ for /n/, or vice versa, we create a

contrast in meaning.
Hence: /m/ and /n/ are in contrastive distribution
in English.
 Contrastive Distribution

When two sounds are in contrastive

distribution:
 They can form minimal pairs.
 We cannot predict the appearance of the two
allophones; they will not have a predictable sound
environment.
 Interchanging the two sounds triggers a change (or
contrast) in meaning.
 Bengali: Voiceless Stops

 Let’s look at some Bengali data below to

determine how Bengali speakers think about
voiceless stops. Look first at [p] and [ph].

Do Bengali speakers consider these sounds to be the

same or different?
 Bengali: Voiceless Stops

 Do Bengali speakers consider these sounds to be

the same or different?

 Same = allophones of the same phoneme

 Different = allophones of different phonemes

(minimal pairs)
 Same/Different Exercise: Bengali

 What does this tell us about how Bengali

organizes [p] and [ph]? Are the two sounds in
contrastive distribution?
 Same/Different Exercise: Bengali

 Does Bengali organize [ph] and [p] in the same

way that English does?
 Which diagram represents the way Bengali
organizes these phones?

[p]
A. phoneme /p/
[ph]

B. phonemes /p/ /ph/

 Same/Different Exercise: Bengali

 Now look at the other voiceless stops in Bengali.

 So how does the organization of voiceless stops in Bengali differ

from that of English?

Voiceless stops and their aspirated counterparts are in

contrastive distribution in Bengali, not so in English.
 Complementary Distribution

 Minimal pairs can help us to determine that two sounds are

allophones of separate phonemes, but the lack of a minimal
pair does not necessarily indicate that two sounds belong to
the same phoneme. For that, we need more proof.

 In order to determine whether two sounds are allophones of

the same phoneme, we need to look for complementary
distribution, where they complement each other by staying
out of the other’s space.

 If two allophones belong to the same phoneme, each

allophone in the set will have a predictable environment.
Complementary Distribution
 Complementary Distribution

 Given the sound environment, we can predict

which of the two allophones will appear.

For example, think back to the Japanese example

with [s] and [ʃ].

These sounds are in complementary

distribution in Japanese; we can predict that [ʃ]
will always appear preceding an [i], and [s] never
will.
 Complementary Distribution

When two sounds are in complementary

distribution:

 These sounds will never form minimal pairs.

 If we swap one for another, the word will end up
sounding weird but no change in meaning will
result.
 Each allophone will appear in a predictable
phonetic environment.
 Distinctive and Nondistinctive Features

Let’s go back to the data from Hindi:

[kap] cup [phəl] fruit

[kaph] phlegm [pəl] moment

 What feature separates the two sounds [p] and [ph]?

 In Hindi, does changing this feature trigger a change in

meaning?

 This means aspiration is a distinctive feature in Hindi, a

feature so important that it distinguishes one phoneme from
another.
 Distinctive and Nondistinctive Features

Is this feature as important in English as it is in Hindi? Do

we use aspiration to distinguish one phoneme from
another?

 If not, it is a nondistinctive feature in English.

 Nondistinctive features do not distinguish one phoneme
from another.
 Instead, they distinguish the various allophones of a
single phoneme from one another. [phɪt] [spɪt]
 Nondistinctive features are the ones that are
predictable and their distribution can be captured by
writing a rule.
 Distinctive and Nondistinctive Features

If we know that in English, voiceless oral stops

are always aspirated in initial position, then we
can predict where and when we will find
aspirated voiceless oral stops in English.

Aspiration is a nondistinctive feature in

English.
 Writing Rules

Determine the base form (the phoneme):

Is one of the allophones more common? Does it
occur in more (and varied) environments? If so, this
is your base form, or the elsewhere allophone.
 Example: We know that [ʃ] appears ONLY
before [i] in Japanese, while [s] appears
elsewhere.
 The behavior of [ʃ] is predictable; it appears in a
very specific environment.
 [s] is not as predictable; it appears everywhere else,
in many different environments.
 Rule Writing Practice

Write the rule:

Describe when the base form surfaces as the
other allophone.
/s/ becomes [ʃ] when it is before [i].

/s/ [ʃ] / ___ i

phoneme becomes allophone when environment

 A Look At Some Other Features

 Consider the two English words: [zIp] and [sIp]

 What is the phonetic difference between the initial

sounds in these two words?

 [z] is [+voice] and [s] is [-voice]

 Is this a crucial feature to a native speaker of

English? How do you know?

 Is voicing [voice] a distinctive feature in English?

 A Look At Some Other Features

 Tenseness: beat [bit] versus bit [bIt]

[i] is [+tense] and [I] is [-tense]

 Is tenseness [tense] a distinctive feature in

English?
 A Look At Some Other Features

 Length of vowels/consonants [long]

[raaaaaaaak tʃaaaaaaaak] vs. [rak tʃak]

 Is length [long] a distinctive feature in

English?
 A Look At Some Other Features:
Crosslinguistic Variation
 Whether or not a feature is distinctive will
vary from one language to another. You need
to consider the data before you.

 Do you remember this data set from Danish?

 Is length a distinctive feature in Danish with
respect to vowels?

 Danish
[vilə] wild [viilə] rest
[menə] remind [meenə] mean
Incorporating features/Natural class into rules

 So far we have focused on identifying

consonants and vowels according to their
features:

[k] voiceless oral velar stop

[i] high front tense unrounded

Each individual sound is a collection of features,

which, taken together, separate it from all other
sounds.
Incorporating features/Natural class into rules

 We can also look at a group of sounds, as we did

in Phonetics, in order to determine shared
features.

 Review: what features do the below groupings

have in common?

[k] [g] [ŋ]

[i], [ɪ], [u], [ʊ]
[f], [θ], [s], [ʃ], [h]
 Incorporating natural class into rules

 What we will find is that sometimes rules will be

very specific, and target only a single sound, but
other times they affect an entire group of
sounds.

 When a rule does target a group of sounds, we will

want to incorporate the idea of natural class into
our rule-writing.

 Random groups of sounds do not pattern together;

we will never see a rule that targets [p], [z], [h].
 Incorporating natural class into rules

 Phonology is not random. Phonological rules

operate systematically within languages.

 When multiple sounds undergo the same change

in the same environment, we will always be
able to pinpoint the phonetic features shared by
all the sounds involved.

 Those shared features have something to do with

the phonological pattern we are observing.
 Incorporating natural class into rules

 This could include any of the natural classes we

talked about in phonetics. You may see:

 all voiced stops behaving similarly

 all bilabial sounds behaving similarly

 all nasals behaving similarly

 all back vowels behaving similarly

 All high vowels behaving similarly

 all palatals behaving similarly

 et cetera, et cetera
SOLVING PHONOLOGY
PROBLEMS
Zinacantepec Tzotzil: I will lead
67

[pim] ‘thick’ [k’a] ‘horse’

[k’ok] ‘fire’ [pus] ‘jail’
[bikil] ‘intestines’ [p’in] ‘pot’
[ka] ‘particle’ [kok] ‘my leg’
[nopol] ‘nearby’ [bik’it] ‘small’
[p’us] ‘hunchback’ [p’ol] ‘to multiply’
Are [p] and [p’] allophones of the same phoneme,
or allophones of different phonemes?
Step 1: Look for minimal pairs.
Can you find two words that differ only by
the sounds [p] and [p’]?
Zinacantepec Tzotzil
68

[pim] ‘thick’ [k’a] ‘horse’

[k’ok] ‘fire’ [pus] ‘jail’
[bikil] ‘intestines’ [p’in] ‘pot’
[ka] ‘particle’ [kok] ‘my leg’
[nopol] ‘nearby’ [bik’it] ‘small’
[p’us] ‘hunchback’ [p’ol] ‘to multiply’

[p] and [p’] are allophones of

different phonemes and they are
in contrastive distribution.
Zinacantepec Tzotzil
69

/p/ /p’/

[p] [p’]
Zinacantepec Tzotzil: You find them!
70

[pim] ‘thick’ [k’a] ‘horse’

[k’ok] ‘fire’ [pus] ‘jail’
[bikil] ‘intestines’ [p’in] ‘pot’
[ka] ‘particle’ [kok] ‘my leg’
[nopol] ‘nearby’ [bik’it] ‘small’
[p’us] ‘hunchback’ [p’ol] ‘to multiply’

Are [k] and [k’] allophones of the same

phoneme, or allophones of different
phonemes?
Zinacantepec Tzotzil
71

[pim] ‘thick’ [k’a] ‘horse’

[k’ok] ‘fire’ [pus] ‘jail’
[bikil] ‘intestines’ [p’in] ‘pot’
[ka] ‘particle’ [kok] ‘my leg’
[nopol] ‘nearby’ [bik’it] ‘small’
[p’us] ‘hunchback’ [p’ol] ‘to multiply’
Zinacantepec Tzotzil
72

[pim] ‘thick’ [k’a] ‘horse’

[k’ok] ‘fire’ [pus] ‘jail’
[bikil] ‘intestines’ [p’in] ‘pot’
[ka] ‘particle’ [kok] ‘my leg’
[nopol] ‘nearby’ [bik’it] ‘small’
[p’us] ‘hunchback’ [p’ol] ‘to multiply’

[k] and [k’] are allophones of

different phonemes and they are
in contrastive distribution.
Zinacantepec Tzotzil
73

1. /k/ 2. /k’/ 3. /k’/ /k/

[k] [k’] [k] [k’] [k’] [k]

Old English: In GROUPS!
74

[fæst] ‘fast’ [fædɛr] ‘father’

[hlavɔrd] ‘lord’ [fɪʃ] ‘fish’
[wif] ‘woman’ [æftɛr] ‘after’
[wivʊŋg] ‘wedlock’ [wivɛl] ‘weevil’
[ɔvɛr] ‘over’ [ʧæəlf] ‘calf’
Are [f] and [v] allophones of the same phoneme
or allophones of different phonemes?
 Old English [f] & [v]
75

 Step 1: Look for minimal pairs.

 Can you find two words that differ only by the sounds [f] and [v]?

 Step 2: If no minimal pair is found, start to list the environments that

each sound occurs in.
Old English
76

[f] [v]
 Old English [f] & [v]
77

 Step 3: Determine whether the sounds are in

complementary distribution. MAKE A CHART! 

 Do they occur in overlapping environments? If not, they

are in complementary distribution.

 Step 4: If they are in complementary distribution, find

the elsewhere allophone, the one with less restrictive
distribution.

 Step 5: Write a rule to account for their distribution.

Old English
78

/ /

[ ] [ ]
/ /  [ ]/ _________
 Modern English: Together
79

 Are [f] and [v] allophones of the same phoneme or

allophones of different phonemes? Why?
[fes] ‘face’ [fεlt] ‘felt’
[fet] ‘fate’ [ves] ‘vase’
[ven] ‘vane’ [vel] ‘veil’
[vεnt] ‘vent’ [fel] ‘fail’
 [fes] vs. [ves]; [fel] vs. [vel]
 Are [f] and [v] in contrastive or complementary
distribution in Modern English?
 Modern English
80

/ / / /

[ ] [ ]
 Japanese
81

 Are [h] and [ɸ] allophones of the same phoneme or

allophones of different phonemes?
 [ɸ] is a voiceless, bilabial fricative.

[joho:] ‘forecast’ [to:ɸu] ‘tofu’

[ɸuku] ‘clothes’ [hoʃi] ‘star’
[hen] ‘strange’ [hako] ‘box’
[tehon] ‘model’ [ɸukai] ‘deep’
[saiɸu] ‘wallet’ [ʃihai] ‘control’
[eɸu] ‘F’

 How do we tell if these sounds are allophones of separate phonemes?

 Are they allophones of separate phonemes?
 What is the next step?
 Japanese
82

 [h]:
 [ɸ]:

Are these allophones of the same phoneme?

Which is the “elsewhere case”?
/ /

[ ] [ ]
 Japanese
83

 Is [ç] an allophone of /h/ or a separate phoneme?

 [ç] represents a voiceless, palatal, fricative

[joho:] ‘forecast’ [hoʃi] ‘star’

[çito] ‘person’ [ko:çi:] ‘coffee’
[hen] ‘strange’ [hako] ‘box’
[tehon] ‘model’ [ku:çi] ‘waste’
[ʃihai] ‘control’ [kiçin] ‘grace’

 What is the next step?

 Japanese
84

 [ç]:
 [h]:
 [ɸ]:

Are all three of these sounds in complementary distribution?

Which is the elsewhere allophone?

 Japanese
85

 [ç] appears before [ ];

 [ɸ] appears before [ ];

 [h] appears elsewhere.

/ /

[ ] [ ] [ ]
 Canadian English
86

[bəɪt] ‘bite’ [taɪ] ‘tie’

[raɪd] ‘ride’ [raɪz] ‘rise’
[rəɪt] ‘write’ [fəɪt] ‘fight’
[baɪ] ‘buy’ [rəɪs] ‘rice’
[faɪl] ‘file’ [ləɪf] ‘life’
[taɪm] ‘time’ [təɪp] ‘type’
[naɪnθ] ‘ninth’ [bəɪk] ‘bike’

 Are [əɪ] and [aɪ] allophones of the same phoneme or

allophones of different phonemes?
 Canadian English
87

 [aɪ]:
 [əɪ]:
 Are [əɪ] and [aɪ] in complementary distribution or
contrastive distribution?
 Which is the elsewhere allophone?

/ /

[ ] [ ]

 / / becomes [ ] when followed by ______________.