Phonemic analysis
involves 2 things:
• Establishing
the set of phonemes in language
Definition of phoneme.
A phoneme is a basic unit of
a language's phonology, which is combined with other phonemes to form
meaningful units such as words or morphemes
Phonemic
• Figure
out the phonological rules that determine the environments where allophone
occur.
Contrastive
• Two
sound are contrastive if interchanging the two can change meaning of the word
English
/p/ - /b/:
cap
[kæp] vs cab [kæb]\
Hindi
/p ͪ/ - /p/
‘fruit’
[p ͪǝl] vs ‘moment’ [pǝl]
Types of
Transcription
• Hindi
/p ͪ/ - /p/
‘fruit’ [p ͪǝl] vs ‘moment’ [pǝl]
Need to represent more than just
the phoneme /p/ to show how contrast works. How?
Add a diacritic [ ͪ ] to the
phoneme symbol [p]
Minimal pair
• Two
(or more) words that differ only by a single sound in the same position and
that have different meanings
- m/n:
[sᴧm] ‘sum’ vs [sᴧn] ‘sun’
- k/g:
[kɪl] ‘kill vs [gɪl] ‘gill’
- s/ʃ:
[mɛsi] ‘messy’ vs [mɛʃi] ‘meshy’
- i/ɪ/ɛ:
[fil] ‘feel’ vs [fɪl] ‘fill’ vs [fɛl] ‘fell’
• While
whole sounds can contrast, so can parts of sounds (i.e., feature):
-
voicing: [tɪl] ‘till’ vs [dɪl] ‘dill’
-
Place: [sᴧm] ‘sum’ vs [sᴧŋ] ‘sung’
-
Manner: [mɛs] ‘mess’ vs [mɛt] ‘met’
Phoneme vs Allophone
• Phoneme:
- A
minimal pair of sound that serves to distinguish meaning between words
- May be composed of a set of
sounds (allophones) that are considered by native speakers to be the ‘same’
sound.
• Allophone:
the different phonetic realization of a phoneme.
Top [t ͪap] Stop [stap] Little
[lɪɾl] Kitten [kɪʔn]
Phoneme: /t/
Allophone: [t ͪ] [t] [ɾ] [ʔ]
Distribution
• Contrastive
distribution: when the sounds can occur in the exact same phonetic environment
(thereby forming a minimal pair), e.g.,
-
Initial: [su] ‘sue’, [zu] ‘zoo’
-
Medial: [bᴧsɪŋ] ‘bussing’, [bᴧzɪŋ] ‘buzzing’
-
Final: [kloƱs] ‘close’, [kloƱz] ‘clothes’
• Complementary
distribution: when two (or more) phonetically similar sounds never occur in
exactly the same environment, but in complementary or mutually-exclusive
environments
– p/p ͪ: [spæt]
‘spat’ [p ͪæt] ‘pat’
[spul]
‘spool’ [p ͪul] ‘pool’
[spik]
‘speak’ [p ͪik] ‘peak’
Sounds in
complementary distribution
• …are
allophones of a single phoneme
• …do
not occur in minimal pair
• …are
noncontrastive
• …are
predictable (based on environment)
Expressing environment
• Phonemic
transcription is written in slash //
• Phonetic
(allophonic) transcription is written in bracket [ ]
• /__
]word
• The
symbol slash “/” means “in the environment”
• A
long underline stands for the place where the allophone occurs relative to its neighbors.
• /__
]word means ‘in the environment
immediately before the end of a word’.
The following expression:
1. /____ θ
is to be read “in the environment
before theta” or for short just “before
theta”
2. / consonant ________
voiceless
is to be read after a voiceless
consonant
3. Square brackets in phonology
mean “and”; hence
consonant
voiceless
means “a segment which is consonant
and is voiceless”
4. Symbol “+”
and “ – “ are used before feature names that a segment either has, or does not
have, the phonetic property that a feature designates.
+
consonant = consonant
-voice
= voiceless
/ +consonant _____
-
voice
5. The notation given below can be
read
“at the
end of a word”.
/ ____ ]word
“at the
beginning of a word” would be
/[word_______
6. The sound
which occurs to the left of the arrow is the underlying phoneme, while the
sound to the right of the arrow is the allophone.
a. /l/
à [ɫ]/___ ]word
b. /l/
à[ˬĪĪ] / +consonant _____
+ voiceless
Formalizing
Phonological Rules
• Rule
of word-final velarization in English
/l/ à [ɫ]/___ ]word
means “the phoneme /l/ becomes [ɫ] in the
environment ____ ] word [faɪɫ, fuɫ, aɫ]
• /l/
à[ˬĪĪ] / +consonant
+ voiceless
The
allophone [ˬĪĪ] occurs after a voiceless consonant [sˬĪĪaɪt, pˬĪĪaƱ, kˬĪĪɪŋ]
Level of
Representation
• Two
level of representation
1.
phonemes=underlying form
2. allophone=surface
form
• General
form
1.
underlying form àsurface
form/ environment
2. /x/ à[Y] / A__B
- The phoneme /X/ is pronounce as the
allophone [Y] when it occurs between A and B
Underlying phoneme
• The
general principle is that it is the sound which appears in the widest variety
of environments which is the underlying phoneme’
• Thus
it is [p], not [p ͪ] which is the underlying phoneme in English
/p/ /k/ /t/
[p] [p ͪ]
[k] [k ͪ] [t] [t ͪ]
·
Example
|
A
|
B
|
||
|
Blue
|
[blu]
|
Plow
|
[pļaw]
|
|
Gleam
|
[glim]
|
Clap
|
[kļæp]
|
|
Slip
|
[slɪp]
|
Clear
|
[kļiɹ]
|
|
Flog
|
[flɔg]
|
Play
|
[pļej]
|
|
Leaf
|
[lif]
|
State the environment in terms of
the natural class: [p k] = voiceless stops
Example: /l/ rules in English
/l/ à
[ļ] after a word initial voiceless
stop
/l/ à
[l] elsewhere
We don’t need 2 rules for voiceless
/l/:
/l/ à
[ļ] /after [p]
/l/ à
[ļ] / after [k]
/l/ à
[ļ] / after voiceless stop
• The
underlying phoneme is usually the sound which occurs in the elsewhere
environment.
Example:
slight
plow
[slaɪt]
[plaƱ] Underlying form
[sˬĪĪaɪt] [pˬĪĪaƱ] /l/ voicing
[sˬĪīaɪʔt] - final glottalization
of
[sˬĪīaɪʔt] [pˬĪĪaƱ] voiceless stop
Surface
form = allophonic/phonetic transcription
File slight wealth listen
/faɪl/ /slaɪt/ /wɛlθ/ /’lɪsən/
Underlying
forms
- sˬĪĪaɪ t - -
/l/ devoicing
- - wɛˬɫθ - /l/ dentalization
faɪɫ - - -
/l/ velarization
[‘faɪɫ] [sˬĪĪaɪt] [‘wɛˬɫθ] [’lɪsən] surface forms
Allophone often have
a phonemic motivation.
• Slight
[sˬĪīaɪt] plow [pˬĪĪaƱ]
• For
example The [ˬĪĪ] allophone may be the result of coarticulation with the
preceding voiceless sound.
• Wealth
[‘wɛˬɫθ] filthy [fɪˬɫθi]
The dental [ˬɫ] also may be
attributed to coarticulation with the following dental sound.
• A
sound which is a phoneme in one language
may be an allophone in another language . Conversely, a sound which is
an allophone in one language may be a phoneme in another
• Consider
the case of the flap [ɾ] in North American English.
• ‘city’
‘sit’
/sɪti/ /sɪt/ Underlying
form
Rules:
sɪɾi - flapping
[sɪɾi] [sɪt] surface
form
The rules is a s follow
/t/
à [ɾ] / [+vowel]
____ +vowel
-stress
The phoneme /t/ is realized as /ɾ/
when it is preceded by a vowel and followed by a stressless/unstress vowel
In North American
English
• [ɾ]
is an allophone of /t/ phoneme. The environment
for [ɾ] is between 2 vowels of which the second is stressless.
Phonemic Phonetic
data /’deɪtə/ [‘deɪſə]
latter /’lætɚ/ [‘læſɚ]
eating /’itɪŋ/
[‘iɾɪŋ]
Ottoman /’ɑtəmən/ [‘ɑɾəmən]
Rhetoric /’ɹɛtəɹɪk/ [‘ɹɛɾəɹɪk]
Automatic /,ɔtə’mætɪk/ [,ɔɾə’mæɾɪk]
• /t/
does not appear as the [ɾ] allophone because it fails to follow a syllabic
sound (tan, guilty, active) or because it fails to precede a syllabic sound
(cat, Atkins).
Phonemic Phonetic
tan /’tæn/ [tæn]
attend /ə’tɛnd/ [ə’tɛnd]
guilty /’gɪlti/ [‘gɪlti]
cat /’kæt/ [‘kæt]
active /’æktɪv/ [‘æktɪv]
Atkins /’ætkɪnz/ [’ætkɪnz]
In Spanish
• /t/
and /ɾ/ are separate phoneme in Spanish. It is demonstrated by minimal pairs
such as the following:
[‘pita]
‘century plant’ vs [piɾa] ‘funeral pyre’
• The
difference between [t] and [ɾ] signals a different in meaning. [t] and [ɾ] are
in contrast and separate phoneme, /t/ vs /ɾ/.
• Comparing
English and Spanish, we see that the [t] vs [ɾ] difference in allophonic
(non-distinctive) for English, but phonemic (distinctive) for Spanish. Thus, in
this area, the two language are phonetically similar but phonologically
different.
• Both
English and spanish have a [d] and a [ð] (the voiced dental fricative).
• In
English we know that the 2 sounds are separate phonemes, because minimal pairs
exist:
die [daɪ]
vs thy [ðaɪ]
bayed
[beɪd] vs bathe [beɪð]
den [dɛn]
vs the [ðɛn]
• But
in Spanish, there are no such pairs. One
can determine that [d] and [ð] are allophonic variants:
[daðo] ‘given’
[deðo] ‘finger’
[usteð] ‘you (polite)’
[donde] ‘where’
[de ðonde] ‘from where’
• From
these data indicate that [ð] occurs only after vowel, while [d] is the
elsewhere allophone of the same phoneme.
• Phonemes:
/t/, /d/, /ɾ/
• Phonological rule: /d/ Spirantization
/d/ à [ð] / [+vowel] ______
The
phoneme /d/ is realized a [ð] when it follows a vowel.
Spirant
is fricative
Spirantization:
change from stop to fricative.
/d/ is
stop and [ð] is fricative
v Derivations
1.
‘given’ ‘you’ ‘where’
/dado/ /usted/ /donde/
underlying forms
ð ð - /d/ spirantization
[daðo] [udteð]
[donde] surface forms
2.
/lado/ underlying form
Rules:
laðo Spirantization
[laðo]
surface form
• The
rule of spirantization in Spanish is really a more general rule applying to all
voiced stops (/b/, /d/, /g/), not just /d/.
+stop à [fricative] / [+vowel]
______
+voice
• The
differences in phonological organization between English and Spanish reflect a
different division of phonetic space.
• Phonetic
space as made up of multiple dimensions.
• We
place [d] at the center of this space, and in different directions show [ð]as
differing from [d] minimally in its fricative character (“continuancy”); [ɾ] differing
from [d] in having short, weak closure, and [t] differing minimally from [d] in
voicing
• [t] Voicing
Closure
[d] [ɾ]
Continuancy
[ð]
Spanish
/t/, /ɾ/, and /d/ with the latter
having 2 allophones [d] and [ð]
English has /d/, /ð/, and /t/, with
the latter having 2 allophones [t] and [ð]
• The
chart shows that the sound systems of languages can differ in their
phonological organization, as well as in the sounds that they contain. In
principle, we could imagine two languages that had exactly the same sounds, but
a radically different phonological organization. This would happen if the two
languages selected different phonetic distinctions to be contrastive vs.
non-contrastive. Using the phonemic method, we would analyze such languages as
having the same set of sounds, grouped into phonemes in two different ways.
Sources:
http://www.linguistics.uscb.edu/faculty/gordon/107/ling107.week1&2.pdf
Hayes2009-02.pdf
Tidak ada komentar:
Posting Komentar