LINGUISTICS 221 LECTURE #9
Introduction to Phonetics and Phonology
Phonemic Analysis
1. PHONETICS AND PHONOLOGY
PHONETICS: The study of the inventory of all speech sounds which
humans are capable of producing.
PHONOLOGY: The study of the sound systems of languages. Out of the
very wide range of sounds the human vocal apparatus can produce (studied
by PHONETICS) only a relatively small number is used distinctively in any
one language. Phonology is concerned with the function of sounds.
The relationship between phonetics and phonology is a complex one:
Statement (a):
Phonetics is concerned with universal properties of speech, studied by
scientific methods; it deals with physical or concrete reality while
phonology is concerned with the linguistic organization of this reality.
Statement (b):
Reversing this argument, phonology may be said to tackle the true
mental reality behind speech, while phonetics handles 'merely' the
concrete manifestations of this reality.
Phonology deals with the systems and structures of speech, while phonetics
focuses more narrowly on articulation and acoustics. But: the boundary
should not be sharply drawn; speech is analyzed by breaking it down into its
several aspects, but the true reality is one of integration.
1
Phonetics: Phonology:
sounds vary with their context hypothesize rules to characterize the
variation;
identifies patterns of the sequencing
and distribution of speech sounds;
relates sounds patterning with other
components of grammar: interfaced
with morphology and syntax.
SPEECH SOUND (=Segment): Any discrete unit that can be identified in
the stream of speech.
can
(3 segments)
science
(6 segments)
Utterances can be represented as sequences of discrete units (=segments).
BUT:
Speech sounds are not produced as a series of discrete segments; rather, they
merge and blend together.
The vocal apparatus does not work like
[k˙] pause [æ] pause [n]
BUT: [k≠˙æ~n]
Speech is DYNAMIC rather than STATIC.
TARGET POSITIONS: Those positions of speech organs that are specified
for a given sound. During speech, these target positions are most of the time
only approximations.
2
A sound begins to be articulated before the completing of the articulation of
the previous sound.
e.g. book [b∑]
labialization
queen [k˙∑]
Although a speech signal is physically continuous, we seem to perceive it as
a sequence of discrete entities.
BUT: Do we have the right to claim that speech is segmentable if the
articulatory and acoustic aspects suggest the opposite?
SPEAKER:
Intention: "can"
The speaker intends to utter a sequence of discrete sounds, but the
vocal apparatus instead functions continually in motion, moving from
one sound to the next.
Assumption: The transitions from one sound to the next are automatic
features of the vocal mechanism -- as such can remain outside the
intention of the speaker.
LISTENER:
Perceives the continuous signals as discontinuous.
EVIDENCE FOR THE RECOGNIZING OF INDIVIDUAL SEGMENTS
IN THE FLOW OF SPEECH:
a. Alphabetic writing
There is a correlation between a sequence of symbols and a sequence of
speech sounds (the correlation may not always be ideal, but it is undeniable
that there is a correlation).
3
in the written word there is a continuum, yet we perceive it as consisting of
three letters just like in the printed form: can
Alphabetic writing teaches us:
(i) that speech consists of linear sequences of unitary, discretely different
sounds;
(ii) that a sound occurring at a given place in one word can be identified
with certain sounds occurring at different places in other words.
b. Slips of the tongue
e.g., you have hissed my mystery lectures;
with this wing I thee red
etc.
Slips of the tongue illustrate the treatment of speech sounds by speakers as
discrete segments which can be shunted from one word to another.
c. Rhymes
e.g., the fat cat in the hat
Rhymes suggest the need for recognizing individual segments.
d. Adding or deleting segments
e.g., r-addition (idear is for idea is)
h-drop (at for hat)
Segments rather than syllables or words are added or deleted.
-------------------------
4
"Same" utterance: can
(i) speaker differences
different!
(ii) the same person never says it in
the same way
BUT: We perceive them as identical utterances even though they are
physically different.
In language the PERCEPTUAL, the SUBJECTIVE, the DISCRETE
take precedence over the PHYSICAL, the OBJECTIVE, the
CONTINUOUS : A BASIC ASSUMPTION OF PHONOLOGY.
2. DISTINCTIVENESS AND CONTRAST
Examine the following words:
a. pin, bin, tin, din, kin, fin, sin
b. beat, bait, bet, bat, boot, boat, bought
In (a) the words differ only in their initial consonant segments.
In (b) the words differ only in their vowel segments.
Segments whose function is to contrast forms are PHONEMES.
Phonetic differences that are linguistically significant are PHONEMIC.
moon
Different meanings: /m/ and /n/ are PHONEMES
noon
5
A phoneme is not necessarily phonetically invariant:
pin /p/ phonetically: [p˙]
bin /b/
spin /p/ phonetically: [p]
PHONEME: A contrastive segmental unit that may have predictable phonetic
variants.
/p/ Phoneme: contrastive
[p] [p˙] Allophones: non-contrastive
pin p˙in Same meaning: [p] and [p˙] are non-contrastive
segments (in English).
3. SOUNDS THAT DO NOT CONTRAST: ALLOPHONES
ALLOPHONES are phonetically conditioned predictable variants
of the phoneme. ↓
On the basis of:
(a) The phonetic quality of neighbouring sounds:
/d/
[Îï]
e.g., drain
(b) Position:
/b/
[b≤]
e.g., tab
6
(c) Influence of prosodic elements:
/d/
[d] [‰]
e.g., deep rider
Additional examples for allophones:
/t/
(fill in!)
ELSEWHERE ALLOPHONES:
Abstract level: /g/ phoneme
Concrete level: /g≠/ /g/ allophones
used before a used environments for allophones
front vowel elsewhere
4. PHONEMES AS CATEGORIES
Allophones, such as [d] and [‰] form an abstract phonological category,
the phoneme /d/.
Phonological categories – the phoneme – used to distinguish between words,
but allophones cannot.
7
5. PHONOLOGICAL RULES
Example: Alveolars become dentals before a dental consonant.
Phonological rules are language specific! They are productive, conform to
the well-formedness condition, and reflect unconscious knowledge – Study
these concept on pp. 26-27!
6. COMPLEMENTARY DISTRIBUTION: The distribution of
allophones, in their respective phonetic environments, is such that one never
appears in the same phonetic context as the other.
/t/
[t] [‰]
stop writer
The distribution of /h/ and /˜ / in English:
head, heart, perhaps, enhance
sing, coming, wing [˜]
There are no English syllables ending in [h], and there are no English
syllables beginning with [˜]. Thus, it would appear that [h] and [˜] are in
complementary distribution, and should be assigned as allophones of the
same phoneme -- wrong!
[h] and [˜] have very little in common (they are both consonants).
PHONETIC SIMILARITY:
Sounds which represent different pronunciations of the same phoneme must
be phonetically similar. We consider two sounds to be phonetically similar
when they share the same place or manner of articulation.
7. FREE VARIATION: When two phonemes appear in the same context
without causing a change in meaning.
economics /i/ or /´/
either /i/ or /aj/
8
FREE VARIATION AMONG ALLOPHONES:
map [p] or [p˙]or [b¬]
mat [t] or [t˙] or [t¬]
8. MINIMAL PAIRS:
beet rip feel
bit rib seal
A minimal pair consists of two forms with distinct meanings that
differ by only one segment found in the same position of the word.
The contrastive function of speech sounds is language specific.
Thai: pàa ‘forest’ /p/
p˙àa ‘to split’ /p˙ /
English: pit [p˙]
Same meaning (see above)!
pit [p]
Malayalam:
kutti ‘peg’ /t/
kuˇˇi ‘child’ /ˇ /
English:
train [t˙]
Same meaning!
train [ˇ˙]
English:
Ben /´/
ban /æ/
9
Turkish:
ben [´] or [æ] Sg1.
Minimal triplets: time, dime, lime
Phonemes, minimal pairs and minimal triplets: Study Tables 2.1 and 2.2
NEAR MINIMAL PAIRS: Segments in nearly identical environments.
e.g., author[†]
either [∂]
Near minimal pairs occur more frequently in languages with large
phonemic inventory and long words.
10
