Lec4: Phonology...The ‘wug’ test A well-known experiment on children’s knowledge of...

Lec4: Phonology

Ashwini Vaidya

HUL 242

Ashwini Vaidya Phonology-I 1 / 26

Figure: c©Jean Berko

Ashwini Vaidya Phonology-I 2 / 26

Figure: c©Jean Berko

Ashwini Vaidya Phonology-I 3 / 26

The ‘wug’ test

• A well-known experiment on children’s knowledge ofmorphological rules by Jean Berko (1958)

• Children internalize the rules of making plurals – and alsogeneralize them to unknown forms

• Children (Ages 4-7), Total of 56 subjects

• Q: Do we memorize the plural form ?/spelling?/sound ?/something else?

Ashwini Vaidya Phonology-I 4 / 26

English (regular) plurals

cab boy

match bus

cuff bar

can bush

spa call

love

butt cat

[kæb] [bOI][mæÙ] [b2s]

[k2f] [bar]

[kæn] [bUS][spa] [kOl]

[l2v]

[b2t] [kæt]

Ashwini Vaidya Phonology-I 5 / 26

English plurals

• Hint: Look for minimal pairs [kæt] and [kæb], [b2t] and [b2s]

• Is there a phonetic difference in the endings?

Ashwini Vaidya Phonology-I 6 / 26

Regular plural formation

• Decompose specific sounds into features

• Add [-z] before ???? segments

• Add [-s] before ???? segments

• Add [-Iz] before ???? segments

Ashwini Vaidya Phonology-I 7 / 26

English (regular) plurals

[kæb] [k2f] [mæÙ]

[bOI] [kæt ] [b2s]

[kOl] [b2t] [bUS][kæn] [b2z]

[spa]

[l2v]

[bar]

Ashwini Vaidya Phonology-I 8 / 26

Phonological rule

Form Environment Example

[s] Word ends in a voiceless (-sibilant) consonant [kæt ] [k2f]

[z] Word ends in a voiced (-sibilant) consonant [kæn] [kæb]

[Iz] Word ends in a sibilant consonant [b2s] [mæÙ]

• Underlying form [-z]

• Rule 1: Vowel insertion θ → [I] / [+sibilant] [+sibilant]

• Rule 2: Voicing assimilation [+voice] → [-voice] / [-voice]

Ashwini Vaidya Phonology-I 9 / 26

Phonological rule

Form Environment Example

[s] Word ends in a voiceless (-sibilant) consonant [kæt ] [k2f]

[z] Word ends in a voiced (-sibilant) consonant [kæn] [kæb]

[Iz] Word ends in a sibilant consonant [b2s] [mæÙ]

• Underlying form [-z]

• Rule 1: Vowel insertion θ → [I] / [+sibilant] [+sibilant]

• Rule 2: Voicing assimilation [+voice] → [-voice] / [-voice]

Ashwini Vaidya Phonology-I 9 / 26

Plural formation rule

• Plural formation for cat-cats, bus-buses, cab-cabs

[kæt] + -z [b2s] + -z [kæb] + z Underlying form

- [b2s] + I - Vowel insertion

[kæt] -s - - Voicing assimilation

[kæts] [b2sIz] [kæbz]

Ashwini Vaidya Phonology-I 10 / 26

Other rules for plural formation

• What about knife-knives wife-wives half-halves loaf-loavesthief-thieves ? (Relic of Old English, f becomes v, in betweenvoiced sounds) (Not for proof-proofs)

• Borrowed words from Latin and Greek follow their own pluralending system (formula-formulae, fungus-fungi (L),schema-schemata (G))

• Many other exceptions! foot-feet etc.

Ashwini Vaidya Phonology-I 11 / 26

English data

illegible impossibleindecent irregular

inharmonious illegitimateimpalatable immortalinarticulate irreparable

Ashwini Vaidya Phonology-I 12 / 26

English data

illegible impossibleillegitimate impalatableirregular immortalirreparable inarticulate

inharmoniousindecent

Ashwini Vaidya Phonology-I 13 / 26

English prefix -in

• The prefix -in changes to -il and -ir before /l/ and /r/ (liquids)

• It changes to -im before bilabials like /p/ and /m/

• For all others it is simply added at the beginning

Ashwini Vaidya Phonology-I 14 / 26

Phonological processes

• Ex1: [æsk] becomes [æks] (Children’s speech aminal insteadof animal)

Metathesis

• Ex2 [glas] becomes [gilas])

Epenthesis (of a vowel)

• Epenthesis often occurs when there is an unfamiliar consonantcluster in a borrowed word [gl] or [lb]

• Assimilation Two sound segments become more similar toeach other, such that there’s a change in feature values

Ashwini Vaidya Phonology-I 15 / 26

Phonological processes

• Ex1: [æsk] becomes [æks] (Children’s speech aminal insteadof animal)

Metathesis

• Ex2 [glas] becomes [gilas])

Epenthesis (of a vowel)

• Epenthesis often occurs when there is an unfamiliar consonantcluster in a borrowed word [gl] or [lb]

• Assimilation Two sound segments become more similar toeach other, such that there’s a change in feature values

Ashwini Vaidya Phonology-I 15 / 26

Phonological processes

• Ex1: [æsk] becomes [æks] (Children’s speech aminal insteadof animal) Metathesis

• Ex2 [glas] becomes [gilas])

Epenthesis (of a vowel)

• Epenthesis often occurs when there is an unfamiliar consonantcluster in a borrowed word [gl] or [lb]

• Assimilation Two sound segments become more similar toeach other, such that there’s a change in feature values

Ashwini Vaidya Phonology-I 15 / 26

Phonological processes

• Ex1: [æsk] becomes [æks] (Children’s speech aminal insteadof animal) Metathesis

• Ex2 [glas] becomes [gilas]) Epenthesis (of a vowel)

• Epenthesis often occurs when there is an unfamiliar consonantcluster in a borrowed word [gl] or [lb]

• Assimilation Two sound segments become more similar toeach other, such that there’s a change in feature values

Ashwini Vaidya Phonology-I 15 / 26

Phonological processes

• Ex1: [æsk] becomes [æks] (Children’s speech aminal insteadof animal) Metathesis

• Ex2 [glas] becomes [gilas]) Epenthesis (of a vowel)

• Epenthesis often occurs when there is an unfamiliar consonantcluster in a borrowed word [gl] or [lb]

• Assimilation Two sound segments become more similar toeach other, such that there’s a change in feature values

Ashwini Vaidya Phonology-I 15 / 26

Phonological processes

• Ex1: [æsk] becomes [æks] (Children’s speech aminal insteadof animal) Metathesis

• Ex2 [glas] becomes [gilas]) Epenthesis (of a vowel)

• Epenthesis often occurs when there is an unfamiliar consonantcluster in a borrowed word [gl] or [lb]

• Assimilation Two sound segments become more similar toeach other, such that there’s a change in feature values

Ashwini Vaidya Phonology-I 15 / 26

Affix and its realizations

[-im] [-il] [-ir]

/-in/

a) Prefix -ina) Prefix -in

[-s] [-Iz] [-z]

/z/

b) Regular plural -z

Ashwini Vaidya Phonology-I 16 / 26

Morpheme and Allomorph

• Abstract form of the negative prefix -in and the plural -s

• This is then realized in a particular form, based onenvironmental constraints

• The abstract form is a morpheme; paired with the meaningof ‘negation’ or ‘plural’

• Its phonologically conditioned variants are allomorphs

Ashwini Vaidya Phonology-I 17 / 26

Morpheme and Allomorph

• Abstract form of the negative prefix -in and the plural -s

• This is then realized in a particular form, based onenvironmental constraints

• The abstract form is a morpheme; paired with the meaningof ‘negation’ or ‘plural’

• Its phonologically conditioned variants are allomorphs

Ashwini Vaidya Phonology-I 17 / 26

Morpheme and Allomorph

• Abstract form of the negative prefix -in and the plural -s

• This is then realized in a particular form, based onenvironmental constraints

• The abstract form is a morpheme; paired with the meaningof ‘negation’ or ‘plural’

• Its phonologically conditioned variants are allomorphs

Ashwini Vaidya Phonology-I 17 / 26

Morpheme and Allomorph

• Abstract form of the negative prefix -in and the plural -s

• This is then realized in a particular form, based onenvironmental constraints

• The abstract form is a morpheme; paired with the meaningof ‘negation’ or ‘plural’

• Its phonologically conditioned variants are allomorphs

Ashwini Vaidya Phonology-I 17 / 26

Phoneme and Allophone

• In a similar way, we can think of phonemes and allophones

[ph]in [b]in ra[p]id ra[b]id la[p] la[b]

[th]ot [d]ot a[t]om A[d]am ma[t] ma[d]

[kh]ap [g]ap jac[k]et jag[g]ed pic[k] pi[g]

Table: Distribution of voiced and voiceless stops in English

Ashwini Vaidya Phonology-I 18 / 26

Phoneme and Allophone

[ph] [p]

/p/ • Phonemes are contrastive /p/ /b/

• Allophones are complementary -i.e.never occur in the sameenvironment

• Phonetic similarity betweenallophones and phonemes

Ashwini Vaidya Phonology-I 19 / 26

Minimal pairs

• Forming minimal pair is an important way of forminginferences about sounds

• E.g. [kæt] [kæp], [k2t] [kæt]

• The sound contrast = change in meaning

Ashwini Vaidya Phonology-I 20 / 26

Finnish data

Are [t] and [d] phonemes or allophones?

[kudot] ‘failures’ [madon] ‘of a worm’

[kate] ‘cover’ [maton] ‘of a rug’

[katot] ‘roofs’ [ratas] ‘wheel’

[kade] ‘envious’ [radon] ‘of a track’

Ashwini Vaidya Phonology-I 21 / 26

German fricatives

Are [x] and [c] allophones ?

nIct ‘not’ ba:x ‘Bach’re:c@n ‘rake’ la:x@n ‘to laugh’SlEct ‘bad’ kOxt ‘cooks’

ri:c@n ‘to smell’ fErsu:x@n ‘to try’haImlic ‘sly’ ho:x ‘high’

rEcts ‘rightward’ SlUxt ‘canyon’kri:c@ ‘to crawl’ fErflUxt ‘accursed’

Ashwini Vaidya Phonology-I 22 / 26

Korean data

State the context for Korean [r] and [l]mul ‘water’ ilkop ‘seven’

mulkama ‘place for water’ mal ‘horse’

mure ‘at the water’ malkama ‘place for horse’

pal ‘foot’ mare ‘at the horse’

pari ‘of the foot’ rupi ‘ruby’

s@ul ‘Seoul’ ratio ‘radio’

Ashwini Vaidya Phonology-I 23 / 26

American English (non) rhoticity

“You’re gonna shoot a [di:’@] ??”

Ashwini Vaidya Phonology-I 24 / 26

American English rhoticity

• Syllable final -r (postvocalic -r) dropped in some varieties ofAmerican English

• William Labov’s 1966 study of New York department storeworkers (social stratification of English)

• ‘fur’ ‘fourth floor’ ; pronunciation of /r/ associated withprestige

Ashwini Vaidya Phonology-I 25 / 26

Suprasegmental features

• Pitch, stress and segment length:- ways to distinguish betweensounds as well

Ashwini Vaidya Phonology-I 26 / 26