Li/Lt6
Phonology and Morphology
Lecture 1
Phonemes and abstract representations
Today’s topics


Two theories of mind
Evidence for abstract mental representations




Competence vs performance
Priming
Phonological alternations
Evidence for phonemes







Neutralization
Language games
Speech errors
Loanword phonology and L2 transfer
Orthography
Perception
Discrimination tasks
Two theories of mind
rationalist
reductionist
learning that
learning to
whole > sum of parts
WYSIWIG
abstract, multi-level
concrete, surface level only
symbol-manipulating rules and
constraints
connectionist network +
statistical knowledge
Turing machine
switch network
categorical, symbolic
gradient, numerical
The reductionist position




Popular lay intuition that we store and manipulate surface
forms
Priming effects by individual voice
“Once we introduce mechanisms for describing word-word
correspondences into the grammar, it may be possible to
dispense with underlying representations altogether (see
proposals by Burzio 1998, Bybee 2001).” (Sumner 2003)
connectionists
Evidence for
abstract URs
Psychological reality:
Sapir 1933

Basic points:
speech sounds are stored in the mind as abstract categories
(phonemes)
These are not always identical to their surface manifestations
((allo)phones)
Speakers think of language in terms of phonemes, not phones
(i.e. phonemes have psychological reality)
1.
2.
3.

oak tree
maple tree
Evidence:
Spelling: Alex Thomas (Nootka)
1.
•
•
writes <ħi, ħu> for [ħε, ħ]
writes <C!> for 2 different things he was taught (T!, ’R)
Intuitions
2.
•
NB Sapir needs
external evidence to
decide if the two
forms he hears as
identical are actually
different phonetically
•
Southern Paiute
•
•
•
Tony says [pABAh] for ‘at the water’
Asked to divide the word into syllables, he says [pA ] [pAh]
/p t k/  [β r γ] / V _; no /β r γ/
Sarcee
•
John Whitney feels that dìníh ‘it makes a sound’ has a final /t/, whereas
dìníh ‘this one’ does not (cf. dìníth - í ‘the one who…’)
TREE
TIN
k!
L1 competence : performance
“by the time infants are starting productive use of language they
can already discriminate almost all of the phonological contrasts of
their native language. While they cannot yet produce adult-like
forms, they appear, in many respects, to have adult-like
representations, which are reflected, among other things, in their
vociferous rejections of adult imitations of their phonologically
impoverished productions” (Faber and Best 1994:266-7)
Did you say
TINk?
“Preferential Looking” paradigm (Hirsh-Pasek
and Golinkoff 1993)
• 19-month old infants’ comprehension of
sentence like “Big Bird is hugging Cookie
Monster. Where is Big Bird hugging Cookie
Monster?”
• Infants (who can’t speak yet) looked
longer at correct video
Cruttenden
1985
No!!
competence vs. performance

intoxicated speech of the captain of the Exxon Valdez
(Johnson, Pisoni, and Bernacki 1990)
 Observed effects:
• misarticulation of /r/ and /l/
• final devoicing
• Deaffrication
Performance problem, not different grammar
attempts at producing adult [phεn] pen collected from a 15month-old child in a 30-minute period (Ferguson 1986):
 [mãə], [v], [dεdn], [hIn], [mbõ], [pHIn], [tHntHntHn],
[bah], [dhau], [buã]


competence vs. performance

Bedore, Leonard, and Gandour 1994





4 yr old girl substitutes dental click for {s z S Z tS
dZ}
She can initially imitate the fricatives, but not produce them
on her own
After a short training session, she starts to produce [s]
correctly
She then immediately gets all the others right
“the rapid rate of change observed in the child’s
phonological system seems consistent with a phonological
learning model in which the child has adult-like underlying
phonological representations” (283)
Priming

Colloquial Dutch optionally neutralizes
obstruent voicing with the clitic -der ‘her’




Do speakers respond faster to targets that
match the voicing of a prime?
Lexical decision task, with following
sequence of events:
1.
2.
3.

PRIME
TARGET
DECISION
(verb+clitic construction)
(verbal infinitive or nonword)
(is the target a legitimate word of Dutch?)
Results


/ık kœs der/ ‘I kiss her’  [ık kœzdər] ~ [ık kœstər]
/ık kiz der/ ‘I kiss her’  [ık kizdər] ~ [ık kistər]
Responses were faster when voicing of prime
corresponded to voicing of UR
Conclusion

Both variants of a verb are not stored in the
lexicon
Jongman, Allard. 2004. Phonological and Phonetic Representations: The Case of Neutralization. Proceedings of the 2003 Texas Linguistics Society Conference:
Coarticulation in Speech Production and Perception, Augustine Agwuele, Willis Warren, and Sang-Hoon Park, eds., pp. 9-16.
Japanese g~ŋ and rendaku
(Ito and Mester 2003)

Rendaku: C  [+voi] / ]+[ __ X]


Lyman’s Law: X does not contain [+voi, -son]


taba ‘bundle’ satsu+taba ‘wad of bills’ (*satsu-daba)
g-weakening: non-initial /g/  [ŋ]



kami-kaze vs. ori-gami
[gai+jiN] ‘foreigner’
[koku+ŋai] ‘abroad’
Rule interaction


R feeds GW: /ori+kami/  [oriŋami]
R counterfeeds GW: /saka+toge/ ‘reverse thorn/  [sakatoŋe]
If GW preceded R it would
change g to ŋ and thereby feed
R, resulting in *sakadoŋe
Evidence for
abstract phonemes
Phonemes are also abstractions
Ambiguity: German devoicing
Rad
Rat
nom.
[
At]
gen.
[Ads]
[Ats]
[] = voiced uvular
fricative cf. French, Hebrew
Perception
Categorical perception: VOT
Categorization functions for synthetic stimuli ranging from [b] to [p]. Open circles indicate the percent of
times that each stimulus was perceived as [b], and the filled circles indicate the percent of times that each
stimulus was perceived as [p]. (Lisker and Abramson 1970)
Hearing Cs that aren’t there I:
The McGurk effect
When hearing the sound BA, while seeing GA:
 most adults (98%) think they are hearing DA

McGurk, H. and MacDonald, J. 1976. Hearing lips and seeing voices. Nature 264:746-8.
Hearing Cs that aren’t there II:
V formants   C 
Identification of
burstless stops
with different
vowels:
transitions
are
all you
need!
Delattre, Liberman, & Cooper (1955) JASA 27, 769-773
Phonemes vs. allophones

t : th in English (allophonic) vs. Thai
(phonemic)
Speakers produce consistent and finely controlled
distinctions between allophones
 When measured experimentally, subjects typically
distinguish allophones at above chance levels, but
much less easily and well than phoneme pairs

• for aspiration in English, see Pegg & Werker 1997,
Whalen et al. 1997, Utman et al. 2000, Jones 2001
L2 transfer
Phoneme vs. allophone


Phonemes can distinguish
meaning (flight : fright)
Allophones don’t
Orthography
Phonological mediation

In semantic disambiguation tasks, the phonology of a word
interfered with the disambiguation process even when the
reader had access to its semantic representations (Van
Orden 1987).
 reader was presented with IS THIS A FLOWER?
 followed by the presentation of ROWS
 Correct response accuracy for such sentence-target pair
was significantly lower when the targets shared a
phonological representation with a word neighbor that
semantically fit the question (i.e., ROSE).
 when the subject had the ability to resolve the task with the
correct answer (NO) phonological information intervened.
Letter priming
 Lee
and Turvey 2003
 forms
with deleted silent letters (SALM,
COLUM) prime better than forms with deleted
pronounced letters (COUSI)
 Conclusion: phonological representations are
activated during visual word recognition.
Speech errors
Types of speech errors
Language games
Backwards English

Cowan & Leavitt 1992 study of one woman
Example: garage [graž] reversed as [žarg]
 Evidence that she reverses phonemes (rather than
letters):

• 1. no silent letters pronounced in reverse forms
• 2. homographs were always pronounced differently (two
<g>'s in garage)

Not functioning as reversed tape recorders:
• Compound units (diphthongs and affricates) were
consistently preserved as units rather than being
reversed.
• choice [tšojs] was reversed as [sojtš] (rather than *[sjošt])
Verlan

Procedure

Invert syllables (in polysyllabic words)
• L’envers ‘the reverse’ → [verlan], femme ‘woman’ → meuf
• Cf. Spanish Vesre, which inverts syllable order, e.g. muchacho → chochamu


What to do with monosyllables?




Include final optional schwa, if there is one
moi [mwa] ‘me’ → ouam [wam]
fou [fu] ‘crazy’ → ouf
Viens chez ouam soir-ce y'a une teuf de ouf, je suis avec l'autre
nasbo, j ai du sky et la race de beuh
‘Come to my place tonight there is a huge party, I'm with this hot chick,
I've got some whiskey and a lot of weed’
 teuf - fête ‘party’, nasbo = bonasse ‘hot chick’, sky = whiskey, beuh =
herbe ‘weed’
Key for us: monosyllables  inversion of phonemes


Drop final vowel
Conclusions
Words are not acoustic signals, but rather abstract
mental representations of language.
 Words in turn are composed of abstract
phonemes, which are again abstract mental
symbols rather than elements of the physical
world.

References
Bedore, L., L. Leonard, and Jack Gandour. 1994. The substitution of a click for sibilants: a case study. Clinical linguistics & phonetics 8.4:283-293.
Bishop, D. and J. Robson. 1989. Accurate non-word spelling despite congenital inability to speak: phoneme-grapheme conversion does not require subvocal articulation. British
Journal of Psychology 80.1:1-13.
Burzio, Luigi. 1998. Multiple Correspondence. Lingua 104:79-109.
Bybee, Joan. 2001. Phonology and Language Use. Cambridge: Cambridge University Press.
Cowan, Nelson & L. Leavitt. 1992. Speakers' access to the phonological structure of the syllable in word games. In M. Ziolkowski, M. Noske, & K. Deaton (eds.), Papers from the
26th Regional Meeting of the Chicago Linguistic Society, Volume 2: The Parasession On the Syllable in Phonetics and Phonology. Chicago: Chicago Linguistic Society.
Cruttenden, Alan. 1985. Language in infancy and childhood, second edition. Manchester: Manchester University Press.
Delattre, P., Alvin Liberman, & F. Cooper. 1955. Acoustic loci and transitional cues for consonants. Journal of the Acoustical Society of America 27.4:769-773.
Faber, Alice & Cathi Best. 1994. The perceptual infrastructure of early phonological development. In R. Corrigan, G. Iverson, & S. D. Lima, eds., The reality of phonological rules,
pp. 261-280. Philadelphia: John Benjamins.
Ferguson, Charles. 1986. Discovering sound units and constructing sound systems: it’s child’s play. In Invariance and variability of speech processes, Joseph Perkell and Dennis
Klatt, eds., 36-51. Hillsdale, NJ: Erlbaum.
Hirsh-Pasek, Kathy, & Roberta Golinkoff. 1993. Skeletal supports for grammatical learning: What the infant brings to the language learning task. In C. K. Rovee-Collier & L. P.
Lipsitt, eds., Advances in infancy research, Vol. 8, pp. 299-338. Norwood, NJ: Ablex.
Ito, Junko and Armin Mester. 2003. Japanese morphophonemics: markedness and word structure. Cambridge, MA: MIT Press.
Johnson, Keith, Dominic Pisoni, and R. Bernacki. 1990. Do voice recordings reveal whether a person is intoxicated? A case study. Phonetica 47.3-4:215-37.
Jongman, Allard. 2004. Phonological and Phonetic Representations: The Case of Neutralization. Proceedings of the 2003 Texas Linguistics Society Conference: Coarticulation in
Speech Production and Perception, Augustine Agwuele, Willis Warren, and Sang-Hoon Park, eds., pp. 9-16.
Krifi et al. 2003: phonological features shared between (phonological correspondents of) graphemes increase priming effects
Lee, Chang and M. Turvey. 2003. Silent Letters and Phonological Priming. Journal of Psycholinguistic Research 32.3:313-333.
Lisker, Leigh and Arthur Abramson. 1970. The voicing dimension: some experiments in comparative phonetics. Proceedings of the Sixth International Congress of Phonetic
Sciences (1967), Prague.
McGurk, H. and MacDonald, J. 1976. Hearing lips and seeing voices. Nature 264:746-8.
Olson, A. and Alfonso Caramazza. 2004. Orthographic structure and deaf spelling errors: syllables, letter frequency, and speech. Quarterly Journal of Experimental Psychology A
57.3:385-417.
Pegg, J. & Janet Werker. 1997. Adult and infant perception of two English phones. Journal of the Acoustical Society of America 102:3742–3753.
Piras, F. and P. Marangolo. 2004. Independent access to phonological and orthographic lexical representations: a replication study. Neurocase 10.4:300-7.
Remez, R., P. Rubin, D. Pisoni, and T. Carrell. 1981. Speech perception without traditional speech cues. Science 212:947-50.
Sapir, Edward. 1933. La réalité psychologique des phonèmes [The psychological reality of phonemes]. Journal de Psychologie Normale et Patholoique 30:247-265. [Reprinted in
English translation in his 1949 selected writings.]
Sumner, Megan. 2003. Testing the abstractness of phonological representations in Modern Hebrew weak verbs. Doctoral dissertation, State University of New York at Stony
Brook.
Van Orden, G. 1987. A ROWS is a ROSE: spelling, sound, and reading. Memory and Cognition 15.3:181-198.