Phonology Phonetics

1
Phonology ? Phonetics

• Understanding Features

2
Richness of the Base

• The source of all systematic cross-linguistic
  variation is constraint reranking. In particular,
  the set of inputs to the grammars of all
  languages is the same. The grammatical
  inventories of a language are the outputs which
  emerge from the grammar when it is fed the
  universal set of all possible inputs (Smolensky
  1996).

3
ROTB

• Constraint ranking is everything!
• Simplistic view of features
• Have specific phonetic definitions.
• Features mapped to phonetics.

4
What about variability?

• Views on variability
• Multiple grammars
• Variable/unranked constraints/rules
• Stochastic OT/processes

5
Enter features

• View from the world of phonological contrast.
• Some featural alternation could be feature
  interpretation, i.e., phonology ? phonetics
  mapping.
• Not all variation is in the phonology.
• Some might be allophonic (in the phonology).
• Example Voicing assimilation in English
• Others might be allo-phonetic (in the phonetics).
• Jeff and I coined the second term. ?.
• Example onset s might be diff. from a coda
  s .

6
Laryngeal Features

• Classic view
• /- voiced
• Any voicing distinction is to be interpreted in
  terms of this feature.
• English/German - b - voiced
• ph - -voiced
• French/Spanish - b - voiced
• p - -voiced

7
Simplified Phonetic Facts 1

• English/German
• a. /orthographic-b/
• b - fully voiced
• (or) p - voiceless unaspirated
• (utterance initially, variably)
• b. /orthographic-p/
• ph - aspirated voiceless
• (or) p - voiceless unaspirated
• (after s, other v.less. frics?)

8
Simplified Phonetic Facts 2

• French/Spanish
• a. /orthographic-b/
• b - fully voiced
• b. /orthographic-p/
• p - voiceless unaspirated
• (after s, other v.less. frics?)
• - variable voicing

9
Further simplified facts

• English/German - /b/ is variable
• /ph/ is consistent
• French/Spanish - /b/ is consistent
• /p/ is variable

10
Laryngeal Realism

• English/German
• /orthographic-p/ is specified for aspiration.
• /orthographic-b/ is unspecified, so it varies (to
  some extent).
• French/Spanish
• /orthographic-b/ is specified for voicing.
• /orthographic-p/ is unspecified, so it varies (to
  some extent).

11
Laryngeal Realism

• English/German
• ph vs. Ø
• French/Spanish
• b vs. Ø

12
Assumption in Laryngeal Realism

• Couched in underspecification theory
• Unspecified features do not play a role in
  phonology.
• No processes use them as triggers (in structural
  descriptions).
• Testable statement
• Seems to be true of laryngeal features.

13
However, we havent talked about features yet!

• English/German
• ph vs. Ø
• If consistently aspirated, then must be specified
  for aspiration spread glottis

14
However, we havent talked about features yet!

• French/Spanish
• b vs. Ø
• If consistently voiced, then must be specified
  for voicing voiced

15
What can be a possible feature?

• Theory of features
• Avery-Idsardi (2001) Innovation
• Articulators, Dimensions, Gestures.
• Articulators group dimensions.
• Dimensions group antagonistic gestures.

16
3D Larynx

• 3D Larynx

17
Dimensional Theory
18
Essence of Dimensional Theory

• Phonology looks as far as Dimensions, no lower.
• To be honest, there are exceptions to do with
  headhood we wont go into that part of their
  theory.

19
Dimensional Theory
Phonological Features
20
However, we havent talked about features yet!

• English/German
• ph vs. Ø
• X vs. X
• GW Ø
• If consistently aspirated, then must be specified
  for Glottal Width.
• Phonetics takes care of aspiration or glottal
  constriction.

21
Dimesional Invariance
Consistently Glottal Width or GW
22
Dimensional Theory
23
French/Spanish quick look

• French/Spanish
• b vs. Ø
• X vs. X
• GT Ø
• If consistently voiced, then must be specified
  for voicing Glottal Tension

24
Dimensional Theory
25
Grand Vision for phonological features
26
What does this mean for nasality?

• /nasal are under the dimension of soft
  palate.

27
Grand Vision for phonological features
28
SP and its gestures
Nasal segment
Oral segment
Antagonistic gestures. One Dimension
Soft Palate (SP)
29
The Dimension SP

• Soft Palate.
• Root X
• Dimension SP
• / \
• Gesture nasal oral

30
However

• Phonology (usually) looks as low as dimensions,
  and no lower!

31
The Dimension SP

• Soft Palate in the phonology.
• Root X
• Dimension SP
• / \
• Gesture nasal oral

Purview of Phonology
32
Implications

• A segment can be marked for the dimension SP in
  the phonology, but no more!
• X
• SP
• (nasal or oral gestures are part of the
  phonetic implementation)

33
The Dimension SP

• You cant mark nasal or oral in the
  phonology!
• Coolest prediction ever!
• If specified for phonological nasality,
  segments can surface with (phonetic) non-nasal
  variants cos the phonetic gesture isnt
  specified!
• Because
• X ? X (or) X (or) X
• SP SP SP SP
• nasal oral Ø

34
Partially-nasal stops

• What are they?
• Segments with both a nasal and an oral portion.
• Recording
• Whole phrase
• Word
• kapan sounds like kapadn

35
Partially-nasal stops

• At least 2 kinds
• Nasal-based
• No different from simple nasals in the phonology.
• Behave like nasals spread nasality
• Voiced-based
• Simple voiced stops in the phonology, unmarked
  for phonological nasality.
• Nasal on the surface for phonetic reasons
  (enhancement).

36
Categorical Phonology

• All types of Partially Nasal Stops (can) show
  surface variability.
• Nasal-based - m or mb or b
• Voice-based - mb or b (or m ??)
• But, they NEVER trigger nasalization variably.

37
Logical Possibilities

• Because Nasal-based PNS show surface variation,
  it is logically possible that
• If the surface alternant is m,
• aham ? aham (nasalization occurs)
• If the surface alternant is b,
• aham ? ahab (no nasalization occurs)
• If the surface alternant is bm,
• aham ? ahabm (no nasalization)

38

• Therefore, expected variation for the same word
• aham ? aham
• (or) ahab
• (or) ahabm

39
Observed fact

• There is NEVER any such variation.
• (If they nasalize adjacent segments, they always
  do!)
• Jambi Malay (Tadmor Yanti 2004)
• (i) ayam ? aya(b)m chicken
• (ii) ayam-e ? ayame his chicken

40
In the phonology

• In the phonology, a nasal-based PNS is a simple
  nasal.
• So, it always triggers nasalisation, if the
  language has a nasalisation process.
• Surface variation is a result of the phonetic
  mapping, NOT phonology.

41
What about Voice-based PNS?

• Observed Variation
• mb or b (or m ??)
• However, in the phonology, they are not specified
  for nasality.
• They NEVER spread nasality.

42
Conclusion

• Contra the claims of Classic OT architecture
  (ROTB), not all the predictable facts are in the
  phonology!
• At least some are to be captured at the
  phonology-phonetics interface.