History of Phonology

1
History of Phonology

• with an emphasis on recent history

2
1900-1930

• Development of Phonetics are a special branch of
  linguistics
• Unlike historical linguistics, also concerned
  with sounds through its preoccupation with sound
  change, phonetics was firmly rooted in synchronic
  analysis
• Articulatory phonetics
• Acoustic phonetics

3
new tools

• spectrograph
• X-ray photos (and films)
• sound recordings

4
Phonology

• Off-shoot of phonetics
• Strictly devoted to those aspects of sound
  structure which are linguistically relevant
• E.g. pitch differences related to tone or accent
  are phonologically important, pitch differences
  related to sex are not
• First International Congress of Linguists in The
  Hague in 1928 is often viewed as the beginning of
  phonology, set off by

5
Prague school

• definition of phoneme
• importance of binary oppositions
• marked vs unmarked member of pair
• neutralization
• languages are systems you cant take out one
  thing and study it separately that way you lose
  information about various contrasts within the
  language

6
Prince Nikolay Sergeyevich Trubetzkoy 1890-1938

7
Roman Jakobson1896-1982
8
Jakobsons accomplishments

• wide-ranging scholar
• worked on Russian case, phonological theory,
  poetics, and numerous other topics
• introduced the Prague school to the USA
• integrated work on language acquisition and
  language loss by aphasia in linguistic theory

9
Generative phonology

• Morris Halle and Noam Chomsky started working on
  phonology in the 1950s
• Culminating in The Sound Pattern of English (1968)

10
Morris Halle
11
Morris Halle, continued

• student of Roman Jakobson
• likewise of Russian (actually, Latvian) descent
• worked primarily on Slavic and English
• in his The Sound Pattern of Russian, Halle
  attacked the classical phoneme
• with Chomsky, developed generative phonology
  (1956-1968 after 1968 Chomsky stopped doing
  phonology)

12
The Sound Pattern of English 1968

• Authors Chomsky and Halle
• Should have been Halle and Chomsky
• Important for its formalization of phonological
  representations, rules, and its methodology
• Discusses many major issues in the phonology of
  English, including phonotactics, phonological
  rules, and stress assignment in underived,
  derived and compound words

13
Segments

• defined as a bundle of features
• e.g. feature-1
• feature-2 -
• feature-3
• feature-4 -
• etc.
• Features have a standard phonetic
  interpretation, in terms of articulation
  (Jakobson had proposed an acoustic
  interpretation)

14
One exception to binary features

• To capture four levels of stress, Chomsky and
  Halle used numeral values for stress features 1
  stress, 2 stress, 3 stress and 4 stress
• So features, in SPE, come in 2 types
• boolean valued features (/-)
• numerically valued features

15
Rules

• context-sensitive rules
• A ? B / C __ D
• however, not involving whole segments, but
  features, or sets of features
• many new notational devices were introduced, to
  formulate rules a notation, curly brace
  notation, etc.

16
Methodology

• economy basic principle
• feature-counting evaluation metric
• highly abstract underlying forms
• complex derivations, involving the phonological
  cycle
• phonotactics done by rule
• synchronic analysis became a mirror of diachronic
  analysis in SPE

17
E.g.

• Dutch has no diphtongs before /r/
• Historical account diphtongization never took
  place before /r/
• Possible synchronic account assume diphtongs are
  underlying monophongs, and diphtongize them
  unless followed by /r/
• Advantages reduces the inventory of underlying
  segments (economy), and derives the phonotactic
  generalization

18
Disadvantages

• Need to use exception features, e.g. for loans
  that came into the language after the sound
  change (minuut, titel)
• Mixes up diachrony and synchrony
• Overly abstract learnability issue

19
Reactions to SPE

• immediate and wide following
• many phonologists embraced the methodology,
  notation and ideas, to describe phonological
  problems in a variety of languages, thus creating
  the field of generative phonology

20
However, there was also an immediate backlash

• Abstractness natural phonology (David Stampe,
  Patricia Donegan, Theo Vennemann, Joan Bybee
  (Hooper))
• Morphology new separation of word-based sound
  regularities from general sound regularities
  (Mark Aronoff, Paul Kiparsky)
• Autosegmental phonology explosion of the segment
  (John Goldsmith, Nick Clements, etc.)

21
Abstractness

• Need for absolute neutralization?
• Absolute neutralization underlying form never
  shows up as surface form
• In SPE, this was a common phenomenon
• Learnability problem only if children use the
  same methodology as Chomsky and Halle, will they
  arrive at the same underlying forms

22
Autosegmental phonology

• originated in the study of tone languages, where
  it was noted that
• tonal features (like High Tone) may stretch over
  many segments, sometimes entire words
• and when they change, e.g. through assimilation,
  all segments bearing the tone change

23
Suggestion (Goldsmith)

• get rid of the absolute slicing hypothesis
• put tonal features on a separate level (called
  tier), and then connect them to the various
  segments bearing the tonal features
• allow the connection to be not one-to-one, but
  many-to-many

24
So,

• One segment may bear two tones (e.g. Hi-Lo, heard
  as falling tone and Lo-Hi, heard as rising)
• And one tone may be connected to many segments

25
Notation
Hi
Lo
Tonal tier
C
Segmental tier
V
C
26
Floating tones

• are tonal features not (yet) associated with a
  segment
• can be linked in the course of a derivation
• may be separate morpheme
• or originate through deletion of a segment