LST 322

Non generative phonology

Group 7

Topic: Autosegmental phonology and metrical phonology

Before autosegmental phonology

The analysis of suprasegmental came late to the generative phonology even though it had been
tackled in American structuralism with the long components of Harris 1944 article" simultaneous
components in phonology" and despite being a particular focus of firthian prosodic analysis. The
standard version of generative phonology of the era (Chomsky and Halle's the sound pattern of English)
made no special provisions for phenomenal that had been labeled suprasegmental or prosodic by earlier
traditions.

What autosegmental phonology is

Auto segmental phonology is a non- linear approach to phonology that allows phonological
processes such as tone and vowel harmony to be independent of and extend beyond individual
consonant and vowels. As a result, the phonological processes may influence more than one vowel or
consonant at a time. Examples: ( Mendes, Sierra Leone)

In an auto segmental analysis of Mende, tone is not a property of individual vowels or syllables but it is a
property of the word as a whole

Tone 1 syllable 2 syllable 3syllable

H ndá mouth ngúlú'tree' kélélé ' fraction'

L kpa ' debt' bèlè' trousers' kpàkàli ' chair'

Hl mbû 'owl'. kényà 'uncle' félàmà 'junction'

LH mbâ 'rice' nàvó 'money' ndàvàlà 'sling'

LHL. mbà 'companion' nyàhâ ' woman' nìkílí 'peanut'

Formal representation:
The framework

In the autosegmental frame work a toneless syllable is assigned the tone of an adjacent syllable
creating a one to many relationship between a tone and a set of tone- bearing units. As a theory of
phonological representation, autosegmental phonology developed a formal account of ideas that had
been sketched in earlier work by several linguists, notably Bernard Bloch (1948), Charles Hockett (1955)
and J. R. Firth (1948). According to such a view, phonological representations consist of more than one
linear sequence of segments; each linear sequence constitutes a separate tier. The co-registration of
elements (or autosegments) on one tier with those on another is represented by association lines. There
is a close relationship between analysis of segments into distinctive features and an autosegmental
analysis; each feature in a language appears on exactly one tier.The working hypothesis of
autosegmental analysis is that a large part of phonological generalizations can be interpreted as a
restructuring or reorganization of the autosegments in a representation.

Who proposed the theory

Autosegmental theory was proposed or introduced by John Goldsmith in his PhD thesis in 1976 at the
Massachusetts Institute of Technology (MIT) in his dissertation to represent tone and other
suprasegmental phenomenal. Goldsmiths intuition, embodied in the term he created was that
autosegments constituted an independent conceptually equal tier of phonological representation with
both tiers realized simultaneously like the separate voices in the musical score.

Goldsmiths theory: Reasons, problem and Solutions to the sound patterns of English

The autosegmental framework captured that tones required a separate tier of representation which was
the concept of tonal stability, since a sequence of elements on one tier can be connected to a single
element on another. This autosegmental model was a natural account of time spreading.

Goldsmiths autosegmental solution was the well formedness condition. Which was at the very core of
autosegmental framework was a rare constraint, posited nearly two decades before optimality theory.
The well formedness condition required that every tone on the tonal tier be associated with some
segment on the segmental tier and vice versa. Tones thus spread more or less automatically to segment
lacking.

The problem of too many association and spreading onto adjacent elements which are not confined to a
tonal characteristics emanated.

The notion of tiers representation soon gave way to a model with one basic tier connected to tiers for
particular kinds of articulation, tone and intonation, nasality, vowel features and others. This lead to the
hierarchical representation of phonological features in current models of feature geometry and so on.

Autosegmental representations and processes also provide a means of representating non

What is metrical phonology?

Metrical phonology is one of the main approaches devoted mainly to the study of phonological analysis
of the structure and acoustic of stress. Metrical phonology is the branch of linguistic theory concerned
with stress phenomenal in natural language. It is distinguished from previous approaches because it
proposes a hierarchical structure reminiscent of the structures used in traditional discussion.

What are metrical trees

Metrical trees shows hierarchical structure in the syllables of a word in a binary branching
structure strong- weak (sw) or weak-strong ( ws). Stress is thus seen as a relational property, the
relationship a node had with it's sister node determines its status.( Kager 1995:378) for a word the
prosodic hierarchy is shown below:

Pw - prosodic word ( word at it rythmic level)

F - foot ( the next larger rhyth

O - syllable ( a unit of pronunciation)

U - mora ( element in the rhyme of a syllable)

There are times were the metrical trees may show only the syllables and word level. The quality of the
vowel determines to a large extent if the syllable is strong or weak. In other words, when drawing the
metrical tree of a word the vowel quality has to be considered as an important part in relation that a
node ( syllable) has with the node(syllable) next to it determines if the syllables will be labeled strong or
weak.

Similarities between Autosegmental phonology and metrical phonology

1. AP and MP addressed different deficiencies of the (sound pattern of English) Though they focused on
different phenomenal, but their goal was to correct the deficiencies of the sound pattern of English by
Chomsky and Halle.

2. They are both major contributions to the field of phonology.

3. Both studies the sound structure of language.

4.Deals on syllable and tones, stress, etc

5. Both uses branching node

6. They are both a non-linear approach to the study of phonological processes.

Difference between autosegmental phonology and metrical phonology

1. Autosegmental phonology or AP uses association lines known as the skeletal tiers for its theoretical
framework while MP makes use of metrical trees and grid.

2. AP account for tones phonogical processes like assiilation-nazalisation, vowel elision, etc while MP
account for stress and syllables, linguistics prominence.

3. Autosegmental phonology could not account for primary stress unlike metrical phonology that can
account for both primary and secondary stress and also stress prominence in a word.

Advantages of autosegmental and metrical phonology

1. It correctly predicts the ambiguity, between broad and narrow focus. There are two possible metrical
patterns for two words phrase: strong- weak( sw) and weak-strong ( w-s). For example : Gusskied can be
pronounced either as Gus skied which is strong-weak or Gus Skied which is weak- strong.

2. It is consistent with the pattern of deaccenting in which accents can shift from both left and right. This
is because swapping S and W nodes will cause stress to move left if the S nodes was originally at the
right and move right if it was originally on the left.