2.1 Discuss the different analytic approaches in the context of Phonetic and Phonological Analysis
Different Analytic Approaches
Speech is a continuum with few points in the stream which constitute
- a) Natural breaks – to breath in, swallow, and cough.
- b) Breaks that show an articulatory, auditory or acoustically steady state that could serve as a basis for analytical segmentation into real phonetic units. For instance, the series of words – ‘the red pen’ has breaks that allow analytical segmentation.
The segmentation that enables the view of real phonetic segment of speech is mostly imposed. Human beings do not say distinct sounds but a continuum of speech sounds with anticipatory articulation of each sound in the series. However, for a detailed study of human speech sounds, segmentation is prerequisite.
Two chief approaches to segmentation
The parametric approach
In the parametric approach there are parallel segmentations of speech. Each component of the vocal performance is treated as a parameter whose value is in the state of constant change. The components could be
a) Auditory such as pitch of voice or loudness of the voice, for example, the series of words
Do you really mean that?
could be said in many different ways. For instance it could be said in two ways with the same pitch-pattern but due to the variation in loudness of voice one ends up being a shout and the other a murmur.
b) Initiatory such as syllable structure. In the initiatory component, the potential values
of a feature are continuous but have a measuring system to limit them. For example,
the syllable structure in each language determines the number of sounds that can
occupy each of the parts of the syllable.
The linear approach
The linear approach is a serial segmentation in which the speech continuum is divided into units of varying durations. For example, in the analyses of the degrees of stricture, the following durations are established
Maintainable duration – fricative
Momentary duration – flap/tap
Each of the speech units can then be characterized in terms of representative values shown during the production of that unit by the individual phonetic components making up the performance.
In the linear approach, any phonetic feature can have only a limited number of values, which are referred to as categories. For instance, pitch is a phonetic feature, which has three values only, namely; high, mid, low.
The linear categories are thus discrete. This makes the linear featural categories convenient for use in the spoken and written description of the speech continuum. This explains why the linear analysis is the one taken in most textbooks on phonetics.
The representation of the speech continuum
When the speech continuum is analysed, either parametrically or linearly the speech continuum is presented in terms of the following:
a) Symbols and diacritics
b) In terms of phonological rules (algorithms) such as /r/ [ŗ] / t, p
This rule would be rewritten as “/r/ becomes devoiced when it immediately follows the
voiceless stops /t, p/.”
c) Phonotactics – distribution statements.
d) Metrical trees
The analyses of the speech continuum include the analyses of phenomena attributed to ‘strings of sounds’ such as stress, the syllable, pitch and rhythm. Each one of these is attributed to measurements in metrical analysis. These measurements are presented in metrical trees. For example, the syllable structure tree presented here below
e) Metric numbers
Some phonological phenomena can be calculated and assigned a numerical value. For instance, stress can be assigned a numerical value because words have relative degrees of stress.
Example the presentation of numerical metrical numbers for stress as illustrated here