Reading quiz qns

1. Who is the linguist who first proposed that distinctive features should be binary?
   1. Roman Jakobson
2. Sounds that are [+cons] are [-son]. T/F?
   1. False (there are sonorant consonants)
3. Which of the following sounds poses a potential contradiction to the hypothesis that features and feature definitions are universal?
   1. Some languages use [ATR] and some use [tense]
   2. [l] problem
4. Which of the following sounds tend to form a natural class, crosslinguistically: dentals, labiodentals, bilabials?
   1. Bilabials + labiodentals
5. The feature [high] has only been used for vowels. T/F?
   1. False (palatals)

Distinctive Feature theory

The theory of the set of features that combine to define all possible human language sounds.

The features are hypothesised to be the same across all languages.

There are 3 different versions of DF theory (of which we will mostly go with the one from Zsiga’s textbook).

If a combination of segments do not form a natural class, this means that no feature combination will isolate them from all the other segments.

Language hypotheses

1. Phonological Alternations target only natural classes
   1. i.e. all segments in the same class should undergo same alternation under a rule
2. Contrasts, classes and alternations are closed sets of phonetic parameters same across all languages.
   1. These parameters are called distinctive features.
   2. Definition of natural class: set of phonemes that share certain distinctive features.
3. Features are binary oppositions (+/-) for every segment
4. Every lexical item has a UR
   1. Underlying Representation (UR) = combination of distinctive features
5. UR $\rightarrow$ SR: via feature changing rules

Notes

Structuralism: paradigm that language is a

• self-contained
• self-regulatic
• semiotic (making meaning) system
• it’s elements are defined in relation to other elements within the system.

If the question asks “Which two form a natural class”: make sure the union of their features doesn’t form a set that include other sounds in the language.

BUt just cos a set of sounds for a natural class doesn’t mean that there is some phonological process/rule that tarrgets that class.

e.g. Obstruent assimilation in Russian may not apply to all natural classes

Major Class and Manner Features

Binary features

1. [+/- Syllabic] Can form syllable nucleus?
   1. [+syl] Vowels
   2. [-syl] All others
2. [+/- Consonantal] Significant vocal tract constriction?
   1. [-cons] Vowels and glides [j, w, ɰ, ɥ]
   2. [+cons] All other sounds
3. [+/- Sonorant] Build up of pressure behind oral constriction
   1. [-son] obstruents (plosives, fricatives, affricates)
   2. [+son] All other sounds
4. [+/- Continuant] complete closure in the oral cavity?
   1. [-cont] plosives, affricates and nasals.
   2. [+cont] All other sounds

Every sound has a value for each of the binary features.

Unary Features

The lack of evidence for targeted classes such as [−nasal] and [−lateral] have led to the suggestion that at least some features are unary rather than binary. In particular, unary features for place of articulation based on active articulator have proven to be useful

Sounds don’t form a natural class ever amongst sounds without nasal/lateral/DR. Hence we don’t give the opportunity of creating the class that way.

There is no phonological alternation that only classifies the non-nasal sounds, or the non-lateral sounds. Languages don’t have target non-nasals together.

1. [+/- Nasal]
2. [+/- Lateral]
3. [+/- Delayed release]
   1. natural classes and alternations: affricates are [+delayed release]; stops and all other segments are [−delayed release]

A lack of complete featural specification of a given segment is termed underspecification. Absence of a unary feature (as when non-nasal consonants simply have no [nasal] specification) is one type of underspecification

Problems

Definitional problems: [l] and [+/- cont]. Should it be complete closure or does closure in the centre only count as [-cont]?

Problems with Jakobson’s method: sounds’ manner features may not be binary (e.g. affricates are both [-cont] at the stop and [-cont] at the fricative articulation)

Laryngeal Features

1. [+/- voiced]
2. [Constricted Glottis]
   1. ejectives, glottal stop and creaky voice
3. [Spread Glottis]
   1. aspirated sounds

Major place distinctions

Features based on active articulator

WE need the bigger classifications.

1. [labial]
2. [coronal]
3. [dorsal]
4. [pharyngeal]
5. [laryngeal]

What if just (dentals, alveolars) as a natural class? We don’t have a way. This is why we hve subsidiary place features.

Subsidiary place distinctions

Coronals only (Place):

1. 1. [+ant] in front or on alveolar ridge
   2. [-ant] otherwise
2. [+/- distributed]:
   1. [-dist] apical (tongue tip) – Alveolar and Palatal
   2. [+dist] laminal (tongue blade) – Dental, Post-Alveolar and Retroflex

Fricatives/affricates only (Manner)

1. 1. [+str] Labiodental, Alveolar and Post-alveolar (Sibilants), Uvular fricatives & affricates)
   2. [-str] otherwise

High and Low (post-alveolar to palatal, palatal to uvular

1. [+/- high]
   1. [+ high] post-alveolar and retroflex
   2. [- high] palatal
2. [+/- low]
   1. [+ low] all guttaral consonants (palatal to uvular), the dorsals + palatal
   2. [- low]

No language ditinguishes bilabial vs labiodental, or post-alveolar vs palatals, or velars vs uvulars for non-fricatives/affricates

Languages don’t contrast between bilabial and labiodental OR post-alveolar and palatal or velar/uvular for NON-FRICATIVES/AFFRICATES.

But there are languages that contrast velar and uvular stops.

Vowel features

1. [+/- round]
2. [+/- high]
3. [+/- low]
4. [+/- back]
   1. [+back] central and back
   2. [-back] front
5. [+/- advanced tongue root (ATR)]
   1. [+ATR] Articulatory: tongue is pulled forward
   2. [+ATR] Acoustic: Lowered F1 compared to [-ATR] counterpart

Backness: only 2 [+/- back]

• What about english? Then ADD ONE MORE
• [+front] = front
• [-front] = central & back
• [+front, -back], [-front, -back], [-front, +back]

Rounding: only 2 [+/- round]

Backness * Rounding: 4