Phonetic / phonological typology

1. Phonetic / phonological typology

Parameters of variation,
principles of generalization
Based on
Maddieson 2010, 2015

2. Components of typological study: a reminder

Limits to variation
Parameters of variation
Correlations between parameters
– AKA implicational universals
Explanatory approach

3. Limits to elaboration (pulmonic consonants)

IPA Chart, http://www.internationalphoneticassociation.org/content/ipa-chart, available under a Creative Commons
Attribution-Sharealike 3.0 Unported License. Copyright © 2005 International Phonetic Association.

4. Limits to elaboration (non-pulmonic and vowels)

IPA Chart, http://www.internationalphoneticassociation.org/content/ipa-chart, available under a Creative Commons
Attribution-Sharealike 3.0 Unported License. Copyright © 2005 International Phonetic Association.

5. Attestedness vs. frequency

The idea behind IPA – a universal
alphabet for transcription
Not all of these segments are equally
frequent, and some are rare

6. Attestedness vs. frequency

phoible.org (2155 lgs, unbalanced sample)
– m present in 85 percent of languages, x in 18
– u present in 87 percent, e present in 68, ø in 2
an example of rare sounds – glottalized
(laryngealized) sonorants
– individual systems attested in Mesoamerica, one
among Austronesian and one in Daghestan

7. Consonant inventories: ‘size principle’

basic set ~
p, b, t, d, k, g, ʔ, tʃ, m, n, ŋ, f, s, ʃ, l, r, w, j, h
typically involve place of articulation x voicedness
– though many lgs lack voicedness altogether
“consonants with places and manners of
articulation outside those represented in this set
… tend only to occur in inventories with a larger
total number of consonants” Maddieson 2010
– see below for why

8. Vowel inventories

Basic set ~ i e a o u, which involve contrast in
backness, roundedness, height
– but in what precedence?
Triangular (most common)
… vs. vertical systems --- Northwest
Caucasian, Arandic, Chadic
– height comes first

9. Vowel inventories

No language contrasts roundedness alone
A small minority of languages have front
rounded or back unrounded vowels
Height > Backness > (?) Roundedness
additional properties – nasalization,
pharyngealization, length – are added to
plain vowels – cf. ‘size principle’ for
consonantal inventories

10. ‘size principle’: why?

Note that it is exactly the size principle
effects that give rise to implicational
universals
Various explanations: ontogenetic,
philogenetic, system-internal etc.
Maddieson: “Pervasive pragmatic
requirements for efficient communication”;
“better balance between … relative ease of
articulation and relative perceptual salience”

11. Explanation: voiced fricatives

Maddieson 2010: out of 637 lg sample, 62%
have voiced obstruents and only 35% have
voiced fricatives. Why?
Generation of a high velocity airstream
required for fricatives is in conflict with closure
of vocal folds required for voicing
On the other hand, the airflow required for
vocal folds vibration is impeded by the oral
constriction

12. Explanation: voiced fricatives (ctd)

Maddieson 2010: But why voiced fricatives are
preferred in systems where there are already voiced
obstruents? Cf.:
Plosive voicing
Yes
No
Total
Fricative voicing
Yes
No
177
218
44
198
221
416
Total
395
242
637

13. Explanation: voiced fricatives (ctd)

Principle of economy – that system is more
compliant to the principle of economy which
uses less gestures.
“The use of a given feature in several
different consonants reduces the number of
distinct motor and perceptual patterns that
must be mastered by a speaker, compared to
a situation in which every consonant would
have a set of features unique to itself”.

14. Vowel harmony: types

Best known: front/back & rounding --- Altaic,
Uralic and other – least attested
Also attested: height harmony --- Itelmen,
some Bantu, Nez Perce
Most widespread: “cross-height” / ATRharmony --- Niger-Congo, Nilo-Saharan,
probably Mongolian and Tungusic

15. Vowel harmony: ATR

Most widespread: “cross-height” / ATR-harmony -
-- Niger-Congo, Nilo-Saharan, probably
Mongolian and Tungusic
Cf. standard Yoruba
u
i
e
o
ɛ
ɔ
a
“The size of pharyngeal cavity at
the back of the mouth”; green
vowels are produced “by pulling
the root of the tongue forward and
often also lowering the larynx”

16. Consonants vs. vowels

“there is a balance between vowel and consonant
size inventory” a widespread belief
Consonants
Vowels
Small Average
Large
Total
Small
47
153
65
265
Average
34
105
98
237
Large
34
87
57
178
Total
115
345
220
680
a widespread belief is false

17. Levels of pho-typology

Unlike morphology and syntax, and like
semantics, phonetics relates to many levels
of linguistic segmentation – syllable, word,
utterance
Syllables are of course decomposed in
smaller segments – but note the term!

18. A prosodic typology: syllable

What syllables are more natural of a human
language?
– Not an easy question when starting from a relatively
complex language like Russian or English
– strengths
Typologically, CV seems to be the default
structure. What kind of arguments may be used?

19. A prosodic typology: syllable

Typologically, CV seems to be the default
structure. What kind of arguments may be used?
–
–
–
types of syllables present in all languages --- CV
types of syllables readily absent from languages --- V, VC, CVC
(frequency) distribution of syllable types with individual
languages
The latter is an illustration of an important
typological principle – typological patterns may
sometimes be shadowed by variation within
individual languages

20. A prosodic typology: rythm

stress-timed
– one syllable in the word is prominent, other are
reduced in duration, speech rhythm is determined by
‘beats’ – stressed syllables --- German or English
syllable-timed
– syllables are near-equal in duration, speech rhythm is
determined by the total number of syllables --- French
or Spanish
mora-timed
– bi- vs. monomoraic syllables, the mean count of moras
per min is constant --- Japanese (CV 1 mora, CV: or
CVC 2 moras)

21. A prosodic typology: stress

stress – syllable prominence within word
no stress --- 135 our of 461 --- Yoruba
predictable position --- Korean, stress falls on
the second mora (?) --- 195 --- what is the
function of the stress?
demarcation
contrastive/lexical stress --- Russian --- 131

22. A prosodic typology: tone

“pitch tied to particular lexical or grammatical
forms” --- 153 lgs out of 461
level tone --- from 2 to 5 --- Africa (typically 2)
and North / South America
contour tone --- South East Asia and
Mesoamerica

23. A prosodic typology: stress and tone?

Stress may combine with tone!
Tone
Stress
Lexical
Predictable No stress Total
18
34
101
153
Non-tonal 113
161
34
308
Total
195
135
461
Tonal
131

24. Conclusions