Parametric description and long-domain features

1. Segmental vs. parametric views of speech

In studying the IPA alphabet, we may become so focussed on the linear sequence of consonants and vowels that we forget the fundamental fact that speech is produced by six independently-controlled organs (lungs, larynx, jaw, velum, tongue and lips) all moving simultaneously and continuously.

Multiple parallel articulatory parameters

The figures above show the up-and-down movements of small metal pellets placed on several articulators (two on the tongue) during two utterances of the phrase "perfect memory". The raising of the tongue dorsum is labelled κ, of the tongue tip τ and of the lower lip β. It can be seen that in the upper version, in which the two words are clearly separated, movements κ and τ are strongly overlapping, whereas β comes later. In the lower version, in which the two words are run together, movements κ, τ and β are all rather strongly overlapping.

2. What can be a prosody? Virtually anything!

2.1. Laryngeal features

In addition to tone and intonation, several other laryngeal features are prosodic in some languages. In Cœur d’Alène (Schitsu’Umsh, Skitswish; Reichard 1938), a severely endangered Salishan language spoken in northern Idaho by few speakers (“less than a handful” in 1966), glottalization (probably a combination of creaky voice, glottal stops and glottalic consonants) is used as a diminutive morpheme:

mar-marím-EntEm-ilts	vs.	m’-m’ar’-m’ar’ím-En’tem’-il’ts
\___________________/		\__________________________/
not glottalized		glottalized
“they were treated one by one”		“they little ones were treated one by one”

A complex of creaky voice, pitch and loudness in Danish (stød) distinguishes words in a way that parallels the use of contrastive tones in Swedish or Norwegian, e.g. [bœn] ‘beans’ vs. [bœnʔ], or [t^hɑŋgṇ] ‘the thought’ vs. [t^hɑŋʔgṇ]. (See Laver 1994 and Grønnum 1998 for further examples.) Although I have transcribed these using a glottal stop symbol, the glottalization is more dispersed, and may involve creaky voiced vowel, creaky voiced sonorant consonants, falling pitch, and glottal closure. It is therefore usually analysed as a prosodic feature, rather than a segmental feature. Fischer-Jørgensen (1987) is a very thorough study (part of it conducted here in the Oxford Phonetics Laboratory in 1981! A copy is in our library.)

Voicing assimilation is another (perhaps less spectacular) example of a feature not staying nicely confined within a segment. For example, in Russian, there is a maximum of 1 voicing distinction in consonant clusters. A cluster takes on the voicing of the last consonant in the sequence, without regard to boundaries:

Morphemic boundary:	/gorod+k+a/	[gorotka]	‘little town’
Clitic boundary:	/mtsensk# bi/	[mtsenzgbɨ]	‘if Mcensk’
Word boundary:	/mtsensk## bil/	[mtsenzgbɨl]	‘it was Mcensk’

Sonorants e.g. [m] are not distinctively voiced, and they are transparent to voicing assimilation:

iz # mtsensk+a		[is mtsenska]	‘from Mcensk’
ot # mzd+i		[od mzdɨ]	‘from the bribe’

Some French examples:

anticipation of [+voice]:		anticipation of [-voice]:
je passe vite [paz vit]		chemin de fer [t fɛːʁ]
la tête droit [tɛd dʁwat]		coup de pied [ku t pje]
avec vous [avɛg vu]		esprit de corps [ɛspʁi t kɔ:ʁ]
place d’armes [plaz daʁm]		tout de suite [tu t syit]
		rez-de-chaussée [ʁe t ʃose]

2.2. Velum movement

Terena/Tereno (southwestern Mato Grosso, Brazil) (Bendor-Samuel 1960). Note that y = IPA [j].

3rd person possessives (non-nasal)		1st person possessives (nasal)
eˈmoʔu ‘his word’		ẽˈmõʔũ ‘my word’
ˈayo ‘his brother’		ˈãỹõ ‘my brother’
ˈowoku ‘his house’		ˈõw̃õŋgu ‘my house’

Sundanese (Austronesian, W. Java) (Robins 1957):

ɲaian [ɲãĩãn], to wet
miasih [mĩʔãsih], to love
kumaha [kumãhã], how?

2.3. Tongue root prosodies

2.3.1. Spread of velo-pharyngealization in Tashlhiyt Berber. This language (like various other Afroasiatic languages, such as Arabic) has a contrast between pharyngealized and non-pharyngealized coronal consonants, as well as some pharyngeal consonants. In words containing these consonants, entire syllables or even the whole word becomes pharyngealized, as you can hear from these audio clips:

lʁwrˁd ‘hill’

zʁwrˁ ‘to grease’

zrˁ ‘to watch’

tˁnˁgdˁtˁ ‘you drowned’

nnqqwrˁtˁ ‘silver’

2.3.2. [ATR] harmony

In a number of languages of West Africa, a particular pattern of vowel harmony is found in which the vowels fall into two sets: more advanced and higher vowels {i, e, o, u} and correspondingly less advanced and lower {ɪ, ɛ, ɔ, ʊ}. [a] is typically "neutral", and can co-occur in words with vowels of either set. In the more advanced and higher vowels, the tongue root has been found to be more advanced, and hence they are given the feature [+Advanced Tongue Root] or [+ATR]: the non-advanced vowels are consequently labelled [-ATR]. See Laver (1994: 289-291) and references therein, and Tiede (1996) for MRI images of Akan [i] vs. [I]. The vowel system of Ega (Kwa, Côte d’Ivoire) (Connell, Ahoua and Gibbon 2002) is quite typical:

High, [+ATR]	i			u
High, [-ATR]		ɪ			ʊ
Mid, [+ATR]	e			o
Mid, [-ATR]		ɛ			ɔ
Low			a

Examples of words containing vowels in the two sets (note that Ega also has tones, not transcribed here):

[+ATR]		[-ATR]
efi	‘eye’	ɔvɛ	‘silk cotton tree’
oji	‘cold’	ʊvɛ	‘dog’
efe	‘taking flight’	ɛɲɛ	‘arrival’
uje	‘bundle of wood’	ɔsɪ	‘woman’
emo	‘smelling’	ɛzɔ	‘quarrel’
ize	‘wood’	ɛcI	‘laughing’

2.4. Tongue body prosodies; lip prosodies

Vowel harmony systems involving frontness vs. backness, openness vs. closeness or lip-rounding vs. lip-spreading are widely documented in the phonetics and phonological literature, and exemplify the prosodic use of tongue body and lips.

2.5. Tongue tip prosodies

Finally, a number of languages have harmony systems involving tongue-tip consonants: all the coronal consonants of a word are either more advanced. [+anterior], or less advanced, [-anterior]. In some languages this distinction is dental vs. alveolar, in others alveolar vs. post-alveolar. In Coleman (2003), I present evidence that English also exhibits this kind of [anterior] harmony to some extent. For example:

Chumash (Chumashan, California (extinct))	k-sunon-us	‘I obey him’
	k-ʃunon-ʃ	‘I am obedient’
Tahltan (Athabaskan, NW Canada)	ɛsdɑn	‘I’m drinking’
	dɛθk^wʊθ	‘I cough’
Zayse (Omotic, Ethiopia)	zatst	‘lead’
	ʔiʃitʃ	‘five’
English	ʌt^həsɑi	‘utter sigh’
	ʌt^həʃɑi	‘utter shy’

2.6. Manner of articulation (degree of stricture)

In vowels, this would be e.g. height or [ATR] harmony. In consonants, long-domain fortis vs. lenis patterns. Are there any? Perhaps gemination in Malayalam (Local and Simpson 1998), e.g.:

‘it increased’	ʌ̟ð̞ə̠	k̠˕^xöˑɽɽ^ə̠ɨ̤˕	lenis/lax
‘he increased it’	ʱɐ̠v̞ɛ̈n^ə a̟ð^əə̟	k̟ˑ̄ṵ˖̟ʈ^ə̟ˑ̄ḭ˗	fortis/tight

3. Prosody reconsidered

The preceding examples have shown that many more features than duration and pitch are (or can be) prosodic. What, then, do we mean by ‘prosody’ if not just tone, stress, intonation, and loudness, the traditional ‘suprasegmental’ features?

a) Features (or groups of features) that not located at a single place in the sequence of consonants and vowels.

b) For example, (groups of) features associated with a whole syllable, word or phrase.

c) Also, features of the boundaries of syllables and words (e.g. assimilation, liaison, absence vs. presence of sounds in particular syllable or word positions). ‘Grenzsignale’ (Trubetzkoy 1969: 273-297). Recognition of the non-segmental behaviour of features, and the close relationship between features and the specific places in syllable or word position in which they occur, led to the origin in the mid 1930’s to the London school of "prosodic phonology", under J. R. Firth, which presented a thorough critique of and offered a theoretical alternative to phonemic phonology. Some of the ideas of the prosodic school have influenced the contemporary mainstream of phonological theory, especially via the framework of Autosegmental Phonology.

References

Bendor-Samuel, John T. (1960) Some problems of segmentation in the phonological analysis of Tereno. Word 16, 348-355.

Coleman, John (2003) Discovering the acoustic correlates of phonological contrasts. Journal of Phonetics 31, 351-372.

Connell, Bruce, Firmin Ahoua and Dafydd Gibbon (2002) Ega. Journal of the International Phonetic Association 32, 99-104.

Fischer-Jørgensen, Eli (1987) A phonetic study of the stød in Standard Danish. Annual Report of the Institute of Phonetics, University of Copenhagen vol. 21. 55-265.

Grønnum, Nina (1998) Danish. Journal of the International Phonetic Association 29, 99-105.

Laver, John (1994) Principles of Phonetics. Cambridge University Press.

Local, John and Adrian Simpson (1988) The domain of gemination in Malayalam. In David Bradley, Eugénie J. A. Henderson and Martine Mazaudon, eds. Prosodic analysis and Asian linguistics: to honour R. K. Sprigg. Pacific Linguistics, C-104, 33-42.

Reichard, Gladys (1938) "Cœur d’Alène", in Franz Boas (ed.) Handbook of American Languages, Bureau of American Ethnology, Bulletin 40, Washington.

Robins, Robert H. (1957) Vowel nasality in Sundanese: a phonological and grammatical study. Studies in Linguistic Analysis. The Philological Society. 87-103.

Tiede, Mark (1996) An MRI-based study of pharyngeal volume contrasts in Akan and English. Journal of Phonetics 24, 399-421.

Trubetzkoy, N. S. (1969) Principles of Phonology. (English translation of Grundzüge der Phonologie.) University of California Press.