1. Terminology

The IPA provides a group of symbols for stress, length, intonation, syllabification and tone under the general heading “suprasegmentals”, reflecting a conceptual division of speech into “segmental” and “suprasegmental” parts. However, as we shall see, this division is not very clean, with phonetic correlates of stress, intonation etc. often manifest in the consonants and vowels, i.e. at the segmental level. Therefore, rather than referring to “segmentals vs. suprasegmentals”, I prefer the more general term “prosody” for these phenomena (among others).

2. Stress and its phonetic correlates

In English, the term “stress” is used in reference to the degree of prominence of individual syllables of single words (lexical stress), e.g. elephant (stressed - unstressed - unstressed) vs. elephantine (unstressed? - unstressed - stressed - unstressed). But not every word is actually stressed when it occurs in a sentence: some lexical stresses are picked out for sentential stress - perhaps better called “accent”. E.g. “Albert went to the zoo.”

IPA symbols for primary stress [ˈ] and secondary stress [ˌ] may be placed before vowels, or equally well at the beginning of each syllable (thus also providing some information about syllable boundaries): elephant [ˈelɪfənt], elephantine [ˌelɪfˈanˌtɑɪn].

Ladefoged (1993: 249), like many introductory phonetics textbooks, says “the nature of stress is fairly well understood. Stressed sounds are those on which the speaker expends more muscular energy ... so that there is an additional increase in pitch.” Davenport and Hannahs (p. 23), similarly, write “Stressed syllables are produced with more muscular effort, and are louder or longer than unstressed syllables.” There are some problems with this view:

a) In some languages (e.g. Welsh, see Williams 1982; Chamorro, see Chung 1983) and even in some dialects of English (e.g. Birmingham, Tyneside, Welsh English) the pitch may be lower on stressed syllables than unstressed syllables, and in Welsh at least, stressed syllables may be shorter and quieter than unstressed syllables, as shown by Briony Williams in her PhD thesis and in several articles arising from it (Williams 1982, 1986, 1999). In a perception test, two English and one Welsh listener judged the placement of stress in a list of 20 isolated words read by a native speaker. Their judgements were correlated with the following acoustic parameters:

It can be seen from this figure that Welsh vs. English speakers seem to associate quite opposite acoustic features with stress; in particular, syllables perceived as stressed by Welsh listeners have shorter vowels, lower amplitude within the vowel, monotonous pitch (i.e. no pitch movement) on the stressed vowel. The shorter vowels of stressed syllables may however be associated with a longer following consonant (Williams 1998: 8). Although stressed syllables in Welsh are not associated with pitch movement or peaks, it has also been noted that final syllables in Welsh are often associated with a pitch peak, despite being unstressed (Buczek-Zawiła 2014).

Audio examples: siarad /ˈʃarad/ "talk", felen /ˈvelen/ "yellow", yr ogof /ər 'ogov/ "the cave"

Figure 2: Waveform and power plot of yr ogof

b) Vowel and consonant qualities may be phonetic correlates of stress too. Cf. English:

In these examples, the stressed syllables have longer, closer, more rounded vowels and initial [h], whereas unstressed syllables have shorter, more central, less rounded vowels and somtimes initial [h] is absent. The phonetics of stress, then, includes not just loudness, pitch and duration, but also vowel quality, liprounding. Vowel quality differences are especially important markers of stress in English. Unstressed vowels are often said to be “reduced” to [ə] or [ɪ]. Some English vowels, e.g. [ɒ], are only found in syllables with  (primary or secondary) stress: tonic [ˈtɒnɪk] vs. [təˈnɪsɪtɪ]. The aspiration of voiceless aspirated stops is stronger in stressed syllables than unstressed syllables: e.g. [ˈp^hhet^hrəl] vs. [p^həˈt^hhroʊl]

c) The phonetic realization of stress is rather language-specific. Trubetzkoy (1939) pointed out that in (West) Bulgarian the contrast between close /u/ and mid /ɔ/ found in stressed syllables is neutralized in unstressed syllables, where only /u/ - the closer phoneme - occurs. (/u/ may be pronounced [o] in unstressed syllables, but the point is that it is closer than [ɔ], and so unstressed vowels have closer vowel qualities. Similarly, unstressed /a/ is [ə], rather like in English.)

In Russian, the contrast between open /a/ and mid /o/ is neutralized in unstressed syllables, where /o/ is unrounded and lowered to /a/.

Hence:

Similarly, in Chamorro, mid vowel (allophones) are found in stressed closed syllables and only close vowel (allophones) occur in unstressed syllables, e.g. [ˈsoŋsuŋ] “village” vs. [i suŋˈsoŋɲa] “his village”; [ˈtsoʔtsuʔ] “word” vs. [i tsoʔˈtsoʔmu] “your (sg.) work”; [ˈɡwesɡwis] “to brush”, [ˈsensin] “flesh”, [ˈneni] “baby”, etc. This is the opposite of the correlation between stress and vowel height in the English examples above, in which stressed [ˈu] is close and unstressed [ə] is mid. It appears then as if stress is not a well-defined phonetic feature, but an abstract phonological feature whose phonetic realization is i) manifold and ii) language-specific.

3. Duration

Length contrasts can be transcribed by doubling of letters (for simple long vs. short duration), or using length marks: [ː] (in print, or, just as good, [:]), [ˑ] (or [·]). E.g. Luganda [k̩̀kúlà] “treasure” vs. [kúlà] “grow up”, Tashlhit Berber [ttgg^wa], Italian nonno [ˈnonnɔ] “grandfather” vs. nono [ˈnonɔ] “ninth”, or even English [kkju] “thank you/f*ck you” (depending on intonation and context of use) vs. [kju] “cue/queue”. There is also an IPA diacritic for “extra short” sounds, e.g. [ă].

Avoid confusing phonetic duration (measured e.g. in milliseconds) with phonological length (e.g. long vs. short vowels and consonants). Many durational differences in English are purely allophonic. Note that the duration of the short vowel /a/ can easily exceed that of the long vowel /iː/ because more open vowels have a longer duration in English than closer vowels.

References

Chung, S. (1983) Transderivational relationships in Chamorro phonology. Language 59 (1). 35-66.

Ladefoged, P. (1993) A Course in Phonetics. (Third Edition) Harcourt Brace Jovanovich.

Trubetzkoy, N. S. (1939) Grundzüge der Phonologie. Travaux du cercle linguistique de Prague 7.

Williams, B. (1982) The Problem of Stress in Welsh. Cambridge Papers in Phonetics and Experimental Linguistics Vol. 1. Department of Linguistics, University of Cambridge.