Preaspiration in the Nordic Languages: Synchronic and Diachronic ...

Preaspiration in the Nordic Languages: Synchronic and Diachronic ... Preaspiration in the Nordic Languages: Synchronic and Diachronic ...

www2.lingfil.uu.se
from www2.lingfil.uu.se More from this publisher
26.07.2013 Views

noise becomes audibly postalveolar in character, which can be seen as a compromise between the places of articulation of the vowel and the stop. Likewise, in CK’s production of sticker ‘juts out’ (Figure 4–12) a period of breathy voice is followed by palatal friction, as one might expect in the context of a sequence of a close front vowel and a prevelar stop. In Figures 4–13 and 4–14, we see examples of bilabial friction in FS’s production of the words tupp ‘rooster’ and hjortarna ‘the deer (pl.)’. In the first case, the friction can be attributed to the combined effect of rounding and the imminent bilabial closure. In the second case, involving a postalveolar stop closure, the labiality is due only to the rounding of the vowel. p i Figure 4–17. Subj. MP: […den] djupa viken o[ch…] ‘…the deep bay and…’ As we saw in the durational analysis, GT and MP tend not to preaspirate fortis stops to the same degree as do CK and FS. Nevertheless, the voice offset preceding a fortis stop can be a correlate of the fortis vs. lenis distinction in their speech. This can be seen by comparing MP’s production of a fortis stop in Figure 4–15 and a lenis stop in Figure 4–16. In Figure 4–15, the glottis appears to start spreading approximately 25 ms before the oral closure, and one can detect a slight breathiness just before the closure is made. This is in contrast with the transition between vowel and stop in Figure 4–16, in which the vowel retains full modal voice throughout its production. The presence of this slightly early voice – 130 –

offset is perceptually salient when contrasted with a vowel with full modal voice. All the CSw subjects tend to produce friction in the transition phase between a close front vowel and an occlusion. In Figure 4–17, MP produces the sequence djupa viken ‘deep bay.’ In the first word, there is a rapid voice offset before the stop closure that does not cause any breathiness in the vowel. In the second word, the narrow escape channel of the close front vowel causes a friction noise to be produced before the closure of the prevelar stop. The narrow escape channel causes air velocity to increase, which facilitates voice offset. Thus a brief interval of breathiness may be present, which is obscured by the friction noise. It should be noted that such intervals of voiced friction count as preaspiration in the measurements of the CSw data. 4.2.3.5 Postsonorant fortis stops In the CSw data, fortis stops in word-medial, postsonorant position (e.g. in words such as valpar ‘puppies,’ vinter ‘winter’ and björk ‘birch’) are fairly rare. Sequences of nasal + fortis stop are the most common, and the data contain a total of 66 such occurrences in stressed syllables in content words (see Table 4–8). Data from the function words inte ‘not’ and sånt ‘such things’ are not included. For lateral + fortis stop sequences, there are 39 such occurrences (here the function word alltså ‘so, that is’ and the phrase helt enkelt ‘simply’ are excluded). For r-sound + fortis stop there are 21 such occurrences in the data (there are no function words in the data with such sequences). Table 4–8. Frequency of occurrence of sonorant + fortis stop sequences occurring after stressed vowel in content words for the four CSw subjects. Sequences of r-sound + fortis stop do not include /rt/ sequences, which are instead treated as postalveolar stops. GT CK MP FS Total Nasal + fortis stop 8 20 20 18 66 Lateral + fortis stop 6 12 10 11 39 R-sound + fortis stop 1 3 11 6 21 Total 15 35 41 35 126 – 131 –

offset is perceptually salient when contrasted with a vowel with full<br />

modal voice.<br />

All <strong>the</strong> CSw subjects tend to produce friction <strong>in</strong> <strong>the</strong> transition phase<br />

between a close front vowel <strong>and</strong> an occlusion. In Figure 4–17, MP produces<br />

<strong>the</strong> sequence djupa viken ‘deep bay.’ In <strong>the</strong> first word, <strong>the</strong>re is a<br />

rapid voice offset before <strong>the</strong> stop closure that does not cause any breath<strong>in</strong>ess<br />

<strong>in</strong> <strong>the</strong> vowel. In <strong>the</strong> second word, <strong>the</strong> narrow escape channel of <strong>the</strong><br />

close front vowel causes a friction noise to be produced before <strong>the</strong> closure<br />

of <strong>the</strong> prevelar stop. The narrow escape channel causes air velocity<br />

to <strong>in</strong>crease, which facilitates voice offset. Thus a brief <strong>in</strong>terval of breath<strong>in</strong>ess<br />

may be present, which is obscured by <strong>the</strong> friction noise. It should be<br />

noted that such <strong>in</strong>tervals of voiced friction count as preaspiration <strong>in</strong> <strong>the</strong><br />

measurements of <strong>the</strong> CSw data.<br />

4.2.3.5 Postsonorant fortis stops<br />

In <strong>the</strong> CSw data, fortis stops <strong>in</strong> word-medial, postsonorant position<br />

(e.g. <strong>in</strong> words such as valpar ‘puppies,’ v<strong>in</strong>ter ‘w<strong>in</strong>ter’ <strong>and</strong> björk ‘birch’)<br />

are fairly rare. Sequences of nasal + fortis stop are <strong>the</strong> most common,<br />

<strong>and</strong> <strong>the</strong> data conta<strong>in</strong> a total of 66 such occurrences <strong>in</strong> stressed syllables <strong>in</strong><br />

content words (see Table 4–8). Data from <strong>the</strong> function words <strong>in</strong>te ‘not’<br />

<strong>and</strong> sånt ‘such th<strong>in</strong>gs’ are not <strong>in</strong>cluded. For lateral + fortis stop<br />

sequences, <strong>the</strong>re are 39 such occurrences (here <strong>the</strong> function word alltså<br />

‘so, that is’ <strong>and</strong> <strong>the</strong> phrase helt enkelt ‘simply’ are excluded). For r-sound<br />

+ fortis stop <strong>the</strong>re are 21 such occurrences <strong>in</strong> <strong>the</strong> data (<strong>the</strong>re are no<br />

function words <strong>in</strong> <strong>the</strong> data with such sequences).<br />

Table 4–8. Frequency of occurrence of sonorant + fortis stop sequences<br />

occurr<strong>in</strong>g after stressed vowel <strong>in</strong> content words for <strong>the</strong> four CSw subjects.<br />

Sequences of r-sound + fortis stop do not <strong>in</strong>clude /rt/ sequences, which are<br />

<strong>in</strong>stead treated as postalveolar stops.<br />

GT CK MP FS Total<br />

Nasal + fortis stop 8 20 20 18 66<br />

Lateral + fortis stop 6 12 10 11 39<br />

R-sound + fortis stop 1 3 11 6 21<br />

Total 15 35 41 35 126<br />

– 131 –

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!