Printer Friendly

Quantity in Leivu.

1. Introduction

Leivu was an Estonian linguistic enclave in North Latvia. There are no more Leivu speakers left (Nigol 1988). Researchers of Leivu have pointed out that the grammatical structure of Leivu resembles that of Hargla subdialect of South Estonian Voru dialect. There are similarities in the vocabulary and sound structure (Nigol 1955 : 149; Pajusalu, Hennoste, Niit, Pall, Viikberg 2002 : 190-191). As to the phonetic features of Leivu, Salme Nigol (1955 : 149) has for example drawn attention to the first half-long component of late diphthongs, where the second component is a raised vowel (e.g. soi? 'wolves', mail 'hill, adess.sg.', saida? 'to set', paida? 'to escape', laimbede 'closer to', cf. Estonian soeD, mael, seada, paGeDa, lahemale).

On the other hand, a strong influence of Latvian on Leivu phonetics has been observed. (1) The diphthongisation of short mid vowels (e.g. tiera 'grain, seed ', kuoda 'hall ', ei uole 'it is not ', cf. Estonian tera, koDa, ei ole), (2) labialisation of a (a > a > a > uo, e.g. vohN 'old ' , kange ' strong'), diphthongisation of mid vowels in Q2 words (e.g. kiele ' language, gen.sg. ' , puora 'grinding wheel, gen.sg.', skuolli' school, gen.sg. ', cf. South Estonian kele, pora, koli), (3) and diphthongisation of long high vowels (e.g. leim ' glue' : leimi ~ limi ' glue, gen.sg. ' ,mour 'wall' : mouru ~ muru 'wall, gen.sg. ', sour ' big ' : soure ~ sure 'big, gen.sg.') have been named as due to Latvian influence (Ariste 1931;Niilus 1935; 1937a; Nigol 1955; Tauli 1956; Suhonen 1989; Vaba 1997; cf. Gaters 1977; Rudzite 2005).

What concerns consonants, voicing of short plosives, quality change l > l before back vowels, s > s, z (especially in intervocalic position or before i, e.g. suzi ' wolf', mi massi ' pay, imperf. 1st pl. ',pussu ' gun, nom.pl. '), have been seen as Latvian influences on Leivu pronunciation. Also, the loss of h in the weak grade words (e.g. liehm 'cow' : lie(m)ma ' cow, gen.sg.' , taht 'star' : taije ' star, gen.sg.') has been named as a Latvian influence (Ariste 1931; Nigol 1955; Niilus 1935; 1937a; Tauli 1956; Suhonen 1989; Vaba 1997; cf. Gaters 1977; Rudzlte 2005). According to Lembit Vaba (1997 : 41) the loss of h from this position is a late phenomenon (in 1920s h can be found in transcriptions of Leivu). In late 1920s examples about transcription of words both with and without h can be found, but the transcriptions from 1930s point to a wider loss of short h (Niilus 1936). Unlike in Voru South Estonian the word-initial h has been lost in Leivu (Niilus 1936). (4) h has also been completely lost at the syllable boundary of non-initial syllables (Niilus 1936).

In most examples, the loss of short h has caused st0d at the syllable boundary, e.g. ra'a ~ ra 'money ', va'amb 'less ', na'a ' skin, gen.sg. ', pa'a 'head, illat.sg.', ta'a 'I want', tu'a ? 'ash, nom.pl.', vi'ma? 'rain, nom.pl. ' (Ariste 1931; Nigol 1955; Niilus 1935; 1937a; Tauli 1956; Suhonen 1989; Vaba 1997; Winkler 1999). This can be compared to st0d in Latvian and Livonian. (5) However, the intervocalic short h can also be lost completely or replaced by the approximant j (e.g. ria 'rake', puaba 'Sunday', jaije 'chilly', vaijer maple , Niilus 1936).

S. Nigol (1955) has pointed to the Latvian influence on Leivu quantity relations. Valter Niilus (1935) mentions that in Leivu, the vowels of the short first syllable are pronounced longer than in Standard Estonian (e.g. muna 'egg' , kana'chicken', pieza 'nest', cf. Standard Estonian mun?, kana, peza). His transcriptions of Leivu show variation in vowel durations of this type of words: sezar, sezar 'sister', sezara 'sister, gen.sg.' (Niilus 1937b). According to V. Niilus (1935), the lengthening of vowels can also be found in other word types: e.g. eng 'fishhook', kiirm 'secluded place', perv 'brink'.

There are no earlier acoustic phonetic studies of Leivu phonetics. A preliminary analysis of the speech of one speaker (Teras 2007) showed considerable variation in syllable duration ratios of Q1 words. The duration ratios of Q2 and Q3 words had some overlap, but the fundamental frequency turning point in the first syllable of Q2 words was late and of Q3 words early. While the speech of only one speaker was analysed a question arose whether similar tendencies also occur in the speech of other Leivu speakers.

In the following, the acoustic phonetic characteristics (duration ratios and fundamental frequency movement) of Leivu Q1, Q2 and Q3 words in the pronunciation of two Leivu speakers will be analysed. Answers to the following questions will be sought:

1) What are the duration ratios of the first two syllables in words in spontaneous Leivu?

2) Is there any difference in pitch contours associated with differences in syllable ratios?

3) What are the acoustic characteristics of words where stod is expected?

The results will be compared to studies on quantity of Estonian and South Estonian (cf. Lehiste 1960; 1997; Liiv 1961; Krull 1993; Asu, Lippus, Teras, Tuisk 2009; Pajusalu, Parve, Teras 2001; Parve 2003), and Latvian and Livonian (Lehiste, Teras, Ernstreits, Lippus, Pajusalu, Tuisk, Viitso 2008).

2. Material and method

Spontaneous speech of two male speakers of Leivu was analysed. The speaker Peeter Melec (PM) was born in 1867. He lived in Soosaare (Suzari) village and was recorded by Valmen Hallap in 1956 (tape EMH0003a in the archive of Estonian dialects at the Institute of the Estonian Language). The speaker Anton Bok (AB) was born in 1908. He lived in Pajusilla (Karklupe) village. He was recorded in 1971 by Paulopriit Voolaine (tape F-158 in the archive of Estonian dialects and related languages at the University of Tartu). His mother tongue was Leivu and he acquired Latvian at school. He has been called the last Leivu speaker; he died in 1988 (Nigol 1988).

Disyllabic quantity 1 (Q1), quantity 2 (Q2) and quantity 3 (Q3) words were selected from spontaneous speech. The analysed material consisted of 309 words in total (Q1 141 words, Q2 89 words, Q3 79 words). When the first syllable was long (Q2 and Q3 words), it contained either a long monophthong or a diphthong as a syllable nucleus or a short vowel followed by a voiced consonant (the first part of a geminate consonant or consonant cluster). The analysed words were in phrase-initial (79 words), internal (166 words) or final (64 words) position. All words carried sentence-level stress. Some examples of analysed words: tare 'room, farmhouse', tera 'grain, seed', eza 'father', nane 'woman, wife', skuolin 'school, iness.sg.', talve 'winter, gen.sg.', peima 'milk, part.sg.', lamba' sheep, gen.sg.', sanna 'sauna, part.sg. '.

The recordings were analysed using the Praat software for speech analysis (Boersma, Weenink 2007-2009). The duration of all segments was measured. Syllable durations and duration ratios were calculated. When the first syllable (S1) is open, syllable duration equals that of the syllable nucleus. When the syllable is closed, the duration of the coda consonant is added to the duration of the syllable nucleus. The second syllable (S2) duration equals the duration of the second syllable vowel. Fundamental frequency measurements were taken at the beginning and end of each syllable, and at the peak or turning point of F0 curve within the first syllable. The location of the F0 peak relative to the beginning of the first syllable was also established and will be given in percentages.

3. Duration and fundamental frequency in Q1, Q2 and Q3 words

3.1 Durations and duration ratios in Q1, Q2 and Q3 words

Average syllable durations and duration ratios of syllables in Q1, Q2 and Q3 words are given in Table 1 and Table 2. In Table 1, syllable durations and duration ratios of all words independent of their position in the sentence are presented. In Table 2, words in the phrase-initial (79 words), phrase-internal (166 words) and phrase-final (64 words) position are given separately.

In Q1 words, average durations of the first and second syllable are 125 ms and 113 ms, and the average duration ratio is 1.21 (s.d. 0.45) (see Table 1). Depending on the position of the word in the phrase, the average duration of the first syllable is 122 ms (phrase-initial), 121 ms (phrase-internal), 137 ms (phrase-final), and of the second syllable 119 ms, 107 ms, 133 ms respectively. The average duration ratios of syllables in Q1 words are 1.15, 1.24 and 1.13 respectively (see Table 2). The influence of phrase-final lengthening on vowel duration can be noticed--the duration of the second syllable vowel is longest in phrase-final words. Standard deviations show that there is a large variation in the duration ratios of Q1 words. The average duration ratio in words with a short open first syllable can vary between 0.76 and 1.66 (see Table 1).

Due to variation, words expected to be in Q1 (short first syllable both in Standard and South Estonian) were divided into two groups. The first group consists of words where the ratio was less than or equal to one (the second syllable vowel was longer than the first syllable vowel, or both vowels were of equal length, 54 words). The second group consists of words where the ratio was larger than one (the first syllable vowel was longer than the second syllable vowel, 87 words). Average syllable durations and duration ratios of syllables in these two groups of words are given in Table 3.

In the first group, the average syllable durations are 108 ms (S1) and 139 ms (S2) which gives a duration ratio of 0.81 (see the first part of Table 3). This ratio is similar to that of Estonian Q1 words. In the second group, the average syllable durations are 138 ms (S1) and 98 ms (S2), and the duration ratio is 1.46 (see the second part of Table 3). This ratio is much bigger than in Estonian Q1 words and resembles that of Estonian Q2 words where the ratio is 1.5 (cf. Lehiste 1960; 1997; Liiv 1961). An ANOVA shows that the difference is significant at p < 0.0001 level.

In Leivu Q2 words, the average syllable durations are 191 ms (S1) and 108 ms (S2), and the average duration ratio is 1.87 (see Table 1). Average syllable durations in different sentence positions are as follows: 180 ms (S1) and 105 ms (S2) (phrase-initial), 186 ms (S1) and 101 ms (S2) (phrase internal), 225 ms (S1) and 137 ms (S2) (phrase-final) (see Table 2). The average duration ratios are 1.78, 1.94, and 1.76 respectively. The influence of phrase-final lengthening can also be noticed in Q2 words. The duration of both S2 and S1 is longest in this position which may point to the influence of sentence stress on syllable duration. Standard deviations of duration ratios show that variation is also quite large in Q2 words. Standard deviation shows that the average syllable ratio in Q2 words varies between 1.24 and 2.5 (see Table 1).

In Q3 words, the syllable durations are on an average 224 ms (S1) and 96 ms (S2), and the duration ratio 2.44 (see Table 1). Average syllable durations in different sentence positions are 232 ms (S1) and 88 ms (S2) (phrase initial), 201 ms (S1) and 90 ms (S2) (phrase-internal), 261 ms (S1) and 111 ms (S2) (phrase-final) (see Table 2). The average duration ratios are 2.65, 2.35, and 2.26 respectively. Phrase-final lengthening is also present in Q3 words: the duration of S2 is longest in this position. Standard deviation shows that the average duration ratio varies between 1.75 and 3.13 in Q3 words (see Table 1).

Standard deviations of duration ratios of Q2 and Q3 words indicate that there is some overlap of duration ratios in these words. In both quantities, duration ratios smaller than or equal to two and larger than two can be found. Table 4 and Table 5 present average syllable durations and duration ratios in these two groups of Q2 and Q3 words.

64% of the Q2 words and 29% of the Q3 words have a duration ratio smaller than two (1.47 and 1.7 respectively). 36% of Q2 and 71% of Q3 words have a ratio larger than two (2.54 and 2.76 respectively). Although the ratio varies, it can be seen that Q2 words are characterised by a smaller and Q3 words by a larger ratio. It is quite probable that the location of F0 turning point can differentiate such Q2 and Q3 words where the duration ratios are similar. Fundamental frequency contours of Q1, Q2 and Q3 words will be dealt with next.

3.2 Fundamental frequency contours of Q1, Q2 and Q3 words

Average F0 values in the beginning and end of each syllable and at the turning point (or peak) are given in Table 6. The location of F0 turning point in relation to the total duration of the first syllable was also calculated and is given in percentages. Words in all quantities were divided into two groups: the first group consists of words where the turning point was in the first half of the syllable (an early peak) and the second group includes such words where the turning point was in the second half of the syllable (a late peak).

Q1 words are characterised by a late F0 turning point which is located at 77% of the total duration of the first syllable. F0 is falling in the second syllable. In 18% of the analysed Q1 words, F0 was falling during the whole word and the F0 turning point occurred on average at 28%. The F0 turning point was also late in most Q2 words (at 76% of the total duration of the first syllable). However, in 25% of the analysed Q2 words there was an early F0 turning point (at 35%). Q3 words are characterised by an early F0 turning point occurring at 28% of the total duration of the first syllable (in 23% of the analysed words there was a late F0 turning point occurring at 71% of the total duration of the first syllable). Since F0 is falling already during the first syllable, it reaches its lower values in the end of the first syllable. In Q2 words with a late F0 peak, the F0 value in the end of S1 is 179 Hz and in S2 the F0 values are 175 and 165 Hz. In Q3 words with an early F0 peak the corresponding values are 165, 163 and 155 Hz.

As all the analysed Q1, Q2 and Q3 words could be divided into two groups according to their duration ratios, F0 contours of these groups will be analysed separately. The results are given in Tables 7-9 (where Q1, Q2 and Q3 words are presented separately).

Q1 words where the duration ratio was 0.81-1.46 (see Table 3) are characterised by a late F0 turning point which occurred both in words with a smaller and larger duration ratio (see Table 7). It can be seen that Q1 words with the duration ratio larger than one are phonetically similar to Q2 words with the duration ratio smaller than two (see part 4 of Table 7 and part 3 of Table 8).

Table 8 shows that the duration ratio of Q2 words can also be quite large, whereas the F0 turning point still occurs in the second half of the first syllable in most cases. In words where the duration ratio was larger than two, the F0 turning point was late in 73% of the cases, and in words with the duration ratio smaller than two it was late in 75% of the cases.

Like Q2 words, also Q3 words had a varying duration ratio. However, the analysis of F0 contours indicates that Q3 words, regardless of their duration ratio, are characterised by an early F0 turning point. When the duration ratio was smaller than two, only two words had a late F0 peak (see part 3 of Table 9). When the duration ratio was larger than two, 73% of the words had an early F0 turning point.

The data contained only one disyllabic word where the loss of short h between vowels has caused laryngealization: ra'aca 'money, comit.sg.' (Speaker PM, see Figure 1). The durations of syllables are 254 ms (S1) and 89 ms (S2), the duration ratio 2.85. The duration of the laryngealization period is 32 ms. There is an early F0 peak occurring at 13% of the first syllable. In the speech of speaker AB, the word pat' tell, imperf. 3rd sg.' occurred two times. Although the loss of j had not caused laryngealization, there was an early F0 peak in both cases.

4. Discussion

The average duration ratio (1.21) in Leivu Q1 words is on average much larger than in Estonian Q1 words where it is around 0.6 (cf. Lehiste 1960; 1997; Liiv 1961). Standard deviation shows considerable variation in the duration ratio of Q1 words. Q1 words show two kinds of tendencies in their duration ratios: the second syllable vowel can be pronounced longer or as long as the first vowel, or the first syllable vowel is pronounced longer than the second syllable vowel (see Figure 2). The second type is like a mirror image of the first type. The pronunciation where the first vowel is longer than the second vowel can point to influences from Latvian. In Latvian, when the two syllables of a disyllabic word are short then the first syllable is pronounced longer than the second syllable (duration ratio 1.2-2.0, cf. Lehiste, Teras, Ernstreits, Lippus, Pajusalu, Tuisk, Viitso 2008). A correlation analysis shows a very small positive correlation (r = 0.1) between the duration of the first and the second syllable vowel: when the first syllable vowel lengthens then also the second syllable vowel is longer.

Although there is no significant difference in the duration ratio in Q1 words in the second type and Q2 words with a duration ratio smaller than two (the third pair of columns in Figure 2, the duration ratios 1.46 and 1.47), there is a significant difference both between the duration of S1 and S2 of these two types of words (p < 0.001). Significantly shorter durations of syllables in Q1 words with a larger duration ratio (138 and 98 ms) than in Q2 words with a smaller duration ratio (174 and 121 ms) may cause a word to be recognised as a Q1 word.

There is an overlap of syllable durations of Q2 and Q3 words (see Figure 2). This can be due to spontaneous speech. Such an overlap has also been found in Estonian spontaneous speech (Asu, Lippus, Teras, Tuisk 2009). However, an ANOVA shows that there is a significant difference between the duration ratios of Q2 and Q3 words at p < 0.0001 level. The average duration ratios (1.87 and 2.44) in these words are comparable to those found in South Estonian spontaneous speech: 1.6 and 2.9 in the Hargla sub-dialect of Voru dialect and 1.5 and 3.08 in Setu dialect (Pajusalu, Parve, Teras 2001, Parve 2003). They are also comparable to the duration ratios of Q2 and Q3 words in spontaneous speech of Standard Estonian: 1.72 and 3.21 (Krull 1993), 1.7 and 2.3 (Asu, Lippus, Teras, Tuisk 2009).

The characteristic fundamental frequency contours of Leivu Q1, Q2 and Q3 words (see Figure 3) are also comparable to the F0 contours found in South Estonian (cf. Parve 2003) and Standard Estonian (Asu, Lippus, Teras, Tuisk 2009).

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

[FIGURE 3 OMITTED]

5. Conclusion

The Leivu sound system has similarities with that of the South Estonian Hargla sub-dialect. It also gives evidence of language contacts with Latvian. In the present study spontaneous speech of two male speakers of Leivu was analysed. Standard deviations of syllable duration ratios in disyllabic Q1, Q2 and Q3 words showed considerable variation. In words expected to be in Q1, the syllable duration ratio was 0.8-1.7. The smaller ratio is similar to the syllable ratio in Estonian Q1 words. Although the bigger ratio is characteristic of Estonian Q2 words, these words have a significantly shorter first syllable than Q2 words. The bigger ratio may indicate a Latvian influence on Leivu pronunciation. In Latvian, when both syllables in a disyllabic sequence are short the first vowel is pronounced longer than the second syllable vowel. There was an overlap in syllable durations of Q2 and Q3 words (the ratio is 1.2-2.5 and 1.8-3.1 respectively) in Leivu. However, a ratio under 2 was characteristic of Q2 words and a ratio over 2 of Q3 words. The fundamental frequency analysis showed that Q1 and Q2 words were characterised by a late F0 peak in the first syllable and Q3 words by an early F0 peak. Even if the duration ratios in Q2 and Q3 words overlapped, the two quantities were differentiated by the location of F0 peak. Only one word pronounced with laryngealization was found in the present data. Thus, further research including more recordings of Leivu is needed to investigate this aspect.

doi: 10.3176/lu.2010.01

Acknowledgements

I am grateful to Ilse Lehiste, Tiit-Rein Viitso, Karl Pajusalu, and Eva Liina Asu for their valuable suggestions and comments. This research was partly supported by grant of the Estonian Science Foundation "Prosodic correlates of southern BaltoFinnic and Volga languages" (No. 6983).

REFERENCES

Ariste, P. 1931, Lisandeid Koiva murrete vokaalide haalikuteloole.--EK X, 175-179.

Asu, E. L., Lippus, P., Teras, P. Tuisk T. 2009, The Realization of Estonian Quantity Characteristics in Spontaneous Speech.--Nordic Prosody.

Proceedings of the Xth Conference, Helsinki 2008, Frankfurt, 49-56. Boersma, P., Weenink, D. 2007-2009, Praat 4.6-5.1 [Computer program]. http://www.praat.org/.

Gaters, A. 1977, Die lettische Sprache und ihre Dialekte, The Hague (Trends in Linguistics 9).

Krull, D. 1993, Word-Prosodic Features in Estonian Conversational Speech: Some Preliminary Results.--Experiments in Speech processes, Stockholm (PERILUS 17), 45-54.

Lehiste, I. 1960, Segmental and Syllabic Quantity in Estonian.--American Studies in Uralic Linguistics, Bloomington (UAS 1), 21-82.

--1997, Search for Phonetic Correlates in Estonian Prosody.--Estonian Prosody: Papers from a Symposium, Tallinn, 11-35.

Lehiste, I., Teras, P., Ernstreits, V., Lippus, P., Pajusalu, K.

Tuisk, T., Viitso, T.-R., 2008, Livonian Prosody, Helsinki (MSFOu 255).

Liiv, G. 1961, Eesti keele kolme valtusastme vokaalide kestus ja meloodiatuubid.--KK, 480-490.

Nigol, S. 1955, Markmeid matkalt leivu keelesaarele.--ESA, 147-151.

-- 1988, Anton Boks (1908-1988), viimane Leivu.--KK, 755. Niilus, V. 1935, Leivu murret. Haalikutelooline ulevaade I-III (AES 179; Manuscript).

-- 1936, Leivu (Koiva) murde h-haaliku arengust.--EK XV, 36-41.

-- 1937a, Heikki Ojansuu louna-eesti murdekogudest.--EK XVI, 140-146.

-- 1937b, Valimik leivu murdetekste. Choix de textes dialectaux leivu, Tartu (Akadeemilise Emakeele Seltsi toimetised XXXI).

Pajusalu, K., Hennoste, T., Niit, E., Pall, P., Viikberg J. 2002, Eesti murded ja kohanimed, Tallinn.

Pajusalu, K., Parve, M., Teras, P. 2001, On the Main Characteristics of the Prosody of South Estonian Dialects.--CIFU IX, pars VI, 9-13.

Parve, M. 2003, Valted lounaeesti murretes, Tartu (Dissertationes Philologiae Estonicae Universitatis Tartuensis 12).

Rudz Ite, M. 2005, Darbi latviesu dialektologija, Riga.

Suhonen, S. 1989, F. J. Wiedemann ja Heikki Ojansuu leivun murteen tutkijoina.--Ferdinand Johann Wiedemannin muisto, Helsinki (Castrenianu min Toimitteita 31), 39-52.

Tauli, V. 1956, Phonological Tendencies in Estonian, K0benhavn (Det Konglige Danske Videnskabernes Selskab, Historisk-filologiske Meddelelser 36, 1).

Teras, P. 2003, Lounaeesti vokaalisusteem. Voru pikkade vokaalide kvaliteedi muutumine, Tartu (Dissertationes Philologiae Estonicae Universitatis Tartuensis 11).

-- 2007, Leivu haaldusjoontest.--Tartu Ulikooli Louna-Eesti keele- ja kultuuriuuringute keskuse aastaraamat, Tartu, 19-30.

Vaba, L. 1997, Uurimusi lati-eesti keelesuhetest, Tallinn-Tampere.

Viitso, T.-R. 2009, Livonian and Leivu: Shared Innovations and Problems.--LU XLV, 269-282.

Winkler, E. 1997, Katkeintonatsiooni tekkimisest laanemeresoome keeltes.--Odagumeresoomo veeremaaq. Laanemeresoome perifeeriad, Voro (Voro Instituudi Toimotiseq), 201-206.

PIRE TERAS (Helsinki-Tartu)

(1) Leivu linguistic enclave was in the area where Latgalian High Latvian is spoken (Viitso 2009; cf. Gaters 1977; Rudzlte 2005).

(2) The diphthongisation of short e can also be found in High Latvian dialects (Rudzite 2005) and the diphthongisation of short o is characteristic of Standard Latvian. However, S. Suhonen (1989) finds that the diphthongisation of short e and o is not necessarily a Latvian influence. According to T.-R. Viitso (2009) the breaking of short mid vowels took place by analogy with the breaking of long mid vowels.

(3) In Q3 words, as in other South Estonian dialects, mid vowels are raised, e.g. kil 'language', pur ' grinding wheel' , skull 'school' (see Teras 2003).

(4) V. Niilus (1936) finds that the word-initial h has already been lost since the 19th century.

(5) In addition to laryngealization Leivu, like South Estonian Voru dialect, has a glottal stop. A sporadic loss of glottal stop has also been noticed (e.g. Vaba 1997).

Address:

Pire Teras

University of Helsinki

University of Tartu

E-mail: pire.teras@ut.ee
Table 1

Average syllable durations (in ms), duration ratios and standard
deviations (s.d.) of Q1, Q2 and Q3 words (N--number of measurements,
Q1/Q2/Q3)

Speaker                    N            Q1                  Q2
                                        S1    S2    S1/S2   S1    S2

PM        Average           75/48/52    131   126   1.15    212   111
          s.d.                           35    37    0.54    61    36
AB        Average           66/41/27    119   101   1.27    169   105
          s.d.                           28    36    0.36    58    35

All       Overall average   141/89/79   125   113   1.21    191   108
                                         32    37    0.45    60    35

Speaker                     Q2      Q3
                            S1/S2   S1    S2    S1/S2

PM        Average           2.07    250   104   2.55
          s.d.              0.74     55    33    0.78
AB        Average           1.67    199   87    2.33
          s.d.              0.53     61    22    0.60

All       Overall average   1.87    224   96    2.44
                            0.63     58    28    0.69

Table 2

Average syllable durations (in ms), duration ratios
and standard deviations (s.d.) of Q1, Q2 and Q3 words
in phrase-initial, phrase-internal and phrase-final position
(N--number of measurements, Q1/Q2/Q3)

Speaker                     N          Q1                  Q2
                                       S1    S2    S1/S2   S1    S2

PM        Phrase-initial    12/15/8    124   138   1.04    208   111
          s.d.                          37    55   0.53     80    43
AB        Phrase-initial    28/10/6    121    99   1.27    152   100
          s.d.                          26    26   0.28     39    21
All       Overall average   40/25/14   122   119   1.15    180   105
                                        32    40   0.40     59    32
PM        Phrase-internal   45/27/26   128   120   1.16    207   108
          s.d.                          36    29   0.58     58    32
AB        Phrase-internal   30/24/14   114    93   1.33    166   95
          s.d.                          29    38   0.41     54    28
All       Overall average   75/51/40   121   107   1.24    186   101
                                       32     34   0.49     56    30
PM        Phrase-final      18/6/18    143   132   1.18    246   124
          s.d.                          32    39   0.45     66    40
AB        Phrase-final        8/7/7    132   134   1.07    204   150
          s.d.                         31     45   0.39     87    42
All       Overall average   26/13/25   137   133   1.13    225   137
                                        32    42   0.42     76    41

Speaker                     Q2      Q3
                            S1/S2   S1    S2    S1/S2

PM        Phrase-initial    2.02    261    95   2.82
          s.d.              0.75     72    17   0.81
AB        Phrase-initial    1.54    203    81   2.48
          s.d.              0.42     77    20   0.47
All       Overall average   1.78    232    88   2.65
                            0.58     74    19   0.64
PM        Phrase-internal   2.08    232   103   2.40
          s.d.              0.82     52    33   0.80
AB        Phrase-internal   1.80    171    77   2.30
          s.d.              0.53     46    15   0.74
All       Overall average   1.94    201    90   2.35
                            0.68     49    24   0.77
PM        Phrase-final      2.12    270   110   2.65
          s.d.              0.75     42    38   0.74
AB        Phrase-final      1.41    253   112   2.28
          s.d.              0.60     45    19   0.46
All       Overall average   1.76    261   111   2.46
                            0.68     43    28   0.60

Table 3

Average syllable durations (in ms),
duration ratios and standard deviations (s.d.)
in Q1 words presented in two groups
(N--number of measurements)

Speaker   Duration ratio    N    S1    S2    S1/S2

PM        S1/S2 < 1         33   104   148   0.73
          s.d.                    18    37   0.17
AB        S1/S2 < 1         21   112   131   0.89
          s.d.                    28    44   0.14
All       Overall average   54   108   139   0.81
                                  23    40   0.16
PM        S1/S2 > 1         42   152   109   1.47
          s.d.                    31    27   0.50
AB        S1/S2 > 1         45   123    87   1.45
          s.d.                    28    21   0.29
All       Overall average   87   138    98   1.46
                                  29    24   0.39

Table 4

Average syllable durations (in ms),
duration ratios and standard deviations (s.d.)
in Q2 words presented in two groups
(N--number of measurements)

Speaker   Duration ratio                    N    S1    S2     S1/S2

PM        S1/S2 [less than or equal to] 2   24   188    132   1.46
          s.d.                              36    30   0.27
AB        S1/S2 [less than or equal to] 2   33   160    110   1.49
          s.d.                              60    37   0.39
All       Overall average                   57   174    121   1.47
                                                  48     33   0.33
PM        S1/S2 > 2                         24   237     90   2.67
          s.d.                                    72     30   0.52
AB        S1/S2 > 2                          8   206     87   2.42
          s.d.                                    29     17   0.34
All       Overall average                   32   221     89   2.54
                                                  51     24   0.43

Table 5

Average syllable durations (in ms),
duration ratios and standard deviations (s.d.)
in Q3 words presented in two groups
(N--number of measurements)

Speaker   Duration rat                      N    S1    S2    S1/S2

PM        S1/S2 [less than or equal to] 2   14   222   136   1.68
          s.d.                                    50    41   0.24
AB        S1/S2 [less than or equal to] 2    9   158    91   1.72
          s.d.                                    58    27   0.26
All       Overall aver                      23   190   114   1.70
                                                  54    34   0.25
PM        S1/S2 > 2                         38   260    93   2.87
          s.d.                                    53    20   0.65
AB        S1/S2 > 2                         18   220    84   2.64
          s.d.                                    55    20   0.48
All       Overall aver                      56   240    89   2.76
                                                  54    20   0.57

Table 6

Average fundamental frequency (in Hz) and standard deviations (s.d.)
in Q1, Q2 and Q3 words (N--number of measurements, TP--turning point)

Quantity         Speaker   N     S1beg.   TP    %    S1end   S2beg.

Q1, early peak     PM       13    203     212   38    194      188
                                   48      51   13     39       35
                   AB       13    186     184   18    147      148
                                   41      44   15     42       51
                   All      26    194     198   28    171      168
                                   44      48   14     40       43
Q1, late peak      PM       62    191     204   76    195      188
                                   24      28   15     26       23
                   AB       53    154     161   79    155      155
                                   23      26   15     23       26
                   All     115    173     182   77    175      172
                                   23      27   15     25       25
Q2, early peak     PM       14    194     204   41    185      180
                                   13      17    8     15       14
                   AB        8    185     188   30    145      138
                                   45      44   13     26       23
                   All      22    190     196   35    165      159
                                   29      31   10     20       19
Q2, late peak      PM       34    185     202   75    193      187
                                   19      24   16     22       22
                   AB       33    161     175   78    164      163
                                   33      35   15     31       25
                   All      67    173     188   76    179      175
                                   26      30   15     26       23
Q3, early peak     PM       42    201     219   34    197      192
                                   28      34   10     30       30
                   AB       19    183     185   23    134      133
                                   38      37   14     25       23
                   All      61    192     202   28    165      163
                                   33      36   12     27       27
Q3, late peak      PM       10    189     211   64    197      192
                                   18      20    9     20       24
                   AB        8    149     163   79    155      160
                                   30      34   15     32       27
                   All      18    169     187   71    176      176
                                   24      27   12     26       25

Quantity         Speaker   S2end

Q1, early peak     PM       173
                             38
                   AB       137
                             50
                   All      155
                             44
Q1, late peak      PM       173
                             22
                   AB       149
                             29
                   All      161
                             25
Q2, early peak     PM       168
                             13
                   AB       130
                             22
                   All      149
                             17
Q2, late peak      PM       179
                             19
                   AB       150
                             30
                   All      165
                             24
Q3, early peak     PM       187
                             32
                   AB       123
                             24
                   All      155
                             28
Q3, late peak      PM       182
                             19
                   AB       154
                             26
                   All      168
                             23

Table 7

Average fundamental frequency (in Hz) and standard deviations (s.d.)
in Q1 words divided into two groups
(N--number of measurements, TP--turning point)

Speaker   Duration ratio         N    Q1, early peak
                                      S1beg.   TP    %    S1end

PM        S1/S2 [less than       7      193    197   37    179
          or equal to] 1
          s.d.                           34     37   14     21
AB        S1/S2 [less than or    4      171    175   31    139
          equal to] 1
          s.d.                           46     47   29     43
All       Overall average        11     182    186   34    159
                                         40     42   21     32
PM        S1/S2 > 1              4      224    232   36    212
          s.d.                           75     79   14     58
AB        S1/S2 > 1              9      179    176   11    140
          s.d.                           41     41   11     45
All       Overall average        13     201    204   23    176
                                         58     60   13     51

Speaker   Duration ratio         N    Q1, late peak
                                      S1beg.   TP    %    S1end

PM        S1/S2 [less than or    26     193    197   78    192
          equal to] 1
          s.d.                           27    25    16     25
AB        S1/S2 [less than or    17     159    167   73    160
          equal to] 1
          s.d.                           27    29    17     28
All       Overall average        43     176    182   76    176
                                         27    27    17     27
PM        S1/S2 > 1              38     190    209   72    199
          s.d.                           22    29    15     27
          S1/S2 > 1              36     154    161   81    154
          s.d.                           23    26    14     23
          Overall average        74     172    185   76    176
                                         23    28    15     25

Speaker   Duration ratio         Q1, early peak
                                 S2beg.   S2end

PM        S1/S2 [less than         173     165
          or equal to] 1
          s.d.                      16      16
AB        S1/S2 [less than or      133     114
          equal to] 1
          s.d.                      45      45
All       Overall average          153     140
                                    30      30
PM        S1/S2 > 1                207     192
          s.d.                      56      68
AB        S1/S2 > 1                136     133
          s.d.                      46      50
All       Overall average          171     163
                                    51      59

Speaker   Duration ratio         Q1, late peak
                                 S2beg.   S2end

PM        S1/S2 [less than or      187     170
          equal to] 1
          s.d.                      24      21
AB        S1/S2 [less than or      163     151
          equal to] 1
          s.d.                      38      33
All       Overall average          175     161
                                    31      27
PM        S1/S2 > 1                189     175
          s.d.                      22      22
          S1/S2 > 1                155     151
          s.d.                      26      30
          Overall average          172     163
                                    24      26

Table 8

Average fundamental frequency (in Hz) and standard deviations (s.d.)
in Q2 words divided into two groups
(N--number of measurements, TP--turning point)

Speaker   Duration ratio     N    Q2, early peak
                                  S1beg.   TP    %    S1end   S2beg.

PM        S1/S2 [less than    8   197      211   39   189     184
          or equal to] 2
          s.d.                     14       11   10     6       6
AB        S1/S2 [less than    6   177      178   28   141     131
          or equal to] 2
          s.d.                     42       41   14    28      23
All       Overall average    14   187      194   34   165     158
                                   28       26   12    17      14
PM        S1/S2 > 2           6   189      196   43   180     174
          s.d.                      9       21    4    21      20
AB        S1/S2 > 2           3   211      216   34   156     157
          s.d.                     60       56    6    22      15
All       Overall average     9   200      206   39   168     165
                                   35       38    5    21      17

Speaker   Duration ratio     N    Q2, late peak
                                  S1beg.   TP    %    S1end   S2beg.

PM        S1/S2 [less than   16   185      201   81   194     192
          or equal to] 2
          s.d.                     14       23   15    23      24
AB        S1/S2 [less than   27   161      174   77   163     161
          or equal to] 2
          s.d.                     33       37   14    31      26
All       Overall average    43   173      188   79   178     176
                                   24       30   14    27      25
PM        S1/S2 > 2          18   185      202   69   192     183
          s.d.                     23       26   15    22      19
          S1/S2 > 2          6    162      175   81   173     172
          s.d.                     37       31   18    27      20
          Overall average    24   173      189   75   183     177
                                   30       28   16    25      20

Speaker   Duration ratio     Q2, early peak
                             S2end

PM        S1/S2 [less than   171
          or equal to] 2
          s.d.                11
AB        S1/S2 [less than   123
          or equal to] 2
          s.d.                22
All       Overall average    147
                              17
PM        S1/S2 > 2          164
          s.d.                15
AB        S1/S2 > 2          150
          s.d.                 1
All       Overall average    157
                               8

Speaker   Duration ratio     Q2, late peak
                             S2end

PM        S1/S2 [less than   183
          or equal to] 2
          s.d.                21
AB        S1/S2 [less than   148
          or equal to] 2
          s.d.                31
All       Overall average    165
                              26
PM        S1/S2 > 2          177
          s.d.                17
          S1/S2 > 2          157
          s.d.                24
          Overall average    167
                              21

Table 9

Average fundamental frequency (in Hz) and standard deviations (s.d.)
in Q3 words divided into two groups
(N--number of measurements, TP--turning point)

Speaker   Duration ratio      N    Q3, early peak
                                   S1beg.   TP    %    S1end   S2beg.

PM        S1/S2 [less than    14   197      213   35   193     187
          or equal to] 2
          s.d.                      29       36   12    30      30
AB        S1/S2 [less than     7   167      167   26   131     131
          or equal to] 2
          s.d.                      31       26   15    18      16
All       Overall average     21   182      190   31   162     159
                                    30       31   13    24      23
PM        S1/S2 > 2           29   203      221   34   199     194
          s.d.                      27       33   10    30      31
AB        S1/S2 > 2           12   193      196   21   136     134
          s.d.                      40       40   14    29      27
All       Overall average     41   198      208   28   167     164
                                    33       37   12    29      29

Speaker   Duration ratio      N    Q3, late peak
                                   S1beg.   TP    %    S1end   S2beg.

AB        S1/S2 [less than    2    174      178   59   166     161
          or equal to] 2
          s.d.                      61       81   10    77      55
PM        S1/S2 > 2           9    187      211   63   198     193
          s.d.                      17       21   10    22      25
AB        S1/S2 > 2           6    141      158   85   152     160
          s.d.                      13       15    9    15      20
All       Overall average     15   164      184   74   175     176
                                    15       18    9    18      23

Speaker   Duration ratio     Q3, early peak
                             S2end

PM        S1/S2 [less than   176
          or equal to] 2
          s.d.                26
AB        S1/S2 [less than   119
          or equal to] 2
          s.d.                17
All       Overall average    147
                              22
PM        S1/S2 > 2          192
          s.d.                33
AB        S1/S2 > 2          126
          s.d.                28
All       Overall average    159
                              30

Speaker   Duration ratio     Q3, late peak
                             S2end

AB        S1/S2 [less than   159
          or equal to] 2
          s.d.                49
PM        S1/S2 > 2          181
          s.d.                20
AB        S1/S2 > 2          152
          s.d.                22
All       Overall average    167
                              21
COPYRIGHT 2010 Estonian Academy Publishers
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2010 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Teras, Pire
Publication:Linguistica Uralica
Article Type:Report
Geographic Code:4EXES
Date:Mar 1, 2010
Words:6428
Previous Article:In memoriam Arvo Eek.
Next Article:Qualitative changes in phraseology: general principles and catalysts.
Topics:

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |