Atvirojo ir uždarojo skiemens balsių trukmė pietinių vakarų aukštaičių tarmėje

LITHUANIAN KALBOTYROS QUESTIONS XL (1998) Asta KAZLAUSKIENĖ Vytauto Didžiojo universitetas, Kaunas THE LENGTH OF OPEN AND CLOSED SKIEMENS IN THE SOUTHWESTERN AUKSTAITIA TARGET 1. V. Vaitkevičiūtė (1960: 207-208), describing the phonetic circumstances determining the length of vowels in Lithuanian common language, states that in the closed syllable vowels and diphthongs are shorter than in the open syllable, because part of the energy is also devoted to the consonant; in the closed end, she believes, the vowels are longer than in other positions, but shorter than the vowels of the open end. Thus, according to shorter — in the closed ending, shortest — inside the word. However, more recent studies of the length of the internal and open vowels of the South-West Highland Kaunas dialect (hereinafter — PVA) (Kazlauskienė 1998) show that the dialect, on the basis of which the common language was created, is characterized by a quantitative reduction of the terminal vowels. Thus, the open-end vowels in the PVA dialect are certainly not longer than the internal vowels. Now it remains to find out whether the length of vowels (especially the ending) in the PVA dialect"' depends on the nature of the syllable (openness, closure). 1.1. The length of the closed and open vowels of the end was studied in the words“: renkas „renkas“ —renka“, „renką“, rankas „rankas“ —ranka „ranka“, reeįj.kas (abbreviated as 1. 3 person „renkasi“) „renkas“ — renj.ka „renka“, lenka-s,lenkias“ —lenke., „Ienkią“, leq.kce.s „lenkęs“ —len.kee. (sick. r. biti. k. I. Participant dgs. vard.) ‘lenkę’, silkces ‘silkes’ —silkce (vns. įnag.) ‘silke’, lefj. kes (abbrev. es. 1. 3 person. „lenkiasi“) „lenkias“ —lerj.kee,lenkia®, ni.kštis „thumbs“ —ni.ksti, „mykšty“, kd.rti.s „kartys“ —kd.rti. „karti“, ru-kštis „rūgštis“ —ru“kšti „rūgšti“, bd.rtis „bartis“ —bd.rti „barti“, ri-kstu.s'This work is based only on those western Kaunas highland dialects, which according to the prevailing opinion (e.g.: Salys 1946: 51-52; Zinkevičius 1966: 118 and below; LKT 1970: 28-29; LKA 2 1982), do not shorten the end vowels. * When transcribed examples, it was assumed that the front-row lower-case vowel [e] before hard consonants, back-row vowels and at the end of the word is wider, e.g. resto. „rūsto“, rėsti „rūsti“, retū „retu“, sar.tee. "Sarte", butter? (vs. loc.) ‘svieste- ’. Of course, this is not the case in all PVA talks. Approximately from Lekėčiai, Višakio Rūda, Kazlų Rūda, Marijampolė to the east, most often (regardless of the subsequent softness of the consonant) only the long or half-long [e] with stem accent is spread. Since only two speakers came from these places, the variant showing the pronunciation of the majority of speakers was chosen. In the presentation of the research words, only the softness of consonants used before the vowels of the back row is marked. Since it has already been established (Kazlauskienė 1998) that in the PVA dialect unpronounced long endings and even stem vowels are abbreviated (except for vowels used before the short ending, e.g.: dra'sa „courage“, rupi „care“, ticki „tick“, treši „fertile“, ke-dėz „chair“), these vowels are marked as semi-long. 113 ‘acid’ — ri-kStu. „rūgštų“, ru-kštūs „rūgštus“ — ru-kštū „rūgštu“, būstus „būstus“ — bii-stu ‘would’. The length of ir internal vowels in closed and open syllables was analyzed. The following minimum word pairs are selected for this: sūktas (v.s. vard.) ‘twiste- — d’ twist (abbreviated. es. I. 3 asm. ‘turns’) ‘turns’, ‘turned’ (dgs. gal.) “twiste- — d” brush (success) We'll be "screwed". I. 3 asm.) ‘sukas’, kdė:pta (unspecified (g) "baked" — k@&-pa "Baked", fried (nearly. r. be. 1. dgs of participant. vard.) ‘kepti’ — kept ‘kepi’. For the experiment, we chose examples in which the number of vowels used before one mų explosive against the blowing and ir explosive consonants was compared: suskiū „suskiu“ —Suku „Šukiu“, sūskis „suskis“ —sūkis (verb „suktis“ of the imperative pronoun unit 2 person) „sukis“. Furthermore, we wanted to compare the length of vowels used before one blowing and blowing and explosive consonants. Selected words for this: pest ‘peštu’ — pest "pešu", walking (total) "pešti" — pawn (we'll be. 1. 2 asm.) ‘on foot’. 1.2.These words were used to form short sentences of clear meaning and similar rhythm, in the middle of which were the research words. The sentences were read by 8 (2 women, 6 men) representatives of (4 Kaunas dialect kapsai ir 4 zanavykai). Jų average age — 62 m. 1.3. The duration of sounds was measured on an IBM-type personal computer using the computer speech signal processing program “KALBAM 43” (programming language TURBO-PASCAL, v. 6.0)* developed by P. Kasparaitis at Vilnius University. The words read by Diktoriy were recorded in the computer memory and an oscillogram prepared by the program is displayed on the screen. After listening, it can be broken down into individual sounds, the lengths of which are measured by the application itself The results of the measurements were processed by mathematical statistical methods: calculated arithmetic mean (x), standard deviation (s), coefficient of variation (v), 95% confidence interval (+). The significance of the results was checked by the Student criterion. Statistical analysis of the data was performed using the program “STUDENT” (programming language TURBO-PASCAL, v. 7.0) developed by A. Girdenis. The formulas used by the program are well known from other experimental phonetic works." 2. The results of the study show that the ratio of the number of vowels in the open and closed ends is approximately as follows: [-]/[—#I: [-/[—5] = 1: 1,2; [=/[—#]: FM—s] =1: 4,15 T-/—-24]:: [-/1—3]:=1721,35 PU—#]: P/N—s] = 1: 1,3. Thus, the vowels of the closed ends are on average 1.2 times longer than those of the open ends. In comparison, the difference in the length of the short open and closed vowels is greater. 2.1. The ratios of the number of vowels in the fixed-end words at the positions of examination are as follows: [a-)/[—#]: [a')[—s] = 1: 1,1; [€-]/[—#]: [e)[—s] = 1: 1,1; [i)/[—] #]: [i')/[—s] = 1: 1,2. If we compare the circumfleksiniy and other vowels (see section 2), we see that this is the smallest difference in length. Probably this was influenced by the studied examples (renkas —renka“, lenkde-s —lenkee-, ni.ksti-s — ni.kšti“), because two pairs were made up of present tense participants of the active kind, which, unfortunately, are used less and less often, and the open ending of the third pair * For a brief description see (Girdenis et al. 1995: 15-21). * All sound measurements and statistical calculations were performed in the Department of General Linguistics at VU. “ This symbol indicates the position of the vowel: [-] —long uncreated, [—] —short unaccented, [=] —quaternium, [+] —short accented vowel, [—#] —open ending, [—5] —closed ending, [—C] —inner position of the word. 114 secondary (ni.kSti-„- thumby“- ), it, a possible thing, is in itself longer than the primary. On the other hand, the length of the vowels of the rigid end may be less dependent on the nature of the syllable. "The 100 Greatest Musicians of All Time" (PDF) (PDF). "The 100 Greatest Musicians of All Time" (PDF). "The 100 Greatest Musicians of All Time" (PDF). "The 100 Greatest Musicians of All Time" (PDF). t < ¢ (ms) (ms) (%) (ms) Milio [reenka-- s] 19 | 1615 36,6 22,6 143,9+179- ,2 0,88<tv05- =2,03 Įrenka) 20 | 150,9 38,7 25,6 132,8+169- ,0 [lenkde-s) 14 | 186.5 39.8 21,3 163,5+209.- ,5 1,03<1t405=2,06 [lenkee-] 14 | 170.8 40,7 23,8 147,3+194- ,3 [ni.kšti“- s] 18 | 1235 34,8 28,2 106,2+140- ,8 1,52<1t405=2,03 [ni.kšti“] 18 | 105,4 36,6 34,7 87,3+123,6 [=)/[—s] 51 | 1550 44,2 28,5 142,5+167- ,4 1,62<t05=- 2,01 [=)/[— #] 52 | 140.5 46,4 33,0 127,6+153,4 (le. kee.s] 10 | 1295 210 162] 1145+1445 0,78<t905- =2,10 [ler. kee.] 10 | 121.8 23.1 18.9 105.3+138- .3 [kd.rti.s] 18 83.2 15.8 19.0 75.3+91.0 3.76>t9001= 3.61 [kd.rti.] 17 64,2 13,9 21,7 57,1+71,4 Int.s) 18 94,2 17,2 18,2 85,6+102,7 5,70> 14.001 =3,60 Int.] 18 64,7 13,6 21,1 57,8+71,5 [-)/[—s] 46 97,5 24,7 25.3 90,2+104,9 3,52>1t9001= 3,52 Įrankas) 23 89,9 18,3 20,3 82,0+97,8 2,69>t901= 2,69 Įrankas) 23 73,8 22,1 29,9 64,3+83,4 [silkces) 21 | 100,7 22,5 22,3 90,5+111,0 2,40>1905- [-/[— #] 45 77,2 28,9 37,4 68,5+85,9 =2,02 [silkee] 21 83,7 23,4 27,9 73,1-+94,4 Įru:-kštis) 22 73,6 18,6 25,3 65,3+81,9 4,04>1)00,- =3,54 [ru-kst] 22 54,1 12,8 23,7 48,4+59.8 [rue-kšties) 22 78,5 16,4 20,9 71,2+85,7 4,85 >t0.001= 3,55 [ru-kštū] 21 55,6 14,5 26,0 49.0+62,2 1s] 88 85,6 214 25,0 81,0+90,1 5,67:>to.001= 3,41 V—#] 87 66,8 22.3 33.3 62.1+71.6 [ren.kas) 19 78.0 19.7 25;3 68.5+87.5 2.85 >t001=2.- 72 Įrerj.ka) 19 59.9 19.4 32.3 50.6+69.2 [leįj.kae.- s] 16 91.6 232 253 79.3+104.0 2.11>1.05- =2.04 [len.kas] kee") 16 76,3 17,5 22,9 67,0+85,6 © Values of the symbols used in the statistical score tables: n — number of measurements, X — arithmetic mean, s — standard deviation, v — coefficient of variation, 9596 p. Interv. — 9596 confidence interval, f, —calculated value of the Student criterion, 7. —critical value of the Student criterion, a. —significance of the results. 115 Examples n x S V 95% include. t <¢ (ms) (ms) (%) interv. (ms) i § [bd.rtis) 19 61,2 9,3 15,2 56,7+65,7 5,80>19.001=3,59 [bd.rti] 18 43,8 8,9 20,3 39,4+48,3 [housing] 20 61,7 16,5 26,7 54,0+69,4 4.20>t0001=3.57 [Bii-stu] 20 42,9 11,3 26,3 37,6+48,2 [V[—s] 74 72,2 21,3 29,4 67,3+77,1 5.13>1t0001=3.43 [—/|— #] 73 54,9 19,7 35,9 50,3+59,5 As you can see iš 1 According to the results of the statistical analysis presented ir in the table, the differences in the duration of open-closed vowels are statistically insignificant. Nevertheless, the differences in the average lengthy of the vowels in comparison are not small ([a-]/[|— 5] —[a:V/[—#] = 10,6 ms; [ė-]/[—5] —[€:/[—#] = 15,7 ms, [i-/[—5] —[1/[—] #] = 18,1 ms), and, increasing the number of realizations, they would become more prominent. 2.2. The ratios of the unaccented ir long" open-closed vowels are approximately as follows: [e-]J/[—#]: [e-]/[—85] = 1: 1.1; [i-}/[—#]: [i-}/[—s] = 1: 1.3; [)/[—#]: [w]/[—s] = 1: 1.5. The statistical significance of these vowels is slightly different from that of the strong vowels. The differences in the duration of upper vowels [i-, u-] in the investigated positions are statistically significant: the established confidence intervals do not even touch, and the calculated values of the Student criterion exceed the highest critical values of this criterion by 0.001. Only the difference in the duration of the vowel [e:] is small ([e-]/[—5] —[e')/[—#] = 7.7 ms) and statistically unreliable. 2.3. The pronunciation of the short open-end vowels and the corresponding closed-end vowels is approximately as follows: [4]/[—#]: [a]/[—s] = 1: 1,2; [e]/[— #]:[eV[—s] = 1: 1,2; [I/[—#]: [(/[—5] = 1: 1,4; [a)/[—#]: [A/[—5] = 1: 1,4. The differences in the duration of all short pronounced open and closed vowels are statistically significant because the confidence intervals of the duration almost do not touch and the calculated values of the Student criterion are greater than the critical values of this criterion. 2.4. Unaccented short vowels are also shorter in the open ends than in the closed ones. Their ratios are as follows: [a]/[—#]: [a]/[—s] = 1: 1.3; [e]/[—#]: [e]/[—5] = 1: 1.2; [i]/[—#]: [i]/[—85] = 1: 1.4; [u)/[—#]: [u)[—s] = 1: 1.4. The differences in the duration of these vowels are statistically reliable: the confidence intervals of the low vowels touch little, of the high vowels — not at all, and the calculated values of the Student criterion exceed the critical values of that criterion. As already mentioned, in unaccented syllables long vowels are shortened, so the term long is used here conditionally. 116 3. Although the hard-toed and short-toed (and accentuateir d, unaccentuated) vowels of closed ends are longer than those of open ends, but they are shorter than the corresponduž ing internal vowels” (see section 4.4). 2 slow. ). 2 slow The proportions of some ir internal open and closed vowels” Vowels Position and proportions P:U Open end-Closed end |PreElast no end syllable [a] 1 1.1 1.3 1.3 [e-] 1 1.1 1.2 1.1 [i] 1 1.2 1.5 1.3 [= 1 1,1 13 12 [a] 1 1.2 1.4 12 [ė] 1 12 1.3 1,2 [i] 1 1.4 1.8 1.3 [it] 1 1.4 1.5 1,1 [¥] 1 1,3 1,5 1,2 [a] 1 1.3 1.5 1.2 [e] 1 1.2 1.2 1 [7] 1 1,4 1,7 1,2 [u] 1 1.4 1.8 12 [1] 1 1,3 1,5 1,2 Table 2 shows that in the comparative positions the ratios of upper vowels are higher than those of lower vowels (cf. [V'V[—#]: [V'/[—s]: [VV [— C—]=1:13:15 and[V)Y[—- #]: [V “/[—85]: [V“/[—C€] = 1: 1,1: 1,2). Thus, the length of upper vowels depends more on the position. This is not an unexpected result, since, as Packer’s study of the perception of vowels in common language (1982: 96-102) shows, the quality is more important for the opposition of upper long and short vowels, and the duration of lower rising vowels. Therefore, the length of *Vidiniy vowels is measured in words: ma-sto. ‘bridge’ ([a‘] —201.4 ms), goes Stu. "drill" ([@'] —208,7 ms), gri-Stu. "return" ([i*] — 154,0 ms); pé.rgree- .stu. ‘perforated’ ([@-] —92,0 ms), pé.rgri.stu. ‘re-grounded’ ([i] —88,1 ms), pcé.rpu.- stu. "Blown" ([u-] —80,8 ms); find (general), find ([a] —103,0 ms), rust ([it] —83,2 ms), tremble ([e] —123,1 ms), stir (general) stir ([!] —98,4 ms); ratū „ratu“ ([a] —91,8 ms), krutū „krutu“ ([u] —76,4 ms), retū „retu“ ([¢] — 93,3 ms), ritū „ritu“ ([i] —73,8 ms). Speakers, recording, duration measurement methodology is the same (see sections 1.2 and 1.3). The average length of the vowels is given in parentheses. ? * P: U — the ratio of the length of the penultimate syllable to the length of the vowels of the closed ending. 117 the number of upper vowels may be more position-dependent than the number of lower vowels. Very interesting results were obtained by ir comparing the duration of long unpronounced stem open and closed vowels. Upper stem only [i-] is longer than the [i-]-type vowels of the open and closed endings (their quantity ratios are approximately as follows: [i-]/[—#]: [i-J/[—85]: []/[—€] = 1: 1.3: 1.4). The upper stem vowel [u“] is longer than the open end [u-], but shorter than the closed end [wu] ([1:]/[—#]: [u-]/[—C€]: [w)[—s] = 1: 1,2: 1,5). The lower case vowel [e:] in the penultimate syllable of the subjunctive is shorter than both the closed and even the open ending [e] ([e-]/[—C]: [e-J/[—#]: [e-]/[—5] = 1: 1,3: 1,4). This shows that the pokirtic Zemutiniai vowels in the PVA dialect are significantly shortened. 4.0. It would be useful to compare the length of the end and internal vowels in open and closed syllables. But exactly this is not easy to do, because it is difficult to find suitable pairs. Furthermore, when the syllable boundaries are determined by the phonological syllable model (STR), the number of closed internal syllables is further reduced. Because the closed endings are only with the consonant [s], and this consonant is attached to the next syllable inside the word. Three sample groups were selected for the internal vowel experiment: in the first group, comparable vowels were used against one (V-7V) and two explosive consonants (VT-TV), in the second group — against a blown and explosive (V-STV) and against one explosive consonant (V-7V), in the third group — against a blown and explosive (V-STV) and against one blown consonant (V-SV). 4.1 The results of the measurement and statistical evaluation of the length of the first group of examples (the investigated vowel is used before one and before two explosive consonants) show that the length of the internal vowels, which do not form diphthongs, is not significantly affected by the nature of the syllable (open or closed), a possible object (see Figure 4.1). 3 slow. ). Only in the words ke pta and kée-pa the difference in the length of the internal vowels is statistically significant: the confidence intervals touch little, the calculated value of the Student criterion at the level of 0.05 slightly exceeds the critical value. The other three pairs (suktas —sūkas, suktas —sukas, kepti —kepi) have very similar mean lengths of internal vowels and their differences are statistically insignificant. It seems that the internal vowels in an open syllable should be longer than in a closed syllable. But only the turned — the turned one can see it." "About it see (Girdenis 1985: 5-11; 1995: 115-121; Pakerys 1995: 286-288). "! By the way, in the word sukas the syllable boundary coincides with the morphemy (the prefix suand the verb kasti the future tense root kas). 118 3 table Duration (ms) of internal open and closed syllable vowels and statistical evaluation of this duration Examples | n xX S V 9596 p. 11, (ms) | (ms) | (%) | interv. (ms) [hit] 15 659] 128] 194 | 588+73,0| 0,02<tos=2- ,05 [hit] 15| 657] 168] 255] 56,4+75,0 [turned] 15 51,61 93] 179| 46,5+56,7| 1,73<toos=- 2,05 [turned] 15| 585] 123] 210] 51,7+653 [turned] 12| 159,8| 17,1 | 10,7 | 149,0+170,7 | 2,26>t00s=- 2,07 [turned] 12| 177,1| 20,1 | 11,4 | 164,3+189,9 [baked] 15) 72,3| 259| 359| 57,9+86,6| 0,18<tv0s=- 2,05 [baked] 15| 741] 28,2| 380| 585+89,7 [siskis] 12 83,8| 9,9| 11,8] 77,6+90,1| 3,32>t0m0s- =3,19 [turn] 12] 684] 12,7| 18,6| 60,3+76,5 [inserted] 6| 728] 54| 74] 67,2+78,5 | 3,93>t000s- =3,58 [šukū] 6(61,8| 43| 69| 57,4+66,3 [pesti] 12 1027) 11,3] 11,0 | 955+109,8 | 1,49<t05=2,- 07 [pesi] 12] 958] 11,5] 120 | 884+103,1 [pesti] 9] 794 11,1] 139 70,9+88,0| 0,95<tyos=- 2,11 [pest] 9 744] 11,3] 151] 658+83,1 Such contradictory results (one difference in length statistically significant, three insignificant, in which the tendency to pronounce longer vowels of the open syllable can only barely be seen), I think, can be explained only by the fact that perhaps the long accented vowels in the open syllable are longer than in the closed, And the number of short vowels is not affected by the nature of the syllable”. On the other hand, as A. Girdenis (DLKG 1994: 36-37) states, Lithuanian common language (and therefore probably PVA) is characterized by a tendency to open syllables, so in a cohesive text closed syllables can become open. The study of the length of these examples seems to suggest that either we have phonetically open syllables here, or the number of short vowels is not affected by the nature of the syllable. 4.2 The length of internal vowels before a blown and explosive (V-STV) and before a single explosive (V-7V) differs slightly more, as shown by the statistical analysis of their length differences. 12 According to some speakers, the word kepta itself is very rarely used or not used at all. The vowel in it may therefore have been pronounced shorter than usual in this position. Perhaps this was influenced by the very common word fried "fried". 119 In examples with Ski — The differences in the duration — of the internal vowels suku, sūskis sūkis are statistically significant: the established confidence intervals do not touch, the calculated values of the Student criterion exceed the critical values. 0,005 Longer vowels in these examples before the blown ir explosive consonant (V-STV). If the internal open ir closed syllable (before one ir before du explosive consonants) the length of the vowels would differ, it would be possible to say that in suškū type words the syllable boundary goes between the blowing and explosive consonant. In that case, the open internal syllables would be shorter and the closed ones would be longer (as would the limbs). Unfortunately, the length of internal open and closed syllable vowels does not differ. Therefore, we have to look for other reasons why the vowel before the blown and explosive (V-STV) is longer than before the single explosive (V-7V). In this case, the longer vowel length was probably influenced by the subsequent blowing consonant, as the results of the study show that the vowels in that position are longer. When pronouncing a blowing consonant, the mouth cavity becomes tight, and when pronouncing an explosive consonant, the mouth cavity becomes closed (LKG 1965: 68-88; Mikalauskaitė 1975: 38-50; Pakerys 1995: 88). The latter requires more energy to overcome than the earlier ones, which is why adjacent vowels may be shorter." 4.3 In the examples pešti —peši, pėsti —pėši, the differences in the length of the vowels are small, and there is no statistical basis to say that the number of these vowels differs. However, there is a tendency that vowels are longer before a blown and an explosive consonant than before a single blown consonant. As previous research has shown, the vowels of the open syllable (V-TV) are slightly longer than those of the closed syllable (VT-TV). Therefore, in the examples pešti, pėsti the syllable boundary cannot be between the blowing and the explosive consonant, because then there would be a longer vowel of the closed syllable, which would contradict the results of the study of the length of open and closed syllables, for which the boundary is not in doubt. Thus, it can be concluded that in these words we have only open internal syllables, and the boundaries of the syllables are: pe-šti, pe-3i, pėšti, pė-ši. Thus, the experiment certainly supports the statements of phonology theory (Girdenis 1995: 115-121) that the ST group does not close the syllable. Such statistically insignificant differences in the length of the vowels of the open and closed stem syllables would show that A. Girdenis (1985: 6; 1995: 115—16) and A. Pakerys (1995: 282-283) were probably right when they stated that there are no phonetic signs signalling the beginning and the end of the syllable." * On the other hand, when comparing the researched words, it appears that suški „suskiū“ and sūskis „suskis“ are not so neutral. Although the lyrics were recorded in a neutral intonation, they could be pronounced more emotionally than the comparable words suka „šukiu“ and sūkis „sukis“. "Perhaps for the same reason, the syllable boundaries of Lithuanian have not been experimentally investigated. It is true that there have been attempts to discuss them (LKG 1965: 122-126; Mikalauskaitė 1975: 71-75; Pakerys 1995: 281-288), but the specificity of the works made it necessary to do so. Only common languages (Strimaitienė, Girdenis 1978: 61-68) and Northern Samogitian (Kliukienė 1983: 50-60; 1992: 38-51) have been studied in more detail. 120 PVA internal open ir closed syllable vowel quantity analysis. Therefore, in the jų opinion (DLKG 1994: 36-37; The 1995: 282-283), distinction between open and ir closed syllables in the common Lithuanian language is not appropriate. 5.1. In summary, the following conclusions can be drawn. 1. In the PVA dialect, the closed-end vowels are longer than the už open-end vowels. Perhaps the final consonant of the closed ending protects the syllable vowel from abbreviation. On the other hand, the effect of the length of open-ended vowels may result in a smoother ir transition of the vowel iš pause. į 2. The length of upper vowels depends more on the position. 3. Pats The fact that the vowels of the PVA dialect end are abbreviated is confirmed once again. In almost all cases, the limbs (and open, even closed) vowels ir are shorter than the corresponding internal vowels. 4. The length of už internal vowels, which do not form diphthongs, is not affected by syllable closure. In open syllables (V-TV) the vowels are only slightly longer than in closed syllables (VT-TV). 5. internal vowels before a blown and explosive consonant (V-STV) are longer than before one explosive consonant (V-7V). Since the length of the vowels in certain closed (VT-TV) and open (V-TV) syllables is almost identical, the slightly longer length of the inner vowel before the blown and explosive consonants can only be explained by the influence of the blown consonant. 6.The length of the internal vowels before the blown and explosive consonant (V-STV) and before the single blown (V-SV) differs little. It is therefore not possible to say that the consonant ST group closes the syllable." LITERATURE DLKG -Grammar of the Contemporary Lithuanian Language, Vilnius: Mokslo ir enciklopedijų leidykla, 1994. Girdenis A. 1985: The boundary of the phonological syllable: construct or reality? —Kalbotyra 36 (1), Girdenis A. 1995: Theoretical foundations of phonology, Vilnius: Petro ofsetas. Girdenis A. et al. 1995: Prosodic phenomena and allophones analysis of Lithuanian language and its dialects. Scientific report, Vilnius: VU. Kazlauskienė A. 1998: Relationships of vowel quantity in the South-West Highland dialect. — Linguistics 47 (1) (print). internal open junctions. It seems that syllable boundaries can only be determined on the basis of phonological analysis. The syllables of Lithuanian are described in detail in this respect (Girdenis 1995: 113-120; also cf. Girdenis 1985: 5-11). * In closing, I would like to thank 5-11. Prof. Dr. A. Girdenis for his valuable advice and encouragement to carry out this research. 121