Printer Friendly

Survey of previous research on Livonian prosody/[TEXT NOT REPRODUCIBLE IN ASCII].

Livonian prosody is unique among the Uralic languages in contrasting at the same time (a) short and long monophthongs, diphthongs, and triphthongs, (b) prevocalic single and geminate consonants and word-final short and long consonants, and (c) lexical tones (at least the plain tone and the broken tone or stod) in stressed stem-initial syllables and in using (d) different coda weight to multiply the number of possible sound patterns. Moreover, all the features are actively used in inflectional morphology. In addition, Livonian is unique among Finnic languages in contrasting final short and long vowels in stressed monosyllables. The wide spectrum of rare features has made Livonian prosody both a hard and interesting field of research for all linguists who have ever dealt with it.

1. In the Livonian grammar written by F. J. Wiedemann (1861) the writing of vowels was somewhat inconsistent. F. J. Wiedemann clearly noticed the existence of different tones but explained them as specific vowel or consonant lengths similar to quantity contrasts that were discovered in Estonian at the end of the 1850's (1861 : 11). Such an explanation was based on the fact that, long nuclear vowels with the broken tone are shorter than those with the plain (level or raising) tone, and that coda consonants or coda vowels following a short nuclear vowel, are shorter in words with the broken tone than in words with the plain tone. Livonian broken tone or stod as such was first recognized by the Danish linguist Vilhelm Thomsen (1890) who noticed this feature, shared with the Danish language, when he interviewed a Livonian sailor for Germanic loanwords in the Copenhagen harbor during 4-5 hours.

2. Livonian tones were acoustically first studied by Lauri Kettunen. In the phonetic introduction (18 pp.) of his book "Untersuchung uber die livischen Sprache I" (Kettunen 1925) he writes about DER BRUCH DER STIMME or "STOSSTON". According to L. Kettunen, the essence of that tone is breaking of the voice in the middle or at the end of a sound: the vocal cords can produce a complete closure but most often only a momentary weakening is produced. He also found a similar breaking in post-tonic syllables with a long vowel, which he identified as half-long. In this book, L. Kettunen has no specific term for the plain tone. In the book, there are five pages of kymograms exhibiting the broken tone, word-final stops, geminates, consonant clusters, and vowels. The words in kymograms are segmented and the duration of segments is presented; diphthongs and triphthongs are presented as wholes for understandable reasons. L. Kettunen used his data to improve the phonetic transcription of Livonian. In total his book contains 130 pp. of texts with translation into Estonian.

3.1. Eleven years later a pupil of Lauri Kettunen, Lauri Posti who had studied Livonian tones in Riga together with the Latvian phonetician Anna Abele, turned his attention to the similarities of Livonian and Latvian tones (Posti 1936). L. Posti used the equivalents of terms used for Latvian tones, (1) cf. Finnish NOUSUINTONAATIO 'rising intonantion' and KATKOINTONAATIO 'broken intonation' and Latvian STIEPTA INTONACIJA and LAUZTA INTONACIJA. L. Posti presents the hitherto longest list of minimal pairs with the broken and plain tone. As Latvian has also a third tone, namely the falling tone (KRITOC SA INTONACIJA), L. Posti claimed to have found it also "in words with a late irregular lengthening" (at(o) (2) [ate] ~ [ate] 'they are', jarandiz 'away' (3)) and hoped that it occurred in the first syllable of words such as aiga [aiga*] 'edge', raiga [raiga'] 'thigh', kuoigid [kuoigi'D] 'ships'. Unlike L. Kettunen who transcribed the words vaztob 'makes tired' and jogst (4) 'river (elative)' as [va'sB] and [jo'kst] L. Posti claimed that the first consonant after a short vowel is always voiced in a stressed syllable with stod and the "intonation" in a long unstressed syllable is never glottal. Although L. Posti stressed the similarity of Livonian and Latvian (actually Low Latvian) tones he nevertheless admitted that Anna Abele when having first heard the Livonian word [liepa] (5) 'alder' (i.e. a word with a short stressed initial syllable followed by a so-called half-long vowel considered long in the modern Livonian orthography, cf. liepa), thought that the word has stress on the 2nd syllable. (In reality, Livonian has several sound patterns whose prosodic structure conflicts with that of Latvian).

3.2. As Lauri Posti (1936) presented several new ideas concerning phonetic transcription, Lauri Kettunen (1936) in his reaction to the article found them unnecessary and also corrected some word forms presented by L. Posti. Among L. Kettunen's corrections there is, however, an incorrectly pronounced form kuo'igid alongside the correct form kuoigid 'ships'.

3.3. Lauri Posti's article (1937a) is a response to L. Kettunen's reaction. The most important part of it is the set of 12 kymograms made together with Aarni Penttila. Four of the kymograms concern his claim about the falling intonation in Livonian. As it becomes clear from the kymograms that words such as laiga 'broad' and kuoigid 'ships' are not pronounced with the falling tone, and L. Posti thinks that "this intonation must be regarded as the short intonation, the intonation that is characteristic of a short single vowel" (1937a : 193). As the pitch contour of the remaining two words, pernai 'landlady' and jouguz 'yew' (6), cannot be considered falling, L. Posti explains that the falling intonation begins with a rise and ends in a fall. When speaking about broken intonation, L. Posti similarly divides it into two parts: the rise and the fall.

3.4. Lauri Kettunen (1937) in his reaction to Posti (1937a) first notes that in L. Posti's kymograms for rising intonation the pitch in the second syllable is lower than in the first syllable. Later he adds that the pitch is mostly plain but in the second syllable there is a sharp drop. This means that in contrast to L. Posti whose discussion of "intonation" remains in the framework of the lexical tone in the first syllable, L. Kettunen was either unable or unwilling to distinguish between "syllable intonation" and sentence intonation. Even more interesting is L. Kettunen's claim about there being no other falling intonation in Livonian than the one that rose when stod is absent. Supposedly he had in mind an alternative use of words both (a) with the rising-falling pitch combined with intensive glottalization and (b) with prevailingly falling pitch combined with weak glottalization.

3.5. Lauri Posti (1937b) in an answer to Kettunen 1937 repeats his conviction that there is a special falling intonation in Livonian.

4. In the grammatical introduction to his Livonian dictionary, Lauri Kettunen (1938) presents a detailed section on intonation. The section begins with a short description of Latvian "intonations" based on J anis Endzelins's grammar. Later L. Kettunens accepts the Latvian German terms also for Livonian. The plain tone is called GEDEHNTE INTONATION 'stretched intonation'. The broken tone is STOSSINTONATION or BRUCHINTONATION; still "depending on a person (especially in West Livonian), the STOSSLAUT (7) (or <<BRUCHLAUT>>) and probably also the weakening of the expiratory stress can be totally absent, and then by ear only a falling intonation can be perceived that differs from the stretched intonation also by shorter syllable quantity" (Kettunen 1938 : xii).

In case of syllables ending in a voiceless stop or s instead of b, d, g, z, cf. agsto [a'kste] 'to harrow', kadst [ka'tst] 'from the hand', abito [a'pt'e] 'to help' vs. aggoz [a'ggeZ] 'harrow', kadud [kaduD] 'hands', ab [a'B] 'help', "the syllable is sonorously short and when the break of voice cannot be heard, the syllable intonation is hardly different from in a syllable with a short vowel + voiceless consonant" (Kettunen 1938 : xxi). Interestig enough, L. Kettunen's examples about the hardly different syllable intonation concern the words ligtAo [li'kte] 'to soak (tV)'and likto [likte] 'to move (itr.)', liktab [liktaB] 'he moves (tV)' which are derived from intransitive verbs liggo [li'gge] 'to soak, to get soaked', 3sg. lig ub and likkAo [likt>> 1/2] 'to move', 3sg. likub [likkuB]. His conclusion "One can then well speak about a short intonation, that according to its character might be falling" probably contains an evaluation of both the length of the domain of the tone of the first syllable and of the characteristic pitch movement of the second syllable.

The falling intonation (proper) occurs in the first syllable ending in what L. Kettunen calls an overshort (actually half-short) consonant or vowel, e.g. lazret [lazret] 'military hospital', pernai [p.ernai] 'landlady', aiga [aiga] 'edge', laiga [laiga] 'broad', and in words ato 'they are' and jarandiz 'away'. (Cf. fn. 3).

5. In 1941 Aarni Penttila and Lauri Posti published an instrumental study of the plain and broken tone. Here 56 oscillograms and data about duration of segments were published. As the segment boundaries are not indicated in oscillograms, hardly anybody except Kalevi Wiik has ever tried to use the data.

6. Fanny de Sivers (1965) applies the description of stod in Latvian, Livonian, and Danish by Janis Endzelins in his Latvian grammar, according to which for Stosston after a resounding beginning, a momentary glottal stop (which is very often replaced only by the weakening of voice) enters in the middle of the syllable. As also Lauri Kettunen (1938 : xxxv) has explained that the word ro 'money' is actually pronounced as ro'o or ro'o, F. de Sivers maintains that stod in Livonian is not a tone but le coup de glotte 'the glottal stop', i. e. a segment. To support this she shows that there exist counterparts to the Livonian words with stod in Finnish and Estonian: e.g. raha 'money'; the consonant h has been recognized as one of the sources of the Livonian stod. F. de Sivers hopes that a complete phonological analysis would explain the phonological role of the glottal stop. (8)

7. Marilyn May Vihman (1971) in her unpublished doctoral dissertation, which contains an interesting approach to the historical phonology of Livonian, dealt also with instrumental study of stod in Danish and Livonian. Here she explains the acoustical essence of stod as a special type of wave forms.

8. In addition to L. Kettunen (1925) and A. Penttila and L. Posti (1941), data on the duration of segments in different sound patterns have been presented by Seppo Suhonen (1982), Hille Pajupuu and Tiit-Rein Viitso (1986). S. Suhonen studied the total duration of monophthongs, primary and secondary diphthongs, and triphthongs in words in ten patterns with either the plain or broken tone. H. Pajupuu and T.-R. Viitso dealt with the duration of components of all diphthongs and triphthongs in similar patterns. The data analyzed by S. Suhonen are at least partially selected from the recordings by three different East Livonian speakers representing two different subdialects. The data analyzed by H. Pajupuu and T.-R. Viitso were read by one East Livonian speaker.

9. In addition to instrumental studies, problems of historical phonetics/ phonology have been dealt with by Lauri Posti (1936; 1937a, and especially 1942), Lauri Kettunen (1938; 1942; 1947, and especially 1960), and most systematically and dynamically by Marilyn May Vihman (1982).

There exist special studies of the rise of metaphony in Livonian by Mikko Korhonen (1969), and of stod by Kalevi Wiik (1989), Seppo Suhonen (1994), and Eberhard Winkler (2000). Unlike L. Kettunen and L. Posti who distinguish several ways of the rise of stod, K. Wiik presents the Syllable Boundary Theory, which is used to give an attempt to describe the formation of Livonian stod as a remnant of a former syllable boundary. Eberhard Winkler argues against the autochtonous nature of the Livonian stod and explains the rise of both Livonian and Leivu South Estonian stod with Latvian influence.

Problems of Livonian phonology were dealt with by Tiit-Rein Viitso (1974; 1975) and reanalyzed in 1981. In Viitso 1981 : 16-22 main types of tone and quantity patterns were established. A set of important patterns participating in Livonian gradation, called quantity alternation by L. Posti, was presented already in Posti 1942 : 296-301.


Airila, M. 1944, [Rev. of Posti 1942] Livische Lautgeschichte.--FUFAnz. 28, 212-227.

Endzelin, J. 1922, Lettische Grammatik, Riga.

Kettunen, L. 1925, Untersuchung uber die livischen Sprache I, Tartu (ACUT B [VIII.sub.23]).

--1936, Eraista liivin kielen merkintakysymyksista.--Vir., 487-494, 502.

--1937, Viela liivin intonaatioista.--Vir., 313-317, 380.

--1942, [Rev. of Posti 1942] Liivin kielen aannehistoriaa.--Vir., 317-330.

--1947, Hauptzuge der livischen Laut- und Formengeschichte, Helsinki.

--1960, Suomen lahisukukielten luonteenomaiset piirteet, Helsinki (MSFOu 119).

Korhonen, M. 1969, Paapainottoman tavun vokaalin ja j:n regressiivisesta vaikutuksesta liivissa, Turku (Turun yliopiston fonetiikan laitoksen julkaisuja 6).

Pajupuu, H., Viitso, T.-R. 1987, Livonian Polyphthongs.--Estonian Papers in Phonetics. EPP 1984-1985, Tallinn, 96-131.

Penttila, A., Posti, L. 1941, Uber die steigende und. sog. Stossintonation im Livischen.--FUFAnz. XXVII, 235-272.

Posti, L. 1936, Liivin kielen intonaatioista.--Vir., 314-326, 384.

--1937a, Lisaa liivin intonaatioista.--Vir., 189-200, 263.

--1937b, Vielakin liivin intonaatioista.--Vir., 448-451.

--1942, Grundzuge der livischen Lautgeschichte, Helsinki (MSFOu 85).

--1945, Liivin aannehistorian alalta.--Vir., 205-208, 302.

--1973, Alustava ehdotus liivin yksinkertaistetuksi transkriptioksi.--FU-transkription yksinkertaistaminen, Helsinki (Castrenianumin toimitteita 7), 38-41.

Setala, E. N. 1934, Zur livischen phonetik: eine kurze bemerkung.--FUF XXII, 176.

Sivers, F. de 1965, Le coup de glotte en letton et en live.--Cel,i XII, 51-56.

Suhonen, S. 1982, Uber die Quantitat der livischen Vokale.--Voces amicorum Sovijarvi. In honorem Antti Sovijarvi septuagesimum annum agentis die XXII mensis aprilis anno MCMLXXXII, Helsinki (MSFOu 181), 295-306.

--1994, [Rev. of Wiik 1989] Liivin kielen aanneoppia.--Vir., 657-662.

Thomsen, V. 1890, Beroringer mellem de finske og de baltiske (litauisk-lettisske) Sprog, Kobenhavn.

Vihman, M. M. 1971, Livonian Phonology, with an Appendix on Stod in Danish and Livonian. Dissertation submitted in partial satisfaction of the requirements for the degree of Doctor of Philosophy in Linguistics in the Graduate Division of the University of the University of California, Berkeley (MS).

Viitso, T.-R. 1974, On the Phonological Role of Stress, Quantity, and Stod in Livonian.--SFU X, 159-169.

--1975, Outlines of Livonian Phonology.--Symposion Phonologische Analyse der uralischen Sprachen Berlin 17.--20. September 1974, Berlin (Akademie der Wissenschaften der DDR. Zentralinstitut fur Sprachwissenschaft. Linguistische Studien. Reihe A. Arbeitsberichte 22), 83-113.

--1981, Laanemeresoome fonoloogia kusimusi, Tallinn.

Wiik, K. 1989, Liivin katko, Turku.

Winkler, E. 2000, Zum Stosston im Ostseefinnischen.--Aspekte baltischer Forschung, Essen (Schriften des Instituts fur Baltistik. Ernst-Moritz-Arndt-Universitat Greifswald 1), 344-362.


* This study was funded by the Estonian Science Foundation (grant no. 6528) and the Estonian Ministry of Education and Research (targeted research project no. 0182572s03).

(1) Note that the Latvian and Lithuanian phonetic tradition never distinguished between tone and intonation.

(2) As the phonetic transcription used by the referred authors depends on the author, here also the current orthographical forms are presented. Letters in parentheses as in at(o) reflect the existing free variation of shorter and longer forms.

(3) Actually at(o) belongs to allegro forms of the longer form umato and maintains the vowel length of the former second syllable; jarandiz, on the other hand, was at least mostly pronounced as [jarandiZ]. Both Lauri Kettunen and Lauri Posti have systematically ignored or mistreated the West Livonian and Ira (or Central Livonian) short vowel lengthening from full-short to half-long in open stressed syllables and the parallel shortening of the long vowel of the second syllable. This characteristic change was first described by E. N. Setala already in 1913 in an unpublished note, see Viitso 1981 : 38.

(4) The form jogst is characteristically a West Livonian form, cf. West Livonian jo'g, GSg jo'g, PSg jo'ggo, Ira jo'ig, GSg jo'ug, PSg jo'ugo, ElaSg jo'ugst, East Livonian jo'ug,GSg jo'ug, PSg jo'ugo, ElaSg jo'ugst. Still it is unclear whether the Livonian words analyzed by Lauri Posti (and Aarni Penttila) all represent West Livonian. Lauri Kettunen's data represent East Livonian. Characteristically, it remains unknown whether the pre-war researchers L. Kettunen, L. Posti, and A. Penttila have used one or more speakers for producing their kymograms and oscillograms.

(5) In a voiced environment, shorter strong and voiceless obstruents are transcribed as half-long single obstruents (e.g. ate and liepa) by L. Posti, and as short geminate obstruents (e.g. a.tte and lie.ppa) by L. Kettunen.

(6) The two words are compounds or at least former compounds. The word pernai 'landlady' was a rare, probably West Livonian shortened form of the usual Estonian borrowing perinai, cf. Estonian pere|naine 'family' + 'woman, wife'. As an apocopated form of the former disyllabic stem, *pere or '*eri, per would be pronounced with the broken tone. For the word jouguz ~ jovguz, the form jovguz is pronounced with two stresses [joe'ugu:Z] and folk etymologically analyzed as 'flour spruce' or 'meal spruce'. Anyway, the tone of the first syllable in the words pernai and jouguz may be a characteristic of a type of compounds.

(7) The term Stosslaut, Finnish katkoaanne has provoked several linguists to look for a sound (German Laut, Finnish aanne), i.e. a segment, which was done ignoring the characteristic pitch movement.

(8) It should be noted that it is only the partitive and illative forms of monosyllabic vocalic stems with stod ro [ro'] and vo [vo'] 'beeswax; foam' which serve as the strongest argument against the segmental nature of stod: both words take the partitive singular ending -dAo and the secondary illative singular ending -zo, which similarly to the corresponding Estonian case endings -d and -ha/-he/-hu occur only with monosyllabic vocalic stems, cf. rodo [ro'de] and rozo [ro'ze], in Ira and West Livonian [ra'de] and [ra'de]. See also the partitive and illative forms modo [mo'de] and mozo [mo'ze] of mo [mo] 'land, country; soil'; in the illative form the secondary ending was added to the former illative form [mo'] < 'maha(n); the former illative form has been best preserved in Ira and West Livonian adjective [ma'] 'down'.


COPYRIGHT 2007 Estonian Academy Publishers
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2007 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Viitso, Tiit-Rein
Publication:Linguistica Uralica
Date:Mar 1, 2007
Previous Article:Livonian orthography/[TEXT NOT REPRODUCIBLE IN ASCII].
Next Article:Quantity in Livonian: preliminary results/[TEXT NOT REPRODUCIBLE IN ASCII].

Related Articles
China's future nuclear submarine force: insights from Chinese writings.
Chinese evaluations of the U.S. Navy submarine force.
Some comments about Paul Ariste's doctoral dissertation on phonetics of Hiiumaa Estonian dialects/[TEXT NOT REPRODUCIBLE IN ASCII].
The Finno-Ugric Prosody Project/[TEXT NOT REPRODUCIBLE IN ASCII].
Quantity in Livonian: preliminary results/[TEXT NOT REPRODUCIBLE IN ASCII].
Livonian gradation: types and genesis/[TEXT NOT REPRODUCIBLE IN ASCII].
Niina Aasmae, Stress and Quantity in Erzya/[TEXT NOT REPRODUCIBLE IN ASCII].
Rhythm related effects in Erzya/[TEXT NOT REPRODUCIBLE IN ASCII].
Prominence and gemination in Ingrian/[TEXT NOT REPRODUCIBLE IN ASCII].
Livonian and Leivu: shared innovations and problems.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters