„Eesti grammatika“ on koguteos, mis kirjeldab enam kui tuhandel leheküljel eesti keele ehitust alates häälikutest kuni lausete ja dialoogideni. Raamatu viis osa annavad ülevaate eesti keele foneetikast ja fonoloogiast, sõnamuutmisest ja sõnamoodustusest, kirjalikust süntaksist ning esmakordselt ka suulise keele süntaksist ja dialoogi põhiehitusest. Sissejuhatus tutvustab keele tasandeid ja keelekirjelduse põhimõisteid. Grammatika koondab ennekõike selle sajandi eesti keele uurimise tulemusi, pakkudes kasutajale ühtede kaante vahel olulist teavet nüüdiseesti ühiskeele kõigi peamiste aspektide kohta. Käsitlus toetub tänapäevase keelekasutuse andmetele ning toob esile ka keelekasutuse varieerumise ja muutumise tendentsid. Raamatu lõpus on põhjalik eesti keele alaste uurimuste valikbibliograafia. „Eesti grammatika“ on vajalik infoallikas kõigile, kelle töö, õpingud või huvid seostuvad eesti keelega.

This study focuses on the temporal properties of Inari Saami tetrasyllabic words that consist of two metric feet of different quantity degrees. In this study, we analyze the durational structures of primary and secondary stressed feet. The material comes from four elderly native speakers of Inari Saami who read carrier sentences with tetrasyllabic test words. The durations of all segments in the test words were measured and the durations and duration ratios of the feet were analyzed. The temporal properties of tetrasyllabic words are compared to earlier findings on di- and trisyllabic words in Inari Saami and to Estonian and Livonian which both exhibit a three-way consonantal quantity distinction similar to Inari Saami. The results showed that tetrasyllabic words are mainly divided into two disyllabic feet where the first two syllables form a primary stressed foot and the secondary stress is on the third syllable. However, a pattern where a trisyllabic foot is followed by a stressed fourth syllable also occurs in our data. The phonological three-way distinction in consonants is exhibited in primary stressed feet; in secondary stressed feet the opposition is binary as consonants are either short or long.

The article focuses on Estonian words of two and more feet, comparing the temporal structures of the feet in words of four to six syllables that consist of short primary- and secondary-stressed syllables. A phonological introduction to the problem of secondary stress in Estonian is followed by the treatment of lengthening of vowels in secondary-stressed feet. The article then analyses duration ratios on the basis of phonetic data. The pronunciations of the western and eastern peripheries of Estonia — western Saaremaa and eastern Võrumaa — are compared to follow possible regional differences in Standard Estonian. It appears that only in Saaremaa the unstressed vowel in a secondary-stressed foot has generally lengthened, and the duration ratios of the secondary-stressed foot are similar to that of the primary-stressed foot. In both areas the temporal structure of feet depends on the general structure of the prosodic word.

This paper focuses on the phonetic realisation of the broken tone in the highly endangered Finnic languages – Leivu and Livonian. An investigation of the temporal and tonal characteristics of words with the plain tone and broken tone consisting of a long first syllable and a short second syllable in spontaneous speech was carried out. The changes in pitch alignment, durations, and duration ratios of the syllables in disyllabic words were analysed and the role of intensity was studied. The analysis revealed a difference in the observed acoustic characteristics in words with either plain tone or broken tone. Also, differences between Leivu and Livonian were observed. With respect to tonal characteristics, two distinct patterns are evident. In words with broken tone, the pitch peak in the primary-stressed syllable is earlier, while in words with plain tone the peak is late. In Livonian, the duration of the syllable nucleus is slightly shorter in broken tone words than in plain tone words, in Leivu it can be longer.

Leivu is one of the South Estonian dialects historically spoken in eastern Latvia and influenced by Latvian. One likely influence is broken tone or stød, which was developing in Leivu mainly as a result of the loss of /h/ in first quantity degree words. The aim of this study is to determine what characterises the pronunciation of CV’V-words (lost intervocalic /h/) and differentiates these from CVV-words. Sound durations, F0 and intensity contours of the syllable rhyme were analysed. Vowel duration in CV’V-words tends to be longer than in CVV-words. In CV’V-words, a short drop in intensity can occur between two identical or two different vowels, with the first vowel often being longer than the second one. In some cases, the second vowel in CV’V words was laryngealised. In CV’V-words, an early F0 turning point where F0 starts to fall occurs more consistently than in CVV-words where F0 can also be rising. Kokkuvõte. Pire Teras: Katketoon leivu CV’V-sõnades. Leivu on üks lõunaeesti murretest, mida räägiti Ida-Lätis ja mida mõjutas läti keel. Üks neist mõjudest on tõenäoliselt katketoon, mis oli tekkimas peamiselt esmavältelistes sõnades /h/ kao tulemusel. Liivi keeleski on katketooni kujunemise üheks põhjuseks peetud just /h/ kadu. Selle töö eesmärk on välja selgitada, mis iseloomustab leivu /h/-kaoliste CV’V-sõnade hääldust ja mis eristab neid kolmandavältelistest CVV-sõnadest. Analüüsiti häälikukestusi, põhitooni- ja intensiivsuskontuure. Vokaalikestus on leivu CV’V-sõnades veidi pikem kui CVV-sõnades. CV’V-sõnades võib toimuda intensiivsuse langus ja tõus või järsk intensiivsuse langus nii kahe ühesuguse kui ka erineva vokaali vahel: esimene vokaal on enamasti kestuselt pikem kui teine. Ka CVV-sõnade hilistekkelistes diftongides on esimene osis sageli pikem kui teine. Mõnel juhul larüngaliseerub CV’V-sõnades silbituuma lõpuosa. Põhitoonikontuuris on CV’V-sõnades palju järjekindlamalt varane pöördepunkt ja langev põhitoon kui CVV-sõnades, kus tuleb ette ka hilise pöördepunktiga tõusvat põhitooni.

This paper studies the phonetic realisation of the ternary length opposition of consonants in Inari Saami disyllabic feet focussing on the distinction between half-long and long geminates. Inari Saami is compared with other Saami and Finnic languages. Segmental durations, fundamental frequency and intensity are analysed. The results show a three-way distinction in consonant duration most robustly after a short stressed syllable vowel. In the case of a long stressed syllable vowel there is no difference between half-long and long geminates. In all of the analysed word structures, segments are inversely related to each other, as with an increase in consonant duration the durations of the surrounding vowels decrease. The fall in the fundamental frequency contour is less steep in words with a long geminate than in those with a short consonant or a half-long geminate. The intensity of the unstressed vowel decreases with the increasing quantity of the intervocalic consonant. The results indicate that the three-way quantity distinction in Inari Saami is primarily manifested by durational properties in a disyllabic foot, while fundamental frequency and intensity have a less important role. (C) 2018 Elsevier Ltd. All rights reserved.

In the present paper, the pronunciation of word-initial /h/ is considered. Previous research has pointed to the influence of the communicative situation: in informal speech /h/ is frequently dropped, but while reading texts and in formal speech this rarely happens. In addition, other factors influencing the pronunciation of /h/ have been found. However, no attention has been paid to how the same speaker pronounces the word-initial /h/ in different communicative situations. This analysis of the pronunciation of eight speakers shows that all of them retained /h/ more often in formal than in informal speech. The main factor influencing the pronunciation of /h/ is accentuation: /h/ is pronounced significantly more often in accented than in deaccented words. In the case of accented words, differences between speakers were found: two speakers dropped /h/ significantly more often (/h/ was more often retained after a pause, after a vowel-final word and in not frequent words), others pronounced it almost always in formal speech and in informal speech they used it more often in utterance initial and final words. In deaccented words, /h/ was significantly more often pronounced in formal speech (except for two speakers) and in informal speech after a pause.

Selles artiklis käsitletakse sõnaalgulise /h/ hääldamist. Varem tehtud uurimused on osutanud /h/ häälduse olenemisele suhtlusolukorrast: argisuhtluses jääb /h/ tihti hääldamata, aga teksti lugedes ja avalikus suhtluses juhtub seda pigem harva. Lisaks suhtlusolukorrale on leitud muidki /h/ hääldust mõjutavaid tegureid. Tähelepanu ei ole aga pööratud sellele, kuivõrd varieerub /h/ hääldus samal kõnelejal eri suhtlusolukorras. Kaheksa keelejuhi häälduse analüüs näitab, et kõik hääldavad /h/-d rohkem avalikus kui argisuhtluses. Peamiseks /h/ hääldamist mõjutavaks teguriks osutus sõna lauserõhulisus: /h/-d hääldatakse oluliselt sagedamini rõhulistes kui rõhututes sõnades. Rõhuliste sõnade puhul oli keelejuhtide vahel oluline erinevus: kaks keelejuhti hääldas /h/-d teistest oluliselt vähem (enim rõhulistes sõnades pärast pausi ja vokaaliga lõppevat sõna ning harva esinevates sõnades), teised hääldasid /h/-d pea alati avalikus suhtluses ning argisuhtluses enim lausungialgulistes ja -lõpulistes sõnades. Rõhututes sõnades hääldas enamik keelejuhte /h/-d oluliselt rohkem avalikus suhtluses ning argisuhtluses hääldasid kõik kõnelejad /h/-d oluliselt rohkem pärast pausi. Abstract. Pire Teras: Pronunciation of word-initial /h/ by same speakers in formal and informal speech. In the present paper, the pronunciation of word-initial /h/ is considered. Previous research has pointed to the influence of the communicative situation: in informal speech /h/ is frequently dropped, but while reading texts and in formal speech this rarely happens. In addition, other factors influencing the pronunciation of /h/ have been found. However, no attention has been paid to how the same speaker pronounces the wordinitial /h/ in different communicative situations. This analysis of the pronunciation of eight speakers shows that all of them retained /h/ more often in formal than in informal speech. The main factor influencing the pronunciation of /h/ is accentuation: /h/ is pronounced significantly more often in accented than in deaccented words. In the case of accented words, differences between speakers were found: two speakers dropped /h/ significantly more often (/h/ was more often retained after a pause, after a vowel-final word and in not frequent words), others pronounced it almost always in formal speech and in informal speech they used it more often in utterance initial and final words. In deaccented words, /h/ was significantly more often pronounced in formal speech (except for two speakers) and in informal speech after a pause. Keywords: phonetics, word-initial /h/, variation, Estonian

The present paper aims to take a closer look at the phonetic variation of short intervocalic /h/ in spontaneous Estonian. Disyllabic CVhV-words were extracted from the Phonetic Corpus of Estonian Spontaneous Speech. Visual inspection of spectrograms shows, that the most frequent variant of short intervocalic /h/ is the voiced variant (70% of the cases). In 21% of the cases, voicing has resulted in the loss of /h/, and consequently disyllabic words can be pronounced as monosyllabic CVV-words. In 3% of the cases, the loss of /h/ is accompanied by vowel shortening whereas /h/ is pronounced as voiceless in 6% of the cases. Some potential factors conditioning variation are discussed in the paper. The percentage of voiced variant increases in accented words, in formal situation and it is favoured by female speakers. Loss of /h/ occurs most frequently in deaccented words, and in informal situation, but in the case of male speakers it also occurs in the formal situation. Loss of /h/ accompanied by vowel shortening occurs only in deaccented words in informal situation. The voiceless variant occurs more often in formal situation, and in accented words pronounced in isolation.

There are 9 vowels in Estonian: /i, u, u, e, o, o, o, a, a/. Unlike other vowels the classification of the vowel /o/ has been problematic. In the phonetic transcriptions /o/ has been transcribed as [gamma] or [w] (IPA) or [e] (Finno-Ugric transcription). What are the possible reasons for this variance? This paper gives an overview of the results of experimental phonetic research focusing on the quality of /o/: articulation, acoustics and perception. All these results, as well as phonetic transcriptions of Estonian dialects, point to variation in the quality of /o/. In the articulation of different speakers the configuration of the body of the tongue and its height are similar to the close back vowel /u/ or to the close-mid back vowel /o/, but /o/ is unrounded ([w] or [gamma]). In the acoustic analyses /o/ has rather been compared to the close-mid vowels than close vowels, admitting that /o/ can be closer than other close-mid vowels. Perception tests have shown that /o/ occupies the space corresponding to both the close-mid and close vowels. The pronunciation of /o/ can also vary due to the speaker's regional background. Phonologically, /o/ has been classified as an unrounded close-mid central or back vowel or unrounded close back vowel. Considering the phonetic variation of /o/, a compromise in the classification of /o/ can be that it is an unrounded non-open back vowel.

The paper investigates the variation of pitch and intensity as a feature of the quantity contrast in Inari Saami. Previous studies [1, 2] have shown that in the context of short vowels, an intervocalic consonant has three degrees of length, while after a long vowel the distinction is binary, and that there is also a compensatory lengthening mechanism between the segments within the foot. The research here is based on the same data that was used for studying the temporal characteristics of quantity. We analyze fundamental frequency of the whole disyllabic foot as well as the intensity of the intervocalic consonant and surrounding vowels. In other Finno-Ugric languages such as Estonian, Livonian and Lule Saami, these characteristics have been shown to be important in the realization of the quantity contrast. Our results indicate that in Inari Saami the interaction between the overall falling pitch contour and quantity is not as strong as it is in Estonian. However, in words with long geminates the pitch fall is shallower than in other word structures. Similar to Lule Saami, in the stressed first vowel the intensity increases with consonant quantity, while in the unstressed second vowel it decreases.

The sound system of Inari Saami exhibits a threeway quantity contrast which is similar to some other Finno-Ugric languages – Estonian and Livonian – in several respects. In all these languages the domain of ternary quantity is a disyllabic foot where the first syllable is stressed and the second syllable is unstressed. In this study we investigate the realisation of the ternary quantity contrast in Inari Saami disyllabic words with phonologically short and long vowels or diphthongs as the nucleus of the first syllable. Our results show that in consonant duration the ternary distinction occurs after a short vowel. The opposition is binary after a short diphthong (consonant is either short or long) and after a long vowel or a long diphthong (consonant is either short or half-long). A vowel in the unstressed second syllable compensates for the length of the preceding vowel and consonant within the foot.

äesoleva uurimuse eesmärgiks on iseloomustada eesti keele palatalisatsiooni eripära, nagu see ilmneb eesti keele spontaanse kõne lindistustes

The three-way distinction of quantity occurs in several Finnic and Saami languages. The paper focuses on the length contrast of consonants in Inari Saami. Similarly to Estonian and other Finno-Ugric languages where three quantities are described, in Inari Saami the distinction between single consonants, short geminates or consonant clusters, and long geminates or consonant clusters appears only on the boundary of a stressed and unstressed syllable of a disyllabic foot. Our results show that in Inari Saami the duration of consonants is inversely related to the duration of both preceding and following vowels, and there is a tendency towards foot isochrony. The results are in line with previous studies on quantity opposition in Inari Saami and in other Finnic languages, showing the ternary distinction of consonant quantities as a foot-level feature of the language. Index Terms : Inari Saami, geminates, three-way quantity 1. Introduction The sound system of Inari Saami reveals three phonologically distinctive quantities. The ternary duration contrast occurs in several Saami languages, including North Saami which has a central position in the Saami language area [1]–[3]. Inari Saami is an eastern Saami language spoken by about 200 native speakers in northern Finland. The Inari Saami phonology is characterized by left-headed feet, word-initial primary stress, and a distinction between short and long vowels and consonants both in stressed and unstressed syllables. The three-way distinction of quantity appears only with consonants in primary stressed feet that are left headed. The ternary contrast is realized by the distinction of single consonants, short and long geminates (traditionally called half-long and long consonants), or consonant clusters on the boundary of the stressed syllable and the following unstressed syllable, e.g.

The three-way distinction of the Estonian quantity is a feature of the primary stressed disyllabic foot. The quantity degrees are realized by different temporal patterns of the segments within the foot. Additionally, other phonetic features appear to vary depending on the quantity, such as pitch contour and vowel quality. In this paper, the quantity-related variation of segment duration, pitch, and vowel quality was investigated in spontaneous Estonian in words occurring in sentence-medial position. While a disyllabic foot can be made up of a number of segmental combinations, the data showed that the quantity opposition operates independently of the segmental structure of the foot. The effect of the analyzed features was evaluated with a multinomial logistic regression model. The model showed that all the variables which were included had a significant effect. Besides the inverse relation of the stressed and unstressed syllable rhymes, which is traditionally used to describe the three-way opposition, syllable onset duration also had a weak effect. Additionally, vowel quality was significant for the opposition of short (Q1) and long (Q2), and pitch alignment was significant for the opposition of long (Q2) and overlong (Q3) quantities.