Th is paper examines the code-switching of verbs in the speech of two children bilingual in Eston... more Th is paper examines the code-switching of verbs in the speech of two children bilingual in Estonian and English (aged 3 to 7). Verbs typically have lower rates of code-switching than nouns, due to their central role in argument structure, lower semantic specifi city, and greater morphological complexity. Th e data examined here show various types of morphological mixing, and include examples which violate the prediction from the literature that only fi nite verbs bear infl ectional morphology from the other language , suggesting that children do not adhere to the same constraints as adults when code-switching.
This paper examines the phenomenon of lability in Estonian. We describe the types of lability fou... more This paper examines the phenomenon of lability in Estonian. We describe the types of lability found in our database and the distribution of verbs according to formal (derivational) and semantic verb classes, and we propose an explanation for the spread of lability in the Estonian verbal lexicon. Estonian displays a wealth of labile verbs, compared with Finnish, for instance, a closely related language which uses similar (derivational and morphosyntactic) valency changing devices. We argue that Estonian lability has been co-conditioned by the following factors: a) intensive contact with German, a language rich in labile verbs, b) fluctuations in the productivity and regularity of the valency-decreasing derivation in the recent history of Estonian, c) the phonetic merger of different derivational suffixes leading to misinterpretation of the valency patterns of their derivatives, and d) form-driven analogy. In relation to the last factor we introduce the notion of “clustered lability,” which designates the expansion of the labile pattern over verbs sharing the same root.
This article investigates variation in constructions expressing need in Estonian, based on materi... more This article investigates variation in constructions expressing need in Estonian, based on material from the Corpus of Spoken Estonian, the 1990s Fiction Subcorpus of Written Estonian, and the Estonian Dialect Corpus. The predicates
included in the study, vaja/tarvis olema ‘need’, form two basic constructions: one takes a nominal complement and expresses a need for something or a lack of something (Mul on raamatut vaja. 'I need a book'); the other takes an infinitival complement and conveys modal semantics (Mul on vaja raamatut lugeda. ‘I need to read a book’). These constructions are characterised by considerable variability,
conditioned by language-internal (both syntactic and semantic) factors on the one hand, and on the other, language-external factors, primarily language contact. In
this paper, we focus on those aspects of variation in the ‘need’-constructions which have clear parallels with the closest neighboring contact languages, Russian and
Latvian. These include the choice of modal adverb (tarvis or vaja), case-marking on the nominal complement (nominative or partitive), the choice of copula verb (olema ‘be’, tulema ‘come’, minema ‘go’ or copula ellipsis), and case-marking of
the experiencer argument (adessive, allative or omitted argument). We describe parallel developments for each of these elements in the corresponding constructions
in Russian and Latvian, which partially account for the variation in Estonian.
The emergence and later fading of two phonological templates - a 'palatal' template and consonant... more The emergence and later fading of two phonological templates - a 'palatal' template and consonant harmony - are investigated in the first 500 words produced by a child acquiring Estonian and English. Throughout the period the child's use of palatal forms, in particular, considerably exceeds their frequency in Estonian, the child's dominant language. Regression in accuracy is also traced, both overall and in individual word forms. Changes in frequency of use of the template patterns are related to growth in the size of the lexicon, the consonant inventory, and the length in syllables of words attempted. Articulatory difficulty is found to play at best a minor role in motivating pattern use, which is ascribed instead to the challenges of planning and recall.
This paper examines the nominative use of the Estonian pronoun ise ’self’ in texts culled from in... more This paper examines the nominative use of the Estonian pronoun ise ’self’ in texts culled from internet fora, where a remarkable number of uses of ise are to be found without any antecedent. The main question posed is whether ise is systematically used here in place of personal pronouns as a subject. More specifically, we investigate (a) whether ise stands in place of the subject by taking the subject position when there is no other subject in the utterance; (b) whether ise behaves like a topic as typical, utterance-initial personal pronouns do, and if so, whether it then loses its emphatic, focussing function; and (c) is ise used primarily to refer to first person referents.
Results from a corpus analysis show that ise is used without
an antecedent in over half of all sentences with ise in our data, and that this phenomenon is frequently correlated with first-person reference. Ise also seems to follow the tendency of subjects to undergo inversion due to a V2 preference in Estonian. However, preverbal ise is also shown not to have completely lost its emphatic function. The use of ise in subject position in place of personal pronouns is likely to be associated with impersonalisation as a negative (or distancing) politeness strategy: the need to avoid an emphatic or contrastive first person pronoun is likely to have conditioned the frequent use of ise in subject position in internet forum texts.
This paper analyses impersonal constructions in spoken Estonian. The impersonal constructions fou... more This paper analyses impersonal constructions in spoken Estonian. The impersonal constructions found in two spoken corpora differing in levels of formality are examined and classified according to five readings of the implicit argument: universal, vague and specific existential, corporate and a future-hypothetical reading. The analysis focuses on the implicit impersonal actor referent, investigating differences in its interpretation as well as overt expressions of the implicit argument. We account for the diverse readings of the implicit actor as being derived through pragmatic means. The impersonal construction is taken to bear a unified semantics, which specifies only that the implicit argument has the default semantic features of human, plural, and actor. The argument receives its full interpretation through pragmatics and the surrounding discourse context.
Th is paper examines the code-switching of verbs in the speech of two children bilingual in Eston... more Th is paper examines the code-switching of verbs in the speech of two children bilingual in Estonian and English (aged 3 to 7). Verbs typically have lower rates of code-switching than nouns, due to their central role in argument structure, lower semantic specifi city, and greater morphological complexity. Th e data examined here show various types of morphological mixing, and include examples which violate the prediction from the literature that only fi nite verbs bear infl ectional morphology from the other language , suggesting that children do not adhere to the same constraints as adults when code-switching.
This paper examines the phenomenon of lability in Estonian. We describe the types of lability fou... more This paper examines the phenomenon of lability in Estonian. We describe the types of lability found in our database and the distribution of verbs according to formal (derivational) and semantic verb classes, and we propose an explanation for the spread of lability in the Estonian verbal lexicon. Estonian displays a wealth of labile verbs, compared with Finnish, for instance, a closely related language which uses similar (derivational and morphosyntactic) valency changing devices. We argue that Estonian lability has been co-conditioned by the following factors: a) intensive contact with German, a language rich in labile verbs, b) fluctuations in the productivity and regularity of the valency-decreasing derivation in the recent history of Estonian, c) the phonetic merger of different derivational suffixes leading to misinterpretation of the valency patterns of their derivatives, and d) form-driven analogy. In relation to the last factor we introduce the notion of “clustered lability,” which designates the expansion of the labile pattern over verbs sharing the same root.
This article investigates variation in constructions expressing need in Estonian, based on materi... more This article investigates variation in constructions expressing need in Estonian, based on material from the Corpus of Spoken Estonian, the 1990s Fiction Subcorpus of Written Estonian, and the Estonian Dialect Corpus. The predicates
included in the study, vaja/tarvis olema ‘need’, form two basic constructions: one takes a nominal complement and expresses a need for something or a lack of something (Mul on raamatut vaja. 'I need a book'); the other takes an infinitival complement and conveys modal semantics (Mul on vaja raamatut lugeda. ‘I need to read a book’). These constructions are characterised by considerable variability,
conditioned by language-internal (both syntactic and semantic) factors on the one hand, and on the other, language-external factors, primarily language contact. In
this paper, we focus on those aspects of variation in the ‘need’-constructions which have clear parallels with the closest neighboring contact languages, Russian and
Latvian. These include the choice of modal adverb (tarvis or vaja), case-marking on the nominal complement (nominative or partitive), the choice of copula verb (olema ‘be’, tulema ‘come’, minema ‘go’ or copula ellipsis), and case-marking of
the experiencer argument (adessive, allative or omitted argument). We describe parallel developments for each of these elements in the corresponding constructions
in Russian and Latvian, which partially account for the variation in Estonian.
The emergence and later fading of two phonological templates - a 'palatal' template and consonant... more The emergence and later fading of two phonological templates - a 'palatal' template and consonant harmony - are investigated in the first 500 words produced by a child acquiring Estonian and English. Throughout the period the child's use of palatal forms, in particular, considerably exceeds their frequency in Estonian, the child's dominant language. Regression in accuracy is also traced, both overall and in individual word forms. Changes in frequency of use of the template patterns are related to growth in the size of the lexicon, the consonant inventory, and the length in syllables of words attempted. Articulatory difficulty is found to play at best a minor role in motivating pattern use, which is ascribed instead to the challenges of planning and recall.
This paper examines the nominative use of the Estonian pronoun ise ’self’ in texts culled from in... more This paper examines the nominative use of the Estonian pronoun ise ’self’ in texts culled from internet fora, where a remarkable number of uses of ise are to be found without any antecedent. The main question posed is whether ise is systematically used here in place of personal pronouns as a subject. More specifically, we investigate (a) whether ise stands in place of the subject by taking the subject position when there is no other subject in the utterance; (b) whether ise behaves like a topic as typical, utterance-initial personal pronouns do, and if so, whether it then loses its emphatic, focussing function; and (c) is ise used primarily to refer to first person referents.
Results from a corpus analysis show that ise is used without
an antecedent in over half of all sentences with ise in our data, and that this phenomenon is frequently correlated with first-person reference. Ise also seems to follow the tendency of subjects to undergo inversion due to a V2 preference in Estonian. However, preverbal ise is also shown not to have completely lost its emphatic function. The use of ise in subject position in place of personal pronouns is likely to be associated with impersonalisation as a negative (or distancing) politeness strategy: the need to avoid an emphatic or contrastive first person pronoun is likely to have conditioned the frequent use of ise in subject position in internet forum texts.
This paper analyses impersonal constructions in spoken Estonian. The impersonal constructions fou... more This paper analyses impersonal constructions in spoken Estonian. The impersonal constructions found in two spoken corpora differing in levels of formality are examined and classified according to five readings of the implicit argument: universal, vague and specific existential, corporate and a future-hypothetical reading. The analysis focuses on the implicit impersonal actor referent, investigating differences in its interpretation as well as overt expressions of the implicit argument. We account for the diverse readings of the implicit actor as being derived through pragmatic means. The impersonal construction is taken to bear a unified semantics, which specifies only that the implicit argument has the default semantic features of human, plural, and actor. The argument receives its full interpretation through pragmatics and the surrounding discourse context.
Uploads
included in the study, vaja/tarvis olema ‘need’, form two basic constructions: one takes a nominal complement and expresses a need for something or a lack of something (Mul on raamatut vaja. 'I need a book'); the other takes an infinitival complement and conveys modal semantics (Mul on vaja raamatut lugeda. ‘I need to read a book’). These constructions are characterised by considerable variability,
conditioned by language-internal (both syntactic and semantic) factors on the one hand, and on the other, language-external factors, primarily language contact. In
this paper, we focus on those aspects of variation in the ‘need’-constructions which have clear parallels with the closest neighboring contact languages, Russian and
Latvian. These include the choice of modal adverb (tarvis or vaja), case-marking on the nominal complement (nominative or partitive), the choice of copula verb (olema ‘be’, tulema ‘come’, minema ‘go’ or copula ellipsis), and case-marking of
the experiencer argument (adessive, allative or omitted argument). We describe parallel developments for each of these elements in the corresponding constructions
in Russian and Latvian, which partially account for the variation in Estonian.
Results from a corpus analysis show that ise is used without
an antecedent in over half of all sentences with ise in our data, and that this phenomenon is frequently correlated with first-person reference. Ise also seems to follow the tendency of subjects to undergo inversion due to a V2 preference in Estonian. However, preverbal ise is also shown not to have completely lost its emphatic function. The use of ise in subject position in place of personal pronouns is likely to be associated with impersonalisation as a negative (or distancing) politeness strategy: the need to avoid an emphatic or contrastive first person pronoun is likely to have conditioned the frequent use of ise in subject position in internet forum texts.
included in the study, vaja/tarvis olema ‘need’, form two basic constructions: one takes a nominal complement and expresses a need for something or a lack of something (Mul on raamatut vaja. 'I need a book'); the other takes an infinitival complement and conveys modal semantics (Mul on vaja raamatut lugeda. ‘I need to read a book’). These constructions are characterised by considerable variability,
conditioned by language-internal (both syntactic and semantic) factors on the one hand, and on the other, language-external factors, primarily language contact. In
this paper, we focus on those aspects of variation in the ‘need’-constructions which have clear parallels with the closest neighboring contact languages, Russian and
Latvian. These include the choice of modal adverb (tarvis or vaja), case-marking on the nominal complement (nominative or partitive), the choice of copula verb (olema ‘be’, tulema ‘come’, minema ‘go’ or copula ellipsis), and case-marking of
the experiencer argument (adessive, allative or omitted argument). We describe parallel developments for each of these elements in the corresponding constructions
in Russian and Latvian, which partially account for the variation in Estonian.
Results from a corpus analysis show that ise is used without
an antecedent in over half of all sentences with ise in our data, and that this phenomenon is frequently correlated with first-person reference. Ise also seems to follow the tendency of subjects to undergo inversion due to a V2 preference in Estonian. However, preverbal ise is also shown not to have completely lost its emphatic function. The use of ise in subject position in place of personal pronouns is likely to be associated with impersonalisation as a negative (or distancing) politeness strategy: the need to avoid an emphatic or contrastive first person pronoun is likely to have conditioned the frequent use of ise in subject position in internet forum texts.