We kindly inform you that, as long as the subject affiliation of our 300.000+ articles is in progress, you might get unsufficient or no results on your third level or second level search. In this case, please broaden your search criteria.

Download

Towards sophisticated writing

Author(s): Krista Kerge,Hille Pajupuu,Pilvi Alp,Halliki Põlda,Anne Uusen / Language(s): English Issue: 16/2014

In the framework of the Natural Models1 approach to learning Estonian (see Kerge, Uusen, Põlda 2014; Pajupuu et al.2010), examples of highly educated non-philologist native adult speech and writing, rather than ideal edited standard language use, are considered the benchmarks to be striven for1. To improve both teaching and assessing writing, the vocabulary parameters of creative writings of Estonian L1 students from grades 5, 7, 9, and11 and the vocabulary parameters of writings of L2 writers with certified A2-, B1-, B2-, and C1-level proficiencies are compared to this benchmark. It is found that, although most adults must have more experience than young students, even at the C1-level, the L2lexicon of L2 writers is strikingly poorer than the lexicon of native educated adults or even upper secondary students.

More...

Download

External factors and the interference of L1 Estonian on L2 English pronunciation

Author(s): Merilin Miljan,Kristiina Ader / Language(s): English Issue: 11/2015

This study focusses on external factors of second language learning (L2) and their effect on L2 sound production. The aim was to find out whether young adults whose first language (L1) is Estonian speak L2 English with less accent than older speakers of L1 Estonian. Prior studies have claimed that more exposure to the target language lessens the effect of L1 interference (internal factors) (e.g. Piske et al. 2001, Muñoz, Llanes 2014). An apparent-time study was implemented in order to test whether a change in learner setting (i.e. limited exposure to the target language vs. abundant exposure) shows any evidence which can be explained by such external factors. Data was elicited from the speech samples of 97 speakers of L1 Estonian and L1 interference effects were analysed sound by sound. The results show that the younger generation displays less L1 Estonian interference than the older generation. Both greater exposure to the target language and earlier age of L2 learning onset correlate with the ability to produce more sounds target-like.*

More...

Download

Statistilised meetodid ühendverbide tuvastamisel tekstikorpusest.

Author(s): Aedmaa Eleri / Language(s): Estonian Issue: 11/2015

The present article compares lexical association measures (AMs) for automatic extraction of Estonian particle verbs from the newspaper part of the Estonian Reference Corpus. The main purpose of this study is to ascertain the best symmetrical AM for Estonian particle verb extraction. The central focus lies on the impact of the corpus size on the performance of the compared symmetrical association measures. In addition, asymmetrical AMs have been included in the study to observe their suitability for Estonian particle verb extraction. Five symmetrical association measures have been used, namely the t-test, log-likelihood, ǒ2, mutual information, and minimum sensitivity, as well as two asymmetrical association measures, namely FRQGLWLRQDO SUREDELOLW\ DQG Ʃ3 7KH DVVRFLDWLRQ PHDVXUHV ZHUH FRPSDUHG DJDLQVW the co-occurrence frequency of verb and verbal particle. The analysis of the comparison reveals that the t-test achieved the best precision values and the corpus size has an impact on the performance of the AMs. As the corpus size increased, the performances of the t-test, log-likelihood, ǒ2 and minimum sensitivity increased and the precision of mutual information decreased. The performance of (simple) frequency did not change significantly as the size of the corpus increased. The comparison of symmetrical and asymmetrical AMs revealed that asymmetrical association measures are suitable for the task of Estonian particle verb extraction and provide slightly different and more detailed information about the extracted particle verbs. The results presented in this article confirm that further study of asymmetrical AMs is necessary and more experiments are needed to broaden our knowledge of the performance of asymmetrical AMs.

More...

Download

Considering Well, Accurately, Completely: Lithuanian Lexical Verb Placement on the Syntactic Tree

Author(s): Julija Korostenskienė / Language(s): English Issue: 26/2015

The present article is concerned with the position of the Lithuanian lexical verb on the syntactic tree. As is known, Pollock (1989) postulates the verb raising parameter and examines the differences between a verb raising language like French and a verb non-raising language like English using the position of low adverbs relative to the lexical verb as a criterion. Adverb placement since then has come to be regarded as a powerful diagnostic tool. In addition, by exploring the relation between manner adverbs and stative and eventive verbs, Katz (2003) elaborates on the structure of the syntactic tree immediately above the verb phrase vP/VP and posits the Aspectual Phrase (AspP) as immediately dominating the vP/VP of eventive verbs. The present article examines the verb raising parameter in light of the distributional behaviour of threemanner adverbs: gerai, tvarkingai and visiškai and posits two objectives: to find additional support for the low placement of the Lithuanian lexical verb on the syntactic tree and to examine the area immediately above the verb phrase vP/VP. It is argued that native Lithuanian prefixes are to be regarded as the lexically manifest head of the aspectual phrase AspP which is positioned above the verb phrase.

More...

Download

Protoeurooplaste keelepärandist

Author(s): Mari Kendla,Jüri Viikberg / Language(s): Estonian Issue: 61/2015

The article discusses problems of the hypothetical substrate language during the past half century, taking into account the existing vocabulary and results of etymological studies. While in the 1950s it was ascertained that Finno­Ugrians have lived in our region for at least 5000 years, later a hypothesis was elaborated that people of the Kunda culture (6th millennium BC) most probably did not speak any Finno­Ugric languages. According to Paul Ariste they were among the early inhabitants of Europe, i.e. Proto­Europeans. As there are numerous words that are common to the Balto­Finnic languages but not known in other Finno­Ugric languages, we can assume that at least part of them belong to the substrate vocabulary and derive from Proto­European. Ariste assumed that traces of such substrate can be found particularly in ancient Estonian place names (e.g. Peipsi), Estonian words for landscape features (e.g. allikas ‘fountain’, neem ‘cape’), fish names (e.g. ahven ‘perch’, haug ‘pike’, rääbis ‘vendace’, siig ‘lavaret’), and somatic vocabulary (e.g. higi ‘sweat’, huul ‘lip’). The substrate hypothesis has been put forward by other Balto­Finnic ethnogenesis and vocabulary researchers, such as Huno Rätsep, Kalevi Wiik, Kaisa Häkkinen, Tapani Lehtinen, Janne Saarikivi, Petri Kallio, and Jaakko Häkkinen. Over the last few decades, much research has been done on the contacts of the western Uralic languages with the Indo­Iranian and Baltic languages, in the course of which many words of hitherto unknown origin have been defined as loanwords, e.g. haug ‘pike’, hõlm ‘hem’, kiitma ‘praise’, koha ‘pike perch’, kube ‘groin’, lõug ‘chin’, mets ‘forest’, nahk ‘skin’, Ruhnu, siig ‘lavaret’, teib ‘dace’, vimb ‘vimba bream’. But there are still a number of Balto­ Finnic words which have no equivalents in distant cognate languages and are not etymologically identified as loanwords. We can still assume that the words eile ‘yesterday’, helmes ‘bead’, jänes ‘hare’, liha ‘meat’, must ‘black’, saar ‘island’ and soo ‘marsh’, as well as the place names Peipsi, Pärnu and Võrtsjärv, are probable Proto­European substrate words.

More...

Download

On the Placement of the Reflexive/ Reciprocal Marker –si- in Lithuanian Verbs

Author(s): Julija Korostenskienė / Language(s): English Issue: 28/2016

Set within the generative syntactic tradition, the present article examines the interchange in the layout of the reflexive/ reciprocal marker (RM) and the verb in prefixed and prefixless Lithuanian verbs, a long-standing historical puzzle. It is first shown that the RM is obligatorily coreferential with the subject of the sentence. Given the generative premise that all subjects are merged within the verb, it is argued that the RM is a physically manifest trace of the subject and forms a binding domain with its antecedent, which stipulates the layout of morphemes within the verb. In addition, the position of the RM also depends on whether its antecedent is an agent, experiencer, or theme, since these have different merging, i.e. original, positions. After the relationship between the RM, the sentence subject and the verb has been defined, prefixes are examined since they form the left boundary of the RM in prefixed verbs. Given their resultative meaning, verbal prefixes are argued to occupy an aspectual position AspP, placed immediately above the verbal complex, i.e. v/VP and the RM within it. Since in non-contrastive contexts, the negative clitic is attached to the verb and consequently affects the position of the RM in otherwise prefixless verbs, inducing change just as the resultative prefixes, the binding domain is determined for the following types of the morphological composition of the verb: prefixed and non-prefixed positive and negative verb forms.

More...

Download

Latviešu valodas lokāmie divdabji sekundāri predikatīvā funkcijā

Author(s): Agita Kazakeviča / Language(s): Latvian Issue: 7/2016

The article focuses on secondary predicative uses of Latvian declinable participles. Participial clauses, despite their being very widespread, have been excluded from this study due to the fact that only two kinds of participles are capable of entering into this sort of syntactic construction, while the main objective of the study has been to see whether the claim expressed in the “Grammar of Latvian” (Latviešu valodas gramatika, 2013) to the effect that all Latvian participles are capable of functioning as secondary predicative components also holds true for contemporary periodicals and literary prose texts, as well as to identify and analyse functionally borderline cases.The data confirms that all Latvian participles (declinable, in this case) can function as secondary predicative components, although, importantly, declinable present tense participles, especially those ending in ošs, oša, are very rarely used in this way, attributive use being by far more typical. This can be explained by the fact that adjectival meaning is actually more strongly expressed in these participles than their verbal semantics (as is well known, declinable participles have both (Lokmane 2006, 226)). Ever since the “Grammar of Modern Standard Latvian” (Mūsdienu latviešu literārās valodas gramatika, 1959, 639) declinable past tense active voice participles have been recognized as having the highest frequency of use among participles in Modern Latvian. The same can be confidently stated about their frequency of use as secondary predicative components in the 21st century (see, e.g., Nītiņa 2013, 573).

More...

Download

Lietuviško balso sintezatorių kokybės vertinimas

Author(s): Pijus Kasparaitis / Language(s): Lithuanian Issue: 28/2016

Text-to-speech synthesis of most popular languages is widely used for several decades, while the Lithuanian text-to-speech synthesis breakthrough occurred only in recent years. Six new Lithuanian synthetic voices appeared in 2013-2015. Therefore, there was a need to evaluate the newly created Lithuanian text-to-speech synthesizers. This paper presents a chronological review of the current Lithuanian text-to-speech synthesizers. Unit selection algorithm that was implemented in recent synthesizers SINT.AS and LIEPA and selection procedure of announcers are described in more detail because they were crucial to the synthesized voice quality. The main characteristics of synthetic voice are intelligibility and acceptability; they are assessed by involving human-listeners and the received data are processed by statistical methods. Thus, this paper will investigate nine recent Lithuanian synthetic voices (Regina, Edvardas, Aistė, Vladas, Laima, Marijus, Egidius, Aistis 2, Gintaras) and evaluate what intelligibility of a synthetic voice is actually achieved, which synthesizer is better when comparing them with each other, to give advice to potential synthetic voice application developers when choosing a synthesizer voices and finally to show synthesizer developers the most promising areas of improvement.

More...

Download

Syntactically Coded Corpus of Spoken Lithuanian: Developmental Issues and Pilot Studies

Author(s): Ingrida Balčiūnienė,Laura Kamandulytė-Merfeldienė / Language(s): English Issue: 28/2016

The paper deals with the main methodological issues of development of the Corpus of Spoken Lithuanian with particular attention to its syntactic coding and applications for automatized language analysis. First, we consider a methodology of development of the Corpus as well as the principles of transcribing and coding Lithuanian speech data. The main concepts, such as “utterance” “sentence”, etc. are discussed. Second, we present results of a pilot study in interrogatives that are typical for natural spontaneous spoken Lithuanian. Results of the automatized analysis of interrogatives revealed that a frequency and distribution of the Wh- and yes/ no questions is rather similar. Among the Wh- questions, the questions non-containing the interrogative particle seem to be dominant, while the questions containing the interrogative particle at the beginning ot at the end were much rarer. Among the different functional subtypes of Wh- questions, adverbial ones seem to be the most freequent; among the adverbial Wh- questions, the spatial ones were the most frequent. Certainly, the present study is rather pilot due to the novelty of automatized syntactic approach to the data of spoken Lithuanian, thus much more complex studies still await for future investigations. A use of interrogative sentences will be studied from the perspective of different genres (e.g., monologue vs dialogue), social characteristic of the speakers, and a situation of conversation (e.g., public vs private speech). Generally, we believe that future systematic corpus-based research of spontaneous spoken language will give more possibilities to identify, evaluate, and elaborate the development of the Lithuanian language.

More...

Download

Gramatiskās sistēmas aprakstos (19.–21. gs.) lietotie jēdziena ‘verbs’ apzīmējumi latviešu valodā

Author(s): Anita Butāne / Language(s): Latvian Issue: 7/2016

Until the 1860s, the main researchers of the Latvian language were Baltic German clergymen whose works on Latvian grammar were written in Latin and German. Thus, the linguistic terminology used in these sources was also in Latin and German.Latvian linguistic terminology as a system began developing in the 2nd half of the 19th century with textbooks of German, Russian and Latvian, which included a systemic description of language units in the Latvian language.Originally, many linguistic concepts used to have various denotations, and this diversity existed for several decades. The concept ‘verb’ has had the following Latvian equivalents in the course of time: laikavārds, darīšanas vārds, verbs, saceklis, izteiksmes vārds, runas vārds, darbības vārds, norisenis and stāvokļa vārds.The diversity of linguistic terms in Latvian diminished to a certain extent since early 20th century – after the report by teacher Pēteris Abuls in 1901 about the usage of terminology in Latvian language textbooks, and the discussion of these terms in a 1902 meeting of the Science committee (Zinību komisija) of the Riga Latvian Society (Rīgas Latviešu biedrība). The unification of Latvian linguistic terms rooted out such equivalents of ‘verb’ as laikavārds, darīšanas vārds, saceklis, izteiksmes vārds, while the term darbības vārds was chosen to denote this concept. Later, during 1940s and 1950s, a term norisenis was also suggested to denote the verb, but it did not become widely used.The pair of terms darbības vārds and verbs, which are nowadays both used to denote this concept in Latvian, continue the tradition set in the 19th-century Latvian textbooks – the parallel use of both native and borrowed terms.

More...

Download

Derivatīvās galotnes izmantojums termindarināšanā Emīlijas Soidas uzskatu kontekstā

Author(s): Iveta Pūtele / Language(s): Latvian Issue: 6/2015

The derivational ending is a widely used means of term formation. The derivative functions of endings in the Latvian language is one of the topics, Emīlija Soida dealt with in her research, occasionally also making remarks regarding the role of derivational endings in term formation.The article analysis the views of Soida regarding the use of derivational ending in word formation and term formation, types of terms formed with the derivational ending, their characteristics, along with the discussion of the way the ideas of Soida express themselves in contemporary term formation.Soida paid special attention to the derivations from the verbs of the 2nd and 3rd conjugation, as she believed that to be a new activisation of the derivative functions of endings in the Latvian literary languages, when nouns semantically corresponding to the verbs, but of simpler structure are formed. Soida used a term inverted word formation in regard to these derivations.Analysing the terminological material, a conclusion can be reached, that derivation with the derivational ending from the verbs of the 2nd and 3rd conjugation, especially in term formation, has been more widespread in 1960’s–70’s, as indicated by Soida, while the same method of noun formation was also known previously. Prefixed stems of the verbs of the 2nd and 3rd conjugation are most frequently used in formation of nouns with the derivational ending; the most productive derivational ending for creation of nouns from verb stems is -e; the derivational ending is most frequently used exactly in term formation, as the meaning of such a derivative usually is more narrow and specialised in comparison to the original word.

More...

Download

Latīņu salikteņi un to tulkošanas iespējas latviešu valodā

Author(s): Gita Bērziņa / Language(s): Latvian Issue: 6/2015

Like in many languages, compounding is one of the usual word-formation patterns also in the Latin language. By compounding, the Latin vocabulary is supplemented with lexical units that reflect multiform concepts of complex meaning, explicit translation of which in another language, including Latvian, is often a complicated challenge.The paper discusses the dominant models of Latin compounding and the translation possibilities of various Latin compounds in Latvian. Latin compounds in verbal and in nominal system have developed from different syntactic word groups and various stems of nouns, adjectives, numerals, and verbs, including also adverbs and prepositions which are used for their formation. By combining two or more parts in one lexical unit, the components can either change their form or remain unchanged and the meaning can be modified or it remains as it was.A great part of Latin compounds, including denotations of specific notions, belongs to common vocabulary devoid of any stylistic or emotional expressivity. However, compounding has also a significant stylistic potential manifesting itself convincingly in various texts of different literary genres. The translation of these units is the most difficult challenge.

More...

Download

Ģenitīveņi: gramatiskās sistēmas elementu pāreja leksiskajā sistēmā jeb leksikalizēšanās

Author(s): Anita Butāne / Language(s): Latvian Issue: 6/2015

The descriptions of the so-called ģenitīveņi (indeclinable genitive forms of nouns, used attributively) in Latvian usually stress their productive character, thus proving their permanent importance as an object of research. In this article they are analyzed within the framework of the theory of lexicalization. The key aim is to distinguish the possible ways of lexicalization regarding the indeclinable genitives.From a structural point of view, Latvian linguists traditionally distinguish two types of indeclinable genitives: 1) compound indeclinable genitives, and 2) prepositional or prefixal indeclinable genitives. Taking into account their form, semantics and syntactical functions, they are usually described as having the features of various parts of speech, but are defined as nouns that only have the singular or plural genitive case form, are normally used as attributes.Considering lexicalization as a process of change whereby a language unit – either syntactic construction or word formation – acquires new formal, structural and semantic properties, and as the transition of grammatical elements into the lexical system, we might say that the indeclinable genitives present two types of lexicalization: 1) a case form or word formation > lexeme, e.g. when the archaic Latvian noun vāci (‘the Germans’) took the genitive form vācu (‘of the Germans’) and later became an indeclinable genitive vācu (‘German’); 2) word combination or syntactic construction > lexeme, e.g., the genitive collocation mūsu dienu (‘of our days’) became an indeclinable genitive compound mūsdienu (‘contemporary’).

More...

Download

Darbības vārda varēt modālā semantika un funkcijas latviešu valodā

Author(s): Anete Daina / Language(s): Latvian Issue: 6/2015

In many languages of Europe and others there are many studies in modality and modal verbs, especially in Germanic languages, while in Latvian linguistics the extent of research in this field is not so large. Because of that, the purpose of this work is to explore one of the modal verbs in Latvian – varēt ‘can’, its’ modal semantics and functions. To accomplish the purpose of the work, the author reviewed theoretical literature about modality and modal verbs in Latvian, English and other languages, and excerpted and analyzed examples of the use of modal verb varēt ‘can’, so to determine modal verb’s varēt ‘can’ morphological features and modal meanings. In the result it was found out that modal verb varēt ‘can’ in Latvian can be used in all grammatical categories of verb, with or without infinitive, and in epistemic and deontic modality, expressing such modal meanings like assumption, irony, objective necessity, suggestion, permission, instruction, encouragement, demand. Modal meanings are affected by context and verb mood.

More...

Download

Priedēkļverbi mūsdienu latviešu valodas tekstos un to grupēšanas iespējas

Author(s): Daiga Deksne / Language(s): Latvian Issue: 6/2015

Prefixation of verbs in Latvian is one of the most productive techniques for the verb formation. The number of valid prefixes that can be added to different motivation verbs in derivation process varies significantly. In Latvian linguistics the most generic prefix classification system is proposed by Emīlija Soida (2009). Prefixes are grouped by how they expressed the completeness of the action; how they express fulfillment level of the action; and how they describe the process. So far there are no studies of the common characteristics of the motivation verbs for which the same set of the prefixes can be added. For the purpose of such research, the verbs extracted from the European Union legal text corpus and from the New Latvian English dictionary (2007) are used. After extraction prefixes are separated and motivation verbs are grouped by prefix sets. The purpose of this study is to determine the common features of motivation verbs for which the same set of prefixes can be added and to propose a general verb classification system based on this knowledge.

More...

Download

Ciešamā kārta latviešu un angļu zinātniskajā rakstu valodā

Author(s): Vita Kalnberzina / Language(s): Latvian Issue: 6/2015

The article compares the use of the passive voice in the English and Latvian academic writing, first examining the diachronic research and the traditions of the use of the passive voice in Latvian and English, then the frequency of the use of some of the passive voice verbs in Latvian and English Google data base and finally examining the instructions of Latvian, British, Canadian and the US University guidelines for the writers of dissertations. The findings suggest that the attitude of the English academic writing instructors and editors is in striking contrast to Latvian academic writing instructors, which has resulted in the differences in the frequencies of the use of the verbs in the passive voice in the edited articles data base (Google Scholar). The reasons for the difference would need further research.

More...

Download

Constituents of Spatial Adpositional Phrases in Latvian and Mandarin Chinese

Author(s): Marija Nikolajeva / Language(s): Latvian Issue: 6/2015

The article focuses on the syntactic and semantic aspects of the constituents of spatial adpositional phrases in Latvian and Mandarin Chinese. While studying the elements of the spatial expressions, one can discover that the properties of prepositions and nouns differ in Latvian and Mandarin Chinese. As a result, the constructions formed for describing spatial relations in the two languages are typologically different. The Chinese localizers are equivalent to Latvian prepositions, adverbs and nouns. The position of the spatial phrase affects the grammatical meaning of the sentence in Chinese and the pragmatic meaning in Latvian. In Latvian, prepositions and cases are polysemous – each may perform several semantic roles. Chinese prepositions are not polysemous in this sense.

More...

Download

Trešās personas vietniekvārdi „Līdzsvarotajā mūsdienu latviešu valodas korpusā” un zinātniskajos rakstos angļu valodā

Author(s): Zigrida Vinčela / Language(s): Latvian Issue: 6/2015

The use of the third person pronouns differs in the Latvian and English languages if the pronoun antecedents are the nouns that denote a broad mixed-sex human group. The nouns that refer individually to a broad human group are labelled by Latvian linguists as false generic nouns and they traditionally follow the masculine gender inflection pattern and thus involve the use of generic he (all its forms). In the English language, there is no inflectional category for the gender of nouns, so the gender is signalled by the third person singular pronouns (he, his, him, himself for the male antecedents and she her, hers, herself for the female antecedents). As the English language does not have a sex-neutral third person singular pronoun, the generic he has been traditionally used in the cases when a noun refers individually to a broad mixed-sex group of humans. However, during the last four decades the generic he has been questioned because it is often perceived to be male and thus implies discriminatory pronoun use in favour of males. In order to explore the generic pronoun use in the English language a set of examples was extracted from three open-access journals containing articles in applied linguistics. The analysis of examples shows that the authors have avoided the use of generic he that has been replaced by coordinated pronoun forms (he or she; he/she), the combination s/he or the use they as a generic pronoun. However, the choice of generic he alternatives varies not only across the journals, but also in the articles belonging to one journal. The examples extracted from The Balanced Corpus of Modern Latvian (BCML) show that the nouns used in the contexts in which they denote a broad mixed-sex human group are used in masculine, and consequently serve as antecedents for the pronoun he.

More...

Download

Rezultatīvi latviešu valodā

Author(s): Benita Riaubienė / Language(s): Latvian Issue: 5/2014

Nominal semi-predicative components usually are divided into depictives and resultatives that are formed of verb and adjective, participle or prepositional word-combination. Depictives express a state that holds during the reference time of the event encoded by the main predicate (The sun rose red.). Resultatives express a state which is interpreted as a result of the state encoded by the main predicate (The bridge is painted red.). Nominal predicates also can have a meaning of result. That’s why it is necessary to put forward criterions to separate them from resultative semi-predicative components. Transitive verbs in Latvian are independent, it’s impossible to replace them with a copula. Almost all intransitive verbs and a part of reflexive verbs that are used together with adjectives in the sentence are used as copulas. Verbs which together with adjectives do not have big informative load are used as copulas. Resultative constructions in Latvian usually are made by prefixed verbs, because prefixes indicate that activity has concluded. At the same time unprefixed verbs can be used to express that activity has finished. Verbs of cooking, painting, cleaning, filling and emptying without prefixes are more often used ones.The place of resultative semi-predicative components in sentence is analysed. In active voice sentences their place is not fixed as strongly as it is in passive voice sentences. Resultative semi-predicative components can take a place of attribute.

More...

Download

Vai sintaksē pastāv bināra opozīcija obligāts/neobligāts?

Author(s): Ilze Lokmane / Language(s): Latvian Issue: 5/2014

Obligatory or compulsory components of syntactic structures have been opposed to non-obligatory ones in many syntactic theories. However, this opposition is not obvious at all, as far as many of the illustrating examples in linguistic literature are dubious. Thus the aim of the article is to find out the main factors determining obligatoriness and to decide whether obligatory components can always be distinguished from non-obligatory ones. It is concluded that obligatoriness is closely linked to the lexical meaning of the predicate, or, more precisely, some specific semantic elements in it, as well as syntactic structure of the sentence as a whole. However, the most important factor is the situation of speech and the communicative goal of the speaker who decides which participant roles should be syntactically realized in the utterance. Another conclusion is that there exists gradual transition from obligatory to non-obligatory rather than a clear-cut distinction between the two poles.

More...