Topics in Focus Position in Hungarian, Z Gécseg

Tags: University of Tartu, Oxford University Press, question, Uralic languages, demonstrative pronouns, University of Helsinki, University of Helsinki Language, word order, information structure, Cambridge University Press, Estonian, discourse markers, Tartu, Tuomas Huumo, Mari language, Gerson Klumpp, Finnish Literature Society, Helle Metslang, Samoyedic languages, verbal agreement, Amsterdam, Finnish texts, John Benjamins Publishing Company, possessive constructions, grammatical functions, interrogative sentence, University of Turku, Pappert M. S., Udmurt language, Tomsk State University, University of Tartu Language, Corpus-based Approaches, English Linguistics, Finnish language, Cognitive Grammar, narrow focus, E�tv�s Lor�d University, Maria Vilkuna Meaning, Syntactic variation, Mari ............................................................................................................................................, Tundra Nenets, Usport
Content: Grammar and Context IV: New Approaches to the Uralic Languages 6-8 June, Tartu [email protected] grammar2013.ut.ee Organizing committee: Liina Lindstrцm, Kristel Uiboaed, Tuomas Huumo, Hanna Jokela, Helle Metslang, Maarja-Liisa Pilvik, Helen Plado, Triin Todesk, Virve Vihman Scientific committee: Mбrta Csepregi (Eцtvцs Lorбnd University), Riho Grьnthal (University of Helsinki), MarjaLiisa Helasvuo (University of Turku), Tuomas Huumo (University of Tartu, University of Turku), Petar Kehayov (University of Tartu), Gerson Klumpp (University of Tartu), Magdolna Kovбcs (University of Helsinki), Johanna Laakso (University of Wien), Liina Lindstrцm (University of Tartu), Helle Metslang (University of Tartu), Karl Pajusalu (University of Tartu), Renate Pajusalu (University of Tartu), Ilona Tragel (University of Tartu), Virve Vihman (University of Tartu) Grammar and Context IV: New Approaches to the Uralic Languages is organised with the help and support of: Institute of Estonian and General Linguistics at the University of Tartu Doctoral School of Linguistics, Philosophy and Semiotics Tartu City Government European Union and European Social Fund 1
Contents Workshop: Previous knowledge and utterance design .................................................................. 9 Workshop: Referential devices in Uralic and adjacent languages .............................................. 10 Workshop: Syntax­Information Structure Interface in Finno-Ugric .......................................... 11 Workshop: The aspects of aspect ................................................................................................ 12 Workshop: Uralic Essive............................................................................................................. 13 Plenary talk: Depictives and Focus in Uralic Languages ............................................................ 15 Casper de Groot Plenary talk: The Temporality of Grammar and its Coordination with the Body....................... 16 Leelo Keevallik Plenary talk: Four Functions of Komi Possessive Suffixes ........................................................ 17 Gerson Klumpp Plenary talk: On the so-called so-called Finnish Passive ............................................................ 19 Maria Vilkuna Meaning and usage of the verb kьles in Mari language .............................................................. 21 Elina Ahola Mood and Modality in Hungarian ............................................................................................... 22 Gбbor Alberti, Judit Kleiber Syntactic flexibility and prosody in marking given/new in Finnish............................................ 24 Anja Arnhold, Caroline Fйry Word order versus stress as means of object focusing in Udmurt .............................................. 26 Erika Asztalos Non-specific indefinites and Hungarian verb paradigms ............................................................ 27 Andrбs Bбrбny Context (in)dependence of Hungarian discourse markers. The case of aztбn............................. 30 Viola Baumann, Йva Dцmцtцr Syntactic variation in Mansi ditransitive constructions .............................................................. 32 Bernadett Bнrу, Katalin Sipcz Evidentiality in Khanty dialects (Evidentiaali hantin murteissa) ............................................... 34 Mбrta Csepregi Structure, constructions and the locus of language contact The case of Old Finnish necessives 35 Merlijn De Smit Differential object­verb agreement and information structure ................................................... 36 Katalin Й.Kiss Possessive suffixes as referential devices in Udmurt .................................................................. 38 Svetlana Edygarova 2
Demonstratives and actions: What is the Finnish demonstrative tuo doing in requests?............ 40 Marja Etelдmдki Demoting the subject: A functional distribution of impersonal constructions in Udmurt and Mari ............................................................................................................................................. 42 Nikolett F. Gulyбs The Essive Case in Skolt Saami .................................................................................................. 44 Timothy Feist «Essive» in Eastern Khanty......................................................................................................... 46 Andrey Filchenko Topics in Focus Position in Hungarian ....................................................................................... 49 Zsuzsanna Gйcseg Responding to polar interrogatives in Finnish ............................................................................ 51 Auli Hakulinen, Marja-Leena Sorjonen Estonian illative as a category ..................................................................................................... 53 Petra Hebedovб Functions of detached NPs in Old Finnish texts ......................................................................... 54 Marja-Liisa Helasvuo, Nobufumi Inaba Multiple directives in everyday Estonian interaction.................................................................. 55 Tiit Hennoste, Andriela Rддbis Centering on overt and zero pronouns in Estonian ..................................................................... 57 Helen Hint OV order in Finnish without movement ..................................................................................... 58 Anders Holmberg On the grammaticalization of some converb constructions in Udmurt, Tatar literary language and Mishar Tatar (Joidenkin konverbirakenteiden kieliopillistumisesta udmurtissa, tataarin kirjakielessд ja misддritataarissa) ................................................................................................ 60 Laura Horvбth Temporal frames of reference and the function of delimitatives in Finnish ............................... 62 Tuomas Huumo A Dialect-Based Approach to the Finnish Essive Case .............................................................. 63 Emmi Hynцnen Topic elicitation. A new look at Hungarian multiple questions .................................................. 65 Andrбs Imrйnyi Structural typicalities in advanced learner Finnish: Usage-based approach towards varietyspecific multiword sequences...................................................................................................... 66 Ilmari Ivaska 3
Adverbs of certainty in Finnish ................................................................................................... 68 Minna Jaakola Functions of the essive-translative in Tundra Nenets ................................................................. 69 Lotta Jalava Ugric Possessive Suffixes in Non-Possessive Functions ............................................................ 71 Gwen Eva Janda A case of syntactic variation: Double morphology in adpositional structures (in Erzya­Russian bilingual discourse) ..................................................................................................................... 73 Boglбrka Janurik Finnish demonstrative tдmд from L2 learner's perspective ........................................................ 75 Hanna Jokela The development of complex grams in Estonian ........................................................................ 76 Anni Jьrine Impersonal and generic reference: A cross-linguistic look at Finnish and English narratives... 78 Elsi Kaiser On the interpretation of generic `zero subjects' in Finnish: Syntactic and information-structural considerations.............................................................................................................................. 80 Elsi Kaiser Previous Knowledge and Utterance Design: The Case of Estonian jaajaa ................................ 82 Riina Kasterpalu On mood and modality in language death: Evidence from Finnic .............................................. 83 Petar Kehayov Ablative as a marker of benefaction in Finnish........................................................................... 85 Seppo Kittilд On Komi demonstrative pronouns .............................................................................................. 86 Gerson Klumpp, Nikolay Kuznetsov Ei ole kaikki muumit laaksossa ­ a productive idiom construction in Finnish language (Ei ole kaikki muumit laaksossa ­ produktiivinen suomen kielen idiomikonstruktio) ........................... 88 Kristiina Kortelainen The negativity of Hungarian modal adverbs (aligha, bajosan, nehezen).................................... 90 Nуra Kugler Clausal kus- (`where') interrogatives targeting epistemic congruence in second position ......... 91 Kirsi Laanesoo Finnish Optative Construction and its Grammaticalization as a Directive Construction (Suomen kielen optatiivirakenne ja direktiiviksi kieliopillistuminen)........................................................ 92 Yrjц Lauranto 4
Finnish circumstantials and frame relevance. Time, place, and manner as arguments vs. adjuncts (Suomen kielen sirkumstantiaalit ja kehysrelevanssi: aika, paikka ja tapa argumentteina vs. adjunkteina) ................................................................................................... 93 Jaakko Leino Borrowed function and discourse words in Komi....................................................................... 95 Marja Leinonen Variation in necessive constructions in Estonian dialects........................................................... 96 Liina Lindstrцm, Kristel Uiboaed L2 production of native Finnish speakers, in the light of the negative (and positive) transfer (target language is Hungarian) .................................................................................................... 98 Szilvia Magyar Essive in contemporary Votic ..................................................................................................... 99 Elena Markus, Fedor Rozhanskiy On aspect's place in the Estonian differential argument marking system................................. 100 Helena Metslang Do I guess or ask? Epistemic modality as a source of interrogativity....................................... 102 Helle Metslang, Karl Pajusalu, Kьlli Habicht Previous knowledge and the format of repair-initiating polar questions .................................. 104 Krista Mihkels Heterogeneous case distribution in NPs in Estonian: puzzles for a theory of case ................... 106 Merilin Miljan Making referents accessible in multi-party interaction ............................................................. 108 Chiara Monzoni, Ritva Laury The development of non-canonical argument structures in Finnish ......................................... 109 Maximilian Murmann Question words with predicative function in Tundra Nenets.................................................... 111 Nikolett Mus Intra-textual referring: referential procedures and linguistic devices ­ a contrastive pilot study on Estonian and Finnish news texts .......................................................................................... 113 Dirk Mьller The role of anaphora in topic coding in Northern Mansi .......................................................... 114 Szilvia Nйmeth Usage and development of LEE(NE)- verbs in Finnic.............................................................. 116 Miina Norvik Constructions of continuous aspect in Finnish: Contextual comparison................................... 117 Tiina Onikki-Rantajддskц 5
South Estonian demonstratives: changing or disappearing? ..................................................... 119 Renate Pajusalu The choice of the predicate form of Estonian purpose clause................................................... 120 Helen Plado Infinitival constructions with an unmarked object in old Hungarian and Mari. A cross-linguistic study .......................................................................................................................................... 122 Pйter Pomozi, Fanni Karбcsony Completing and interpreting: Functions of the third person pronoun hдn `he, she' in reference to a co-participant in spoken Finnish ............................................................................................ 124 Katri Priiki Other-initiated self-repair in everyday interaction in Vхro language ....................................... 125 Maike-Liis Rebane, Tiit Hennoste, Sulev Iva Spatial opposition, visual accessibility and contrast ­ an experiment with Estonian demonstratives........................................................................................................................... 127 Maria Reile Some equivalents of the Estonian subordinate clause conditional in Hungarian (Eesti kхrvallause konditsionaali ungari vastetest).............................................................................. 129 Tiina Rььtmaa Expressing temporary state of being in Mari ­ or how to manage without an essive case ....... 131 Sirkka Saarinen Directives in teacher's guide books (Direktiivit opettajan opaskirjoissa) ................................. 133 Minna Sддskilahti Verb-framing vs. satellite-framing clauses and Information Structure in Estonian .................. 134 Heete Sahkai From doubt to assumption: The Finnish verb epдillд as an example of a constructional meaning change ....................................................................................................................................... 136 Jutta Salminen Intonation of different syntactic structures in relation to different semantic contexts .............. 138 Nele Salveste From a proper name to an appellative. Appellativization of given names in Finnish slang compounds (Proprista appellatiiviksi. Etunimien appellatiivistuminen suomen slangiyhdyssanoissa) ................................................................................................................. 140 Maria Sarhemaa Attribute-Noun Agreement in the Uralic Languages: Evidence from Udmurt and Baltic Finnic Languages ................................................................................................................................. 141 Olga Schitz Acquisition of Spanish L2: Estonian referential devices and Spanish articles.......................... 142 Maarja Sepp 6
Different conceptualizations in expressing "X becomes Y" in Finnish .................................... 143 Mari Siiroinen How does the context disambiguate a construction? The case of the affective Finnish abstract motion construction mennд V-mA-An [go V-INF-ILL] `do something unwished' (Miten konteksti disambiguoi konstruktion? Tapaustutkimus suomen kielen abstraktin liikkeen konstruktiosta mennд V-mA-An tehdд jotakin epдtoivottavaa') ...................................................................... 144 Jari Sivonen Depicitive in Nganasan ............................................................................................................. 145 Sбndor Szeverйnyi Deictic and anaphoric grounding by possessive affixes in Nganasan ....................................... 146 Rйka Zayzon The construal of aspect ­ The interplay of boundedness and transitivity in Hungarian ........... 148 Edit Takбcs Deixis and the referential interpretation of conversational narratives....................................... 150 Szilбrd Tбtrai, Krisztina Laczkу The augmentative grade of verbal comparative in Komi .......................................................... 151 Triin Todesk Temporal factors of verbal aspect in Hungarian ....................................................................... 152 Gбbor Tolcsvai-Nagy Functions of Erzya Translative in Nonverbal Predication ........................................................ 154 Rigina Turunen Mixed dialects ­ mixed syntaxes. A study of openly used person forms in Border Karelian dialects (Murteet ja lauseopit tцrmдyskurssilla. Tutkimus rajakarjalaismurteiden avoimista persoonaviittauksista)................................................................................................................ 155 Milla Uusitupa Pick it up: Referential Devices in Estonian Child-Directed Speech ......................................... 156 Virve-Anneli Vihman, Kriste Lauk Something verb-final in Finnish: The case of conditional clauses............................................ 158 Maria Vilkuna Transitivity in Eastern Mansi: An information structural approach.......................................... 160 Susanna Virtanen Exploring the construal of change in Finnish translative expressions ...................................... 162 Eero Voutilainen Intermediate stages in the grammaticalization of Northern Samoyedic objective affixes ........ 164 Melani Wratil Contact information .................................................................................................................. 166 7
WORKSHOPS 8
Workshop: Previous knowledge and utterance design Organizers: Riina Kasterpalu, Krista Mihkels Language: English In interaction, both speakers and recipients do have different access to certain types of information. There is a distinction between the information that persons have rights and obligations to know from their own experience and the information to what they do not have direct access (Labov & Fanshel 1977; Pomerantz 1980). The state of knowledge is a gradient concept ­ it can range from situations in which the speaker may have total access to the information and the co-participant has not, to situations in which both participants have equal access to the information. The state of knowledge plays an important role in the selection of a particular grammatical construction in a particular language. Prior studies have shown that there is a certain correlation between linguistic form and previous knowledge. ,,Epistemic status embraces what is known, how it is known (through what method with what degree of definiteness, certainty, recency, etc.) and persons' rights, responsibilities and obligations to know it" (Sidnell & Stivers 2013: 377, for details see Curl & Drew 2008, Heritage 2012, Hennoste 2012). The theme session aims to explore, how epistemic status is expressed through linguistic design in different kind of social actions such as questions, requests, prohibitions, etc. and in responses to them. The results of contributions can be used as a basis for comparison the connection between utterance design and previous knowledge in Finno-Ugric and other languages. Questions to be discussed in this panel include, but not limited to, in what way grammatical realization is contingent upon previous knowledge of · epistemic asymmetries of interactants · epistemic domains of co-participants · social practices in a given society · inquired and/or acquired information. The theme session focuses on contributions using conversation analysis but also welcomes presentations of empirical language studies with different theoretical perspectives and methods. References Curl, T. & Drew, P. (2008) `Contingency and Action: A Comparison of Two Forms of Requesting'. Research on Language and Social Interaction, 41(2), 129­53. Hennoste, T. (2012). Kьsimuse vorm, episteemiline staatus ja episteemiline hoiak. Keel ja Kirjandus, 8­9, 674­695. Heritage, J. (2012). `Epistemics in Action: Action Formation and Territories of Knowledge.' Research on Language and Social Interaction, 45, 1­29. Labov, W. & Fanshel, D. (1977). Therapeutic discourse: Psychotherapy as conversation. New York: Academic Press. Pomerantz, A. (1980). Telling my side: "Limited access" as a fishing device. Sociological Inquiry, 50, 3­4, 186­198. Sidnell, J. & Stivers, T. (2013). Handbook of Conversation Analysis. Wiley-Blackwell. 9
Workshop: Referential devices in Uralic and adjacent languages Organizers: Gerson Klumpp, Ritva Laury, and Renate Pajusalu Language: English Reference, deixis and anaphora have long been a traditional research area in linguistic pragmatics. More recently, the shift in linguistics towards functional and usage-based paradigms has afforded new perspectives on the study of reference. Some important developments include the following: - the use of authentic conversational data, especially in conversation analysis, and later in interactional linguistics, and in studies in child language acquisition as well, has made it possible to analyse referential expressions in real interaction and has shown the importance of a mutually negotiated referential framework, based on joint attention of the interlocutors (Hanks 1996, Diessel 2006); - studies of the interconnectedness of information structure and referential devices have improved the methodological basis of the contrastive study of pronominal reference in different languages (Gundel et al. 2010) - the development of experimental linguistics has given new statistically grounded data on the use of demonstratives and other referential expressions, for example new ideas on the old debate on spatiality/non-spatiality of demonstratives (Coventry et al. 2008, Piwek et al.2008); - cognitive linguistics has developed semantically motivated interpretations of deixis and epistemic grounding (Langacker 2002) - typological studies have yielded new data on the systems of demonstratives and their use in different language families (Kibrik 2011) The workshop brings together scholars doing research on demonstratives, personal pronouns, determiners and other linguistic means of reference in Uralic languages and their contact languages. Topics to be discussed in this workshop include, but are not limited to: - Demonstrative and personal pronouns in conversation - Experimental approaches to demonstrative and personal pronouns - Reference tracking in narratives - Reference, (non-linguistic) action and joint attention - Acquisition (first and second language) of referential devices - Determiners and definiteness - Discontinuous pronominal phrases References: Coventry, Kenny R., Bernice Valdйs, Alejandro Castillo, Pedro Guijarro-Fuentes 2008. Language within your reach: Near-far perceptual space and spatial demonstratives. ­ Cognition, Volume 108, Issue 3, pp. 889-895. Diessel, Holger 2006. Demonstratives, joint attention, and the emergence of grammar. Cognitive Linguistics 17: 463- 489. Gundel, Jeanette, Mamadou Bassene, Bryan Gordon, Linda Humnick, Amel Khalfaoui 2010. Testing predictions of the givenness hierarchy framework: a crosslinguistic investigation". Journal of Pragmatics 42, 1770­1785. Hanks, William F. 1996. Language and Communicative Practices. Chicago: Westview Press. Kibrik, Andrej A. 2011. Refrence in Discourse. Oxford Studies in Typology and Linguistic Theory. OUP. Langacker, Ronald W.2002: Deixis and subjectivity. In: Brisard, Frank (ed.): Grounding. The Epistemic Footing of Deixis and Reference. Berlin, New York: Mouton. 1­28. Piwek, Paul, Robbert-Jan Beun, Anita Cremers 2008. 'Proximal' and 'distal' in language and cognition: Evidence from deictic demonstratives in Dutch. Journal of Pragmatics, vol 40, issue 4, pp. 694­718. 10
Workshop: Syntax­Information Structure Interface in Finno-Ugric Organizers: Prof Balбzs Surбnyi, Orsolya Tбnczos Language: English Despite the ever-growing volume of the work on the syntax of Finno-Ugric languages, many of even the most elementary empirical and analytic issues remain barely understood. Syntactic research on Finno-Ugric carried out in formal models of grammar is especially fragmented, with relatively few dedicated research centers reaching critical mass, having only sparse interactions with contemporary functionalist approaches, and with sporadic dedicated conferences and other events to enhance and maintain coherence and mutual awareness of latest results and new research directions. In spite of this, the research community has been producing a wealth of outstanding results inspiring much fruitful research not only within the Finno-Ugric family, but well beyond. This workshop session is aimed as a forum to present and jointly discuss current issues in a particularly fruitful field of investigation: the interface of syntax with information structure. In order to facilitate lively interactions at the workshop involving both invited speakers and other participants of the Grammar and Context 2013 conference, the workshop concentrates on three fundamental areas of syntax which information structure permeates: (1) sentence structure and word order, including the complex interrelations between information structure and word order as mediated through sentence prosody, (2) the morphosyntax of agreement, and (3) the syntactic expression of argument structure. 11
Workshop: The aspects of aspect Organizers: Tuomas Huumo, Gбbor Tolcsvai Nagy Language: English The theme session is planned to discuss the questions of aspect in Uralic (Finno-Ugric) languages. The theoretical framework of the theme session is functional cognitive linguistics. The main questions are: how aspect is construed (with semantic, syntactic and pragmatic factors, their functions and interplay), what are the main functional variables within one language, what are the similarities and differences between the Uralic (Finno-Ugric) languages. What new, unknown phenomena are found out? Are there joint characteristics in construing and expressing aspect, or do the genetically related languages use diverse means for aspect? What are the recent results in research work, concerning description, methodology, as well as theoretical framework? Specific topics to be treated in relation to aspect, among others (to be completed by the topics of the participants of the theme session): · the inherent temporality of the verb; · the nominal aspect; · delimitative adverbials; · the complex (composite) verbal structures. 12
Workshop: Uralic Essive Organizer: Casper de Groot Language: English Several languages of the Uralic language family have a marker which is usually referred to as the Essive case, generally defined in the following fashion: Essive carries the meaning of a temporary location or state of being, often equivalent to the English "as a (child)". The Essive seems to be a unique property of Uralic languages, since descriptions of any other language in the world do not mention the existence of an Essive as a case marker. Other languages mostly use adpositions, complementizers, or do not use a marker at all. A first investigation of the Essive in Uralic on the basis of linguistic descriptions reveals that the following six languages have an Essive: Estonian, Finnish, Hungarian, Karelian, Sami, and Votic. Relics of an Essive or alternative use of the Translative in the sense of the Essive are found in: Erzya, Khanty, Mansi, Moksha, Livonian, Selkup, and Vepsian. The investigation, however, also reveals that information about the distribution of the Essive is very scarce. On the basis of a typological linguistic questionnaire an international group of linguists will in 2013 make a start to describe the use of the Essive markers in the Uralic languages. The investigation will take the form as starting point and then describes its function. The questionnaire will be based on descriptions and discussions in the linguistic literature of data from the six languages with an Essive. If an Uralic language under investigation does not have an Essive, the forms/constructions will be described, where other languages may use the Essive. The area in which the Essive will be studied, can be defined in the following fashion: · Referential vs Predicational · Secondary vs Main predication · Essive vs Translative · Depictive vs Manner · Depictive vs Temporal · Time-stable vs Temporally bound · Explicative vs Evaluative identification The theme session is meant for both individuals and participants of the international group. The contributions should consist of papers (i.e. no reports on ongoing research or lists of examples). The topic of the paper should preferably relate to the central theme of the conference, but may also relate to some other theme. 13
PLENARY TALKS 14
Plenary talk: Depictives and Focus in Uralic Languages Casper de Groot, University of Amsterdam Depictive secondary predicates such as raw in Mary ate the fish raw pose interesting challenges in the fields of syntax and semantics as for instance main versus secondary predications, predicates and adjuncts, the role of the controller of the depictive, the stage-level versus the individual-level distinction, and depictives versus oriented adverbs. Also in the field of pragmatics, depictive secondary predicates provide a challenge in that they are claimed to be part of the focus domain of the clause. In my talk I will discuss a number of syntactic and semantic aspects of depictive secondary predicates in Uralic languages in relation to focus theory, such as: · Contexts in which depictives can be found/used · The marking of focus · Depictive markers being focus markers · Copulas in secondary predications · Depictives and focus constructions · Ambiguity between depictive and manner with certain oriented adverbs While discussing these issues, I will take three different perspectives, that of language description, typology and linguistic theory. I will focus on data from as many Uralic languages as possible. The talk relates to the Theme session "Uralic Essive" as part of the conference. 15
Plenary talk: The Temporality of Grammar and its Coordination with the Body Leelo Keevallik, University of Linkцping Grammar has traditionally been researched as an atemporal phenomenon, describable from "a bird's eye view" as a system of hierarchical structures. The current talk subscribes to an alternative approach where grammar is seen as continuously emergent in real time interaction. It argues that at least some syntactic structures, such as the Estonian biclausal correlate pattern, can be explained more adequately when the time constraints on spoken language production are taken into account. Participants hear the pro-form as a projective device, rather than a "replacement" for what is yet to come. Thus, see `this/it' in the following utterance by a telemarketer projects either a noun or another clause before the turn can be considered as complete.
1 M: 2
jah, .h ag- kas sobiks see teile et, `Yeah, but would see suit you' ma helistaksin teile omme tagasi. nдiteks. `if I called you back tomorrow. For example.'
More crucially, the paper shows how the study of co-present interaction can reveal qualitatively new patterns of human communication. The emerging structures indeed make use of grammar but also involve the body that may be deployed simultaneously or interchangeably with the grammar. For example, reference is accomplished multimodally; body movements are quoted similarly to words; embodied demonstrations regularly take the role of obligatory predicates and arguments within the clause.
A Dance teacher: so we have [embodied demonstration]
B Dance teacher: sis tuleb [embodied demonstration] `then comes'
Multimodal syntagmas abound at dance classes, where most of the data for the current talk have been found, but they also occur in many other settings. It is therefore suggested that the structural options of timing an embodied demonstration with incomplete clausal syntax should be systematically described in grammars. Participants in real life encounters orient to these multimodal syntactic-bodily units as complete gestalts that make perfect sense. Embodied demonstrations are treated as yet another type of element in the construction of meaningful action as well as of "correct" grammar.
16
Plenary talk: Four Functions of Komi Possessive Suffixes Gerson Klumpp, University of Tartu
The manifold functions of Komi possessive suffixes have always attracted the attention of resarchers (Budenz 1870, Schlachter 1960, Prokusheva 1992, Nikolaeva 2003, Leinonen 2006, and others). Modern Komi descriptive grammar refers to the category in question with the terms indan-asalan suffiksjas `demonstrative-possessive suffixes' (ЦKKM 2001), resp. opredelennoprityazhatel'naya kategoriya `determinative-possessive category' (SKYa. 1955). This two-part designation is ought to cover separate functions as, e.g., with the suffix of 3rd person singular in the possessive noun phrase juyslцn imys `the name of the river' in (1): in im-ys `its name', the px. may be read as indicating the possessor of the name, but in ju-ys-lцn `of the river', it indicates the possessor as determinate, or given. (In the same function it occurs also in medlцyd im-ys `the most beautiful name' at the end of the example.) The designation may also cover the pragmatic use of the 2nd person singular suffix as, e.g., in the second line of (1) where it follows the caritive converb: gцgцrvo-tцg-yd `without [we] understanding it'. Its function here is not that of indicating the subject on an infinite verb form, but common ground management in the sense that the speaker encodes the information that both interlocutors do not understand the meaning of the Lithuanian river name Sventoji (`the saint') as given. Finally, a function apart from possessor and givenness marking is the focus marking function of the accusative form of 2nd and 3rd person suffixes (-tц, -sц) as, e.g., with mijanly-sц `for us' in in the second line of (1).
(1) Ju-ys-lцn

im-ys
Sventoji, zev esko im.
river-PX3SG-GEN QUOT name-PX3SG S.
very funny name
Mijan-ly-sц,
gцgцrvo-tцg-yd,
esko, a itovec-jas-ly,
we-DAT-PX3SG:ACC understand-CAR-PX2SG funny but Lithuanian-PL-DAT
gaskц i, med-lцyd im-ys.
perhaps also SUP-beautiful name-PX3SG
`The name of the river is Sventoji, a very funny name. Funny for us, who do not understand it, you know, but for Lithuanians it may be even the most beautiful name.' (from Ivan Toropov: "No-o, bia bordajas!", http://www.finnougoria.ru/logos/proza/2228/23934)
The paper traces the development of Komi possessive suffixes from possessor markers to markers of givenness, and further, from markers of given direct objects to focus markers (side tracks included). Special attention is paid to the distinction of focal from topical direct objects in Komi, as well as in other Uralic languages (cf., e.g., Serdolbol'skaya & Toldova 2012, Dalrymple & Nikolaeva 2011).
References: Budenz, Jуzsef 1870. Ugrische Sprachstudien II: Determination des nomens durch affigierten artikel im mordwinischen und in einigen anderen ugrischen sprachen. Pest: Ludwig Aigner. Dalrymple, Mary & Nikolaeva, Irina 2011. Objects and Information Structure. CSIL 131. Cambridge: CUP. Leinonen, Marja 2006. Omistussuhteen ulokkeita: komin possessiivisuffiksin ei-possessiivisista funktiosta. Journal de la Sociйte Finno-Ougrienne 91, 93­114. Nikolaeva, Irina 2003. Possessive affixes as markers of information structuring: Evidence from Uralic. In Suihkonen, P. & Comrie, B. (eds.). International Symposium on Deictic Systems and Quantification in Languages spoken in North and Central Asia. Collection of papers. Izhevsk: Udmurt State University & Leipzig: Max Planck Institute for Evolutionary Anthropology, 130­145.
17
ЦKKM 2001 = Fedyunлva, Galina V. & Ludykova, Valentina M. & Nekrasova, Galina A. & Popova, leonora N. & Tsypanov, Evgenij A. 2000. Цnija komi kyv. Morfologija. Sykytvkar. Prokusheva, Tamara I. 1992. K voprosu ob usloviyah upotrebleniya lichno-prityazhatel'nyh suffiksov v komi yazyke. Permistika 3. Sykytvkar, 110­115. Schlachter, Wolfgang 1960. Studien zum Possessivsuffix des Syrjдnischen. Finnisch-ugrische Studien 3. Berlin. Serdolbol'skaya, Natal'ya V. & Toldova, Svetlana Yu. 2012. Differencirovannoe markirovanie pryamogo dopolneniya v finno-ugorskix jazykax. In: Kuznecova, Ariadna I. (otv. red.). Finno-ugorskie yazyki: fragmenty grammaticheskogo opisaniya. Formal'nyj i funkcional'nyj podxody. Moskva. SKYa. 1955 = Lytkin, Vasilii I. (red.) 1955. Sovremennyj komi yazyk I. Syktyvkar. 18
Plenary talk: On the so-called so-called Finnish Passive Maria Vilkuna, Institute for the Languages of Finland
The title of this presentation echoes the formulation by Shore (1986,1988) of a long-standing controversy concerning the existence of Passive in Finnish, i.e. whether constructions such as (1) and (2) should be called Passive.
(1) Tartossa jдrjestetддn
konferenssi.
Tartu:INE organize:PASS conference
(2) Konferenssi on
aikaisemmin jдrjestetty
kolme kertaa.
conference be:3SG earlier
organize:PASS:PARTIC three times
The phenomenon in question, both as a morphological category and as a clausal construction, has been systematically called Passive in the Finnish grammatical tradition, while this practice has been objected to by linguists with other theoretical backgrounds, for example Shore (op. cit.) and more recently, Blevins (2003), who makes a principled distinction between passive and impersonal constructions and places the Finnish construction among the latter. On the other hand, e.g., Manninen and Nelson (2004) in a generative framework argue for its status a Passive, and it also appears to fulfill the typological criteria of passivehood presented by Siewierska (2005) and Kulikov (2011). Along similar lines, the Comprehensive Finnish Grammar (http://kaino.kotus.fi/visk) opts for the name Impersonal (literally, "unipersonal") Passive. My presentation will discuss the nature and solvability of the Finnish Passive question, a matter that obviously depends on assumptions about language-independent grammatical categories such as Passive and Impersonal. Rather than aiming at a definitive answer, however, I will focus on the extremely wide range of uses of the Finnish Passive inflection. The question could also rephrased as whether there is "a" Passive in Finnish. Investigating uses of the construction in different types of data both from dialectal and Standard Finnish, I will evaluate the suggestion that it should be seen as a continuum between two prototypes (Shore 1986, 1988) and consider where the line should be drawn between two passive constructions, the simple passive and the be passive (cf. Helasvuo 2006). I will also relate the question to what has recently been said Estonian linguists regarding the passive and impersonal in Estonian (e.g. Torn-Leesik 2009, Torn-Leesik & Vihmann 2010).
References: Blevins, James P. 2003. Passives and impersonals. Journal of Linguistics 39. 473­520. Helasvuo, Marja-Liisa. 2006. Passive ­ personal or impersonal? A Finnish perspective. In Marja-Liisa Helasvuo & Lyle Campbell (eds.), Grammar from the human perspective, 234­255. Amsterdam & Philadelphia: John Benjamins. Kulikov, Leonid. 2011. Voice typology. In Jae Jung Song (ed.) The Oxford Handbook of Linguistic Typology, 368­ 398. Oxford: Oxford University Press. Siewierska, Anna. 2005. Passive constructions. The World Atlas of Language Structures Online, http://wals.info/chapter/107. Shore, Susanna. 1988. On the so-called Finnish passive. Word 39. 151­176. Shore, Susanna. 1986. Onko suomessa passiivia? [Is there a Passive in Finnish?] Helsinki: Suomalaisen Kirjallisuuden Seura. Torn-Leesik, Reeli. 2009. The voice system of Estonian. Sprachtypologie und Universalienforschung 62. 72­90. Torn-Leesik, Reeli & Vihman, Virve-Anneli. 2010. The Uses of Impersonals in Spoken Estonian. SKY Journal of Linguistics 23. 301­343.
19
ABSTRACTS 20
Meaning and usage of the verb kьles in Mari language Elina Ahola, University of Turku Language of presentation: English The topic of my presentation is a verb form (kьles) in the language of Mari. I will analyze the meaning and the usage of this verb form from semantic and contextual viewpoint. The verb kьles has a primary meaning of `to need' and `must', and those are also the translation equivalents provided by the dictionaries and most of previous studies on the matter (see Koskinen 1985). This verb expressing necessity has equivalents in several Finno-Ugric languages and it is reconstructed into FU as *kelke (UEW, Koskinen 1993). The verb kьles in Mari language can in some contexts be also used as an expression will, and could therefore also be translated as `to want'.
(1) Mlanem kofe kьles.
I.DAT coffee need.SG.3.P. / I.DAT coffee want.SG.3.P.
I need coffee.
/ I want coffee.
The phenomenon of using the expressions of necessity for expressing will is not that rare. Similar usage can be seen in other (Finno-Ugric) languages both with the equivalents of kьles but also with other verb forms (for examples in Finnish see Kangasniemi 1992 and in Northern Sami see Koskinen 1998). As seen in example 1 the same verb form might have different meaning in different contexts, and the actual meaning of the expressions is constructed through the context. In my presentation I will focus on the meaning of the verb kьles in different contexts. In some contexts the meaning of the verb is quite unambiguous, in other cases only the context will reveal the actual meaning of an expression including the verb form in question. In some cases it is not at all that relevant to distinguish the explicit meaning (for instance between `to want' and `to need'). It is also important to raise the question of inner and outer necessity. Koskinen (1993) considers all the expressions with the verb kьles as some form of necessity, whereas I would like to raise into the focus the relation of necessity and will. I will base my study on different types of texts, either written in Mari or translated into Mari from another language. I will look up all the occurrences of the verb form kьles in these texts, and study how the verb form is used and what is the meaning of each clause including the form kьles. I will also study, how the meaning is constructed within the context, so my aim is to point which components in the context guide towards the intended interpretation. References: ALHONIEMI, ALHO 1985: Marin kielioppi. Apuneuvoja suomalais-ugrilaisten kielten opintoja varten X. Hifsmittel fьr das Studium der finisch-ugrischen Sprachen X. Suomalais-Ugrilainen Seura, Helsinki. KANGASNIEMI, HEIKKI 1992: Modal Expressions in Finnish. Studia Fennica Linguistica 2. Suomalaisen Kirjallisuuden Seura, Helsinki. KOSKINEN, ARJA 1985: Modaalisuuden ilmaiseminen tseremississд. Sivulaudaturtutkielma. Suomalais-ugrilainen kielentutkimus, suomalaisen ja yleisen kielitieteen laitos. Turun ylipisto, Turku. ­­­­­­­­ 1993: Suomalais-ugrilaisen nesessiiviverbin tiet. ­ Sirkka Saarinen ­ Jorma Luutonen ­ Eeva Herrala (toim.), Systeemi ja poikkeama. Juhlakirja Alho Alhoniemen 60-vuotispдivдksi 14.5.1993, s. 55­75. Turun yliopiston suomalaisen ja yleisen kielitieteen laitoksen julkaisuja 42. Turun yliopisto, Turku. ­­­­­­­­­ 1998: Toiminnan vдlttдmдttцmyys ja mahdollisuus. Pohjoissaamen modaalisten ilmausten semantiikkaa ja syntaksia. Suomalais-Ugrilainen Seura, Helsinki. UEW= Rйdei, Kбroly (ed.)1988: Uralisches etymologisches Wцrterbuch I­III. Wiesbaden.
21
Mood and Modality in Hungarian Gбbor Alberti1, University of Pйcs Judit Kleiber, University of Pйcs Language of presentation: English This talk is devoted to a thorough formal semantic and pragmatic analysis of the Modality + Tense + Mood marker combinations of Hungarian verbs. Eszes (2005) sketched results in the area, relying on Kratzer's (1991) possible-worlds semantics. We intend to use another formal semantics, a DRT-based (Kamp et al. 2011) representationalist dynamic discourse-semantic framework (Alberti 2009, Alberti­Kleiber 2012), because we accept Pollard's (2007:33) criticism on the mainstream Kripke/Montagueinspired possible-worlds semantics: "the idea of taking worlds as a primitive of semantic theory is a serious misstep". We follow Pollard (p34) in assuming that "worlds are constructed from propositions ..., and not the other way around", but we intend to work out this approach in a DRT-based framework in order to account for the phenomena concerning referent accessibility which ensure the legitimation of DRT. We claim, however, that our system is devoid of DRT's "extra level" problem due to its novel "architecture": (1) A COMPARISON BETWEEN THE ARCHITECTRUE OF MONTAGUE GRAMMARе / DRTв / OUR APPROACHж:
repr'tion of linguistic form intermediate (logical) representation (to be eliminated?)
repr'tion of linguistic form discourse representation
repr'tion of linguistic form discourse repr'tion repr'tion of human mind
We use (gigantic) DRSs as lifelong representations of interpreters' information states. Embedded DRS boxes (consisting of propositions) serve as ilks that play the role of possible words but they are finite. They are responsible for expressing interpreters' beliefs, desires and intentions (partly concerning to BDI's of each other). In our approach, hence, an interpreter's information state is a labeled tree system of "worldlets" (finite information pools), which is practically the description of his/her brain or "internal world". As internal worlds are considered to be parts of the entire world model, our system requires no extra level of representation. This radically new approach to intensionality can be formally defined by simultaneous recursion as 1 We are grateful to SROP-4.2.1.B-10/2/KONV/2010/ KONV-2010-0002 (Developing Competitiveness of Universities in the Southern Transdanubian Region) for their contribution to the first author's costs at the Tartu Grammar and Context Conference, and for supporting our Research Team ВeALIS in 2012. In 2013 ВeALIS is supported by SROP-4.2.2.C-11/1/KONV-2012-0005 (Well-Being in the Information Society); the final version of this paper is due to this project. 22
an epistemic multi-agent system В where "agents" get information about the world around them. The words below, which typically have at least two modal meanings, serve as an illustration. The formulae in (2a-b), for instance, show labels of two "worldlets". The one reading (deontic) of hazamehetett (2a) pertains to a person x's "lack of intension", i.e. permission, for someone to go home, and speaker s's "maximal belief", i.e. definite knowledge, that the person in question did go home (allowed to go home). The other reading (epistemic) of hazamehetett (2b) can be characterized as follows: speaker s guesses that a person went home; but what s confidently knows is that certain "premises" typically resulting in someone's having gone home are met.
(2) Modality ® ЇMood
Ж + Past hazaЧmegy `home-go'
-nA 'would' + Past
-hAt `can/may' kell `must'
a. hazaЧmeЧhetЧett бx,INT,0с бs,BELMAX,+с
b. hazaЧmeЧhetЧett бs,BELmed,+с бs,BELpremMAX,+с
e. hazaЧmeЧhetЧett volЧna бx,INT,0с бs,BELMAX,-с
f. hazaЧmeЧhetЧett volЧna бs,DESgreat,+с бs,BELMAX,-с
c. haza kellЧett d. haza kellЧett g. haza kellЧett h. haza kellЧett
menЧni(e)
menЧniЧe
volЧna menЧni(e) volЧna menЧniЧe
бx,INT,+с бs,BELMAX,+с
бs,BELalmostAMX,+с бx,INT,+с бs,BELpremMAX,+с бs,BELMAX,-с
бs,DESalmostMAX,+с бs,BELMAX,-с
References: Alberti 2009: ВeALIS: An Interpretation System which is Reciprocal and Lifelong. Workshop `Focus on Discourse and Context-Dependence' (16.09.2009, 13.30-14.30 UvA, Amsterdam Center for Language and Communication), http://www.hum.uva.nl/aclc/events.cfm/C2B8E596-1321-B0BE-6825998CFA642DB2, http://lingua.btk.pte.hu/realispapers. Alberti and Kleiber 2012: Where are Possible Worlds? Acta Linguistica Hungarica 59 (1-2), 3-26. Eszes 2005: World Times and Possibilities ..., LingDok 4., JATEPress, Szeged, Hungary. Kamp et al. 2011: Discourse Representation Theory, Phil. Logic 15, 125­394. Kratzer 1991: Modality, in Semantics:..., de Gruyter, Berlin, 639­650. Pollard 2007: Hyperintensions, ESSLLI 2007, http://www.cs.tcd.ie/esslli2007.
23
Syntactic flexibility and prosody in marking given/new in Finnish Anja Arnhold, Goethe-University Frankfurt am Main, University of Alberta Caroline Fйry, Goethe-University Frankfurt am Main Language of presentation: English It is well-known that information structure influences both word order and prosody in Finnish (e.g., Vilkuna 1995, Vainio & Jдrvikivi 2007). In this presentation, the interplay between syntax and prosody in marking information structure is investigated based on data from two production experiments where the participants either used a fixed SVO word order (exp 1) or were free to choose the most suitable word order (exp 2). The results indicate that prosodic markers of constituents' given/new status were used more pervasively in the first experiment, i.e. when the experimental design precluded the use of syntactic marking. In Experiment 1, 17 native Finnish speakers realised eight three-word SVO sentences as answers to different questions. The questions either induced answers with all-new sentences or they marked one of the constituents as contextually new (and narrowly focused), while the other two constituents were contextually given. Results from 947 analysed sentences indicated that the difference between given and new constituents was consistently marked prosodically: Participants realised new constituents with a larger pitch range and longer duration than given ones. They also realised pauses more frequently after new constituents and additionally employed voice quality for marking the difference. In Experiment 2, 20 native Finnish speakers observed changing layouts of two or three plastic toy animals placed on the table in front of them. Their task consisted in describing the layouts in their own words so that a second native speaker could reproduce them with a second set of identical toys. With each change in the layouts, an animal was removed from the table and another one was added or moved to a new position. Descriptions thus consistently included constituents with systematically varying given/new status. 220 utterances were included in the analysis. Speakers most frequently described the location of a new or relocated animal (the `locatum') relative to a static given one (the `relatum') (68%). Alternatively, they described it as replacing a given animal that was removed from the table during the same turn (28%). In about 5% of the descriptions, a non-relative expression (e.g. `to the far right') was used. The new/given status had a clear influence on word order: When a new animal was introduced, its referent was placed later in the utterance than the locative expression including a given animal in 84% of the sentences. In contrast, constituents referring to a relocated given animal always preceded locative expressions. In addition to syntactic correlates of information status, new words had overall longer durations than given ones. Likewise, voice quality was influenced by information status. However, the use of pauses was less systematic. Interestingly, new constituents did not show an overall larger pitch range. These results demonstrate that Finnish employs both syntax and prosody to mark information structure and, importantly, that both types of marking interact. While prosodic cues were still employed even when word order was free, they were less pervasive when additional syntactic means of marking information structure were available. References: Vainio, M. & Jдrvikivi, J. 2007. Focus in production: Tonal shape, intensity and word order. J. Acoust. Soc. Am. 121(2), EL55-EL61. 24
Vilkuna, M. 1995. Discourse configurationality in Finnish. In Kiss, K. Й. (ed.), Discourse Configurational Languages. New York/Oxford: Oxford University Press, 244-268. 25
Word order versus stress as means of object focusing in Udmurt Erika Asztalos, Eцtvцs Lorбnd University, Udmurt State University Language of presentation: English Up to now, little has been known about the means of focusing in Udmurt, and even the information that we have about it is, to some extent, contradictory. While Vilkuna (1998) claims that, in Udmurt, ,,focus... does not appear to be positionally restricted" and the ,,Udmurt preverbal position seems to be a neutral and frequent focus... position, but this does not prohibit the placement of... exhaustive foci elsewhere" (Vilkuna 1998: 195), Tбnczos (2011) affirms that the focused element occupies the immediately preverbal position in standard Udmurt and the sentence-final position in its non-standard variant. Additionally, Asztalos (2012) suggests that certain speakers of Udmurt also accept a different positioning of at least the focused object, if it is given phonetical stress (they also consider such sentences to be grammatically correct in which the focused object under stress is located between or before two topics preceding the verb, or between two postverbal complements). The aim of the present research is to examine to what extent stress and positioning may be considered as means of object focusing. I concentrate on the following problematics: 1. Do speakers of Udmurt ­ or at least some of them ­ also produce such sentences (besides considering them grammatical) in which the focused object is not located in the focus positions identified by Tбnczos (2011)? 2. Do positions in the sentence structure exist in which the focused object cannot grammatically appear at all? 3. Is there any correlation between certain morphosyntactic and semantic features of the focused object and its possible positions in the sentence structure? The problematics are examined by means of a fieldwork carried out in consultation with native speakers of Udmurt. References: Asztalos Erika Йva 2012. . - . 4. , - « », 712. Ponarjadov, Vagyim Vasziljevics [, ] 2010. - ( ). , . Tбnczos Orsolya 2011. Szуrendi variбciуk йs lehetsйges okaik az udmurtban. Nyelvtudomбnyi Kцzlemйnyek 107: 218228. Vilkuna, Maria 1998. Word Order in European Uralic. In: Constituent Order in the Languages of Europe (ed. by A. Siewierska). Mouton de Gruyter, BerlinNew York, 173233. 26
Non-specific indefinites and Hungarian verb paradigms Andrбs Bбrбny, University of Cambridge Language of presentation: English
1 Background Recent work on Hungarian verb paradigms (cf. Coppock and Wechsler 2012, Coppock 2012) stresses certain interpretive effects of direct objects (DOs) that correlate with paradigm choice. On Coppock's (2012) account, the objective paradigm (OBJ in (1a)) co-occurs with familiar DOs. When the DO appears with the subjective paradigm (SUBJ in (1b)), however, the referent of the DO does not have to be familiar, it is rather, roughly, non-specific, as in (1b).
(1) a. Pбl-nak olvas-t-uk
vers-й-t.
P.-DAT read-PAST-1PL.OBJ poem-3SG.POSS 'We read Pбl's poem.'
b. Pбl-nak olvas-t-unk
vers-й-t.
P.-DAT read-PAST-1PL.SUBJ poem-3SG.POSS 'We read any poem by Pбl.'
The subjective paradigm has been claimed to co-occur only with non-specific readings of possessed direct objects, for example (cf. a.o. Szabolcsi 1994, Bartos 1999). In addition, the non- specific interpretation of possessives has been argued to depend on possessor extraction. This, if correct, provides a structural argument against Coppock and Wechsler (2012) and Coppock (2012), who maintain that only the lexical features of the object trigger both verb paradigms. 2 Proposal I argue that syntactic considerations are in fact crucial to account for the distribution of the paradigms with possessed direct objects in particular. Coppock's (2012) semantic insights should be extended by structural considerations. Problem 1: Coppock makes wrong predictions w.r.t. the distribution of presuppositions, cf. (2). Coppock treats the possessive suffix as a presupposition trigger, which means that (2b) is wrongly predicted to presuppose (2c), since presuppositions should be constant under negation (in (2a), `>> X' means X is presupposed ).
(2) a. macskбja> `cat' x.y.[>> [y CAT(y) POSS(x, y)]] b. Mari-nak nincs macskб-ja. M.-DAT NOT.is cat-3SG.POSS 'Mari doesn't have a cat.' ¬x[CAT(x) POSS(m, x)] c. Mari has a cat.
(Coppock 2012: 21)
Problem 2: with Szabolcsi (1994) and Bartos (1999), I argue that structure, i.e. the choice of possessor, does play a role. The subjective paradigm is only possible with extracted dative possessors, (3a), similar to other non-specific possessive constructions which require are ungrammatical with a local dative or nominative possessor, cf. (3b).
27
(3) a. Pбl-nak / *Pбl vers-й-t
olvas-t-unk.
P.-DAT / P.-NOM poem-3SG.POSS-ACC read-PAST-1PL.INDEF 'We read any poem by Pбl.' (with dative possessor, *with nominative possessor)
b. *Csak Mari-nak macskб-ja
nincs.
only Mari-DAT cat-3SG.POSS NOT.is
intended: 'Only Mari doesn't have a cat.' (cf. Szabolcsi 1994)
These issues can be resolved if the formal feature specification of the direct object depends on the structure in the left periphery of the possessed noun phrase, i.e. the presupposition trigger is not necessarily introduced lexically but is dependent on the syntactic configuration. Local configurations, i.e. nominative and non-extracted dative possessors (in the noun phrase), require a familiar or specific interpretation by triggering the feature [DEF], which is located in D, not on the possessive suffix, contra Coppock and Wechsler (2012), Coppock (2012). (4a) illustrates a local possessor, triggering [DEF], while (4b) shows an extracted possessor, not triggering [DEF].
(4) a. [DP Pбl [D [DEF] [PossP Pбl [NP vers-й-t ]]]]
P.
P. poem-3SG.POSS-ACC
'Pбl's poem'
b. [DP Pбl-nak] ... [D [PossP Pбl [NP vers-й-t ]]]
P.-DAT
P. poem-3SG.POSS-ACC
'a (non-specific) poem by Pбl' or '(non-specific) poems by Pбl'
3 Predictions This approach does not run into problems like (2c) and makes strong predictions. Prediction 1: there should be non-specific direct objects with the subjective paradigm and quantifiers like minden `every', explicitly ruled out by Coppock (2012). Such examples do in fact exist:
(5) Minden problйmб-jб-t
meg-old-unk ...
every problem-3SG.POSS-ACC VM-solve-1PL.INDEF
`We solve all your [polite form] problems.'
(http://bit.ly/RC1jnH, 22/02/13)
Prediction 2: if the subjective paradigm and non-specific readings are closely related, this approach makes strong predictions about where non-specific DOs can appear in the clause. This, in turn, can deepen the understanding of how information structure and Hungarian clause structure are related, in particular in light of recent work on this topic, cf. Gйcseg and Kiefer (2009). Thus, relocating Coppock's (2012) feature [DEF] to D and making it dependent on syntactic structure accounts for a wider range of data and makes testable predictions about the structure of the Hungarian clause.
References: Bartos, Huba. 1999. Morfoszintaxis йs interpretбciу: A magyar inflexiуs jelensйgek szintaktikai hбttere [Morphosyntax and interpretation: the syntactic background of Hungarian inflection]. Doctoral Dissertation, ELTE, Budapest. Coppock, Elizabeth. 2012. A Semantic Solution to the Problem of Hungarian Object Agreement. Ms. Accessed: 22 March, 2013. URL http://eecoppock.info/nls.pdf. Coppock, Elizabeth, and Stephen Wechsler. 2012. The objective conjugation in Hungarian: agreement without phifeatures. Natural Language & Linguistic Theory 30:699­740. Gйcseg, Zsuzsanna, and Ferenc Kiefer. 2009. A new look at information structure in Hungarian. Natural Language & Linguistic Theory 27:583­622. Szabolcsi, Anna. 1994. The Noun Phrase. In The Syntactic Structure of Hungarian, ed. Ferenc Kiefer and Katalin Й. 28
Kiss, volume 27 of Syntax and Semantics, 179­274. New York: Academic Press. 29
Context (in)dependence of Hungarian discourse markers. The case of aztбn Viola Baumann, Pбzmбny Pйter Catholic University Йva Dцmцtцr, Pбzmбny Pйter Catholic University Language of presentation: English The paper addresses the issue of discourse markers (DMs) which have been grammaticalized from the distal demonstrative az ('that'). Discovering the functional spectrum of aztбn (< azutбn < az utбn 'after that') it proposes an interpretation of the multifunctionality of the demonstrative DMs in general. It claims that the meaning of a DM is independent of the context in the sense that it should be taken unaffected by the meaning of the words around it. On the other hand, the meaning depends on the context in the sense that the discourse marker's contribution to the meaning of the utterance is determined by the type of grammatical or pragmatic construction it is part of. The proposal is based on some of the principles of grammaticalization theory and the deixis and DM approaches: 1. The meaning of demonstrative DMs derives from their original demonstrative-deictic function (Diessel 1999, Janssen 2002). 2. Grammaticalization takes place as a complex (semantic, pragmatic and structural) change of a phrase or construction, not of an individual item (Traugott 2003). 3. The seemingly various meanings can be reduced to the few syntactic and pragmatic patterns in which they occur, ie. to the clause, sentence or speech act types (Gдrtner­Gyuris 2012). The corpora of the investigation are Hungarian National Corpus (HNC), Hungarian Historical Corpus (HHC) and CHILDES. The paper also shows the ratio of aztбn in different context types. Some types of the contexts in which aztбn as a DM is contained (in the translations the words in bold are not equivalent to aztбn but have approximately the same pragmatic functions in the English sentences): In declaratives or exclamatives aztбn stands after the topic of the sentence as a kind of intensifier: (1) Hбt mйg ez a Melinda! Ez aztбn megйri a pйnzйt! (1989) ('And this Melinda! She really is worth her weight in gold!') In wh-interrogatives aztбn has the initial position, often immediately before the (first) whpronoun: (2) Aztбn kinek mi kцze arra? (1827) (So what's anyone got to do with that'?) In directives the DM aztбn is compatible with the matrix subjunctive, but not with the imperative, which requires inversion in Hungarian (eg. gyere haza 'come home' vs. haza gyere 'home come' in (3)) and are compatible only with the nonDM aztбn 'after that': (3) Petrбss mйrnцk ъr nehйz szнvvel bъcsъzott leбnyбtуl: ­ Aztбn haza gyere, mikor a dolgodat elvйgezted. (1926) 30
('Petrбss engineer said good bye to his daughter with heavy heart: - Do mind to come home, when you're done.') The paper demonstrates a context-dependent analysis of the uses of aztбn defining contexts as syntactic and pragmatic constructions. References: Diessel, Holger 1999. Demonstratives. Form, Function, and Grammaticalization. John Benjamins Publishing Company, Amsterdam­Philadelphia. Gдrtner, Hans-Martin ­ Gyuris, Beбta 2012. Pragmatic markers in Hungarian: some introductory remarks. Acta Linguistica Hungarica 59 (4), 387­426. Janssen, Theo A. J. M. 2002. Deictic principles of demonstratives, pronominals and tenses. In: Brisard, Frank (ed.): Grounding. The Epistemic Footing of Deixis and Reference. Mouton de Gruyter, Berlin­New York, 151­ 96. Traugott, Elizabeth C. 2003. Constructions in Grammaticalization. In: Joseph, Brian D. ­ Janda, Richard D. (eds.) The Handbook of Historical Linguistics. Blackwell Handbooks in Linguistics. Blackwell, Malden­Oxford­ Melbourne­Berlin: 624­47. 31
Syntactic variation in Mansi ditransitive constructions Bernadett Bнrу, University of Szeged Katalin Sipcz, University of Szeged Language of presentation: English
In linguistic typology, ditransitive constructions have recently become the focus of international research: many international projects have studied this area of syntax (Malchukov et al. 2007; Malchukov, Andrej / Haspelmath, Martin / Comrie, Bernard 2010). Smaller Uralic languages are entirely missing from these studies. It is typical for the Ob-Ugric languages to express the ditransitive situation in two ways. Cf. (1), (2):
Khanty
(1) lц mдnд wikд
jants
he I-DAT coat-NOM sew-PastSg3 'He sewed me a coat.'
(2) lц mдnt vikдt he I-ACC coat-INSTR 'He sewed me a coat.'
jants sew-PastSg3
In constrast with the traditional analysis of these constructions (cf. Munkбcsi 1894: 9, 194, 263 Beke 1905: 186, Lavotha 1953, Liimola 1963: 47, Honti 1969), later researches examine the phenomenon in a more complex way, taking into account pragmatic aspects as well (Rombandeeva 1962, 1973, 1979, Honti 1999: 37). More recent researches have connected this phenomenon to the dative shift and the passivization techniques of the Ob-Ugric languages (cf. Kulonen (1990, 1999, Skribnik 2001). As mentioned above, Mansi belongs to languages having more than one ditransitive constructions. One of them is the indirect object construction, in which the Theme (T) of the ditransitive construction, similarly to the Patient of the monotransitive construction, occurs as either a marked or an unmarked object, while the recipient (R) is encoded with the lative suffix n:
(3) ...taw tinala - s - te
ka - te
he sell-Sg3Obj./Sg3 wife-PxSg3 (T) 'He sold her wife for ,,belly" Vaka.'
on belly
waka - n. Vaka-DAT (R)
The other construction used in Mansi is the secundative object construction, in which the Recipient of the ditransitive construction occurs similar to the Patient of the monotransitive construction (in Mansi it occurs as a marked or an unmarked object), whereas the Theme occurs in a different form, with the -l Instrumental marker:
(4) Pi - mn
manr nam - l
pinil - mn?
boy-PxDu1 (R) what name-INSTR (T) put-Du1Obj./Sg3 'Which name shall we (2) give to our son?'
In our presentation we aim to discuss the rules setting the use of the two types of ditransitive constructions, and to analyze which structures are used when and what the choice depends on. In this respect, Mansi behaves unusually: in contrast to cross-linguistic patterns, it seems that in Mansi any ditransitive verb can appear in both constructions. The other interesting phenomenon is that the group of the Mansi verbs occurring in both ditransitive constructions is rather large.
32
In accordance with recent Ob-Ugric syntactic analysis it can be claimed that passivization and the usage of determinate and indeterminate conjugations are strongly influenced pragmatically. (Nikolaeva 1999, Skribnik 2001) We reckon similarly the choice of the ditransitive constructions is determined by the information structure of the sentence. (Cf.: Virtanen 2012) Although the literature mentions it, the detailed investigation of the ditransitive constructions of Northern Mansi has not been done yet. Bibliography: Beke Цdцn (1905): A vogul hatбrozуk [Adverbs in Vogul]. NyK 35: 71-100, 165-193. Honti Lбszlу (1969): A tбrgy jelцlйse a vogul nyelv tavdai nyelvjбrбsбban [Object marking in the Tavda dialect of the Vogul language]. NyK 71: 113-121. Honti Lбszlу (1999): Ugor alapnyelv: tйves vagy reбlis hipotйzis [Ugric protolanguage: false or real hypothesis?] In: (szerk.: Bakrу-Molnбr-Salбnki-Sipos) Budapest Urбli Mhely I., Ugor Mhely. MTA Nyelvtudomбnyi Intйzet. 19-42. Kulonen, Ulla-Maija (1990): Obinugrilaisten kielten syntaksin yhtдlдisyyksistд. Vir. 94: 49-56. Kulonen, Ulla-Maija (1999): Object Marking in the Ugric Languages. In: (ed. Bakrу-Molnбr-Salбnki-Sipos) Budapest Urбli Mhely I., Ugor Mhely. MTA Nyelvtudomбnyi Intйzet. 63-72. Lavotha Цdцn (1953), A tбrgy jelцlйse a manysiban [Obejct marking in Mansi]. NyK 54: 200-218. Liimola, Matti (1963): Zur historischen Formenlehre des Wogulischen I. Flexion der Nomina. MSFOu 127. Malchukov et al. (2007): A. Malchukov ­ M. Haspelmath ­ B. Comrie, Ditransitive constructions: a typological overview, first draft, September 2007, http://email.eva.mpg.de/~haspelmt/DitransitiveOverview.pdf Malchukov, Andrej / Haspelmath, Martin / Comrie, Bernard (ed.), Studies in Ditransitive Constructions. A Comparative Handbook. De Gruyter, 2010 Munkбcsi Bernбt: A vogul nyelvjбrбsok szуragozбsukban ismertetve [Mansi dialects. Conjugation and declination in focus].Ugor Fьzetek II. sz. (1894) Nikolaeva, Irina (1999): Object agreement, grammatical relations, and information structure. Studies in Language 23: 341-386. Rombandeeva, E. I. (1962): Kauzativnye formy mansijskogo glagola. In: (ed. Cincius, V. I.), Ucenye zapiski. Tom 167. Leningrad. 71-84. Rombandeeva, E. I. (1973): Mansijskij (Vogul'skij) jazyk. Nauka. Moskva. Rombandeeva, E. I. (1979): Sintaksis mansijskogo (vogul'skogo) jazyka. Nauka. Moskva. Skribnik, Elena (2001): Pragmatic Structuring in Northern Mansi, CIFU IX/6: 222-239. Virtanen, Susanna (2011): Ditransitiivien vaihtelua mansin itдmurteissa. [Variation of ditransitive in Eastern-Mansi dialect.] FUD 18: 165-177 33
Evidentiality in Khanty dialects (Evidentiaali hantin murteissa) Mбrta Csepregi, Eцtvцs Lorбnd University Language of presentation: Finnish Evidentiality is marked grammatically in the northern Khanty dialects Obdorsk, Synja, and Kazym (Nikolaeva 1999, Kaksin 2008). Verbs that express evidential modality take the same form as the verbal participle (the derivational morpheme t marks the present participle, and m marks the past participle) with a possessive suffix. When it appears as a predicate, the verb becomes finite. It is used in modern spoken Khanty when the speaker is not an eyewitness but is informed of an event from secondhand sources. Evidentiality modality is marked by the morpheme t in the present and m in the past. In eastern dialects participles appear as predicates only in songs, not in the spoken language. In mythical songs as well as other songs, verbal participles appear with possessive suffixes as finite verbs, with t marking present participles and m marking past participles. These verb forms do not express evidentiality; they simply mark a stylistic difference between literary usage (in songs) and everyday language. The participle as a finite verb appears in even the oldest Khanty folklore texts from the 19th century (Jбszу 1969, 1976), but since there is no information about the spoken language of that time, we cannot know whether these forms were used to express evidentiality. Using Eastern Khanty as a comparison, it can be inferred that the verbalization of participles began in songs, and in Northern Khanty it became used to express the evidential mode. It is unclear why this process did not occur in Eastern Khanty. In any case, it bears consideration that in Eastern Khanty the participle + possessive suffix structure serves more functions than in Northern Khanty. Perhaps avoidance of homonymy is the reason the participle + possessive suffix structure did not develop a predicate function in addition to its adjectival and adverbial functions. In the eastern dialect Surgut, evidentiality is expressed by the use of a postpositional form of the participle. The participle + possessive suffix + postposition 'place' structure appears sentence-finally, in the position of the predicate, and the process of postpositional agglutination has already begun. References: A. Jбszу Anna 1969: Mellйknйvi igenevek verbum finitumkйnt az osztjбk nyelv szigvai nyelvjбrбsбban. Nyelvtudomбnyi Kцzlemйnyek 71: 123­127. A. Jбszу Anna 1976: Megjegyzйsek a participiumbуl alakult verbum finitumok mondattanбhoz az йszaki osztjбkban. Nyelvtudomбnyi Kцzlemйnyek 78: 353­358. Kaksin, A. D. 2008: [. . ] . -: . Nikolaeva, Irina 1999: The Semantics of Northern Khanty Evidentials. Journal de la Sociйtй Finno-Ougrienne 88: 131­59. 34
Structure, constructions and the locus of language contact The case of Old Finnish necessives Merlijn De Smit, University of Turku Language of presentation: English
Old Finnish (1540-1809) syntax is characterized by a deep-going, prolific internal diversity which is often hard to explain in a structural fashion. For example, the co-occurrence of genitive, partitive and locative for dative functions, or the co-occurrence of object-like and subject-like case marking with passives. The aim of this presentation is to discuss the roles of both the "constructional" (specific lexical items, frequency) and the "structural" level (overarching syntactic principles) as well as the role of foreign model patterns with one specific case of such variation, namely the subject of necessive constructions. Necessive constructions in Standard Finnish take a genitive (1) or a nominative (2) subject with the latter restricted to existential constructions:
(1) Minu-n pitдд lдhteд I-GEN must go (2) Varastossa pitдд olla hyvд In the storage must be good-NOM
tuuletus ventilation-NOM
In dialectal Finnish, the dividing line between (1) and (2) runs along agentive vs. non-agentive subjects. The same principle can be shown to be at work in Old Finnish as well, but I will argue that in a number of Old Finnish texts, the case-marking pattern is best explained as accusativelike (-n in singular, -0 in plural). This accusative-like pattern has roots in both "internal" language contact (namely, the accusative-like agent of an Old Finnish participial construction used with verba sentiendi and dicendi), as well as "external" language contact: the Latin accusative and infinitive used with the verb oportet 'must'. Questions raised by this hypothesis, which form the main subject of my presentation, are:
(1) Finnish accusative marking is sensitive to number (sg. -n, pl. -0) and lexical category (pronouns are always marked with -t (St.F.) or -n (dialectal and Old Finnish)). How much of the accusative-like pattern we find with necessive subjects in then due to the application of an overarching syntactic pattern ('accusative'), how much due to the entrenchment or freezing of particular kinds of lexical items with a particular case?
(2) Unlike in Standard Finnish, there is no single "necessive construction" in Old Finnish with a single case-marking pattern. Rather, individual necessive auxiliaries behave differently. The specific accusative-like pattern of the subject of pitдд is thus tied to a specific lexical item. How much of the postulated analogy between the necessive construction and the Old Finnish participial constructions, as well as that of Latin oportet, can be characterized as analogy between syntactic patterns, and how much is due to the association between specific lexical items?
35
Differential object­verb agreement and information structure
Katalin Й.Kiss, Research Institute for Linguistics of the Hungarian Academy, Pazmany P. University Language of presentation: English
The talk will provide evidence that verbal agreement with definite objects arose in Hungarian as the marker of the topic status of the object. This hypothesis was put forth on the basis of indirect evidence by Marcantonio (1985). It has also been supported by Uralic parallels; differential object-verb agreement in Hanti, Mansy, and the Samoyedic languages has been shown to encode the secondary topic role of the object (Nikolaeva 1999, 2001, Dalrymple & Nikolaeva 2011, Skribnik 2001). A non-agreeing object is interpreted as focus. The evidence to be discussed is based on the gaps in Hungarian object-verb agreeement: whereas 3rd person personal pronominal objects do, 1st and 2nd person pronominal objects do not trigger verbal agreement. Cf.
(1) a. Pйter lбt-ja-Ш
t .
Peter see-OAgr-3SG him 'Peter sees him/them.'
b. Pйter lбt- Ш tйged/titeket/engem/minket. Peter see-3SG you / youPL / me /us 'Peter sees you/me/us.'
This fact has led some analysts, most recently Coppock & Wechsler (2012), to conclude that 1st and 2nd person pronouns, as opposed to 3rd person pronouns, are indefinite. However, as will be shown, the verb does agree with a 2nd or 1st person object in the presence of a 1st person subject:
(2) a. Йn lбt-l-ak
/ *lбtok tйged.
I see-OAgr-SAgr / see-SAgr you 'I see you.'
b. Йn minket is beveszem
/*beveszek a nйvsorba.
I us
also include-OAgr.SAgr / include-SAgr the list-in
'I include us, too, in the list.'
The gaps in object-verb agreement have been derived by Й. Kiss (2005) from the Inverse Agreement Constraint, a formal, semantically unmotivated constraint observed by Comrie (1980) in Chukchee, Koryak and Kamchadal, forbidding object-verb agreement if the object is higher in the following ,,animacy" hierarchy than the subject:
(3) Animacy hierarchy: 1SG > 1PL > 2SG > 2PL > 3SG > 3PL
This talk will claim that the Inverse Agreement Constraint is, in fact, a constraint on information structure. What it requires is that a secondary topic, i.e., the object in an SOV sentence, be less animate, less topic-worthy, than the primary topic. An object more topic-worthy than the primary subject-topic can only be construed as a focus. The interpretation of the Inverse Agreement Constraint as a constraint on information structure will be supported by evidence from Koryak and Kamchadal. Evidence of various kind shows that Proto-Hungarian was a strict SOV language with a morphologically unmarked object (Й. Kiss 2013). If we assume that it displayed the same interplay of grammatical functions and information structure that has been preserved in the ObUgric languages, i.e., if the subject also had the role of primary topic, and the object had the role of secondary topic or focus, depending on whether or not it elicited agreement on the verb, then Proto-Hungarian must have observed the Inverse Agreement Constraint, and the apparently 36
idiosyncratic gaps in object-verb agreement are the fossilized consequences of that discoursemotivated constraint. The proposal also explains the lack of verbal agreement with 1st and 2nd-person objects in the Samoyedic languages. 37
Possessive suffixes as referential devices in Udmurt Svetlana Edygarova, University of Helsinki Language of presentation: English Possessive suffixes (PX) have been a popular topic in Uralic studies. Traditionally, the possessive and non-possessive functions of PX are distinguished (Kokla 1963: 187-8; Feoktistov 1963: 116-165; Majtinskaja 1974: 268-271; Kim 1986: 86-121, etc.). In particular, identification, definition and/or determination are defined as secondary or non-possessive functions of PX. Furthermore, Nikolaeva (2003) considers the non-possessive functions of PX to be that of discourse-specific markers. Recent investigation of PX in Udmurt proposes a new classification of PX: (a) they can be treated as possessee specifiers and (b) possessor/referent specifiers. This classification is based on grammatical features specific to each function. In particular, when PX are used to specify the possessee, they can be considered specifiers which have an optional character. Their emergence in this case is caused not by grammatical (except possessive constructions with genitive and ablative) but by discursive settings. For example, in (1), the speaker knows that the book belongs to the interlocutor and in the question he or she expresses this knowledge. If one were to drop the PX in (1), the sentence would still be grammatical. (1) ma ari kiga-jed? what about book-2SG `What is the book about?'
When PX specify the possessor or referent (in some cases there is no connection with possession, for example in (2)), their primary function is to encode grammatical agreement with the subject or object, and therefore they have an obligatory character and cannot be dropped. This is typical in cases when PX follow gerunds, numerals, pronouns and nouns which express inalienable entities and which function as adverbial modifiers in the sentence. PX in these cases refer to the subject of the sentence, as seen in (2), (3) and (4).
(2) mon, skola-je mi ni -ku-m, ton-e pumita-j.
1SG school-ILL go-GER-1SG you-ACC meet-PST.1SG `On the way to school I met you.'
(3) mon ogna-m ul-iko.
1SG one-1SG live-PRS.1SG `I live alone.'
(4) mon ki-ja-m
kiga vo-iko.
1SG hand-INS-1SG book hold-PRS.1SG `I hold a book in my hand.'
PX may also refer to the direct object in the sentence. PX in these cases follow a noun which expresses an inalienable entity and functions as an adverbial modifier, as in (5) and (6). If one were to drop PX in examples (2) ­ (6) the sentences would be ungrammatical.
(5) Peti r Masa-jez ki-ja-z Peter Masha-ACC hand-ILL-3SG `Peter hit Masha on her hand.'
sukk-i-z. hit-PST-3SG
38
(6) Peti r Masa-jez ki-ti-z Peter Masha-ACC hand-PROL-3SG `Peter took Masha by the hand.'
kut-i-z. take-PST-3SG
The present study describes the functional and grammatical specificities of PX as specifiers of possessor or referent and also presents a new hypothesis about the history of PX in Udmurt.
39
Demonstratives and actions: What is the Finnish demonstrative tuo doing in requests? Marja Etelдmдki, University of Helsinki Language of presentation: English The paper will discuss requests in which the Finnish demonstrative tuo is used for referring to referents that belong to the speaker, and/or are physically closer to the speaker than the recipient at the time of the utterance. In addition, it is the speaker who introduces the referent into the interaction. Thus, in these cases the referent could be claimed to belong to the speaker's sphere in a cognitive, social and spatial sense. However, in previous studies, the Finnish demonstrative tuo has been described as being used to pointing at referents that are spatially distal to the speaker (eg. Larjavaara 1990) or outside the speaker's (and the recipient's) sphere (Laury 1997) (cf. Lao nan, Enfield 2003). The paper draws from the understanding of deictic field as developed by Hanks (1990, 2005), and suggests a semantic description to the demonstratives that is based on the analyses of naturally occurring interactions. The data consist of everyday face-to-face and telephone conversations; the method used is ethnomethodological conversation analysis; the theoretical framework consists of the insight provided by conversation analytic method (see e.g. Schegloff 1997), linguistic anthropology (e.g. Silversteing 1976, Hanks 1990) and cognitive grammar (Langacker 2008). (For combining conversation analysis with cognitive grammar see also Etelдmдki & al. 2009, Etelдmдki & Visapдд forthc.). The paper will argue that the demonstrative tuo is used for construing the recipient as a person who has an independent access to to the referent and who is - together with the speaker an equal conceptualizer of the referent. By giving the recipient a role as an independent conceptualizer of the referent, the demonstrative tuo takes part in creating the action as a bilateral request or a request for advice. The semantic content of the demonstratives is argued to be minimal and underspecified (see also Etelдmдki 2009, Etelдmдki & Visapдд 2009). However, when embedded in a social situation, the demonstratives prove to be powerful in their capability to organize multiple facets of the current interactional context (see Hanks 2005). More generally, the paper will argue that the Finnish demonstratives are used for directing and regulating access and attention to the referent with respect to the on-going activity (see also Etelдmдki 2009, Etelдmдki & Visapдд forthc.). However, the link between demonstratives and actions is argued not to be direct: there is no one-to-one match between any particular action and a particular demonstrative. The interactants need different kinds of access and different types of focus on the referent in different types of activities and different phases of the activity sequence. The paper suggests that the demonstratives only position the participants and the referent in relation to each other, and that way they create different sorts of access, attentional foci and participation roles. When creating access, participation roles and attentional foci they take part in creating and constructing the activities as well. References: Enfield, N. J. (2003) Demonstratives in space and interaction. Data from Lao speakers and implications for semantic analysis. Language 79: 82­117. Etelдmдki, Marja (2009) The Finnish demonstrative pronouns in light of interaction. Journal of Pragmatics 41: 21­ 46. Etelдmдki, Marja & Minna Jaakola (2009) Tota ja puhetilanteen todellisuus ('The particle tota and the reality of speech situation'). Virittдjд 2: 188­212. Etelдmдki, Marja, Minna Jaakola, Ilona Herlin & Laura Visapдд (2009) Kielioppi kдsitteistyksenд ja toimintana. Kognitiivista kielioppia ja keskustelunanalyysia yhdistдmдssд [Grammar as conceptualization and as action. Combining Cognitive Grammar and Conversation Analysis]. Virittдjд 2: 162­187. 40
Etelдmдki, Marja & Laura Visapдд (forthc.) Why blend CA with CG? In Ritva Laury, Marja Etelдmдki & Elizabeth Couper-Kuhlen (eds.), Approaches to Grammar for Interactional Linguistics, Special issue of Pragmatics. Hanks, William F. (1990) Referential Practice. Language and Lived Space among the Maya. Chicago: The University of Chicago Press. Hanks, William F. (2005) Explorations in the deictic field. Current Anthropology 46(2): 191­220. Langacker, Ronald W. (2008) Cognitive grammar. A basic introduction. Oxford: Oxford University Press. Larjavaara, Matti (1990) Suomen deiksis [Finnish deixis]. Helsinki: Finnish Literature Society. Laury, Ritva (1997) Demonstratives in Interaction. The emergence of a definite article in Finnish. Amsterdam: John Benjamins Publishing Company. Schegloff, Emanuel A. (2007) Sequence Organization in Interaction. A Primer in Conversation Analysis I. Cambridge: Cambridge University Press. Silverstein, Michael (1976) Shifters, verbal categories, and cultural description. In K. H. Basso & H. A. Selby (eds.), Meaning in Anthropology. University of New Mexico Press, Albuquerque, pp. 11­55. Zlatev, Jordan A. (2010) Phenomenology and cognitive linguistics. In S. Callagher (ed.), Handbook on Phenomenology and Cognitive Science. Springer, pp. 415­446. 41
Demoting the subject: A functional distribution of impersonal constructions in Udmurt and Mari Nikolett F. Gulyбs, Eцtvцs Lorбnd University Language of presentation: English Although impersonal constructions vary across languages, the domain shows certain crosslinguistic patterns (cf. Malchukov­Ogawa 2011: 51). Following Siewierska (2008: 116), I consider given constructions to be impersonal if they lack a subject bearing canonical properties. Subject is determined here in functional terms; a canonical subject is an argument that is referential, topical, agentive, animate and definite (Malchukov­Ogawa id. 23; for a more detailed explanation, see Keenan 1976, Givуn 2001). Accordingly, the more criteria are fulfilled by a given subject, the less impersonal the construction is. My presentation will give a classification of some impersonals in Udmurt and Mari, using a corpus-based analysis. As has been pointed out by many linguists (Schiefer 1981, Kalinina­Kolomatsky­Sudobina 2006), Mari and Udmurt seem to share morphosyntactic features for the encoding of impersonals. This presentation will focus on four constructions in particular: A) Pl3 impersonals, in which personal verbal morphology is used with an impersonal reference B) causative impersonals, in which the verb has a causative morpheme and requires an accusative-marked subject C) reflexive impersonals, in which the subject is marked non-canonically or is not overt D) impersonal passives, in which the subject is fully demoted My aim is to provide a categorization of the impersonals mentioned above, using the five functional criteria of canonical subjects, since Udmurt and Mari impersonal constructions have not yet been examined using this approach. On the basis of my earlier, questionnaire-based study, a certain hierarchy of these constructions can be outlined, in which the less prototypically impersonal-like construction is type A, while type D seems to be the most typical representative of the domain. The hierarchy is the following: Pl3 impersonal > causative impersonals > reflexive impersonals > impersonal passive. Additionally, since impersonalization is closely related to perceptualization (cf. Siewierska id.: 121) and the windowing of attention, the use of the impersonal construction is motivated by pragmatic factors (for a more detailed explanation, see Givуn 2001). To outline the characteristics of their usage, I will present the results of a corpus-based survey. The corpus is comprised of linguistic material collected from the Pavlik Morozov parallel texts (Luutonen 2010), folklore texts, modern fiction and newspaper articles, containing ca. 500,000 tokens. With reference to the usage of impersonals, my hypotheses are the following: i) impersonals, of all types, will be found more frequently in newspaper and folklore texts than in fiction. ii) Pl3 impersonals will appear the most often in folklore texts. iii) impersonal passives can be typically found in newspaper text. Preliminary data show these suppositions to be correct, since the usage of impersonal constructions shows certain distribution according to textual frequency. My presentation will provide the results of my study in further detail, with a special reference to each type of impersonal construction. 42
References: GIVУN, TALMY. 2001: Syntax. A Functional-Typological Introduction. Vol. I. Amsterdam: John Benjamins Publishing Company. KALININA, ELENA­KOLOMATSKY, DMITRIY­SUDOBINA, ALEXANDRA 2006: Transitivity increase markers interacting with verb semantics. Evidence from Finno-Ugric languages. In: Kulikov, Leonid­Malchukov, Andrej­de Swart, Peter (eds.): Case, Valency and Transitivity. Amsterdam/Philadelphia: John Benjamins Publishing Company. 441­464. KEENAN, EDWARD 1976: Towards a universal definition of "subject". In: Li, Charles (ed.): Subject and Topic. NY: Academic Press. 305­334. LUUTONEN, JORMA 2010: Pavlik Morzov ­paralleelitekstikorpus. Turku: Research Unit for Volgaic Languages, University of Turku. (manuscript) MALCHUKOV, ANDREJ­OGAWA, AKIO 2011: Towards a typology of impersonal constructions. A semantic map approach. In: Malchukov, Andrej­Siewierska, Anna (eds.): Impersonal constructions. A cross-linguistic perspective. Amsterdam/Philadelphia: John Benjamins Publishing Company. 19­56. SIEWIERSKA, ANNA 2008: Introduction: Impersonalization from a subject-centred vs. agent centred perspective. Transactions of the Philological Society 106 (2): 115­137. SCHIEFER, ERHARD 1981: Das anonyme Subjekt in den finnisch-ugrischen Sprachen. In Bereczki GбborMolnбr Jуzsef (szerk.): Lakу-emlйkkцnyv. Budapest. 140157. 43
The Essive Case in Skolt Saami
Timothy Feist, The University of Surrey Language of presentation: English
Skolt Saami is one of a number of Uralic languages which makes use of the essive case. The essive case appears to be exclusively used by Uralic languages and is thus a typologically rare case. In this presentation, I discuss the form and the function of the essive in Skolt Saami and also consider how this is similar to, and how it differs from, the essive in other Uralic languages. The form of the essive case in Skolt Saami is ­n or ­Vn (where V is a vowel which depends on the inflectional class of the stem). This is similar, therefore, to the Finnish essive marker ­ na/­nд and it is possible that the Skolt Saami essive was historically vowel-final, since many word-final vowels have been lost in the language. The essive case, together with the rarely-used partitive case, are unique among the nine cases of Skolt Saami in that they each display only a single form for both singular and plural referents. Consider the following example (Feist 2010), where the essive marker -n on jдlstempхrttвn is clearly being used for a plural referent as indicated by the presence of a plural accusative object, here the plural demonstrative pronoun tхid.
tхid
oummu
ale
laaddвd jiхcceez
jдlstempхrttвn
DIST.PL.ACC person.PL.NOM begin.PST.3PL repair.INF REFL.ILL.3PL dwelling.house.ESS
people started to repair those (barracks) for themselves as residential houses
With regard to its function, the essive case in Skolt Saami is used to indicate a state or mode of existence, as observed in other Uralic languages.
koon
muхr
valdde, tхn
mвa puдldde lebe
REL.SG.ACC tree.SG.ACC take.PST.3PL DIST.SG.ACC later burn.PRS.3PL or
aunnsen хnne material.ESS use.PST.3PL whichever tree they took, they later burnt it or used it as material
(Feist 2010)
In addition to being used to mark nouns, the essive can be used on predicate adjectives, as the following example illustrates.
вlmmredd
lij
jeддд
ruхpsseen
horizon.SG.NOM PRS.3SG in.the.evening red.ESS
in the evening, the horizon is red
(Moshnikoff et al. 2009)
A second function of the essive in Skolt Saami is particularly noteworthy, as this use of the essive case appears to be unique to Skolt Saami. In addition to indicating a state or mode of existence (essive), the Skolt Saami essive is used to indicate a change of state, which in other Uralic languages (such as Finnish) are marked with a separate translative case. The essive case in Skolt Saami might, therefore, be more accurately referred to as the essive-translative case.
tel Emmel
muutti
sijjid lеdden: paarnid cuвnjan
then God.SG.NOM change.PST.3SG 3PL.ACC bird.ESS boy.PL.ACC goose.ESS
44
da niхid
njuhccвn
and girl.PL.ACC swan.ESS
then God changed them into birds: the boys (he turned) into geese and the girls (he turned) into
swans
(Feist 2010)
This use of the essive to express a change of state is not limited to nouns, but can also be used in predicate adjective constructions, where the adjective is an object complement.
ееlm
kдlkkii
pххrt
cappeen
man.SG.NOM paint.PST.3SG house.SG.ACC black.ESS
the man painted the house black
(Feist 2010)
It is hoped that this overview of the form and function of the essive case in Skolt Saami will help to inform the study of the essive case in the Uralic family as a whole.
References: Feist, Timothy. (2010). A grammar of Skolt Saami. Ph.D. diss. The University of Manchester Moshnikoff, S., Moshnikoff, J. & Koponen, E. (2009). Koltansaamen koulukielioppi (Sддmiхl iхllvuдppes skoouli vддras). Inari, Finland: Saamelaiskдrдjдt.
45
«Essive» in Eastern Khanty Andrey Filchenko, Tomsk State Pedagogical University Language of presentation: English
Although Essive is considered an endemic property of Uralic languages, it is nevertheless not explicitly present in all Uralic languages' case paradigms, for instance, Livonian, Vepsian, Erzya, Moksha, Mansi, Selkup and Khanty lack Essive per se understood as per the standard definition as temporary location or state of being (de Groot, 2012 Essive Questionnaire). These languages rather employ other formal means to code Essive senses. In the proposed discussion, the most frequent environments for the Uralic Essive will be reviewed based on the Eastern Khanty data, to identify formal means used for coding the Essive senses (those frequently attested for the explicit Essive use elsewhere). As follows from the last observation, a more appropriate framework for approaching the issue of "Essive" in Khanty is likely to be an onomasiological, meaning-based grammatical analysis of the data viz. how Eastern Khanty grammar codes the Essive senses. Eastern Khanty has an extended case system, characterized by a variety of semantic cases (LOC, ILL, ELA, ALL, PRO, TRANSL, COM, ABES) and lack of structural ones (NOM=Ш, ACC=Ш; GEN=Ш).
Non-verbal predicates In utterances with non-verbal predicates, Eastern Khanty uses either predicate-nominals or predicate-adjectives:
1. Vas.Kh.: m wдs'-`joan ja (ws-w)
1PL Vasyugan-river people (be-1PL)
We (are) Vasyugan people (EKhNeg_008b)
2. Vas.Kh.: o qam-a-k
head grey-EP-PRD
Head is (already) grey (EKhNeg_100)
3. Vas.Kh.: pьlt цl-д
j-дi.
hole big-TRNSL become-PST0.3SG
The hole got big(ger).
In both predicate-nominals (1) and predicate-adjectives (2-3), both individual-level and stagelevel nominal and adjectival predicates are coded by stems uninflected for case but often using a special Predicator affix /-(a)ki/. Copula, a finite existential verb is omissible in the Present context and obligatory in the Past (2, 3). This pattern of non-verbal predication is further illustrated in the Eastern Khanty constructions of the degrees of comparison of adjectives, where using an analytical construction, which also shows the use of the predicator affix.
4. Vas.Kh.: jь mд-n
ni ll-ki.
3SG 1SG -ACC [from] big-PRD
(S)he is older/bigger than me.
This Predicator affix shows formal affinity to the Translative case marker as well as to an adverbializer affix.
46
Translative Among the most frequent Eastern Khanty formal means of coding the Essive senses is the use of Translative (5) typically coding a participant or entity in the event that is a result of some transformation, i.e. an entity that something transforms into, and has marker /-ka, -qa, -a/:
5. Vas.Kh.: jь mustim ajni-kд
ju-s
3SG beautiful young.woman-TRNSL become-PST2.3SG
She became a beautiful young woman.
Adverbials Only a small group of Eastern Khanty adverbials are non-derived and can be considered a closed class of adverbs proper. The majority of Eastern Khanty adverbial senses are expressed by single-stem nominal forms derived with the help of a variety of `adverbializer' affixes (Tereskin 1961, Gulya 1966). Incidentally, among the most productive affixes are: /-(a), -(a)ki, -(a)k/ apparently without distinct semantic restrictions, appearing across adverbial functional range. pesta `fast, quickly' Я pest adj.`fast, quick, sharp'; j `together, jointly' Я j num.`one'; wan `near' Я wan adj.`short, nearest, proximal'; jmki `well' Я jm adj.`good'; atmaki `badly' Я atm adj.`bad'; alnwtki `on mornings' Я alnw adv.`in the morning' semnk 'boldly' Я n. sem 'heart'; adj. semn 'brave'
6. Vas.Kh.: itn-
j
evening-ADV become-PST0.3SG
Evening came.
7. Vas.Kh.: k sem-n-k
tl-a
very heart-ATTR-ADV DET-ILL
Do not come close (here) too bravely
дl NEGImper
jo-tn come-PST0.2DU
Finally, Essive senses may also be coded by dependent participial adverbial clauses, where Perfective Participle takes co-referential possessive marker and LOC-case marker for coding state (8) or ILL-case marker - for dynamic events (9):
8. Vas.Kh.: jpд, qunta mд wer wl-m-am-na, ...
once when 1SG small be-PP-1SG-LOC Once, when I was small, ...
9. Vas.Kh.: mд jn al-n jm-am-a
at-m
1SG ten year-LOC become-PP-ILL father-1SG
jas-t
say-PST0.3SG/SG When I got 10, my father told me: ...
mд-nt-m 1SG-ACC-1SG
Although Eastern Khanty lacks an explicit Essive marker, there is consistent use of Translative and Translative-related formal markers in the contexts identified provisionally for the Uralic languages as Essive senses. As observed at the onset, Essive is not a canonical structural casemarker, but rather a predicative marker, the marker of nominal or adjectival secondary predicates (de Groot 2012 Essive Questionnaire). In this context, association of Translative with
47
Adverbializer and Predicator affixes (fulfilling the Essive functional range in Eastern Khanty) appears to be particularly illustrative. Further Eastern Khanty data analysis is expected to yield, on the one hand, more comparative material for the cross-Uralic typological analysis of Essive. On the other hand, it will provide an interesting perspective on the possible evolution of Eastern Khanty case system. References: de Groot, C. 2012. Uralic Essive Questionnaire. Filchenko A. 2010. Aspects of the Easter Khanty Grammar. Tomsk: TSPU-Press. Gulya, J. 1966. Eastern Ostyak Chrestomathie. Bloomington: Indiana University Press. Tereskin N. 1961. Ocerki dialektov xantyjskogo jazyka: Vaxovskij Dialekt. Leningrad. 48
Topics in Focus Position in Hungarian Zsuzsanna Gйcseg, University of Szeged Language of presentation: English
In Hungarian, a referential expression appearing in the preverbal focus position of the sentence is generally claimed to function as an identificational focus (Й. Kiss 1998, 2002, Kenesei 2006), that is, to exhaustively identify the individuals the property expressed by the backgrounded post-focus part of the sentence holds for. The post-focus part is generally deaccented and denotes presupposed information. There is, however, a class of sentences with structural focus where the post-focus part bears the same accent as the focused constituent and carries new information. Such constructions are sometimes analysed as VP-focus sentences (Kenesei 1998), all-focus sentences (Й. Kiss 1998, 2007), sentential focus (Kenesei 2006) or hocus constructions (Kбlmбn 1985, 2001). Example (1) illustrates this type of structure:(1) Mi tцrtйnt? (1) `What happened?' Az OLASZ CSAPAT nyerte meg a VILБGBAJNOKSБGOT! (Й. Kiss 2007: 78) the Italian team won PRT the world-cup `It was the Italian team that won the world cup.'
Despite the differences in terminology and in the explanation of the phenomenon, all the approaches agree that the expression appearing in the preverbal focus position is also part of the comment (or logical predicate) of the sentence. The main purpose of this talk is to argue for the claim that in a subclass of these constructions (illustrated in (2a)), the preverbal focus position hosts a constituent denoting the pragmatic topic of the sentence. Following Gundel (1978), I define the pragmatic topic of a declarative sentence as an entity the speaker, in using that sentence, intends to increase the addressee's knowledge about. The constructions discussed in this paper share some of their properties with other languages (cf. for instance Delin ­ Oberlander 1995 for a subclass of cleft structures in English). The argumentation presented here is based on referential and cognitive properties of the constituent in focus position (it is typically a deictic or anaphoric expression) as well as contextual and prosodic features of the construction: the preverbal constituent denotes a situationally or contextually determined entity and the sentence can be used as an apparently "incongruent" answer (Krifka 2001) to a question uttered about the referent of the constituent in focus position (2a); this type of answer is often contextually equivalent with a sentence characterized by a classical topic-comment articulation (2b).
(2) Ki maga? `Who are you?'
a) ЙN vagyok az ЙJJELIR.
I am
the night-watchman
`I am the night-watchman'
b) pro/Йn az ЙJJELIR vagyok.
`idem.'
As for the pragmatic motivation of the construction, I argue for the hypothesis that its use is principally motivated by two purposes:
49
i) The (verbal or adjectival) predicate of these constructions imposes a uniqueness condition to one of its arguments, and allows (or sometimes forces) this argument to appear in a position typically used to express exhaustive identification; ii) Although the material following the constituent in structural focus position denotes new information here, this information is trivial or at a certain degree predictable for the hearer and in this respect has analogous properties to that of the `background' part of a focusbackground structure. References: Delin, J. ­ Oberlander, J. (1995): "Syntactic constraints on discourse structure: The case of it-clefts". Linguistics 33, 456-500. Й. Kiss, K. (1998): "Identificational Focus versus Information Focus". Language 74, 245-273. Й. Kiss, K. (2007): "Topic and Focus: Two Structural Positions Associated with Logical Functions in the Left Periphery of the Hungarian Sentence", In: Fйry, C., Fanselow, G., Krifka, M. (eds.): The Notions of Information Structure, 69-81. Gundel, J. (1988): "Universals of topic-comment structure". In: Hammond, M. ­ Moravcsik, E. ­ Wirth, J. (eds.), Studies in syntactic typology. John Benjamins Publishing Company, Amsterdam/Philadelphia, 209­239. Kбlmбn. L. (1985): "Word order in neutral sentences". In: Kenesei, I. (ed.), Approaches to Hungarian 1, JATE Kiadу, Szeged, 13-23. Kбlmбn, L. (2001): Magyar leнrу nyelvtan. Mondattan 1. Tinta Kцnyvkiadу, Budapest Kenesei, I. (1998): "Adjuncts and arguments in VP-focus in Hungarian". Acta Linguistica Hungarica 45, 61-88. Kenesei, I. (2006): "Focus as identification ». In: Molnбr, V. ­ Winkler, S. (eds.), The architecture of focus. Mouton de Gruyter, Berlin­New York, 137-168. Krifka, M. (2001): "For a structured account of questions and answers". In: Fйry, C. ­Steinefeld, W. (eds.), Audiatur vox sapientiae. A Festschrift for Achim von Stechow. Akademie-Verlag, Berlin, 287-319. 50
Responding to polar interrogatives in Finnish Auli Hakulinen, University of Helsinki Marja-Leena Sorjonen, University of Helsinki Language of presentation: English In this paper we are focussing on verb repeat answers to interrogative questions (YNI), which in Finnish are made with verb-initial word order plus a question-clitic attached to it. In a language like Finnish, the verb repeat is one of two default or conforming ways of giving a minimal answer to a YNI ­ an affirming vs. confirming one (Sorjonen 2001). Here is an example from a TV-interview with the Finnish Minister of Finance on the topic of the recently discovered tax paradises (the "-kц" in the interviewer's question is the question-clitic): IR: Pitдisi-kц EU:n nyt puhdistaa nurkkansa. Should EU now clean its corners. MF: Pitдisi. IR: Ja miten. And how. According to Heritage and Raymond (2012), a repeat answer to a yes/no question in English is highly marked. It is found, for example, to exert agency of the answerer, to implicate inappropriateness in the question and lead to disruption in the subsequent talk. These features cannot be generalized to the Finnish question ­ answer sequences. Our aim is to contribute to the discussion on responsive systems in typologically different languages. Our starting point is the following. First of all, we focus on interrogatively formatted questions that request for information. That is, we take into account the grammatical form of the question. Second, we take into account the sequential context in which the question is asked. It is only in the sequential context that the respective epistemic statuses of the participants can be assessed. Third, we also look at instances where the answer is non-contiguous to the question. Our data come from audio and videotaped interactions. They are mostly interactions among friends and family members but some data are drawn also from institutional contexts. In addition to answers that consist of a verb repeat only, we analyse three other kinds of responses to polar interrogatives: a more fully elaborated verb-initial clause; a verb repeat plus a continuation as a separate turn-constructional unit; and a double verb repeat. Each of these addresses different features in the question and/or in the sequential context. When the verb is accompanied by other elements (subject or object NP) in the same clause, there turn out to be different reasons for why more than a verb repeat is needed. One typical reason for a more elaborate clause is the distance between the question and the answer. So, rather than seeing the answer to pointing to a "problem" in the question, the elaborate answer alleviates the problem arisen e.g. because of the distance. When the answerer deals with an implication embedded in the question, more than a single clause is needed. Finally, answers with a double verb repeat treat the question as self-evident. This finding is in line with our previous findings about responses to assessments (Hakulinen & Sorjonen 2009). The double verb speaker orients to or treats the preceding turn ­ whether a question or an assessments, as self-evident at least to its speaker. In conclusion we will take up the more general implications of our study for understanding response systems across languages. 51
References: HAKULINEN, A. & SORJONEN, M.-L. 2009: Designing utterances for action: Verb repeat responses to assessments in Finnish. In M. Haakana, M. Laakso & J. Lindstrцm (eds.), Talk in interaction. Comparative dimensions, pp. 124­151. Helsinki, Finnish Literature Society. HERITAGE, J. & RAYMOND, G. 2012: Navigating epistemic landscapes: acquiescence, agency and resistance in responses to polar questions. In J.P. de Ruiter (ed.), Questions, pp. 179-192. Cambridge, Cambridge University Press. SORJONEN, M.-L.(2001): Simple answers to polar questions: The case of Finnish. In M. Selting & E. CouperKuhlen (eds.), Studies in Interactional linguistics, pp. 405­431. Amsterdam, John Benjamins. 52
Estonian illative as a category Petra Hebedovб, Masaryk University Language of presentation: English The aim of this paper is to argue that illative case in Estonian can be described as a single and compact category. Estonian illative with its traditionally recognized subcategories of short and long illative has been a topic of discussion for some time, which was concerned also with questions of mutual interplay of these subcategories, priority of one of them and their distribution (e.g. Hasselblatt 2000, Kaalep 2009). I would argue that Estonian illative as a grammatical category can be with benefit analyzed accordingly to the Langacker's (1987) view on categorization, by which categorization lies in the categorizing judgment and which employs the notion of prototype along with the idea of abstraction of schemas for the description of the category's structure. In this view category is portrayed as a schematic network, where prototypical is contrasted with peripheral, schematic is compared with specific. In the description of grammatical category such as Estonian illative, means of formation (morphology of illative forms) alongside with the semantics of illative forms as well as distribution have to be taken into account. In such an analysis short illative and long illative are perfectly describable as members of the same category. In this view category is not described by features whose presence is absolutely required, but is seen as gradient with both central and marginal members (cf. Bybee 2007). Patterns for forming illative in Estonian would not fit into a category based on presence or absence of certain features, but these various patterns can be seen as exemplifying several schemas which bare family resemblances to one another. Cognitive grammar also applies the usage based model (Langacker 1987) where grammar or systematization in language arises from language usage, grammatical patterns and schemas are abstracted from more specific units being used. That is why questions of distribution should not be considered apart from morphological considerations. Such idea is expressed by Kaalep (2009), according to whom frequency is a determinant in the distribution of illative patterns. I would argue that the so called long illative can be analyzed as a default pattern (cf. Bybee 2007). In that case the alongside distribution of long and short illative forms is conceivable. References: Bybee, Joan 2007: Frequency of use and the organization of language. Oxford: Oxford University Press. Hasselblatt, Cornelius 2000: Eesti keele ainsuse sisseьtlev on lьhike. Keel ja Kirjandus 11, 796­803. Langacker, Ronald W. 1987: Foundations of cognitive grammar I. Theoretical Prerequisites. Stanford: Stanford University Press. Kaalep, Heiki-Jaan 2009: Kuidas kirjeldada ainsuse lьhikest sisseьtlevat kasutamisandmetega kooskхlas? Keel ja Kirjandus 6, 411­425. 53
Functions of detached NPs in Old Finnish texts
Marja-Liisa Helasvuo, University of Turku Nobufumi Inaba, University of Turku Language of presentation: English
This study focuses on the functions of detached NPs in Old Finnish. Detached NPs are noun phrases which are not part of any clause (cf. Tao 1993, Ono & Thompson 1994, Helasvuo 2001). They include so-called dislocations and other types of detachments. Detached NPs are not used in Modern Standard Finnish, but they can be found in spoken and more informal written texts. The present study shows that detached NPs form a regular construction type in Old literary Finnish. The data for this study come from the Finnish translation of the New Testament by Mikael Agricola (1548). The data have been obtained from a morphosyntactically coded database of Agricola's works (University of Turku, Finland). There are over 300 detached NPs in the data which have been further analyzed in terms of morphosyntactic and semantic features and referential properties. The following example illustrates the construction type:
(1) Teme nyt ombi henen
tactons / ioca minun
lehetti /
this now be.3SG 3SG-GEN will-3PX REL 1SG-GEN send-PST.3SG
ette
Jocainen quin nдke Poian / ia wsko
henen pдlens /
COMP each.NOM REL see.3SG son-GEN and believe.3SG 3SG-GEN on
henelle
pite
oleman
ijancaikinen Eleme .
3SG-ADE must.3SG be-INF-INSTR eternal.NOM life.NOM
`For my Father's will is that everyone who looks to the Son and believes in him shall have eternal life.' (John 6:40, New Testament, NIV)
The detached NP is in line 2 in the example. It contains a pronominal head jocainen `each', followed by a relative clause. In the following clause (line 3), there is a pronominal NP henelle `on(to) him/her' which is coreferential with the detached NP. Textually, line 1 introduces a will (tactons `his will'), and the detached NP (line 2) specifies whom the will concerns. The will is formulated in line 3. The present study reveals that by far the most common type of detached NP in the data are NPs which contain a relative clause (cf. ex. 1 above). Interestingly enough, this type of detached NPs is also common in modern conversational Finnish (Laury & Helasvuo forthcoming). Our paper investigates the functions of detached NPs in the larger textual context and explores the syntactic and textual linking of the detached NP to the preceding and following text. ABBREVIATIONS 1SG = 1st person singular, 3SG = 3rd person singular, ADE = adessive, GEN = genitive, INSTR = instructive, NOM = nominative, PX = possessive suffix, INF = infinitive, PST = past tense, REL = relativizer.
LITERATURE: Helasvuo, Marja-Liisa 2001: Syntax in the Making. Amsterdam: John Benjamins. Laury, Ritva & Marja-Liisa Helasvuo (forthcoming): Detached NPs with relative clauses in Finnish conversations. Submitted. Ono, Tsuyoshi & Sandra A. Thompson 1994: Unattached NPs in English Conversation. In Proceedings of the 20th Annual Meeting of the Berkeley Linguistic Society. Tao, Hongyin 1993: Units in Mandarin Conversation. Amsterdam: John Benjamins. 54
Multiple directives in everyday Estonian interaction Tiit Hennoste, University of Tartu Andriela Rддbis, University of Tartu Language of presentation: English The paper is focused on multiple directives in everyday Estonian. We treat directives as a category which includes requests, commands, proposals etc (Hennoste, Rддbis 2004; ISK 2004: 1161­1162). Directives are social actions where one participant asks or tells another to do something. An action is the primary expected reaction, accompanied often by a verbal reaction. A typical directive contains an imperative form. Besides are directives expressed by declaratives or questions in the indicative or the conditional (e.g. you'll sit and wait; would you sit?). Modal verbs (can, must) are used often. (Metslang 2004; Pajusalu, Pajusalu 2004). We understand multiple directives as repeated actions, expressed by the same participant, formulated identically or modified, and expecting the same reaction,. We will follow Curl & Drew (2008) and Craven & Potter (2010) as a theoretical model. The method used is Interactional Linguistics. The data come from the Corpus of Spoken Estonian of the University of Tartu (Hennoste et al 2009). In our presentation we will analyze the verbal forms used in repetition of the directives. Preliminary analysis has shown that the formats used in repetition are as follow: - repeating the same form; - upgrading the directive; - concretization of the directive; - adding the motivation; - changing the command to the prohibition; - offering the alternative variant of the action; - encourage; - changing from the verbal directive to the physical action.
The repeating of the directives is motivated by - the non-hearing or non-understanding of the directive by the recipient; - non-granting the directive; - granting the directive partly or wrongly; - explicit refusing the granting.
1. T: [sхida] kuskil 'mujal praegu. ((Madisele))
T: drive somewhere else now ((to Madis))
2. (0.8)
3. K: tagavarad olid kхik [{juba}nii 'suured=et]
K: resources were all already so big that
4. T:
[sхida 'seal, (.) kapi 'peal] vхi
T:
drive there, on the cupboard or
/.../ ((19 turns omitted))
5. T: kule mine (.) kхrista kuskil 'mujal [ma=ei] kuule oma 'mхtteidki.
T: PTCL go rattle somewhere else I don't hear even my own thoughts
6. K:
[jah]
K:
yes
7. (0.5)
55
8. T: eks (.) [{--}] ((tхstab Madise ja ta auto koridori)) T: PTCL ((raises Madis and his car into corridor)) In the example we find three commands by the same person (lines 1, 4, 5), which expect the same reaction. As the commands are not granted, T moves from verbally directing Madis to physically moving him (line 8). All verbal directives are formulated in the second person of imperative. The second command is formulated more concretely (somewhere else vs. on the cupboard). In the third command the motivation is added (I don't hear even my own thoughts). The fourth command is a physical action. References: Craven, Alexandra, Potter, Jonathan 2010. Directives: Entitlement and contingency in action. ­ Discourse Studies 12, 419­442. Curl, Traci S., Drew, Paul 2008. Contingency and action: a comparison of two forms of requesting. ­ Research on Language and Social Interaction 41, 129­153. Hennoste, Tiit, Rддbis, Andriela 2004. Dialoogiaktid eesti infodialoogides: tьpoloogia ja analььs. Tartu: Tartu Ьlikooli Sihtasutus. Hennoste, Tiit, Gerassimenko, Olga, Kasterpalu, Riina, Koit, Mare, Rддbis, Andriela, Strandson, Krista 2009. Suulise eesti keele korpus ja inimese suhtlus arvutiga. - Eesti Rakenduslingvistika Ьhingu Aastaraamat V. Toim H. Metslang, M. Langemets, M.-M. Sepper, R. Argus. Tallinn: Eesti Keele Sihtasutus, 111­130. ISK = Hakulinen, Auli, Maria Vilkuna, Riitta Korhonen, Vesa Koivisto, Tarja Riitta Heinonen, Irja Alho 2004. Iso suomen kielioppi. Helsinki: Suomalaisen Kirjallisuuden Seura. Metslang, Helle 2004. Imperative and related matters in everyday Estonian. ­ Linguistica Uralica 4, 243­256. Pajusalu, Karl, Pajusalu, Renate 2004. The conditional in everyday Estonian: it's form and functions. ­ Linguistica Uralica 4, 257­269. 56
Centering on overt and zero pronouns in Estonian Helen Hint, University of Tartu Language of presentation: English This presentation tackles the question how to distinguish the use of zero pronoun and overt 3SG pronoun ta in Estonian spoken narratives. According to different accessibility hierarchies, the choice of a referring expression depends on the assumed attentional/cognitive status of the referent (e.g. Ariel 1988; Gundel et al. 1993). More reduced forms refer to more salient entities, whereas more elaborate phrases are needed for less salient entities. Therefore, zero pronouns should refer to most salient entities and overt pronouns to slightly less salient entities. However, although it is possible to use zero pronouns as anaphoric devices Estonian, it is not a default choice in referring to most accessible referents. Rather, short pronoun ta (s/he) is used in these contexts. Usually zero can be replaced with an overt pronoun; the opposite possibility is rather rare. It is claimed that zero is mainly used in spoken narratives for the nominal subject if it remains unchanged in a longer sequence of sentences (Lindstrцm 2005). Yet, there are no systematic analyses regarding the distinction of zeros and overt 3SG pronouns in Estonian. This presentation aims to fill this gap, providing a more systematic explanation for the distribution of pronominal forms. Centering theory framework (Grosz et al. 1995) is used and each utterance in the data is analyzed in terms of centering transitions to answer two questions. 1) If and how the set of forward-looking centers (Cf) influences the choice of referring expression: is there a difference between the Cf sets preceding the zero and those preceding overt pronoun ta? 2) Is there a difference between the transition types signaling zero and those signaling overt pronoun? The data come from experimentally elicited spoken narratives. Participants ­ 10 female Estonian native speakers ­ are shown the Pear Film. After seeing the film the participant is asked to retell what happened in the film. Each narrative is divided into segments and utterances to enable centering analysis. Preliminary results suggest that the choice of pronoun (overt or zero) depends on the entities in Cf list. Zero can be used only if there are no other active animate subject characters in the preceding Cf set and the entity referred to is the only actor in the whole segment. Otherwise, overt pronoun ta is required. It also appears that preceding and following transition types influence the choice of pronoun. Zeros are strongly related to CONTINUE transitions (a cross-linguistic phenomenon, see Di Eugenio 1998). Overt pronouns, however, combine with other transition types as well and do not show so strong correlation with just CONTINUE transition. Therefore, it is inferred that centering theory can be helpful in explaining the distribution of similar referential forms which seem puzzling at first sight. References: Ariel, Mira 1988. Referring and accessibility. ­ Journal of Linguistics 24 (1), 65­87. DI Eugenio, Barbara 1998. Centering in Italian. - In M. A. Walker, A. K. Joshi, E. F. Prince (eds.), Centering Theory in Discourse. Oxford University Press, 183-197. Grosz, Barbara J., Aravind K. Joshi, Scott Weinstein 1995. Centering: A Framework for Modelling the Local Coherence of Discourse. ­ Computational Linguistics 21 (2), 203-225. Gundel, Jeanette K., Nancy Hedberg, Ron Zacharski 1993. Cognitive status and the form of referring expressions in discourse. ­ Language 69, 274­307. Lindstrцm, Liina 2005. Finiitverbi asend lauses. Sхnajдrg ja seda mхjutavad tegurid suulises eesti keeles. Doktoritцц. Tartu: Tartu Ьlikooli Kirjastus. 57
OV order in Finnish without movement Anders Holmberg, Newcastle University Language of presentation: English Finnish is classified as an SVO language because the unmarked order in transitive clauses is S(Aux)VO, but SOV is possible as well, as a marked alternative. The condition is that a constituent is focused by movement or base-generation in initial position, such as the focusing adverb kyllд in (1); Vilkuna (1988, 1995), Holmberg (2000, 2001). (1) Kyllд minд kahvia juonut olen. indeed I coffee drunk have `I have indeed had coffee.' Without initial focus the order in (1) is sharply ungrammatical. I will assume that the explanation for this condition is: The effect of linearizing the object to the left of V is that it is de-focused (Vilkuna 1995). In the unmarked case the information focus of a sentence is (a part of) VP. With SOV order, VP can't be focus in Finnish. Since a sentence must have a focus, initial focus is then the (only) other option (Holmberg 2001). When the SOV option is taken, the word order is highly flexible. In a sentence with Aux, V, and O, all the permutations are possible except one: (i) Aux V O, (ii) Aux O V, (iii) O V Aux, (iv) O Aux V, (v) ?V Aux O, (vi) *V O Aux. In a sentence with a put-type verb (Aux V NP PP as unmarked order), as many as 20 of the 24 logically possible permutations are acceptable. This raises a number of questions, including the following: (a) How do we explain the fact that all the permutations of the unmarked Aux V O (PP) order are subject to the initial focus rule? (b) How come the order (vi) (the `FOFC-violating' order; Biberauer & al., to appear, Sheehan, in press) is the one word order which is not permitted? The idea that the defocusing of in the Finnish SOV constructions is the effect of moving to a topic or old-information domain (as in Holmberg 2000) will be shown to not account for (a). The theory which I will articulate is based on Sheehan (in press). The main postulates of Sheehan's theory are (a) every terminal node must be linearized relative to every other terminal node, (b) syntactic heads are marked for whether they take a complement to the left or to the right, subject to parametric variation (against Kayne 1994), (c) in cases where the linear order between two terminals cannot be determined by virtue of (b), it is determined by Kayne's LCA (in part following Kayne 1994). The different permutations of VP constituents and Aux are not derived by movement but by universal linearization rules applying to the structures built by Merge. The cross-linguistic variation is due to variation in the linear specification of given heads (not due to special movement properties). If, for example, V is marked to either precede or follow its complement, then VO and OV order are allowed. Inflected forms, too, are derived by linearization rules in this system (without head-movement). References: Biberauer, T., A. Holmberg, I. Roberts (to appear) A syntactic universal and its consequences. To appear in Linguistic Inquiry. Holmberg, A. 2000. The derivation of OV order in Finnish. P. Svenonius (ed.) The derivation of VO and OV order, 123-150. John Benjamins, Amsterdam. Holmberg, A. 2001. The syntax of yes and no in Finnish. Studia Linguistica 55, 141-175. Sheehan, Michelle. In press. Explaining the Final-over-Final Constraint. In theoretical approaches to disharmonic 58
word orders, eds. Theresa Biberauer and Michelle Sheehan, OUP. Vilkuna, M. 1989. Free word order in Finnish. SKS, Helsinki. Vilkuna, M. 1995. Discourse configurationality in Finnish. In: Й. Kiss, K. languages (244-268), Oxford University Press.
(ed.) Discourse- configurational
59
On the grammaticalization of some converb constructions in Udmurt, Tatar literary language and Mishar Tatar (Joidenkin konverbirakenteiden kieliopillistumisesta udmurtissa, tataarin kirjakielessд ja misддritataarissa) Laura Horvбth, Eцtvцs Lorбnd University Language of presentation: Finnish The notion of converb (/gerund/adverbial participle) can be defined as "a nonfinite verb form whose main function is to mark adverbial subordination" (Haspelmath 1995: 3). Despite the adverbial function, the construction composed of a gerund and a finite verb (conjugated or derived as needed in the clause) can be considered as grammaticalized construction on several occasions in the languages in question. For instance, in Tatar and Udmurt there are some constructions, where the converb provides the lexical meaning while the finite verb has more or less lost its original meaning and is a modifier (auxiliary) expressing aspect ­ usually via an Aktionsart meaning: (1a) Tat. - ()(1b) Udm. - - eat-CVB finish- 'eat (PRF)' Converb constructions are widely used in the Volga-Kama area in Mari, Udmurt, Chuvash and Tatar and they occur frequently in the other Turkic languages, too (see, e.g. Johanson 1995), which have had no connections with the Volga­Kama area. Thus, the usage of these constructions is presumably the result of an early Turkic influence on the Finno-Ugric languages: Mari has been influenced by Chuvash and Tatar, Udmurt mainly by Tatar. In my presentation, I plan to examine some grammaticalized converb constructions in Udmurt, Tatar literary language, Mishar Tatar and "Finnish Tatar" (the language of Mishar Tatar immigrants in Finland). Tatar literary language is very close to "Finnish Tatar", because its morphology is based on the Mishar dialect (see, e.g. Berta 1998: 283), but I assume that there can be also differences in aspectuality caused by the influence of Finnish language. Considering the converb constructions in the related languages, there are only adverbial connections between the two elements in many cases, e.g. modal connection: (2) Udmurt (Kel'makov 2006: 206) - '- - women-PL talk-CVB go-PRS.3PL `the women are going while having a conversation' In example (3), the verb `to go' has a rather obscure meaning, expressing imperfectivity: (3) Udmurt (Kel'makov 2006: 212) '- - run-CVB go-INF `to run' (lit. `to go running') 60
Without any textual context, it is possible that we do not know whether the verb could be interpreted as an auxiliary, an aspectual marker or if it is only an adverbial subordination, where the verb maintains its own lexical meaning. For instance, in the following example it is clear from the textual context that the converb construction ( ) is not an adverbial subordination (e.g. a temporal one): it is not a case of someone squeezing out toothpaste and then taking it out somewhere. The verb is an auxiliary here, the text tells about somebody's habits:
(4) Udm. (Matvejev 2005: 58)
- - - -

who tube-ELAT tooth cleaner paste-ACC compress-CVB take.out-PRS.3SG slowly PRT `who squeezes out (lit. `takes out compressing') the toothpaste from the tube slowly'
It is important to note that grammaticalization is a process of constructions with particular items in them, rather than a process concerning single elements (Bybee 2003: 602): an element is grammaticalized always in its syntagmatic context (Himmelmann 2004: 31). A construction becomes more and more frequent and loses specific semantic features, becomes more general in meaning; the repetition leads to the weakening of semantic force by habituation and automatization (Bybee 2003: 602­604). It enables the constructions to be used in new contexts (see, e.g. Himmelmann 2004: 31­33). Concerning the converb constructions, the converb before the modifier is the syntagmatic context, and the more kinds of converbs the modifier can be added to, the more grammatical its meaning is. In other words, a modifier can lose its own original lexical meaning to a different extent in different structures (with different gerunds). In my presentation, I plan to examine the more frequent converb constructions and auxiliaries in the languages in question. I would like to discuss the way these auxiliaries of converb constructions can contribute to the aspectual meaning of sentences and to determine what kinds of converbs these auxiliaries can be added to in what kinds of textual contexts. I will rely on corpus-based research on dialectal texts, prosaic texts, modern novels and texts in newspapers.
References: Berta, Бrpбd 1998: Tatar and Bashkir. ­ L. Johanson & Б. Csatу (eds.): The Turkic Languages. London­New York: Routledge: 283­301. Bybee, Joan 2003: Mechanisms of change in grammaticalization: The role of frequency. ­ B. D. Joseph & J. Janda (eds.): The Handbook of Historical Linguistics. Oxford: Blackwell: 602­623. Haspelmath, Martin 1995: The converb as a cross-linguistically valid category. ­ M. Haspelmath & E. Kцnig (eds.): Converbs in cross-linguistic perspective. Berlin: Mouton de Gruyter: 1­55. Himmelmann, Nikolaus P. 2004: Lexicalization or grammaticization? Opposite or orthogonal? ­ W. Bisang, N. P. Himmelmann & B. Wiemer (eds.): What makes grammaticalization? A Look from its Fringes and its Components. Berlin: Mouton de Gruyter: 21­42. Johanson, Lars 1995: On Turkic converb clauses. ­ M. Haspelmath & E. Kцnig (eds.): Converbs in cross-linguistic perspective. Berlin: Mouton de Gruyter: 314­347. Kel'makov, V. K. = . . 2006: . Izhevsk. Matvejev, Sergej = 2005: . Izhevsk.
61
Temporal frames of reference and the function of delimitatives in Finnish Tuomas Huumo, University of Turku, University of Tartu Language of presentation: English Temporal frames of reference (FoR) have recently become a "hot topic" in cognitive semantics (Boroditsky 2000, Moore 2006, Evans 2003, among others). Temporal FoRs are metaphorical extensions of spatial ("Levinsonian") FoRs that represent spatial relations against an axis system frontal vs. lateral vs. vertical. Since time is conceptualized as metaphorical motion along a path, the FoRs are also relevant in the expression of time, which, however, utilizes dynamic axis prepositions and a moving ground, as in New Year comes after Christmas (a Moving Time metaphor), or We have hard times ahead of us (a Moving Ego metaphor). In Finnish, the opposition between static and dynamic FoRs (in the spatial domain) is based on the opposition between internal and external local case forms of axis postpositions. For instance, the static concept `in front of' is expressed by the internal case forms of the ete- gram (the inessive edessд `in front of', the elative edestд `from front of' , the illative eteen `to front of'), whereas the dynamic concept `ahead' is expressed by its external case forms (adessive edellд `ahead', ablative edeltд `from ahead', allative edelle `to ahead'; see Keresztes 1964 for a detailed Finnish-Hungarian contrastive analysis). However, the use of the external case forms is possible only if both the Ground and the Figure are in motion ­ if the Figure is stationary, then the internal case forms are used in spite of the moving Ground. Both systems are in use in the expression of time. In my presentation I discuss Finnish temporal FoRs and their use with delimitative phrases of time (e.g., `for two hours'). To use terms borrowed from sociolinguistics, I argue that such phrases can measure either "actual" or "apparent" time. The latter term refers to the projected temporal distance between the participants (Figure and Ground), and is illustrated by She is two years ahead of me in her studies (which does not mean that she is living in the year 2015!) In Finnish, the opposition between actual and apparent time is reflected in the construction indicating a temporal FoR. To express actual time, a GENITIVE + ADPOSITION construction is typically used, as in Syntymдpдivд-ni on kaksi viikkoa joulu-n ede-llд [birthday-1SGPX is two week-PAR Christmas-GEN ahead-ADE] 'My birthday comes two weeks before Christmas'. To express apparent time, the complement often takes the partitive case, as in Sikiц-n jala-t ovat kaksi viikko-a muu-ta ruumis-ta ede-llд [fetus-GEN leg-PL are two week-PAR rest-PAR bodyPAR ahead-ADE 'The legs of the fetus are two weeks ahead [in development] of the rest of its body' (a typical expression type in Internet discussion groups on pregnancy and babies). I argue that in a similar vein with the Finnish partitive object (which, unlike the accusative object, can co-occur with delimitative phrases), the partitive construction of a FoR of apparent time exposes its complement to measure. In sum, my presentation starts with an overview of the temporal FoR system in Finnish and discusses the different functions and interplay of delimitative phrases and adpositional FoR constructions. References: Boroditsky, Lera 2000. Metaphoric structuring: understanding time through spatial metaphors. Cognition 75: 1­28. Evans, Vyvyan 2003. The structure of time. Amsterdam: John Benjamins. Keresztes, Kбlmбn 1964. Morphemic and semantic analysis of the word families: Finnish ETE- and Hungarian EL- `fore'. Indiana University Publications, Uralic and Altaic series, Vol. 41. Indiana University: Bloomington. Moore, Kevin E. 2006. Space-to-time mappings and temporal concepts. Cognitive Linguistics 17 (2): 199­244. 62
A Dialect-Based Approach to the Finnish Essive Case Emmi Hynцnen, University of Turku Languge of presentation: English
The objective of my study is to describe what kind of information about the Finnish essive case can be found in data that has been collected from dialectal narratives. This case has earlier been studied from different perspectives. E.g., Leinonen (2008) compared different kinds of Finnish depictives, Pajunen (2000) wrote about essive-case predicate adjectives, Helasvuo (1990) studied the case in the context of conceptual semantics, and Alhoniemi (1967) brought up the syntactic difficulties that Finnish grammarians have had with essive-case adverbials. In Finnish, the essive is a case that productively expresses state which, in most cases, is temporary. It is typologically unusual to express the impermanence of state by a case, particularly when expressing properties. The semantics of essive can most clearly be noticed by comparing nominative- and essive-case predicate nominals.
Pia
o-n
opettaja ~ sairas.
Pia.NOM be-PRES.3SG teacher.NOM ~ ill.NOM `Pia is a teacher ~ ill'
Pia
o-n
opettaja-na ~ sairaa-na
Pia.NOM be-PRES.3SG teacher-ESS ~ ill-ESS. `Pia is (working as) a teacher ~ (temporarily) ill'
To discover the possible variation between the dialects and standard Finnish, I used a corpus that contains all occurrences of essive of Archives of Syntax (University of Turku). This data consists of 9115 essives, and 6406 of them are from dialects. This data has been transferred to a database and classified both syntactically and semantically. The dialectal data reveals that the essive case is used similarly in standard Finnish and the dialects. However, there are at least four exceptions. 1) In the dialects of the Eastern Finland (mostly Savo), the participle form of perfect and pluperfect tense often contains the essive case marker. 2) The "true" essive-case clausal adverbials are rare in dialects. 3) The essive case is used in depictive constructions that also contain a copula and express age. 4) Circumstantial adverbials are rare in dialects. All of these findings can be explained. 1) Larjavaara (1995) suggests that the dialects of Savo have reanalyzed the n-marker of the participles of the dialect of Karelia as an essive case marker. 2) It is possible that the predicational function of the essive has expanded to nonpredicational functions. It is also possible that the structure is a loan from another language (possibly Swedish). 3) This use can reveal the evolution of the essive-case circumstantial adverbials as well as age-denoting depictives. 4) It is possible that the category of circumstantials is a loan from Swedish (Tapani Lehtinen p.c.).
References: Alhoniemi, A. 1967: Tilan essiivin jдsentдmisestд suomen kielen lauseopeissa. In Juhlakirja Kauko Kyyrцn tдyttдessд 60 vuotta 24.11.1967. Tampere: TaY. Helasvuo, M-L. 1990: Identifioiva kenttд. In Suomen kielen paikallissijat konseptuaalisessa semantiikassa. Helsinki: HY. Larjavaara, M. 1995: Savon (n)nA-supiini. ­ In Palander, Savijдrvi & Forsberg (eds.): Murteiden matkassa. Joensuu: JoY. Leinonen, M. 2008: Depictive secondary predicates in Finnish. ­ In Schroeder, Hentschel & Boeder (eds.): Secondary predicates in Eastern European languages and beyond. Oldenburg: BIS-Verlag.
63
Pajunen, A. 2002: Suomen ominaisuuspredikaatio ja tyyppi olla sairaana. In Pajunen (ed.), Nдkцkulmia kielitypologiaan. Helsinki: SKS. 64
Topic elicitation. A new look at Hungarian multiple questions Andrбs Imrйnyi, Jagiellonian University Language of presentation: English
Hungarian multiple questions of the type illustrated in (1) have long been a puzzle to researchers.
(1) A: Ki mit
eszik reggeli-re?
who what-ACC eats breakfast-for 'What is everyone having for breakfast?'
B: Йn rбntottб-t,
Kati mьzli-t, Pйter vajas kenyer-et.
I scrambled.eggs-ACC Cathy muesli-ACC Peter buttered bread-ACC 'I (eat) scrambled eggs, Cathy, muesli, and Peter, bread and butter.'
The key issue is the position and function of all but the final/rightmost interrogative pronoun, which is widely assumed to be focused, behaving in the same way as interrogative pronouns in simple questions. Based on the semantic property of distributivity (i.e. the fact that ki 'who' in (1) calls for the enumeration of multiple members of a set, each associated with a separate proposition), Й. Kiss (2002) argues that non-final interrogative pronouns are in Spec,DistP, a position normally reserved for universal quantifiers such as mindenki 'everybody'. Surбnyi (2007) rejects this account, and notes two properties which these elements share with topics: "they are presuppositional, and they correspond to topicalized phrases in appropriate answers". However, he stops short of analysing them as syntactic topics, at least partly for theory-internal reasons. Revisiting the issue from a functional perspective, I propose to describe the phenomenon with a new functional category called topic elicitator. Whereas by default, the speaker uses a referential expression to establish the topic in the current discourse space, with TE's such as ki 'who' in (1), she may also instruct the listener(s) to do so. However, TE's are only licensed in clauses containing a focus elicitator as well, cf. mit 'what- ACC' in (1). A crucial condition on the use of a TE is that the referents be easily recoverable from situational context. That is, speakers cannot use the construction to learn new facts about any referent that comes to the listener's mind. Often, the referents are the listeners themselves, with each of them expected to supply a fragment of the required information. The distributive reading of TE's follows from their presuppositional character and considerations of relevance: if, in the given set of highly accessible referents, only a single member were to be specified, the speaker should perform this specification herself. On the other hand, the choice of ki 'who' over mindenki 'everybody' is motivated by the fact that the former is more suitable for foregrounding an instructive/elicitating function. In terms of construction grammar (Goldberg 1995), the pattern shares many of its functional and formal properties with the topic construction of Hungarian. However, it also has peculiarities of its own, and thus deserves to be treated as a construction in its own right.
References: Goldberg, Adele 1995. Constructions: A construction grammar approach to argument structure. Chicago: University of Chicago Press. Й. Kiss, Katalin 2002. The syntax of Hungarian. Cambridge: Cambridge University Press. Surбnyi, Balбzs 2007. Focus structure and the interpretation of multiple questions. In: Schwabe, Kerstin & Susanne Winkler (eds.), On Information Structure, Meaning and Form. Amsterdam: John Benjamins. 229­253.
65
Structural typicalities in advanced learner Finnish: Usage-based approach towards variety-specific multiword sequences Ilmari Ivaska, University of Turku Language of presentation: English
The present paper introduces an application of the keystructure analysis (Ivaska&Siitonen 2011; Ivaska: in preparation) as a data-driven corpus method to trace recurring structural patterns typical for advanced learner Finnish. The research questions are twofold: 1) which structural features distinguish advanced learner Finnish (F2) from the native speakers' (F1) variety, and 2) what is the role of the immediate linguistic context and genre-specific contextual features in explaining these differences. Usage-based models of language describe grammar as probable habits of language use that emerge through actual utterances and generalisations based on their similarities and dissimilarities (e.g. Hopper 1987; Bybee&Thompson 1997). Thus, individual utterances do not merely represent the system but constitute parts of a constantly changing dynamic apparatus (Hunston&Francis 2000). The idea is in line with the corpus-driven approach, in which the corpus data is used from the very beginning of the process to sketch the actual research questions (cf. Tognini-Bonelli 2001). In this paper, typical structural features of learner Finnish are traced by comparing the frequencies of the occurring word form unigrams, bigrams and trigrams extracted from the data sets (for similar approach cf. Aarts&Granger 1998; Wiersma et al. 2011). Words are defined based on their morphological forms. Frequencies are counted following the skipgram approach (Guthrie et al. 2006), and the words do not have to follow each other consecutively, for so long as they are in the same order. Maximum distance of the words is not specified but the words need to be in the same clause unit. Thus, following example contains four unigrams (, , and (<-sti>), six bigrams (, , <-sti>, , <-sti> and ), and four trigrams (, <-sti>, <-sti>, <-sti> and <-sti>).
Word Engl. Mrp.
1, merkitykset `meaning'
2. voivat `to can'


`The meanings can change rapidly.'
3. muuttua `to change'
4. nopeasti `rapidly' <-sti>
The comparison between F2 and F1 is done in the spirit of the contrastive interlanguage analysis (Granger 1996), so that the F2 and F1 varieties of written of Finnish are compared to locate typical quantitative differences between the varieties. The studied items are chosen based on their statistical keyness, i.e. significant frequency differences between the data sets (Scott&Tribble 2006). The keyness is measured using Random forests, a measure created for automatic categorisation (Breiman 2001) that can be used to analyse typical linguistic variation (e.g. Tagliamonte&Baayen 2011). Correlation between the frequencies of different elements is also analysed to uncover the gram clusters that probably depict the same phenomenon. Potential keystructures are analysed in terms of their inner and cotextual variation (Francis 1993) to find possible explanations for the differences. Typical co-occurrences may depict recurring patterns, grammatical structures, and differences between genres can show contextual effects in grammar.
66
The corpus used is The Corpus of Advanced Learner Finnish (Ivaska&Siitonen 2009), which is annotated in terms of word lemmas, parts-of-speech, morphological forms and syntactic functions. The corpus contains exam essays, manuscripts of theses, and other writings written by advanced L2-learners of Finnish. A reference data (including the respective text genres) by native speakers of Finnish has also been collected as part of the corpus. The texts have been written during the research subjects' university studies and, thus, they are all part of a natural academic discourse. References: Aarts, J., & S. Granger 1998. "Tag sequences in learner corpora: a key to interlanguage grammar and discourse". In S. Granger (ed.), Learner English on Computer. London: Longman, 132-141. Ivaska, I., & K. Siitonen 2009. "Syntaktisesti koodattu oppijankielen korpus: mahdollisuuksia ja ongelmia". In P. Eslon, & K. Хim (eds.), Korpusuuringute metodoloogia ja mдrgendamise probleemid. Tallinn; Tallinna Ьlikool, 54-71. Ivaska, I., & K. Siitonen 2011. "Avainrakenneanalyysi: tapa tutkia oppijankielen lauserakennetta korpusvetoisesti". ­ AFinLA-e: Soveltavan kielitieteen tutkimuksia 3, 35-47. Breiman, L. 2001. "Random forests". Machine Learning 45 (1), 5-32. Bybee, J., & S. Thompson 1997. "Three frequency effects in syntax". Berkeley Linguistics Society 23, 65­85. Francis, G. 1993. "A corpus-driven approach to grammar ­ principles, methods and examples". in M. Baker, G. Francis, & E. Tognini-Bonelli (eds.), Text and Technology. In Honour of John Sinclair. Amsterdam: John Benjamins, 137-156. Granger, S. 1996. "From CA to CIA and back: An integrated approach to computerized bilingual and learner corpora". In K. Aijmer, B. Altenberg, & M. Johansson (eds.), Languages in Contrast. Lund: Lund University Press, 37-51. Guthrie, D., B. Allison, W. Liu, L. Guthrie, & Y. Wilks 2006. "A closer look at skip-gram modelling". In Proceedings of Fifth international Conference on Language Resources and Evaluation (LREC), Genoa, Italy. Available at: http://www.lrec-conf.org/proceedings/lrec2006/ (accessed November 2012). Hopper, P. J. 1987. "Emergent grammar". Berkeley Linguistics Society 13, 139-157. Hunston, S. & G. Francis 2000. Pattern Grammar. A Corpus-Driven approach to the Lexical Grammar of English. Amsterdam: John Benjamins. Scott, M. & C. Tribble 2006. Textual Patterns: Key Words and Corpus Analysis in language education. Amsterdam: John Benjamins. Tagliamonte, S. & R. H. Baayen, 2011. "Models, forests and trees of York English: Was/were variation as a case study for statistical practice". Language Variation and Change 24 (2), 135-178. Tognini-Bonelli, E. 2001. Corpus Linguistics at Work. Amsterdam: John Benjamins. Wiersma, W., J. Nerbonne, & T. Lauttamus 2011. "Automatically extracting typical syntactic differences from corpora". Literary and Linguistic Computing 26 (1), 107-124. 67
Adverbs of certainty in Finnish Minna Jaakola, University of Helsinki Language of presentation: English The paper analyses expressions of epistemic certainty in written genres. Focus will be on two Finnish adverbs, tietysti 'of course, naturally, certainly' and varmasti 'definitely, certainly', and their usage in argumentative text type. The study aims at giving epistemic adverbs a cognitive semantics description (Langacker 2008, Verhagen 2007) that takes into account the textual and intersubjective aspects of meaning (Simon-Vandenberg & Aijmer 2007, Martin & White 2006). Both lexemes are derivatives: tietysti includes the verb stem tietд- 'to know', whereas varmasti is derived from the adjective varma 'certain, sure, safe, confident'. Semantics of stems motivates much of the usages of these derivatives. For example, by tiettдvдsti writers display evidential justifications based on general knowledge and predictability, whereas varmasti in most cases construes the justification to be based on writer's reasoning. (Cf. VISK 2004: § 1606, 1608.) Analysis is based on the compilation of Helsingin Sanomat newspaper Internet discussions consisting of 86 971 comments. In this corpus there are 2712 occurrences of tietysti and 4092 of varmasti. When interlocutors overtly mark their arguments with adverbs expressing high degree of certainty, they not only express their epistemic stance but also construe the rhetorical and intersubjective organization of a text. In argumentative comments, adverbs of certainty are used to foreground something about the argument. These adverbs also foreground the epistemic domain per se by evoking an image of people negotiating the epistemic status of entities. Therefore, adverbs of certainty are means to construe intersubjectivity and position the reader in relation to the claims and ideologies expressed in the text. References: Langacker, R. 2008: Cognitive Grammar. A Basic Introduction. Oxford University Press. Martin & White 2005: The language of evaluation. Appraisal in English. Palgrave Macmillan. Simon-Vandenbergen, Anne-Marie & Karin Aijmer 2007: The semantic field of modal certainty. A corpus-based study on English adverbs. Berlin: Mouton de Gruyter. Verhagen, A. 2007: Construal and perspectivization. - The Oxford Handbook of Cognitive Linguistics, Geerarerts, Dirk & Hubert Cuyckens (eds.), 48­82. Oxford University Press. VISK = Auli Hakulinen, Maria Vilkuna, Riitta Korhonen, Vesa Koivisto, Tarja Riitta Heinonen, Irja Alho 2004. Iso suomen kielioppi. [Descriptive Gram- mar of Contemporary Finnish.] Helsinki: Finnish Literature Society. Also available at http://scripta.kotus.fi/visk/etusivu.php 68
Functions of the essive-translative in Tundra Nenets
Lotta Jalava, University of Helsinki Language of presentation: English
In addition to the standard case inventory consisting of the basic grammatical cases (nominative, genitive, accusative) and four locational cases (lative, locative, ablative, prosecutive (`via')), Tundra Nenets noun system has an essive-translative (ESS-TR) form with the marker -ж. It has traditionally not been considered as a case but rather a derivative suffix because of its defective inflectional use (e.g. Salminen 1998), but from a functional point of view it could be considered as case marking in the same way as the formal and modal essives kйnt, -ul/-ьl in Hungarian. The essive-translative is typically used in constructions expressing temporary state of being (cp. essive in Finnish and Estonian) and change in state of a noun, that is, becoming something (cp. Finnish translative). In the construction with translative meaning `to become something', the noun marked with the ESS-TR is usually followed by the verb xya `to leave', but other verbs with similar meaning are also used. The paper focuses on the essive meaning of the marker -ж, and its use in expressions of temporary state of being. The essive-translative form cannot act as a nonverbal predicate as such; instead, it uses the auxiliary tara `be necessary, participate' as an obligatory verbal predicate of the construction; see (1). As for other nominal predicates in attributional, equative or proper inclusion (see Payne 1997) clauses in Nenets, they are typically formed by adding the verbal suffixes straight into the nominal stem, see (2).
(1) ma xada-i
uitel-ж
tara-
I
grandmother-POSS1SG teacher-ESS-TR
`My grandmother was a teacher.'
AUX.3SG-PRET
(2) ma Usport-h ter-dm
I
Usport-GEN inhabitant-SUBJ.1SG
`I am an Usport-inhabitant' (inhabitant of the village of Usport)
The need for using the auxiliary tara instead of predicating the noun marked with the essive- translative, arises from the origin of the essive-translative marker, as it appears to originate from a copula verb `be'. In addition to tara-constructions and translative use, the essive-translative in Tundra Nenets is also used as predicate complements or depictives with other verbal predicates, such as expressing temporary states or actions, e.g. `to work (as something)', as numeral depictives, e.g. `the two of them, as two'. The essive-translative is also governed by the verb ume- `name, call' on their noun complement. In this paper, I will analyse the different secondary predicate types of the essive-translative in more detail. The theoretical background of the study lies in functional syntax and grammaticalization. The data consist of published grammars of Nenets (Terescenko 1947, 1973; Salminen 1993, 1997); fieldwork materials from the Taimyr Peninsula including both personal and traditional stories and elicited translation sentences; and, some of the existing text materials such as newspaper texts and the examples of Nenets literature (Susoi 2000-2001), Nenets folklore texts, and the corpus of sample sentences from the big Nenets-Russian dictionary (Terescenko 1965).
References: Payne, Thomas E. 1997: Describing morphosyntax: a guide for field linguistics. Cambridge: Cambridge University press. 69
Salminen, Tapani 1998: Nenets. -- Daniel Abondolo (ed.) The Uralic languages. London: Routledge, 1998. 516­ 547. Susoi 2000-2001 = . . : " . 8/9 . : . Terescenko 1965 = , . : - . : . 70
Ugric Possessive Suffixes in Non-Possessive Functions Gwen Eva Janda, Graduate School Language & Literature Munich Class of Language Language of presentation: English Possessive suffixes are a set of personal markers attached to nominal stems. In most Uralic languages they have been analyzed mainly in their default-interpretation as markers of all kind of possessive relations as is reflected in their terminological denomination. Though it is regularly pointed to possible other fields of usage while describing possessive suffixes, a detailed analysis of these is yet missing in most Uralic languages. However, the usage of possessive suffixes in different constructions not denoting possessive relations at all is considerable. The analysis of such non-possessive constructions in the Ugric languages Mansi, Khanty and Hungarian is the topic of my doctoral thesis: these related languages show significant divergences due to the long-term settlement in different cultural areas and different language contacts. This also concerns functions of the possessive suffixes. In my presentation, which will be held in English, I'd like to present some first results taken from my corpus analysis consisting mainly of narratives and tales from all three languages. This text genre is well qualified to demonstrate the interaction of text structure and the use of possessive suffixes. My analysis is mainly synchronic due to the absence of historical text corpora for Mansi and Khanty. Some features I'd like to present are e.g. discourse-structuring functions like topic shift / turn taking: the re-introduction of a topical referent is achieved by using a nominal phrase with a possessive suffix marking the head of the phrase (in the following example, the respective referent has not been part of the action in the preceding sentences and is re-introduced now):
(01) sujpil
lupta paxwit josa -
xum -ite [...] potrt -i
lingonberry leaf in width of ski ADJZR man SG<3SG tell PRS[3SG]
"The man with skis, small like lingonberry leaves, says"
A change of topics can also be indicated by a converbal clause with a possessive suffix as means of the text coherence (the converbal clause sums up the preceding state-of-affairs, the main clause signals a shift of perspective, often introducing a new referent):
(02) xuj -ima -te akmatert ak eka joxt -s
lie CVB 3SG once
one wife come PST[3SG]
"While he was lying there, a woman came."
It seems that all of these functions in non-possessive contexts do have one thing in common: a coexistent change or loss of parts of the paradigm resulting in changes in the referential system from deictic and anaphoric use to merely indicating the grammatical linking of two constituents in a phrase. Therefore my presentation will also deal with the theoretical aspects that might be employed to explain varying usages and possible paths of grammaticalization.
Selected references: Fraurud, Kari (2001), "Possessives with extensive use: A source of definite articles?" In: Baron, Irиne - Herslund, Michael - Sшrensen, Finn (eds), Dimensions of possession. (Typological Studies in Language 47) Amsterdam/Philadelphia: John Benjamins Publishing Company p. 243-267 Givуn, Talmy (1983), Topic continuity in discourse: A quantitative crosslinguistic study Amsterdam/Philadelphia: John Benjamins Publishing Company Lambrecht, Knud (1994), Information structure and sentence form. Cambridge: University Press Matras, Yaron (2009), Language contact (Cambridge Textbooks in Linguistics) Cambridge: University Press
71
Nikolaeva, Irina A. - Kovgan, Elena V. - Koskarva, Natalia (1993), "Communicative roles in Ostyak syntax" In: Finnisch-Ugrische Forschungen 51 p. 125-167 Schlachter, Wolfgang (1960), Studien zum Possessivsuffix des Syrjдnischen (Finnisch-Ugrische Studien III) Berlin: Akademie Verlag Skribnik, Jelena (2001a), "Pragmatic structuring in Northern Mansi" In: Seilenthal, Tхnu (ed.), Congressus Nonus Internationalis Fenno-Ugristarum 7.-13.8.2000 Pars VI Tartu: Auctores p. 222-239 Suihkonen, Pirkko (2005), "On the Categories and Functions Developed from the Possessive and Deictic Suffixes in Udmurt" In: Hasselblatt, Cornelius - Koponen, Eino - Widmer, Anna (eds), Lihkkun lehkos! Beitrдge zur Finnougristik aus AnlaЯ des sechzigsten Geburtstages von Hans-Hermann Bartens (Verцffentlichungen der Societas Uralo-Altaica Band 65) Wiesbaden: Harrassowitz Verlag p. 401-431 72
A case of syntactic variation: Double morphology in adpositional structures (in Erzya­Russian bilingual discourse)
Boglбrka Janurik, University of Szeged, University of Tartu Language of presentation: English In this paper, I study adpositional structures in the speech of Erzya­Russian bilinguals. I argue that mixed adpositional structures with double morphology occur as a result of the frequent use of bilingual repetitions (Finch 2009). My analysis is based on a corpus of 20 unstructured interviews. Erzya is the default language of the interviews as agreed by the interviewer and the subjects. However, code-switches to Russian occur in all of the recordings. I apply Muysken's (2000) categorization for the analysis of the code-witches in my data. Muysken differentiates between three categories of intrasentential code-switching: insertions, alternations and congruent lexicalization. Insertions and congruent lexicalizations are relevant for my study. In the former case, Russian prepositional phrases are inserted into the utterance which has an Erzya grammatical structure. (Erzya is the matrix language as defined in Myers-Scotton's (2002) Matrix Language Frame Model.) In example (1), we can see a typical case of insertion:
(1) pose armi-i karm-i
robota-mo
after army-GEN start-1SG.PST work-INF `After being in the army, I started to work.'
In the case of congruent lexicalization, two languages set the grammatical structure of the utterance together. These structures involve the use of double morphology as both the Russian preposition and the Erzya case ending occur in the adpositional phrase as in example (2):
(2) da-jut
lamo pibil' da vee-e
give-3PL.PRS many profit to village-DEF.DAT `They give a lot of profits to the village.'
Bilingual repetitions represent an in-between stage. In this case, speakers insert a Russian codeswitch into the Erzya discourse and then also opt for the Erzya equivalent of the phrase using parallel phrases as in example (3):
(3) pek visie- do dvu et very small-PL until two.GEN year.PL.GEN `(They are) very small, up to two years.'
kavto ije-s two year-ILL
In this paper, I argue that repetitions having a clear pragmatic function (clarification, emphasis, etc.; cf. Gumperz 1982) contribute to grammatical changes in the bilingual Erzya­Russian variety, as they lead to gradual conversion of the two languages.
References: Finch, Shannon Barbaradee. 2009. Repetition as Linguistic and Social Strategy in Hindi-English Bilingual Discourse. University of Texas at Austin. http://repositories.lib.utexas.edu/bitstream/handle/2152/ETD-UT-2009-12-472/FINCH- DISSERTATION.pdf?sequence=1 (Date of access: 10 February, 2013) Gumperz, John. 1982. Discourse strategies. Cambridge: Cambridge University Press. Muysken, Pieter. 2000. Bilingual speech: A typology of code-mixing. Cambridge: Cambridge University Press.
73
Myers-Scotton, Carol. 2002. Contact linguistics. Oxford: Oxford University Press. 74
Finnish demonstrative tдmд from L2 learner's perspective Hanna Jokela, University of Tartu Language of presentation: English I focus on Finnish demonstrative tдmд `this' and its relations to the other demonstratives and pronouns in learners' language. In particular, I am interested in the pronoun use of Estonian students learning Finnish. My material is drawn from texts written by Estonian students (mainly intermediate level). As a starting point there is the observation that Estonian speakers have difficulties when choosing between different Finnish demonstratives, and they also may make choices that are non-typical in Finnish system. In particular, Estonian learners more frequently choose demonstrative tдmд instead of se `it' in certain contexts where Finnish native speakers would choose se `it'. In this study my aim is to shed light on demonstrative acquisition and try to explain motivations behind over-used and/or non-typically used tдmд. Further on, I wish to develop some ideas that are applicable in teaching and understanding the pronoun system. Similar pronouns with different referential properties in Finnish and Estonian can be relatively misleading from the learner's perspective. In the learner's mind there might be several overlapping and interrelated options. Finnish tдmд and Estonian third person singular pronoun tema are connected historically and have similar properties. It is difficult to tell Finnish pronouns se and tдmд apart, and as tema is very widely used in Estonian, Estonian learners might find Finnish tдmд more familiar and prefer tдmд instead of se. Animacy, proximity and accessibility are some of the factors to consider when analyzing learners' choices and interpretations. Laury (1997) suggests that tдmд places its referent in the speaker's sphere, while se places its referent in the addressee's sphere. How is this reflected in the learner's interpretation? In addition, Finnish tдmд can refer to the several types of referents, and Estonian tema can refer to both an animate and an inanimate referent (Pajusalu 2005). I assume that close connections between the personal pronouns and the demonstratives may make demonstrative acquisition more challenging. References: Laury, Ritva 1997. Demonstratives in interaction: the emergence of a definite article in Finnish. Amsterdam: John Benjamins. Pajusalu, Renate 2005. Anaphoric pronouns in Spoken Estonian: crossing the paradigms. Ritva Laury (Ed.) Minimal reference: The use of pronouns in Finnish and Estonian discourse, 107-134. Helsinki: SKS. 75
The development of complex grams in Estonian Anni Jьrine, University of Tartu Language of presentation: English
This presentation is concerned with the development of complex postpositions and complex adverbs in Estonian. The ongoing grammaticalization process that sets off from postpositional phrases (1a) and results in complex adverbs and complex postpositions (1b) has been discussed by Habicht, Penjam (2007) and Jьrine (2011). However, so far, the studies have focused on describing the synchronic variation of the emerging complex grams, no attempt has been made to address the problem from the diachronic perspective. The aim of this study is to gain insight into the very beginning of development of complex grams and to describe the grammaticalization paths of body-part related complex grams as they have evolved over time.
Ex. 1.
a) Juukse-d o-li-d otsapidi mehe-Ш
kьlje-Ш all
hair-PL be-PST-PL partly husband-GEN flank-GEN under `Part of my hair was [stuck] under my husband's flank'
b) Цц-de-l
kui laps mehe-Ш
kьlje-Ш all
night-PL-ADE when baby husband-GEN flank-GEN under At nights when the baby [is] under husband's flank
'At [those] nights the baby [is] next to my husband'
The data comes from the corpus of Old Literary Estonian (size 1,550,802 words), The Corpus of 19th Century Texts (size 520,307 words) and The Corpus of Written Estonian 1890-1990. The studied phrases are selja taga (lit. 'behind one's back'), pea kohal (lit. 'above one's head'), kьlje all (lit. `under one's flank'), jalge all (lit. `below one's feet'), nina ees (lit. `before one's nose'), kдe all (lit. `under one's hand'). In Contemporary Estonian, these phrases have lexicalized (c.f. Brinton, Traugott 2005) and, in certain contexts, function as complex items expressing non-literal/abstract senses. This study attempts to trace the beginnings of the development of complex grams and addresses the following questions:
- Does the data confirm the ordering of changes as is suggested by the schema of cyclical development of function words in Estonian (c.f. Habicht, Penjam 2007)? - Is the semantic shift from concrete to (more) abstract meanings and contextual expansion of the complex items traceable in the available data? - Is the emergence of complex grams language internal development, or is there any evidence of a contact-induced change (c.f. Heine, Kuteva 2005)?
The preliminary results suggest that most of the studied phrases occurred in figurative/abstract senses already in their first occurrences in the corpus. However, most of the earliest attestations can not (yet) be considered to be grammaticalized usages of the studied phrases. In some cases, the earliest usages are extremely restricted contextually, and only the further contextual expansion allows us to consider them instances of grammaticalization. In case of kдe all, the data suggests that the development of complex postposition has been induced by German influence. Thus, in some cases we are not dealing with language internal change but rather with a kind of contact-induced grammaticalization.
76
References: Laurel Brinton & Elizabeth Closs Traugott 2005. Lexicalization and language change. Cambridge: Cambridge University Press. Habicht, Kьlli, Pille Penjam 2007. Kaassхna keeleuurija ja -kasutaja kдsituses [Adpositions as viewed by a linguist and by a language user]. Emakeele Seltsi aastaraamat 52 (2006): 51--68. Heine, Bernd, Tania Kuteva 2005. Language contact and grammatical change. Cambridge: Cambridge University Press. Jьrine, Anni 2011. Vormierinevused tдhenduserinevuse ilminguna: kaassхnaьhendite grammatiseerumine konstruktsioonides. [Change in Form as a Manifestation of Change in Meaning: Estonian Adpositional Phrases] ­ Keel ja Kirjandus: 899--916. 77
Impersonal and generic reference: A cross-linguistic look at Finnish and English narratives Elsi Kaiser, University of Southern California Language of presentation: English In linguistic communication, in addition to referring to specific, known referents, we also need to deal with unknown, generic or non-specific referents. I investigated the linguistic constructions used to refer to generic or unknown entities in Finnish and English. Let's start with two English impersonal constructions: third-person impersonal they (ex.1), and secondperson impersonal you (ex.2). (1) My mother was very beautiful once too. Or so they tell me. (Hunger Games, p.3) Дitikin oli aikanaan kaunis. Niin ainakin sanotaan. =>passive (Nдlkдpeli, Finnish translation) (2) Then I glance quickly over my shoulder. Even here... you worry someone might overhear you. (Hunger Games, p.6) Sitten vilkaisen nopeasti olkani yli. Kuulijoita ш pelkдд tддllдkin.... =>zero person (Nдlkдpeli, Finnish translation) Impersonal they lacks a referential antecedent, and excludes the speaker and hearer (Siewierska/Papastathi `11, Cabredo'08)­it means `people in general', or `someone' (identity may be unknown). Impersonal you also means, roughly, `people in general', but can be used to refer to a speaker's experience, while making a generalization (ex.2, cf.Moltmann'06 on `one'). Furthermore, you has modal force and usually dislikes episodic contexts, unlike they (Malamud'06). To explore how these constructions are realized in Finnish, I compared the original English Hunger Games (S.Collins,'08) with the Finnish translation (Nдlkдpeli, translator H.Bьtzow). In Finnish, the two constructions that come to mind as suitable equivalents for they and you are the passive and the zero-person construction (nollapersoona) respectively. The passive is marked by verbal morphology (ex.1), whereas the zero-person has singular third-person agreement but lacks an overt subject (ex.2). The Finnish passive resembles they in allowing for episodic reference, whereas the zero-person tends to be associated with modality, like you (Hakulinen et al.'04:1297). Like you, the zero-person can refer to the speaker's experience and its generalizability (Laitinen'06, Moltmann'06). I conducted a corpus analysis of impersonal you and they in The Hunger Games and their Finnish translations. Results: Impersonal they is translated into Finnish as a passive construction 81% of the time, but impersonal you is translated into Finnish as a zero-person structure only 47% of the time (see ex.1-2). We observe an asymmetry: Whereas impersonal `they' maps fairly stably onto the Finnish passive, impersonal `you' is translated using a variety of different constructions (zero-person, first-person pronouns, lexical forms like ihminen `human'). This suggests that the coverage of impersonal you is broader than the zero-person construction. On the basis of a detailed analysis of corpus examples, we suggest this may be due to complexities inherent in the notion of generalizability/genericity (e.g. what counts as `generalizable enough' to be realized with a zero-person, rather than first person?), indicating that languages can differ in the level of generalizability needed for particular constructions. Time permitting, we will discuss data from the Estonian translation (Nдljamдngud, translator 78
E.Schapel), to further explore the differences between Finnish and Estonian zero-person sentences (Jokela'12). This research furthers our understanding of how reference to generalizable or unknown entities is accomplished in typologically different languages, and suggests that some aspects of reference may be more susceptible to crosslinguistic variation than others. References: Hakulinen, A. et al. 2004. Iso Suomen Kielioppi. Finnish Literature Society. Jokela, H. 2012. Nollapersoonalause suomessa ja virossa. Dissertation, University of Turku, Finland. Laitinen, L. 2006. Zero person in Finnish: A grammatical resource for construing human reference. In Grammar from the Human Perspective - Case, space and person in Finnish. John Benjamins. Malamud, S. 2006. Semantics and pragmatics of arbitrariness. Dissertation, University of Pennsylvania. Moltmann, F. 2006 Generic One, Arbitrary PRO, and the First Person'. Natural Language Semantics, 14, 257-281. Siewierska, A. & M. Papastathi 2011. Third person plurals in the languages of Europe: typological and methodological issues. Linguistics 43 (2): 575­610. 79
On the interpretation of generic `zero subjects' in Finnish: Syntactic and information-structural considerations Elsi Kaiser, University of Southern California Language of presentation: English
This paper explores the syntax and information-structure of the Finnish generic zero-person construction. Finnish has no overt generic pronoun like English one, and uses third-person singular verbs with null subjects (ex1a, Holmberg'10, Vilkuna'92). The `zero' can bind reflexives and antecede agent-oriented adverbials (e.g. Hakulinen & Karttunen'73, Vainikka'89). My talk focuses on (i) the interpretation of the zero and (ii) the syntax and word order of the construction, and their relationship. (1a) Tдssд ш istuu mukavasti. Here ш sit-3rd-sing comfortably. `One can sit comfortably here' (1b) * ш istuu mukavasti tдssд. (1c) Sitд istuu mukavasti tдssд. (1d) (Minд) istun mukavasti tдssд. I sit-1st-sing comfortably here. Holmberg'10 shows that generic zeros cannot satisfy the EPP (also Holmberg/Nikanne'02). In sentences with generic zero subjects (ex.1a-c), spec-TP needs to be filled with an overt element (1a: locative, 1c: expletive sitд it-PART). Ex(1b), the verb-initial zero-person sentence, is unacceptable, which is striking given that referential null subjects (ex.1d) satisfy the EPP. According to Holmberg, generic zeros cannot satisfy the EPP because their features are a subset of the features of T. However, seeming counterexamples to (1b) come from Laitinen (2006, ex.2), who notes that "many examples do not contain any preverbal elements" (p.215), and that in such cases, the zero is usually interpreted as referring to speaker or addressee. (2) ш Tдytyy harjata hampaat. Must-3rd-sing brush teeth. `One/I/you must brush one's/my/your teeth' (Laitinen's translation)
This raises questions not only about the interpretation of the generic zero, but also the syntactic and word-order (information-structural) properties of this construction. I show that applying insights from Moltmann'06 sheds light on the Finnish zero-person construction. Moltmann argues English generic one is licensed by (i) Inference from the first person or (ii) Inference to the first person (ex.3). Case (i) involves first-person experience or the speaker simulating normal conditions/experience. Case (ii) involves an "already established generalization, a law, general requirement or general recommendation" (p.274).
(3a) One can see the picture from the entrance. [inference from first person]
(3b) One should not lie.
[inference to first person]
I discuss how these notions shed light on the interpretation and word order of zero-person constructions. I explore the hypothesis that verb-initial zero-person constructions tend to involve inference to the first person, and that non-verb-initial zero-person constructions tend to
80
involve inference from the first person. For example, I suggest that `inference from the first person' can be used to explain the frequency and unmarked nature of locative and temporal preverbal elements (4a,1a) in zero-person sentences, and why other preverbal elements seem to have stronger information-structural properties (e.g. topicality, ex.4b) (4a) Ovelta nдkee vessaan. Door-ABL ш see-3rd-sing bathroom-ILLAT `From the door [as soon as you step in], you can see into the bathroom.' (4b) Vessaan nдkee ovelta. `You can see into the bathroom from the door' This work aims further our understanding of the interplay of syntactic, semantic and pragmatic factors in the interpretation of the Finnish zero-person construction, and has potential implications for middle constructions, as well as extensions to other languages. References: Hakulinen, A. & L. Karttunen, 1973. Missing Persons: On Generic Sentences in Finnish. CLS9: Papers from the Ninth Regional Meeting, 157-71, Chicago Linguistic Society Holmberg A. (2010) The null generic subject pronoun in Finnish: a case of incorporation in T. In: Biberauer T; Holmberg A; Roberts I; Sheehan M, ed. Parametric variation : null subjects in minimalist theory. Cambridge: Cambridge University Press, pp.200-230. Laitinen, L. 2006. Zero person in Finnish: A grammatical resource for construing human evidence. In Grammar from the Human Perspective: Case, space and person in Finnish. M.-L. Helasvuo and L. Campbell, eds. Pp. 209-232 Amsterdam: Benjamins. Vainikka, A. (1989), Deriving Syntactic Representations in Finnish, University of Massachusetts Ph.D. dissertation. University of Massachusetts, Amherst. Vilkuna, M. 1992. Referenssi ja mддrдisyys suomenkielisten tekstien tulkinnassa. Helsinki: Suomalaisen kirjallisuuden seura 81
Previous Knowledge and Utterance Design: The Case of Estonian jaajaa Riina Kasterpalu, University of Tartu Language of presentation: English Using Conversation Analysis as a methodology, this presentation sheds light on one of the functions of the Estonian response particle jaajaa. While there are some studies into the functions of the single response particle jaa, little is yet known about the functions of its multiple forms in Estonian spoken interaction. The data for this research comes from the Corpus of Spoken Estonian of the University of Tartu. For studying the functions of particle jaajaa, audio recordings of sales negotiations were selected. In those calls, the callers are selling specific knowledge, both parties representing an institution. The negotiation process is started by clarifying the territories of knowledge, possible shared experiences, and key competencies of both parties. While gaining knowledge about their prospect partners, callers ask lots of questions. In a question-answer sequence, the asker of a question expresses his/her lower epistemic status than that of the respondent. This presentation demonstrates the function of response particle jaajaa in the sequentially third position, i.e. as a reaction of the asker of a question after or during listening to the answer and processing provided information. It will be shown how, in my data, two different pitch contours of jaajaa are used systematically for different interactional purposes. jaajaa uttered with a high level-fall intonation is used to mark the pre-existing knowledge of the listener. In contrast, jaajaa with a rise-fall pitch contour is a sign of an understanding acquired during listening to the answer. Regardless of the intonation, in both cases the listeners aim to halt the on-going activity of the speaker. In that respect, Estonian jaajaa resembles the German response token jaja, although intonation contours of the Estonian jaajaa appear to be somewhat different. Moreover, jaajaa as a proposal for the closure of the sequence uttered with different intonation contours is based on different analysis of the previous turn. While jaajaa with a high level-fall contour is proposing that the previous speaker should not proceed with ongoing action, therefore indicating disaffiliation; jaajaa with a rise-fall contour is acknowledging the possible end of an explanation. Furthermore, jaajaa with a rise-fall contour shows that no further explanation is necessary, thus aligning with the action of the previous speaker. References: Barth-Weingarten, Dagmar 2011. Double Sayings of German JA ­ More Observations on Their Phonetic Form and Alignment Function. Research on Language and Social Interaction, 44(2), 157­185. Drew, Paul, Walker, Traci 2009. Going too far: Complaining, escalating and disaffiliation.Journal of Pragmatics, 41, 2400­2414. Heritage, John 2012. Epistemics in Action: Action Formation and Territories of Knowledge. Research on Language and Social Interaction, 45(1), 1­29. 82
On mood and modality in language death: Evidence from Finnic Petar Kehayov, University of Tartu, University of Regensburg Language of presentation: English Studies dealing with the structural characteristics of language obsolescence discuss processes and phenomena applying to several areas of grammar at once, such as reduction (Campbell & Muntzel 1989), generalization and paradigmatic levelling (Campbell & Muntzel 1989), morphotactic transparency and constructional iconicity (Dressler 1981), suppression of marked features in favour of unmarked features (Dimmendaal 1992; Campbell & Muntzel 1989), innovativeness (Gal 1989: 326), loss of redundancy (Dressler 1981), preference for analytic/isolational structures (Dorian 1977: 27; Schmidt 1985: 61; Dimmendaal 1992: 130), and loss of suppletive forms (Sasse 1992). This circumscription of the research on language death to cross-categorial processes is partly understandable, as language decay is often a rapid process affecting different domains of grammar simultaneously and not consecutively. The lack of research focusing on the behaviour of specific grammatical categories in language death has, however, limitative consequences for our understanding of language change in general. For example, I am not aware of any explicit hypotheses with regard to the relative susceptibility of grammatical categories (e.g. tense or aspect) or their values (e.g. pluperfect or progressive) to loss, change and innovation in language decay. This poverty of research is striking when compared to the wealth of studies that examine the behaviour of grammatical categories in "normal" language contact and propose various explicit universals (e.g. in the form of borrowing hierarchies; see Matras 2009: 153­ 165). In this talk I report some preliminary results from an ongoing study on the behaviour of mood and modality (MM) in certain obsolescent Finnic varieties (such Ingrian, Votic and Lude). My method involves a comparison of earlier MM systems (as documented in text collections and by reference grammars and special studies) with fluent speaker and narrowspeakers systems. My focus is on phenomena which cannot be straightforwardly explained as replicas from dominant languages of the area. Several findings of the study are formulated as probabilistic hierarchies showing the relative susceptibility to loss, change and innovation in MM systems. In this form they are easily comparable with data from other transitional stages of human language, where the linguistic competence of an individual is at stake, such as language acquisition (cf. acquisitional hierarchies; Stephany 1995) or language impairment (agrammatism or developmental disorders). General phenomena recurring in my data are multi-functionalization, particlization and de-auxiliarization of modal verbs, as well as MM structures pioneering changes in other verb categories (such as Tense). The typological relevance of these findings should be put in test by a future research. References: Campbell, Lyle, Martha C. Muntzel 1989. The structural consequences of language death. In Nancy Dorian, (ed.), Investigating obsolescence. Studies in language contraction and death. Cambridge: Cambridge University Press, 181­196. Dimmendaal, Gerrit 1992. Reduction in Kore reconsidered. In Matthias Brenzinger (ed.), Language Death. Factual and Theoretical Explorations with special reference to East Africa. Mouton de Gruyter, 117­135. Dorian, Nancy 1977. The problem of the semi-speaker in language death. In Wolfgang Dressler, Ruth WodakLeodolter (eds), Language death. (= International Journal of the Sociology of Language 12.) The Hague: Mouton, 23­32. Dressler, Wolfgang U. 1981. Language shift and language death ­ a protean challenge for the linguist. Folia Linguistica. 1981/1­2, 5­28. 83
Gal, Susan 1989. Lexical innovation and loss: The use and value of restricted Hungarian. In Nancy Dorian (ed.), Investigating obsolescence. Studies in language contraction and death. (= Studies in the Social and Cultural Foundations of Language 7) Cambridge: Cambridge University Press, 313­331. Matras, Yaron 2009. Language contact. Cambridge: Cambridge University Press. Sasse, Hans-Jьrgen 1992. Language decay and contact-induced change: Similarities and differences. In Matthias Brenzinger (ed.), Language Death. Factual and Theoretical Explorations with special reference to East Africa. Mouton de Gruyter. 59-80. Schmidt, Annette 1985. Young people's Dyirbal: An example of language death from Australia. Cambridge: Cambridge University Press. Stephany, Ursula 1995. Function and Form of Modality in First and Second Language Acquisition. In Anna Giacalone-Ramat, Grazia Crocco Galиas (eds.), From pragmatics to syntax: modality in second language acquisition. Tьbingen: Gunter Narr, 105­120. 84
Ablative as a marker of benefaction in Finnish Seppo Kittilд, University of Helsinki Language of presentation: English
Beneficiaries are coded in a number of ways within and across languages. Beneficiaries may, e.g., be marked with case affixes or adpositions, or they may be introduced via applicativization. As for case marking, dative, or a similar case, is usually employed for Beneficiary coding. The same formal treatment is frequently accorded to the Recipient across languages, which is not unduly surprising given the semantic features Beneficiaries and Recipients share. In addition, the polysemy of Beneficiary and Maleficiary is also common. In languages in which this occurs, the use of a Beneficiary marker appears to be the original function of the marker, which has then later extended to coding Maleficiaries. This is rather expected, because benefaction and malefaction rather form a continuum instead of being mutually exclusive notions. It is therefore possible to interpret a Beneficiary marker as a marker of Maleficiary in favourable conditions. By contrast to the frequently attested polysemies noted above, languages in which Source and Recipient and/or Source and Beneficiary are lumped together, and in which the marker of Source cannot code Recipients, are far less common. Especially rare are languages in which a marker originally used for Source coding (such as the ablative case) takes over beneficiary coding as one of its functions. One reason for the infrequent occurrence of this polysemy is probably found in the clear semantic differences between Sources and Beneficiaries. There are, however, few languages in which this polysemy is attested, although these languages are in the minority. Examples include Kuuk Thaayorre (Alice Gaby, p.c.), Gawwada (Mauro Tosco, p.c.), Mekeo (Alan Jones, p.c.) and Finnish. My paper discusses the use of the ablative for coding benefaction in Finnish, as in:
(1) vanhempi pes-i
lapse-lta kдde-t
parent wash-3SG.PAST child-ABL hand-PL `The parent washed the child's hands'
First, I will discuss the semantics of contexts in which the ablative can be used for Beneficiary coding (the use is far from being fully productive); contexts, in which ablative is used for Beneficiary coding, are less frequent than those in which the allative codes benefaction. Basically, ablative can occur in constructions describing grooming actions (as in (1)) or `changing scenes' (`the nurse changed the patient's sheets'). In the former case, the allative is excluded, or it produces a different reading, while in the latter case, the allative may occur, but with a slight change in focus. Second, the paper discusses the reasons for the rare occurrence of the Beneficiary/Source polysemy by comparing it to the frequent Recipient/Beneficiary polysemy. It will be shown that a grammaticalization path that produces this polysemy is less natural than the (rather frequent) grammaticalization process as a result of which the use of dative or allative extends to coding malefaction. Third, the paper also discusses some potential explanations for the emergence of the examined polysemy in Finnish. For example, ambiguity avoidance plays an important role in this process.
85
On Komi demonstrative pronouns
Gerson Klumpp, University of Tartu Nikolay Kuznetsov, University of Tartu Language of presentation: English
The paper introduces first demonstrative pronouns and their different use types according to grammatical descriptions of the Komi (Zyrian) literary language (ЦKK, SKYa., Fedyunлva 2008, and others). Second, this overview will be confronted with the findings of an investigation into the actual use of demonstratives in a particular corpus, namely the texts by one speaker of the Upper Vychegda dialect as assembled in 5th volume of Syrjдnische Texte (Uotila & Kokkonen 2006). The investigation focuses on the pragmatic use types of demonstrative pronouns (exophoric, discourse deictic, anaphoric, recognitional, cf. Diessel 1999), as well as the internal grammar of adnominal demonstrative noun phrases (px-marking on head noun, agreement, word order, adjacency). There are four demonstrative, resp. 3rd person pronouns used both in pronominal (prosubstantive) and in adnominal (proadjective) function in the Komi literary language: tajц `this one', etajц `this, the other one', sijц `that' and esijц `that'; with case inflection and postpositional constructions the short stems ta, eta, sy and esy are used. In terms of proximity in relation to the speaker, tajц and etajц designate an entity close to the speaker, esijц and sijц a remote one. Unlike the others, sijц is used in anaphoric function, it always presupposes a topical referent; in pronominal anaphora function it has a specific plural form (najц `they; these'). Parallel to their spatial values, the demonstrative pronouns tajц and sijц may also express proximity resp. remoteness in time. Forms with the prefix e- are described as "strengthening" indication of a referent: etajц 'this here', esijц ' that there'. In discourse, pronouns of the "distant" deixis function as connectives: sijц has anaphoric and cataphoric function, esijц is rare and has cataphoric function. Regarding the adnominal use of demonstratives, the usual word order is Dem_N, but inversed order does also occur. Usually a demonstrative appears adjacent to the head noun or to a preceding attribute (adjective, or participle), cf. (1). In a number of instances other elements interfere, e.g., the verbal predicate as in (2) where the following head noun reminds of an apposition. A further problem addressed to in the paper is parcing of adjacent demonstrative and noun complexes as in (3) where the demonstrative sije may have a pronominal, or an adnominal reading.
(1) Sцen vyl-ц razed-asni sije
jedzid pyz-a a som-se .
flat cake on-ILL part-PRS3PL this.ACC white flour-ADJ bread dough-ACC3SG `This bread dough made from white flour is parted into flat cakes.' (UK5: 8.22)
(2) Sija pysj-is bцr ur-is.
this run-PST3SG back squirrel-3SG `This squirrel ran back.' (UK5: 56.17)
(3) Vok-jas Ivan-es bara kvait-isni, nцjt-isni sije
su-em-i.
brother-PL Ivan-ACC again grasp-PST3PL beat-PST3PL this.ACC(OBL) say-PP-ELA `Again, his brothers took Ivan and beat him up (i) for saying this / (ii) for this saying it.'
(UK5:
32.22)
References: Diessel, Holger 1999. Demonstratives. Form, Function, and Grammaticalization. TSL 42. Amsterdam & Philadelphia: John Benjamins. Fedyunлva, Galina V. 2008. Pervichnye mestoimeniya v permskih yazykah. Ekaterinburg.
86
ЦKK: Fedyunлva, Galina V. (red.). 2000. Цnija komi kyv. Morfologija. Syktyvkar. SKYa.: Lytkin, Vasilij I. (red.). 1955. Sovremennyj komi yazyk I: Fonetika, leksika, morfologiya. Syktyvkar. Uotila, Toivo E. 2006. Syrjдnische Texte V. Komi-Syrjдnisch: Ober-Vycegda-Dialekt. M. Zikins Texte. Ьbersetzt und herausgegeben von Paula Kokkonen. MSFOu 252. Helsinki: SUS. 87
Ei ole kaikki muumit laaksossa ­ a productive idiom construction in Finnish language (Ei ole kaikki muumit laaksossa ­ produktiivinen suomen kielen idiomikonstruktio)
Kristiina Kortelainen, University of Turku Language of presentation: Finnish
There is one highly productive idiom construction in Finnish colloquial language ­ ei ole kaikki x:t y:ssд (`not all the x's in the y'). Numerous figurative utterances following the construction have been created, and they all carry the meaning of some sort of stupidity or unstable state of mind: someone doesn't have all the necessary constituents in a place they are supposed to be, which causes that the person's mind isn't working properly. For example sillд ei ole kaikki muumit laaksossa (`he does not have all the Moomins in the valley') or sillд ei ole kaikki inkkarit kanootissa (`he does not have all the Indians in the canoe') meaning that the person who the speaker is referring to is somewhat stupid or crazy.1
EI `negative verb' ei ei
OLE
KAIKKI
`to be / to `all'
have'
ole
kaikki
ole
kaikki
X:T nominative case plural muumi + t inkkari + t
Y:SSД inessive case singural laakso + ssa kanooti + ssa
On the Internet one can find colloquialisms in blog texts, Internet conversation sites and in social media. Therefore the research data (238 idioms) has primarily been collected from four different web sourcesi, where people have intentionally collected various metaphorical utterances describing stupidity. Adding to this, idioms that follow the construction were found in real life communication situations of everyday life. The research confirmed that the idiom construction ei ole kaikki x:t y:ssд carries itself the meaning of the whole idiom. Lexemes x and y are the free constituents of the template. They can vary freely and yet the meaning of the construction can easily be understood. In other words, the lexemes x (ie. muumi) and y (ie. laakso) can be chosen quite freely by the language user. On the other hand, one cannot infer from the separate constituents the metaphorical meaning of the idiomatic utterance. This supports the idea of the construction grammar: the construction itself has an independent meaning from its constituents.ii Through the research it turned out that the idiom construction is relatively fixed: the construction cannot be transformed without changing the meaning of the whole utterance. Nevertheless it allows morphological and syntactical changes to some extent, such as using plural partitive case instead of nominative case: ei ole kaikkia inkkareita kanootissa (`does not have all Indians in the canoe'). The grammatical research of the construction is mainly based on the norms of the Finnish descriptive grammar book "Iso suomen kielioppi" (2004).iii I also found out that the base of this productive idiom construction originates from the old Finnish utterance ei ole kaikki kotona (`does not have everything at home') which has its first written source dating back to 1644.iv Via Internet sources I found out that the beginning of the idiom constructions productive usage dates back to somewhere between the years 2005 and 2007, and is still going on. It is to say, that the idiom construction at issue is a relatively novel language phenomenon, and therefore it is not possible to form wide-ranging conclusions yet. 1 In Finnish language the pronoun se ('it') is commonly used to refer to also people. Sillд is an adessive form of the pronoun. 88
Nevertheless, it is possible to use the research as guide lines for the further studies that I have recently started. References: i Blog "Catharina's Journal". Available at: . [Reference made 2.3.2011.] Jatkoaika-conversation: Available at: . [Reference made 10.10.2011.] Mikon kotisivut (Mikko's homepage): Available at: . [Reference made 10.10.2011.] Vauva.fi-conversation (1): Available at: . [Reference made 2.3.2011.] Vauva.fi-conversation (2): Available at: . [Reference made 10.10.2011.] ii Goldberg, Adele E. 1995: A Construction Grammar Approach to Argument Structure. The University of Chicago Press, Chicago ­ London. iii Iso suomen kielioppi. Edited by Hakulinen, Auli ­ Vilkuna, Maria ­ Korhonen, Riitta ­ Koivisto, Vesa ­ Heinonen, Tarja Riitta ­ Alho, Irja. Suomalaisen Kirjallisuuden Seuran Toimituksia 950. Helsinki, 2004. iv Formulae puerilium colloquiorvm 1644. Edited by Hдkkinen, Kaisa. Wanhan suomen arkisto 2. Turun yliopiston suomalaisen ja yleisen kielitieteen laitos, Turku 2009. 89
The negativity of Hungarian modal adverbs (aligha, bajosan, nehezen) Nуra Kugler, Eцtvцs Lorбnd University Language of presentation: English This paper investigates linguistic polarity as it is manifested in the system of modal adverbs. After a review of the specialized literature on modal adverbs and the relationship between affirmation and negation, the author highlights the role of construal in the expression of modal force, something that logico-semantic approaches fail to account for. Finally, she presents her own interpretation of the polarity of modal adverbs, drawing on the results of her previous work. Special attention is paid to how negatively construed etymons affect the behaviour of modal adverbs, and to what general conclusions the data afford. This study is based on the written language corpus of the Hungarian National Corpus (http://corpus.nytud.hu/mnsz/), on historical corpora: Hungarian Historical Corpus (http://www.nytud.hu/hhc/), and Hungarian Electronic Library (http://mek.oszk.hu/), and on questionnaire studies (Kugler 2000: 173­5, 2003: 50). In Hungarian, modal adverbs marking the lowest degree of possibility have grammaticalized into epistemic-inferential markers from expressions of a restrictive or negative character. Typical conceptual domains of etymons include DIFFICULTY, EFFORT, TROUBLE, PAIN, and ILLNESS, as also attested by a variety of genetically and structurally different languages. Epistemic-inferential modal adverbs typically mark increased epistemic distance vis-а-vis the ground (and within it any factual information shared by the speaker and the hearer): they construe a situation or event as accessible only through the inferential processes of the speaker. As a result, modal adverbs with a negative or restrictive meaning behave differently from the negative particle itself, and their interpretation as negative cannot be dissociated from the study of grammaticalization. References: Kugler, Nуra 2000. Egy attitdjelцl szуcsoportrуl (A mуdosнtуszуk a modбlis rendszerben). [A subgroup of attitudinal lexical expressions: Modal adverbs in the modal system]. In: Tamбs Gecs (ed.): Lexikбlis jelentйs, aktuбlis jelentйs [Lexical meaning, actual meaning], 168­78. Tinta Kцnyvkiadу, Budapest. Kugler, Nуra 2003. A mуdosнtуszуk funkciуi. [The functions of modal adverbs]. Nyelvtudomбnyi Йrtekezйsek 152. Akadйmiai Kiadу, Budapest. 90
Clausal kus- (`where') interrogatives targeting epistemic congruence in second position Kirsi Laanesoo, University of Tartu Language of presentation: English This study focusses on kus-initial rhetorical interrogatives in Estonian occurring as second pair parts in yes/no question-answer adjacency pairs. These clausal interrogatives begin with the question word kus that in general asks information about place. I will argue that kus in this position has lost its locational meaning. The question orients to the previous turn as problematic in regard to the information pursued in the yes/no question. Traditionally, these kus-interrogatives have been categorised as rhetorical interrogatives because of their ability to convey assertions that do not make relevant an answer. Even though the kus-interrogatives studied here convey some kind of information, I will show that this information provided with these utterances is not sufficient for closing the question-answer sequence. All kus-interrogatives in my data in the position of second pair parts can be interpreted as disconfirming answers to yes/no questions. These kus-initial questions can be classified as reversed polarity questions that convey assertions from the opposite grammatical polarity as described by Koshik (2005).These questions are ambiguous: they convey assertions and at the same time they challenge the relevance of the preceding questions. V: varsti saab kool `lдbi ka=vд. soon get-3sg school over too Q will the school be over soon H:KUS TA `LДBI SAAB KURAT. where he over get-3sg devil how the hell will it be over The analysis shows that the yes/no question in this sequence is treated as problematic in terms of its askability. The yes/no question preceding kus-interrogative asks about information that its recipient should have primary access to. However the recipients orient to these yes/no questions as asking about something that already belongs to asker's territory of knowledge. And therefore the question is not worth asking. In other words kus-interrogative disputes the askability of the yes/no question, implying that there is a problem in epistemic congruence as Hayano has defined (2011). The kus-interrogatives are treated in different ways. Asker of the yes/no question can interpret the kus-interrogative as challenging by subsequently giving an account for asking the yes/no question. If the recipient does not orient to the kus-interrogative as challenging, the question-answer sequence is nevertheless not closed, as the asker of the yes/no question regularly continues with a related question to get the information. The data come from casual audio-recorded conversations between family and friends in the Corpus of Spoken Estonian of University of Tartu. Data includes both, face-to-face and telephone conversations. The method of study is conversational analysis. References: Hayano, K. 2011. Claiming epistemic primacy: Yo-marked assessments in Japanese. In T. Stivers, L. Mondada, J. Steensig (eds.), The morality of knowledge in conversation. Cambridge: Cambridge University Press, 58-81. Koshik, I. 2005. Beyond Rhetorical Questions. Assertive Questions in Everyday Interaction. Amsterdam/ Philadelphia: John Benjamins Publishing Company. 91
Finnish Optative Construction and its Grammaticalization as a Directive Construction (Suomen kielen optatiivirakenne ja direktiiviksi kieliopillistuminen) Yrjц Lauranto, University of Tampere Language of presentation: Finnish The paper discusses clause constructions that begin with a finite verb conjugated in the 2nd person of the conditional mood but which do not accept a pronoun functioning as the subject, e.g. (1) nдk-isi-t tд-n [see-COND-2PS this-ACC] `I wish you could see this', and (2) tul-isi-t nyt [come-COND-2PS now] `just come with us (please)'. My data consists of everyday interaction and electronically mediated conversations in the Internet. The structures I deal with have been called optative constructions. The meaning of the optatives has been described as the wish of the speaker. The construction of type (1) is clearly an optative structure. The paper discusses one of the linguistic co-texts in which optative constructions can be found. Clauses including the structure of type (1) tend to be used in cotexts where there can be seen some kind of problem during the conversation. A solution to the problem is frequently offered to the receiver with an optative construction (without particle nyt `now'). The construction of type (2) is not used in a similar way. This construction tends to be used as a persuasive expression. It can thus be seen as a grammaticalization of directivity, acting together with the imperative clause, which obviously is used much more widely than just as persuasion. As a sign of grammaticalization can be seen the fact that the verbs used in the construction of type (2) usually refer to action that can be controlled by the subject referent. (This is not the case in the construction of type (1): any verb can be used.) Nonetheless, the Finnish interpersonal particle nyt is widely used in many types of syntactical structures. So, it can also be used in optative constructions. This means that sometimes there is actually no formal difference between the structure of type (1) and the structure of type (2). Key words: optative construction, persuasive construction, directives, grammaticalization, conditional mood, interpersonal particles. 92
Finnish circumstantials and frame relevance. Time, place, and manner as arguments vs. adjuncts (Suomen kielen sirkumstantiaalit ja kehysrelevanssi: aika, paikka ja tapa argumentteina vs. adjunkteina) Jaakko Leino, University of Helsinki Language of presentation: Finnish Circumstantials, (i.e. expressions of time, place, manner, etc. or "temps, lieu, maniиre, etc." in the words of Tesniиre 1959: 103), tend to be syntactically optional, i.e. adjuncts rather than arguments. However, this is not always the case: quite typical circumstantials may also act as core arguments of the verb. Most obvious cases are expressions of location such as Virtakatkaisin sijaitsee koneen takana (`The power switch is located at the back of the machine'), where the adverbial koneen takana (`at the back of the machine') is obligatory: without it the sentence would be incomplete both syntactically and semantically (*Virtakatkaisin sijaitsee, *The power switch is located). Overall, while the distinction between arguments and adjuncts is related to such semantic categories as time and place, it cannot be reduced to those categories. To explicate the semantic grounds of the argument­adjuct distinction, with special reference to Finnish data, I will introduce the notion of frame relevance. It appears that circumstantial meanings become arguments, rather than adjuncts, to the extent that they are relevant with regard to the event or semantic frame described by the sentence. For example, the relevance of the expression of place baarissa (`at the bar') is obviously different for the event types being, dancing, and thinking in Jussi [oli ~ tanssi ~ ajatteli Maijaa] baarissa (`John [was ~ danced ~ though of Mary] at the bar'), and this has a strong effect on its syntactic status as well. It also appears that there are not only "obligatory" argument-like expressions and "optional" adjunct-like expressions in the traditional valency-driven sense, but also intermediate ones. In different traditions, strikingly similar suggestions have been made to the effect that some arguments are brought into the sentence by the verb's valency, while others are added by other mechanisms which augment the valency in one way or another (e.g. Goldberg 1995, Kay 2005, Rбkosi 2006). Essentially, the verb contributes to the overall frame expressed by the sentence, but its contribution is complemented by other elements, and arguments may be participants required--or made relevant--by either the verb or other elements. Ultimately, the analysis based on frame relevance leads to a continuum rather than a binary or tertiary distinction (cf. Croft 2001: 272­280). We may think of this continuum in terms of entrenchment: to the extent that a given circumstantial element is a conventionalized part of an event type (or a conceptually inseparable element which must be expressed when using a given linguistic resource to express that event type), that element counts as an argument rather than an adjunct. Denying the existence of the argument­adjunct distinction would be too strong a conclusion, however. The distinction is not only intuitively correct but also justifiable: the great majority of adverbial modifiers are fairly indisputably either arguments or adjuncts. The more disputable grey area between the clear-cut cases has, time and again, proven difficult to explicate, but the relative relevance of the participants to the overall frame offers a useful, cognitively oriented and theoretically anchored way to tackle the borderline cases. References: Croft, William 2001. Radical Construction Grammar. Oxford: Oxford University Press. 93
Goldberg, Adele E. 1995: Constructions: A Construction Grammar Approach to Argument Structure. Chicago: University of Chicago press. Kay, Paul 2005. Argument Structure Constructions and the Argument-Adjunct Distinction. In Fried, Mirjam & Hans C. Boas (eds.): Grammatical Constructions: Back to the Roots. Constructional Approaches to Language 4. Amsterdam: John Benjamins. 71­98. Rбkosi, Gyцrgy 2006. On the need for a more refined approach to the argument­adjunct distinction: The case of dative experiencers in Hungarian. In Butt, Miriam & Tracy Holloway King (eds.): Proceedings of the LFG06 Conference, Universitдt Konstanz. Stanford: CSLI Publications. Tesniиre, Lucien 1959. Йlйments de syntaxe structural. Paris: Klincksieck. 94
Borrowed function and discourse words in Komi Marja Leinonen, University of Helsinki Language of presentation: English Various scales of borrowability have been suggested, with much the same principles. However, the Finno-Ugrian languages spoken in Russia are generally absent, although they present a wealth of material for contact studies. In this paper, I shall continue my earlier topic of Komi function words borrowed from Russian, where conjunctions only were treated. Discourse markers are found to be among the most common linguistic items susceptible to borrowing. Y. Matras (2007, 2009), who is especially interested in developing a borrowability hierarchy of grammatical function words, presents the following scale: 1) discourse markers and connectors, 2) phasal adverbs and focus particles, 3) indefinites and interrogatives, 4) expressions of temporal and local relations, 5) numerals, 6) place deixis, demonstratives, and personal pronouns, 7) negators, possessors and existentials. For each group, discourse pragmatic and semantic motivations are given; a specific group is utterance modifiers. Komi, representing originally a situation of isolated contact (a population of relatively mobile men and sedentary monolingual women), has developed a situation of unidirectional bilingualism with diglossia. It is comparable in this respect to the languages researched by Matras, which are minority languages in colonial settings (and, specifically, various Romani vernaculars). The difference is that Komi has a status of a written official language, codified in grammars and dictionaries, used in literature and mass media. During several centuries of contact, the language has borrowed a wealth of Russian items from groups 1-2 (da 'yes', vot; conjunctions da, i, kцtґa, byttґц; particles цd, essц, tolґkц...), from group 3 (indefinites; myj-lґibo), from 4 (peripheral local relation bokyn 'by the side of'). Numerals (group 5) are taken from Russian in dates and other institutional environments. In group 7 Komi shows negatives (nґe sцmyn, nґe + infinitive). Perhaps more cases will be found (group 6?), when spoken language and dialect corpora are included. My aim is to check and complement the grouping suggested by Matras with material taken from dictionaries and dialect material. References: Leinonen, Marja 2002: Influence of Russian on the syntax of Komi. Finnisch-ugrische Forschungen, Band 57, Heft 1-3, Helsinki 2002, 195-358. Leinonen, Marja 2009: Russian influence on the Izma Komi dialect. International Journal of Bilingualism, Volume 13, Number 3, 309-330. Matras, Yaron 2007: The borrowability of structural categories. Grammatical borrowing in cross-linguistic perspective, ed. Yaron Matras,, Jeanette Sakel. Berlin, New York: Mouton de Gruyter. 31-73. Matras, Yaron 2009: Language Contact. Cambridge Textbooks in Linguistics. Cambridge University Press. 95
Variation in necessive constructions in Estonian dialects
Liina Lindstrцm, University of Tartu Kristel Uiboaed, University of Tartu Language of presentation: English
This paper sheds light on the variation in necessive constructions with modal predicates tarvis/vaja olema ' to need' in Estonian dialects. In Estonian grammars (EKG II, Erelt, to appear) modal predicates tarvis olema and vaja olema have always listed as synonymous predicates expressing the same meaning and functions. tarvis/vaja olema can form two related constructions in Estonian: either with a nominal complement (1) or with an infinitive (2).
(1) Mu-l on vaja uut I-ADE be.3SG need new.PRT 'I need a new computer' (2) Mu-l on vaja koju I-ADE be.3SG need home.ILL 'I need to go to home'
arvuti-t. computer-PRT minna. go.INF
Estonian dialect syntax is a rather understudied field in Estonian linguistics. The present work is innovative at least in two respects. On the one hand, most of the studies to date mainly describe particular syntactic phenomena in particular (sub)dialects but there is no comprehensive works on syntactic variation in Estonian dialects. On the other hand, previous works usually do not consider corpus or any frequency data. In dialect syntax studies the frequency data has lately gained more and more attention. According to Szmercsanyi, text frequencies better match the perceptual reality of linguistic input than discrete atlas classifications (2013: 4). In the present paper the frequency data is in the central focus. We show that the variational patterns of tarvis/vaja olema diverge remarkably from the picture given in Estonian dialect dictionaries (e.g. VMS) when we consider frequencies. So we demonstrate how natural corpus data can provide completely new insights to syntactic phenomena that easily remain unnoticed in discrete dialect atlas data. Our data is obtained from the Corpus of Estonian Dialects compiled at the University of Tartu (http://www.murre.ut.ee/estonian-dialect-corpus/), it represents spoken dialect speech recorded between 1957-1980. We have detected 351 instances of tarvis or vaja. The data has been coded for the following lexical and constructional features: tarvis or vaja, verb ellipsis, case marking of the Experiencer, ellipsis of the Experiencer, infinitival or nominal complement, dialect area. statistical analysis is conducted based on the coded features. According to preliminary results, there is a clear regional preference to tarvis olema in Western and Northern dialects, while vaja olema is used mainly in Southern and Eastern dialects. This fact differs from VMS where the two predicates show very similar distributions in dialects. Also verb ellipsis occurs more often in Eastern and Southern dialects, esp. in Seto dialect. Marking the Experiencer does not seem to have significant geographical preferences but is connected to the presence of the infinitival complement (like in ex. 2). In our presentation we demonstrate the results of our statistical study and show how frequency data enables us to see a different distributional picture of syntactic phenomena in dialects.
References: EKG II = Mati Erelt, Reet Kasik, Helle Metslang, Henno Rajandi, Kristiina Ross, Henn Saari, Kaja Tael, Silvi Vare 1993. Eesti keele grammatika II. Sьntaks. Lisa: Kiri. Tallinn: Eesti Teaduste Akadeemia Keele ja Kirjanduse Instituut.
96
Erelt, Mati, to appear. Eesti keele lauseхpetus. Sissejuhatus. Цeldis. Manuscript. Szmrecsanyi, B., to appear. Forests, trees, corpora, and dialect grammars. In B. Szmrecsanyi & B. Wдlchli, eds. Aggregating dialectology, typology, and register analysis: Linguistic variation in text and speech. Berlin: Mouton De Gruyter. VMS = Vдike murdesхnastik, http://www.eki.ee/dict/vms/ 97
L2 production of native Finnish speakers, in the light of the negative (and positive) transfer (target language is Hungarian) Szilvia Magyar, Eцtvцs Lorбnd University Language of presentation: English The following lecture is a smaller part of my PhD thesis in which I intend to investigate the negative transfer and its influence on using and learning L2 if the second language is related closely or distantly or if it is not related to the native language at all. The second site of my research was in Finland, in Jyvдskylд. I can summarize my research as follows. If the learner is a Finnish native speaker and the target language is Hungarian that is distantly related to the (Finnish) native language, what kind of problems can be found, is there a negative transfer in the L2 production, if there is, how strong is it, and can the learners mention cases of the positive transfer? (At same time, parallel to the above mentioned research I did the same research with English as a target language but I would like to speak only about the Finnish-Hungarian relationship at this conference.) I started my research with observing language lessons and language output at the University of Jyvдskylд, at the department of Hungarology. My basic point is: if the same mistake emerges in many cases, with different informants, it can be explained by the negative transfer of the native language. In addition to the observations I made interviews as well, and after that, based on the data of the observations, literature and interviews, I built up a questionnaire that was filled in by native Finnish speaker informants who were learning Hungarian. I would like to present the result of this research. Number of informants: approx. 30, age between 18 and 82 years, average language level is B2. The questionnaire contained questions about the possible positive transfer as well. (The questionnaire based on an indirect language test.) Data collecting and analyzing is still under process, final result is expected by the end of March. The theoretical background of the research is based on: David CRYSTAL, A nyelv enciklopйdiбja, Osiris, 1998. Heidi DULAY ­ Marina BURT ­ Stephen KRASHEN, Language 2, Oxford University Press, 1982. Theoretical backgroung and additional informations for building up the questionnaire: Ulla-Maja PASANEN ­ Anssi HIETANEN, Peruskoulun 9. luokan valtakunnallinen englannin kielen koe ja sen kehittдminen, Jyvдskylдn Yliopisto Kasvaustieteen laitos, 1994. KESZLER Borbбla (szerk.), Magyar Grammatika, Nemzeti Tankцnyvkiadу, 2000. Maija-Hellikki AALTIO, Finnish for foreigners, Otava, Keuruu, 1985. Lбszlу KERESZTES, Unkarin kieli, Suomalaisen kirjallisuuden seura, Helsinki, 1974. VARGA Judit ­ SAARINEN Sirkka: Veikц kissa kielen? Finn-magyar frazeolуgiai szуtбr, Pбzmбny Pйter Katolikus Egyetem Bцlcsйszettudomбnyi Kar, Piliscsaba 2000. Marjukka KENTTДLД, Kieli kдytццn, Helsingin yliopisto kielikeskus Ilona SCARCIELLO, Suomea leikiten, Jyvдskylд, 2006. 98
Essive in contemporary Votic1 Elena Markus, University of Tartu, Institute of Linguistics of the Russian Academy of Sciences Fedor Rozhanskiy, University of Tartu, Institute of Linguistics of the Russian Academy of Sciences Language of presentation: English Similar to other Finnic languages, Votic had a set of three cases denoting temporal state or location as viewed from three different directions, roughly "turning into a state/location" (Translative, marked with -ssi), "being in a state/location" (Essive, marked with -n/-nn/-nn), and "leaving a state/location" (Excessive, marked with -nt). In contemporary Votic, Excessive is no longer used (except for a couple of lexicalised forms: kotont `from home', takant `from behind'), Translative is fully productive, and the use of Essive is grammatically and lexically restricted. In this presentation we will give a preliminary overview of the functions of Essive in contemporary Votic. The study is based on the field materials (narratives and elicitations) that we collected in the period 2001-2012 from speakers of Jхgхperд and Liivtsьlд Votic. With Votic being on the verge of extinction, it does not seem possible to collect many new examples of Essive, and to thoroughly study semantic restrictions in its use. A preliminary search through the corpus has shown that Essive occurs in Votic in non-verbal main predications, in secondary predications, and also denotes temporality and location. In the temporal and locative meaning the use of Essive is mainly lexicalised. An interesting point concerns the distribution of Essive and other cases that mark temporality. Essive marks seasons, months, days of the week, and times of the day. Votic allows considerable variation here, but there are preferences depending on a concrete lexeme. For example, in the contemporary Votic `in the evening' is ohtogonn `evening.Ess' or ohtogoss `evening.Elat'; `in the morning' is mniko `morning.Ess' or mnikoss `morning.Adess', but not *mnikonn `morning.Ess'; `at night' is `ll night.Adess' but not *nn `night.Ess'. Essive can encode predicative nominal and predicative adjective in non-verbal main predications and in secondary predications (tдmд e li seppen `He was a smith.Ess', tдmд tahto e ustin `He wanted to be handsome.Ess'). Essive is used both with animates and inanimates (kase linn e li tsьlдn `This town used to be a village.Ess'), and occurs only in the singular form. In predicative constructions, the competition of Essive, Nominative, and Translative depends on a number of factors, including the tense, the verb, and the meaning of the whole construction. In contemporary Votic, there are almost no contexts where Essive could not be replaced with Nominative. There is a tendency to prefer Essive in the past tense, and Nominative in the present (cf. minu bratko on se tamZ `My brother is a soldier.Nom' and minu bratko e li se tamehen `My brother was a soldier.Ess'). In the constructions referring to the future, Essive occurs very rarely. Normally, either Nominative or Translative is used depending on the verb (tдmд ne B papissi `He will become a priest.Trans', tдmд lB pappi `He will be a priest.Nom'). 1 The research was supported by the Russian Foundation for Humanities, project 12-04-00168a. 99
On aspect's place in the Estonian differential argument marking system Helena Metslang, University of Tartu Language of presentation: English There are four argument types in Estonian that, to a certain extent, show case-alternation between the partitive and total cases (the nominative and/or genitive): the direct object, the existential clause NP (e-NP), the object-like degree adverbial (OLDA; cf. Haspelmath 1997) and the predicate nominal. It has been suggested that aspect influences the case-marking of Estonian objects, e-NPs and OLDAs (Erelt et al. 1993; Kont 1962; Metslang 2007; Nemvalts 2000; Tamm 2004). The case-marking systems of these argument types have in most cases been studied rather separately, for example by Erelt et al. (1993; object, e-NP and OLDA); Kont (1963; object); Nemvalts (2000; e-NP); Metslang (2007; 2008 and to appear a; OLDA and eNP), Rannut (1958; OLDA). In the case of Estonian, the factors influencing different arguments' case-assignment have been compared for example by Tamm (2004) and Metslang (to appear b). Huumo (2010) has provided a similar comparison in Finnish and KoptjevskajaTamm and Wдchli (2001) in the context of the Baltic Sea region. This paper compares the case-marking conditions of the object, e-NP and OLDA and the place of aspect among them. These arguments largely depend on the same case-marking system that includes all argument realization levels: referential (NP level) and clausal factors and lexical predicates (cf. Dixon 1994). The set of case conditions studied here includes polarity, inclusivity-related hierarchies (Metslang to appear b), aspect, Aktionsart, temporal meaning, the NP's lexical/semantic properties, the verb's lexical/semantic properties, the impact of the whole construction and the typologically suggested referential hierarchies. However, their application has some significant differences between these arguments. The aspectual bounding of the situation may occur on the lexical level of the predicate but also by the use of other elements of the clause (e.g. object case, adjunct case, particles; Tamm 2004; Metslang 2001; Vaiss 2004). Nemvalts (2000) suggests that also e-NP's case is an expression means of situational bounding. The talk discusses the question of independence vs. the epiphenomenon status of case-assignment factors. It confirms that situational aspect can be the independent factor of object's and OLDA's case but not of e-NP. However, the alternation between bounded and unbounded situational meaning can trigger different case-alternations of the object and the OLDA: between the total cases and the partitive in the case of object and usually between the nominative and the genitive in the case of OLDA (cf. Erelt 2000: 96 and Metslang 2007). The small-scale aspect-focused study is based on the written data from the Corpus of Written Estonian. Preliminary results will be presented on the case-use prototypes (in the sense of Aarts 2010 and Gries 2003) and factor frequencies of each argument type. References: Aarts, Bas (2007). Syntactic Gradience. The Nature of Grammatical Indeterminacy. Oxford: OUP. Dixon, Robert M. W. (1994). Ergativity. Cambridge: Cambridge University Press. Erelt, Mati (2000). Sihitis. ­ Keelenхuanne soovitab 2. Tallinn: Eesti Keele Sihtasutus, lk 91-98. Erelt, Mati; Kasik, Reet; Metslang, Helle; Rajandi, Henno; Ross, Kristiina; Saari, Henn; Tael, Kaja and Vare, Silvi (1993). Eesti keele grammatika II. Sьntaks, lisa: kiri [Estonian grammar II. Syntax (Appendix: Script)]. Tallinn: Estonian Academy of Sciences, Institute of Language and Literature. Gries, S. Th. (2003). Towards corpus-based identification of prototypical instances of constructions. Annual review of cognitive linguistics. Haspelmath, Martin (1997). From Space to Time. Temporal Adverbials in the World's Languages. LINCOM Studies in Theoretical Linguistics 02. Mьnchen / Newcastle: LINCOM Europa. 100
Huumo, Tuomas (2010). Nominal aspect, quantity, and time: The case of the Finnish object. Journal of Linguistics, 46(1), 83-125. Kont, Karl (1963). Kддndsхnaline objekt lддnemeresoome keeltes. Eesti NSV Teaduste Akadeemia Keele ja Kirjanduse Instituudi uurimused IX. Tallinn. Koptjevskaja-Tamm, M., Wдchli, B. (2001). The Circum-Baltic languages. An areal-typological approach, in Dahl, Ц., Koptjevskaja-Tamm, M. (eds.) Circum-Baltic Languages, vol. 2: Grammar and typology, Studies in Language Companion Series 55, Amsterdam: Benjamins, 615-750. Metslang, Helena (2007). Two types of aspectual opposition of estonian duration adverbials? . Monticelli, Daniele; Treikelder, Anu (Toim.). L'aspect dans les langues et les thйories: similitudes et diffйrences. Aspect in languages and theories: similarities and differences (77 - 90). Tartu: Tartu Ьlikooli Kirjastus Metslnag, Helena (2008). Sihitisesarnased mддrused: kui sarnased? Metslang, Helle; Grьnthal, Riho (Toim.). Lдhivertailuja (108 - 143). Helsinki: Suomalais-ugrilainen seura. Metslang, Helena (to appear a). On the case-marking of existential subjects in Estonian. SKY Journal of Linguistics, 25. Metslang, Helena (to appear b). Partitive noun phrases in the Estonian core argument system. Tuomas Huumo, Silvia Luraghi (Eds.). Partitives. Proceedings of the Partitives workshop at SLE 43rd Annual meeting in Vilnius. DeGruyter Mouton. Metslang, Helle (2001). On the Developments of the Estonian Aspect: The Verbal Particle дra, in Dahl, Ц., Koptjevskaja-Tamm, M. (eds.) Circum-Baltic Languages, vol. 2: Grammar and typology, Studies in Language Companion Series 55, Amsterdam: Benjamins, 443-479. Nemvalts, Peep (2000). Aluse sisu ja vorm: alusfraasi kддndevaheldus tдnapдeva eesti kirjakeeles [The subject's meaning and form: subject case alternation in contemporary Written Estonian]. Tallinn: Estonian Language Foundation. Rannut, Lehte (1958). Objektisarnane ajamддrus eesti keeles. ­ Keel ja Kirjandus I:10, lk. 604-609. Tamm, Anne (2004). Relations between Estonian Verbs, Aspect, and Case. PhD thesis, Budapest. Vaiss, Natalia (2004). Eesti keele aspekti vдljendusvхimalusi vene keele taustal (= Means to express aspect in Estonian, on the background of Russian). MA thesis, Tallinn University. 101
Do I guess or ask? Epistemic modality as a source of interrogativity Helle Metslang, University of Tartu Karl Pajusalu, University of Tartu Kьlli Habicht, University of Tartu Language of presentation: English The paper discusses how the interrogative use of epistemically modalized sentences emerged in Estonian and whether there are any signs of modality markers grammatisizing into interrogative markerss. We will also look at the position of clauses that have modalized in different ways on the semantic scale ranging from asking about absent information to asking for confirmation on the speaker's assumption. Epistemically modalized statements like Дkki Anne sхidab Budapesti 'Perhaps Anne will travel to Budapest' and questions based on the latter, e.g. Дkki Anne sхidab Budapesti? `Perhaps Anne will travel to Budapest?' explicitly reveal the subjective component of the question: the uncertainty of the speaker's knowledge. Since modal particles in Estonian have developed from words with non-modal meanings (дkki 'perhaps' < 'suddenly'), this type of questions represents the development nonsubjectification > subjectification > intersubjectification (cf. Traugott 2010). The paper focuses on historical and current grammaticalization processes in Estonian. Our usage-based analysis relies on the Corpus of Old Written Estonian, Corpus of Estonian Literary Language (1890-1990), Corpus of New Media, and Corpus of Spoken Estonian of the University of Tartu. It combines the syntactic, semantic and cognitive approaches to sentences in text (discourse) that is by itself regarded as a major construction containing markers of cross-clausal features and functions. We assume three grammaticalization sequences: in the genesis of utterance types, the sequence statement > question; in the genesis of sentence types, the sequence declarative sentence > interrogative sentence; in the genesis of speech acts, the sequence assertion of a proposition > asking something about the proposition. The development mechanism of the polar question could be explained by the pragmatic-logical structure of the question which has two principal components: 1) presupposition of the question (the truth value of the proposition can be true or not true, i.e. assertion or negation of the proposition) NEC P V NEC (NEG P) POSS P & POSS (NEG P) 2) directive function (asking for a statement of whether the proposition or its negation is true). The presupposition indicating uncertainty of the speaker's knowledge is the subjective component that is directed at the speaker. The directive component is inter-subjective and is directed at the recipient. Different utterances in a conversation can acquire interrogative interpretation if a suitable relation occurs between the knowledge of the speaker and the listener. These utterances may express a suggestion of a possible inference, continuation or alternative of the previous conversation, or an epistemic modalization of the sentence content. Sentences that typically form such utterances can stay fluctuating along a continuum from statements to interrogativity, thus acting as an extensive breeding ground for questions. However, some of them develop into proper interrogative sentence structures, where question markers can be for example particles of coordinative origin, as is the case in Estonian and other languages in the Circum-Baltic region (Metslang et al. 2011). 102
References: Metslang, Helle & Kьlli Habicht, Karl Pajusalu 2011. Developmental paths of interrogative particles: the case of Estonian. ­ Folia Linguistica Historica 32, 2011, 149­188. Traugott, Elizabeth Closs 2010, (Inter)subjectivity and (inter)subjectification: A reassessment. - Davidse, Kristin, Lieven Vandelanotte, Hubert Cuyckens (eds.), Subjectification, Intersubjectification and Grammaticalization. (Topics in English Linguistics 66.) Berlin / New York: Walter de Gruyter, 29-74. 103
Previous knowledge and the format of repair-initiating polar questions Krista Mihkels, University of Tartu Language of presentation: English When people talk together they frequently encounter problems of hearing, speaking and understanding. According to conversation analysis, there exists an organised set of practices, the repair organisation, through which participants in conversation address and resolve problems of speaking, hearing or understanding.Repair can be initiated either by the speaker of the troublesource or by the co-participant (Schegloff, Jefferson, Sacks 1977; Schegloff 2000;Sidnell 2010). The verbal techniques that co-participants use to initiate the repair rely on many forms of utterance, among them polar questions. The main purpose of the presentation is to analyse the interactional function of differently formed polar questions during the other-initiated repair sequences in Estonian conversation. Secondly, the correlation of the selection of specific question format and epistemic statusof the participants(for details see Heritage 2012) will be discussed. The research is based 1) on a corpus of video-recordings of 1­4 grade lessons of Estonian and Mathematicsgiven in schools in Tartu, Estonia and transcripts of those tapes; 2) on a selection of naturally occurring situations from other institutional settings and conversations between family members and friends taken from the Corpus of Spoken Estonian, University of Tartu.The principal theoretical and methodological framework of the present study is conversation analysis. On the basis of formal question marking, the polar questions that co-participants used to initiate a repair can be divided into following groups. · Questions with preceding particle kas`whether` kaskaelkirjakustolitдnases loos juttu QUES giraffe-ELA was today-INE story-INE talk-PART was there talked about giraffe in today`s story · Questions formed by verb inversion on=suligalpildil is you-ADE every-ADE picture-ADE have you on every picture · Questions with preceding particle et `that` and/or final particle jah `yes` or vхi`or` vхimendusvхi amplifieror etsul=oli tore pдev=sis vanaema=juures=jah THAT PRT you-ADE was nice day then grandmother-GEN POSTPOS YES you had a nice day by your grandmother then · Questions with a construction you mean X pьsivдrvemхtletejah permanent colour-PART mean-PL2YES you mean permanent colours · Prosodically marked questions @ vaal, @ 104
whale · Declarative questions without question marking Erki=on=sinu=`vend Erki is your brother In the presentation I will demonstrate that differently formed polar questions perform different tasks during the other-initiated repair sequence. The form of the polar question is dependent upon the specific interactional function the question bears and the participants` territories of knowledge. · Locating and highlighting the error or unexpected item in the in the prior turn was implemented by prosodically marked repetitions of the trouble-source. · Repetitions of the trouble-source+particlesvхi`or` and jah`yes` tended to be connected with other problems, theydid not react to the error in the prior turn. · Candidate understandings about previous turns were offered by declarative questions without question marking; questions with preceding particle et `that`; final particlesjah `yes` and variants of particle vхi`or`; construction you mean X. · An error in the whole prior turn was indicated by questions with preceding particle kas`whether`or questions formed through verb inversion. References: Heritage, J. 2012.Epistemics in Action: Action Formation and Territories of Knowledge.­ Research on Language and Social Interaction 45, 1­29. Schegloff, Emanuel A. 2000. When "others" initiate repair. ­ Applied Linguistics 21, 205­243. Schegloff, Emanuel; Jefferson, Gail; Sacks, Harvey 1977.The preference for self­correction in the organisation of repair in conversation. ­ Language 52 (2), 361­382. Sidnell, Jack 2010. Conversation Analysis.An Introduction.Wiley­Blackwell. A John Wiley&Sons, Ltd. Publication. 105
Heterogeneous case distribution in NPs in Estonian: puzzles for a theory of case
Merilin Miljan, University of Tartu Language of presentation: English
It is well known that case assignment in numeral constructions in Finno-Ugric and Slavonic languages is difficult to analyse. In Estonian, for example, numerals other than `one' have their complements in partitive (genitive in Slavonic) when occurring in nominative (1) or what is often referred to as accusative in the object position (2). The complement of the numeral is not marked for plural, although semantic plural is assumed. If the numeral bears a different casemarker than nominative, then all the adnominal elements have the same, homogeneous casemarking (3). As noted by, e.g. Brattico (2011), this heterogeneous case distribution, as in (1) and (2), poses a problem for those theories whereby syntactic case is first assigned to a whole phrase and then distributed to the adnominal elements within a noun phrase by some special concord rule (e.g. Chomsky 2001) or feature-sharing rule (e.g. Malouf 2000 in HPSG).
(1) kaks
pliiatsit
two.NOM.SG pencil.PART.SG `two pencils'
(2) Ma
leidsin
[kaks
pliiatsit].
1SG.NOM find.PST.1SG two.NOM.SG/ACC.SG `I found two pencils.'
pencil.PART.SG
(3) kahele
teravale
pliiatsile
two.SG.ALLATIVE sharp.SG.ALLATIVE `onto two sharp pencils'
pencil.SG.ALLATIVE
Brattico (2011) offers an account where case concord is a `byproduct' of one-to-many case assignment. Yet this case assignment mechanism falls short in the light of another type of data in Estonian, viz. adjectives which do not agree with the noun head in any case but occur in partitive instead (4) and (5).
(4) kollast
vдrvi
pliiats /
pliiatsid /
yellow.PART.SG colour.PART.SG pencil.NOM.SG / pencil.NOM.PL / `a yellow pencil/ yellow pencils /onto yellow pencils'
pliiatsitele pencil.PL.AD
(5) lambanahka sheep.skin.PART.GS `a sheepskin coat'
kasukas coat.NOM.SG
Further, differential case-marking may occur in the complement position of a quantifier in quantificational constructions (6), where partitive alternates with elative case.
(6) osa
poisse /
poistest
part.NOM.SG boy.PART.PL / boy.PL.ELATIVE `some boys / some of the boys'
The data above pose several questions, such as how to account for (i) the differential case assignment in (6); partitive in certain adjectival phrases, and (iii) any heterogeneous case distribution apart from numeral constructions. In this paper, a rather radical hypothesis is examined which, rather than offering yet another system of case assigning mechanism, tries to
106
account for the problematic data via rethinking the concept of case. If abandoning the idea that case is a passive realisation of other morpho-syntactic properties of a construction (i.e. being assigned), and assuming that case-markers independently bring information to a clause or construction in which they appear, the facts in (1) to (6) can be explained easily. Evidence for the new analysis is presented and its explanatory potential, which partly relies on inferences made in context, demonstrated. References: Chomsky, Noam. 2001. Derivation by phase. In Michael Kenstowicz (ed.) Ken Hale: A life in Language, 1­52. Cambridge, MA: MIT Press. Brattico, Pauli. 2011. Case assignment, case concord, and the quantificational case construction. Lingua 121, 1042- 1066. Malouf, Robert. 2000. A head-driven account of long distance case assignment. In Ronnie Cann, Claire Grover & Philip Miller (eds.), Grammatical Interfaces in HPSG, 201­214. Stanford: CSLI Publications. 107
Making referents accessible in multi-party interaction Chiara Monzoni, University of Helsinki Ritva Laury, University of Helsinki Language of presentation: English During face-to-face interaction speakers are engaged in both verbal and non-verbal activities which take place in a physical context, The context may include concrete entities which can be referred to, topicalized and brought about into the conversation through different interactional activities (like offering, asking questions, making assessments, etc.). In earlier research on information structuring, it has often been assumed that referents are cognitively accessible to participants in the conversation merely through their physical presence in the context (e.g. Ariel 1990: 58-61; but see Chafe 1994: 79). However, our research shows that participants often engage in preparatory work prior to making verbal reference to concretely present entities. We will focus on some of the ways in which referents in the physical context are introduced in interaction and referred to in Finnish and Italian multi-party conversational data. Finnish and Italian make an interesting language pair for this study, because both allow predications without explicit mention of referents being predicated on (`zero'), but in both languages, pronouns are also used to index referents, and verbs show person marking. Using a conversation analytic approach (e.g. Sacks, Schegloff & Jefferson 1974; Hakulinen (ed.) 1989), the study focuses both on the linguistic resources and especially on the multi-modal resources (such as gaze, gesture and body posture) speakers employ in order to introduce a `new' referent in the talk. We will demonstrate that linguistic resources may be differently distributed depending on the kind of activity that is being done. Among the linguistic resources we will discuss are marked syntactic constructions and zero reference. Most importantly, we will show that multi-modal resources are an integral part of the interactional `referential work' being done in conversation. Multi-modal resources are in fact employed for a series of different `functions', such as: achieving and overtly displaying access to co-present referents in order to isolate and focalize in the context the potentially upcoming referent in the talk. In other words, it seems that it is not sufficient for speakers to indicate a referent through linguistic means, but they often do preparatory work both to display their own access to the referent and to make that referent accessible also to other participants before or concurrently with verbal activities. In other words, having access to co-present referents is made salient in the interaction through multi-modal resources, thereby engaging recipients in a focalization activity before the actual verbal reference to the co-present object is produced. These multi-modal resources are also employed to contextualize the upcoming verbal activities. 108
The development of non-canonical argument structures in Finnish
Maximilian Murmann, Ludwig Maximilian University Munich, University of Helsinki Language of presentation: English
The Finnish language is equipped with a wide array of constructions that do not conform to canonical argument structures. They are instantiated by different kinds of psych verbs, denoting perception, cognition and emotion. Although their exact status remains to be determined, these constructions can be divided into semantically specific and semantically general constructions. Whereas the latter have been the focus of attention, the former, such as `locative' object constructions, have generated less interest within the linguistic community (cf. Sands & Campbell 2001: 286). In my research I will consider both compositional combinations of units and seemingly irregular constructions. The number of idiosyncrasies within and across languages is certainly one reason for the limited interest in some of the non-canonically case-marked arguments, but it inevitably leads to the question how and when these patterns came into being. I consider the emergence of argument structures as the result of constant reorganization of the constructional inventory, which has to be seen in light of cognitive aspects and language contact. The methodological framework of this research is based on Construction Grammar, which has only recently been extended to diachronic issues. It remains to be seen how insights from the grammaticalization approach can be incorporated into this framework (Noлl 2007: 186). I will discuss whether the different ways of using semantic cases (and adpositions) as argument markers in Finnish should be considered as inherited from Proto-Finnic or whether they developed independently. Despite the essential differences between, for example, Finnish and Estonian argument structures, one also has to acknowledge the amount of overlap across the Finnic languages (Schlotthauer 2011: 153), as in example (1). Furthermore, examples (3) and (4) illustrate that Old Literary Finnish featured a postpositional argument structure, presumably a loan translation, which is still common in contemporary Estonian.
(1) a. Hдn usko-o
om-i-in
kyky-i-hin-sд [Finnish]
PN.3SG believe-3SG own-PL-ILL ability-PL-ILL-POSS.3SG `S/he believes in her/his own abilities' (Nissilд 2001: 90)
b. Ta usu-b
oma vхime-te-sse [Estonian]
PN.3SG believe-3SG own ability-PL-ILL `id.' (Nissilд 2001: 90)
(2) a. Hдn usko-o
Jumala-an [Finnish]
PN.3SG believe-3SG God-ILL `S/he believes in God' (Nissilд 2001: 90)
b. Ta usu-b
Jumala-t [Estonian]
PN.3SG believe-3SG God-PTV `id.' (Nissilд 2001: 90)
(3) a. Minu-un voi-t luottaa [Finnish]
PN.1SG-ILL can-2SG trust `You can trust in me' (Nissilд 2001: 46)
b. Minu
pea-le vхi-d loota [Estonian]
PN.1SG.GEN on-ALL can-2SG trust `id.' (Nissilд 2001: 46)
109
(4) a. Parambi on
vskalta Herra-n pд-le quin lootta Inhimis-ten
better be.3SG take.refuge Lord-GEN on-ALL than trust human-PL.GEN
pд-le
on-ALL `It is better to take refuge in the Lord than to trust in humans' (A-III-387-[Ps-118:8])
b. Parambi on
uscalda Herra-n pдд-lle cuin luotta ihmis-i-jn
better be.3SG take.refuge Lord-GEN on-ALL than trust human-PL-ILL 'id.' (B1-[Ps-118:8]-318c)
c. On parempi turvata Herra-an kuin luottaa ihmis-ten
apu-un
be.3SG better take.refuge Lord-ILL than trust human-PL.GEN help-ILL 'It is better to take refuge in the Lord than to trust in (the help of) humans'
(PR 1992-[Ps-118:8])
Due to the lack of common historical evidence, one has to take advantage of the comparative method to shed light on the earliest development of Finnish argument structures. As there is no principled distinction between syntax, morphology, semantics and other levels of language within Construction Grammar, cognate argument structures can be identified through case marking and a comparison of the lexical predicates instantiating them. Historical data from Finnish will help to clarify later developments, as in example (4). The argument constructions will be systematically ordered, analyzed and presented in semantic maps to capture implicational hierarchies and relations between grammatical categories (cf. Barрdal et al. 2012). In my talk I will go into methodological aspects of reconstructing argument structures and discuss the development of some `locative' object constructions in Finnish.
References: Barрdal, Jуhanna ­ Smitherman, Thomas ­ Bjarnadуttir, Valgerрur ­ Danesi, Serena ­ Jenset, Gard B. ­ McGillivray, Barbara 2012: Reconstructing Constructional Semantics: The Dative Subject Construction in Old NorseIcelandic, Latin, Ancient Greek, Old Russian and Old Lithuanian. Studies in Language 36(3), 511-547. Nissilд, Leena 2001: Verbirektiosanasto suomi-viro ­ Verbirektsioonide sхnastik soome-eesti. Helsinki: Finn Lectura. Noлl, Dirk 2007: Diachronic construction grammar and grammaticalization theory. Functions of Language 14(2): 177-202. Sands, Kristina ­ Campbell, Lyle 2001: Non-canonical subjects and objects in Finnish. In: Aikhenvald, Alexandra ­ Dixon, Robert M.W. ­ Onishi, Masayuki (eds.): Non-canonical marking of subjects and objects (Typological studies in language 46). Amsterdam ­ Philadelphia: John Benjamins, 251-306. Schlotthauer, Susan 2011: Kontaktinduzierter Sprachwandel im Bereich der estnischen Verbrektion? Teil II: Verbkomplemente in Form von Adpositionalphrasen. SKY Journal of Linguistics 24, 145-179. Language data A-III = Mikael Agricolan teokset III. Nдkцispainos. Helsinki ­ Porvoo: WSOY, 1987. B1 = Biblia, Se on: Coco Pyhд Ramattu, Suomexi. Stockholmis 1642. Nдkцispainos. Porvoo: WSOY, 1971 PR 1992 = Pyhд Raamattu. Suomen evankelis-luterilaisen kirkon kirkolliskokouksen vuonna 1992 kдyttццn ottama suomennos. Helsinki: Suomen kirkon sisдlдhetysseura, 2002.
110
Question words with predicative function in Tundra Nenets
Nikolett Mus, University of Vienna Language of presentation: English
According to the literature (eg. Kuprijanova et al. 1957: 197-202) the structure of Tundra Nenets (Northern Samoyedic, Uralic) language allows sentences with nominal predicates, or some noun forms in the predicate position. In these clauses, the predicate noun/adjective occurs in the sentence closal position and the agreement (and past tense) marker is directly attached to it without any overt copulative verb (cf. 1a and 2a). Although the structure of the non-verbal predicates in Tundra Nenets is relatively well described, there aren't any descriptions of the question words with predicative function. As my initial results show the so-called nominal and adjectival question words can be used predicatively under the same grammatical circumstances (cf. 1b and 2b).
(1) a. ma Lamdo teta-dm. 1SG propr land.owner-1SG.PRED 'I am farmer Lamdo.' (Labanauskas 2001: 99) b. pidar xib'a-n? pr.2SG who-2SG.PRED 'Who are you?' (Labanauskas 2001: 99)
(2) a. uku jaa jiba. this day warm.3SG.PRED 'Today is warm.' (Nenyang 2005: 128) b. uku jaa num xurka? this day weather what.kind.3SG.PRED 'What is the weather like today?' (Nenyang 2005: 128)
This traditional description, however, focuses mainly on the equative, (proper) inclusive and attributive clauses (on the basis of Payne 1997: 111). In Tundra Nenets, nevertheless, there are locative and existential clauses as well (furthermore possessive1), that also contain a predicative nominal element, but differ from the previously illustrated clauses (cf. 3a­b).
(3) a. locative kiga stol- i-a a. book table-GEN on-LOC be.3SG 'The book is on the table.' (EL: 2012)
b. existential
iki pжdara-xana xo- taa-.
that forest-LOC
birch-PL be-3PL
'There are birch trees in that forest.'
(Almazova 1961: 39)
In these clauses the nominal element of the predicate is a noun inflected in a local case, a local postposition modified by a noun or a spatial adverb. As these elements cannot take any verbal markers, the use of a copulative verb (either BE-type or EX-type) is always needed. A content question can also appear with a locative or existential predicate. However, there aren't any analysis and/or description of these predicate structures. In my presentation, held in English, the following questions arise and are answered: (i) Which question words/phrases can fulfill the function of a predicate in a simple existential and/or locative content question? (ii) What kind of grammatic parameters do these questions have (with a special focus on the word order parameters)? (iii) What kind of copulas are there available for these types of content questions? (iv) Do they have different encoding strategies from those clauses, in which the non-verbal part of the predicate is not a question word?
1 In the frame of this presentation I don't deal with the so-called possessive content questions and/or the possessive clauses. 111
References: Almazova, A. V. 1961: Samoucite neneckogo jazyka. Leningrad: Ucpedgiz. [Nenets selfthaught]. EL 2012 = Evdokija Lampay Kuprijanova Z. N. et al. 1957.= Kuprijanova Z. N., Khomic, L.V. & Scerbakova, A.M. 1957. Neneckij jazyk [Nenets language], Leningrad. Labanauskas, K. I. 2001. Jamidihi" lahanaku" [Ancient Stories]. Moskva: Izdateltsvo Russkaja Literatura. Nenyang, M. A. 2005. Russko­Neneckij razgovornik [Russian­Nenets Phrasebook]. Sankt-Peterburg: Drofa. 112
Intra-textual referring: referential procedures and linguistic devices ­ a contrastive pilot study on Estonian and Finnish news texts Dirk Mьller, Berlin/Brussels/Vaasa University Language of presentation: English In the context of a larger research project in the field of contrastive text linguistics involving three linguistic and cultural areas - German, Finnish and Estonian ­ this paper will discuss textuality creation issues. There are various intra-textual referential procedures and relations that contribute to the creation of textuality, one of the main factors being recurrence relations. The research project is based on a broader concept of co-reference which goes beyond the level of mere grammatical forms, proceeding on the assumption of referring to mental structures in the textual space and, hence, investigating (intra-textual) referring also from a contents-orientated and logical point of view. As an introduction, various intra-textual reference processes will be identified, described and classified. Particular attention is paid to deictic and phoric procedures and the linguistic means used for their implementation. To this end, a short overview of corresponding linguistic expressions of Estonian and Finnish will be presented, focusing on their characteristics, contribution and on differences in the two language systems. The main part of the paper, however, will concentrate on language in use. The corpus of the pilot study is made up of original (= not translated) Estonian and Finnish news texts which cover similar topics and which have been published in online editions of newspapers. In order to define and limit the object of investigation in the present pilot study, only persons will be taken into consideration and analyzed as referents of intra-textual referential processes. The findings from the qualitative and quantitative corpus analysis will be presented, discussed and contrasted - the various linguistic devices, their distribution, and in particular their contribution to the thematic organization of the text. Presumable specificities of the languages (and cultures) concerned in the field of intra-textual reference processes, however, cannot be attributed solely to language system based divergences. Instead, the important role and impact of differing standards of verbalization of facts and different conventions for creating textuality will have to be pointed out. A summarizing description of such differences in news texts in relation to the two languages under investigation will conclude the paper. In future studies, the same research methods will be also applied to other types of texts, to finally be able to draw more general conclusions about conventions in the field of intra­textual referring which later may find application in language training, for instance to improve future translators' text competence. 113
The role of anaphora in topic coding in Northern Mansi Szilvia Nйmeth, Eцtvцs Lorбnd University Language of presentation: English Languages can be organized along a spectrum in terms of their grammatical marking of information structure, with topic prominence at one end and subject prominence at the other. Since no language displays pure topic or subject prominence, the terms weak and strong are used to refer to topic or subject prominence. According to this classification, Mansi, like the rest of the Uralic languages, constitutes a strong topic-prominent language (Skribnik 2001:222). A diverse range of definitions of topic can be found in the literature; in my research, I use the definition provided by Givуn, who considers the topic (T) to be the element that marks old information in the sentence (Givуn 1979:28) and examines it strictly in discourse. Important features of the topic are anaphoric referential accessibility, as well as cataphoric thematic continuity (Givуn 1990:740). In terms of the latter feature, T can be categorized into the following three types: chain-initial (the first occurrence of T), chain-medial (T is continuous with both the previous and the following items) and chain-final (T is continuous with the text that precedes it but not with the text that follows it). In discourse the most important, most continuous topic, known as the primary topic, tends to be marked by the grammatical subject, and the second most continuous topic (secondary topic) is marked by the grammatical object (Givуn 1984:138). A great deal of research has investigated the relationship between anaphora and the topic (e.g., Nikolaeva-Kovgan-Koskareva 1993), and Skribnik (2001) has looked at this area in Mansi in particular. In her research, Skribnik distinguishes the discourse topic (T1) and the paragraph topic (T2) and found that they differ in their coding strategies. T1 is expressed with zero anaphora in 80% of chain-medial (and with an NP or pronoun in the remaining 20%). In contrast, only 50% of case of T2 are marked with zero anaphora, and 39% with an NP and 11% with a pronoun (Skribnik 2001:232). In earlier corpus analysis I found that these statistics do not reflect newer published texts or translations: in these texts, T1 is characterized by the restatement of the pronoun or noun. The texts I examine differ from those of Skribnik not only in style but also in their date of origin, and for this reason it is worth examining the differences from a diachronic perspective as well. In my research, I examine T1-2 and anaphoric relationships in northern dialects of Mansi. The corpus, which has about 5,000 clauses, is made up of texts from two periods and three "genres": (a) prosaic folklore texts, (b) newspaper articles, (c) translated texts about literature. My hypotheses are the following: 1. Groups A and B differ will greatly from group C in terms of anaphoric T-coding. 2. Group C will show Russian influence, as can be confirmed by an analysis of the original Russian texts. 3. Russian influence will also be evident in the texts of group B, although to a lesser degree than in group C. References: Givуn, Talmy 1979: On Understanding Grammar. New York ­ San Francisco ­ London. -- 1984: Syntax. A Functional-typological Introduction. Volume I. Amsterdam ­ Philadelphia. -- 1990: Syntax. A Functional-typological Introduction. Volume II. Amsterdam ­ Philadelphia. Nikolaeva, Irina & Kovgan, Elena & Koskareva, Natalia 1993: Communicative roles in Ostyak syntax. ­ FUF 51: 125­167. 114
Skribnik, Jelena 2001: Pragmatic structuring in Northern Mansi. ­ Congressus Nonus Internationalis FennoUgristarum 7. Tartu: 222­239. 115
Usage and development of LEE(NE)- verbs in Finnic Miina Norvik, University of Tartu Language of presentation: English
The presentation concentrates on the verbs that go back to the Proto-Finnic root *l(ne)(hereinafter referred to as LEE(NE)- verbs) but that have their origin in Proto-Finno-Ugric. There are at least traces of these verbs in all the Finnic languages, and typically, their usage is associated with futurate and modal meanings. (Majtinskaja 1973; Metslang 1996) This presentation discusses the development and usage of LEE(NE)- verbs in different Finnic languages. As LEE(NE)- often expresses the future time reference (FTR), the focus is on FTR. Separate attention is paid to LEE(NE)- as a simple predicate and LEE(NE)- in a complex predicate. It has been argued that in connection with FTR, modal, aspectual and temporal meanings tend to intertwine (e.g. Dahl 2000: 313). Thus, one aim is to compare the rise of these meaning elements in the case of LEE(NE)- in simple predicates vs. complex predicates. Another aim is to propose a grammaticalization path for LEE(NE)- verbs. It will be shown that LEE(NE)- expresses FTR primarily in minor Finnic languages (e.g. Livonian, Tver Karelian, Veps and Votic). Furthermore, even though in these languages LEE(NE)- verbs can enter into complex predicates (see example 1), they most typically function as future copulas (see example 2). The different meaning elements (temporal, modal and aspectual) seem to be dependent on the construction and the broader context.
(1) Votic (Ariste 1977: 105)
kana avvo, senes munas tule b
mato.
hen brood.3SG this.ELA egg.ELA come.3SG snake
mitд siд tahoD sitд tдmд leep sillхх kantamaa
what 2SG want.2SG this 3SG lee.3SG 2SG.ALL carry.SUP `The hen broods, from this egg a snake will come. It will bring you whatever you wish.'
(2) Tver Karelian (Virtaranata 1990: 78)
sie muata
laskietse
2SG sleep.INF go_down.IMP
huomneksella nouzet
in kaikki
ieu valmi
morning.ADE wake_up.2SG so everything lee.3SG ready `You go to sleep, when you wake up in the morning everything will be ready.'
The material is partly collected from written sources (different text collections), partly elicited from recordings made during field-work trips. The newest material to be used is collected by me in 2009 in Tver Karelia and in 2012 during a field-work trip to the areas of Central Lude dialect.
References: Ariste, Paul 1977: Vadja muistendeid. Tallinn: Valgus. Dahl, Цsten 2000: The grammar of future time reference in European languages. - Цsten Dahl (ed.), Tense and Aspect in the Languages of Europe. Berlin and New York: Mounton de Gruyter. 309-329. Majtinskaja 1973 = , . 1973: - . ­ Soviet Finno-Ugric Studies 2: 81­90. Metslang, Helle 1996: The Development of the Futures in the Finno-Ugric Languages. ­ Erelt, Mati (ed.), Estonian: Typological Studies I. Tartu Ьlikooli eesti keele хppetooli toimetised 4. Tartu: Tartu Ьlikooli Kirjastus. 123-144. Virtaranta, Helmi & Virtaranta, Pertti 1990: Karjalan kieltд ja kansankulttuuria. 1, Tverinkarjalaisia kielennдytteitд. Helsinki: Suomalais-ugrilainen seura.
116
Constructions of continuous aspect in Finnish: Contextual comparison Tiina Onikki-Rantajддskц, University of Helsinki Language of presentation: English In this paper, contextual functions of different predications construing continuous aspect will be handled comparing contextual tendencies between predicates formed by the same stem in Finnish. The data comes from the online version of the heading daily newspaper in Finland Helsingin Sanomat: domestic news of a year's period and their comments written by the readers of the online version. The hypothesis concerning the division-of-labour between different types of predicates are formed on the basis of my earlier studies, which compared predicates denoting a posture with three stems: istu- `sit', maka- `lie', and seiso- `stand' forming different predicates such as 1) (hдn) istuu lit. `s/he sits', 2) (hдn on) istumassa 3sg be+3sg sit-INF-INE `s/he is sitting', 3) (hдn on) istuallaan 3sg be+3sg sit-ADE-PX `s/he is sitting/in a sitting posture' 4) (hдn on) istuvassa asennossa 3sg be+3sg sit-pctp-INE posture+INE `s/he is in a sitting posture' 5) (hдn on) istunut 3sg be+3sg sit-pctp 's/he has been sitting' Earlier findings indicate differences on four levels: a) There are structural differences between the different predicate constructions. For example, the locatives of state such as 3 represent ergative type absolute construal in the system of stative predicates, and the NUT-participle (5) construes a past tense. b) The tendencies of clause structure or constructions differ with these five predication types. Locatives of state (3) are often used also as manner adverbials and participles (4, 5) as attributes. MA-infinites (2) often occur together with perception verbs and such. The single predicate verb (1) tends most clearly to have adjuncts of place with it. c) There is a scale of agentivity between these expressions, in which the single verb (1) is most agentive and the locatives of state (3) and the attributive use of the participle (4) represent the other, least agentive end. The expressions also differ by their schematicity in their contextual uses. The predicate verb (1) is most polysemous or vague, but the locatives of state represent another kind of schematicity reflecting the absence of modifiers. d) There are also contextual functions in the wider context surpassing the limits of the clause structure. Preliminary findings concerning the use of the MAinfinitive or progressive construction indicate that imperfective constructions are not only backgrounding, orientative, and framing. The progressive construction (2), opening an internal time-window to the state-of-affairs, gives a potential to stop the story-line for moment of description, and this might even be the high peak or conflict of the story line. In addition, the progressive construction is often used for describing a task-like activity, and to bring the narrator's own experience to the fore. They are also used as parts of argumentation. 117
On the basis of the findings in the HS-corpus, I will refine the preliminary results and hypothesis concerning the contextual tendencies of different constructions of continuous aspect in Finnish. References: Ayano, Seiki 1998: The Progressive in Japanese and Temporal Advancement in Narrative. ­ Belgian Journal of Linguistics (BJL) 12: 1-19. Bertinetto, Pier Marco & Karen H. Ebert & Casper de Groot 2000: The progressive in Europe. ­ Dahl, Цsten (ed.), Tense and aspect in the languages of Europe pp. 517­558. Berlin: Mouton de Gruyter. Dowty, D. 1986: The effects of aspectual class on the temporal structure of discourse: Semantics or pragmatics? ­ Linguistics & Philosophy 9(1): 37­62. Heinдmдki, Orvokki 1995: The progressive in Finnish: Pragmatic constraints. ­ Bertinetto, Pier Marco & Valentina Bianchi & Цsten Dahl & Mario Squartini (eds.), Temporal reference, aspect, and actionality. Vol 2: Typological perspectives pp. 143­154. Torino: Rosenberg & Sellier. Hatcher, Anna Granville 1951: The use of progressive form in English. ­ Language 27: 254-280. Langacker, Ronald W. 1991: Concept, image and symbol. The cognitive basis of grammar. Berlin: Mouton de Gruyter. Onikki-Rantajддskц, Tiina 2006: Metonymy in locatives of state. ­ Helasvuo, Marja-Liisa & Lyle Campbell (eds): Grammar from the human perspective: Case, space and person in Finnish. Pp. 67-100. Amsterdam: Benjamins. ---- 2005: 3. infinitiivin inessiivi ja mддrдpaikkaisuuden arvoitus. ­ Herlin, Ilona & Laura Visapдд (eds): Elдvд kielioppi: Suomen infiniittisten rakenteiden dynamiikkaa. Pp. 173-193. Helsinki: SKS. `3rd infinitive inessive and the mystery of specified location' ---- 2001: Sarjoja: Nykysuomen paikallissijaiset olotilanilmaukset kielen analogisuuden ilmentдjinд. Helsinki: SKS. Dissertation. 'Patterns and analogy ­ a case study of state-denoting local case constructions in modern Finnish. Tamm, Anne (forthc.): Cross-categorial spatial case in the Finnic non-finite system: focus on the absentive TAM semantics and pragmatics of the Estonian inessive m-formative non-finites. Tommola, Hannu 2000: Progressive aspect in Baltic Finnic. ­ Dahl, Цsten (ed.), Tense and aspect in the languages of Europe pp. 655­692. Berlin: Mouton de Gruyter. 118
South Estonian demonstratives: changing or disappearing? Renate Pajusalu, University of Tartu Language of presentation: English In my presentation I focus on demonstratives in present-day South Estonian, especially in its Vхro variety (Vхro language) which is the strongest present regional language in the area. Most of Vхro people are bilingual Vхro and Estonian speakers and Vхro is becoming more similar to Estonian in many aspects (Karl Pajusalu 2009). On the other hand, demonstratives (especially distal tuu) are sociolinguistically a very prominent feature of South Estonian which may be preserved as an indication of difference with Estonian. In Finnic languages the number of demonstrative pronouns can vary from three (Finnish and Karelian) to one (Livonian) (Laanest 1982: 197-199). Standard Estonian has two demonstrative pronouns: see `this' (traditionally seen as proximal) and too `that' (traditionally seen as distal). However, too is actually rare in Standard Estonian and is productive only among speakers from southern Estonia. Thus, North Estonian spoken varieties have only one demonstrative pronoun see. In addition to one- and two-term systems of demonstrative pronouns, there is a three-term system in South Estonian (sjoo~seo, taa, and tuu). Among older speakers of Vхro language, the middle term taa refers to the hearer's sphere. It means that in addition to the distance-based system of Standard Estonian, South Estonian uses (to some extent at least) systems based on participant spheres as well. The South Estonian three-term person-oriented system is unstable even in the speech of older people at the moment. (Renate Pajusalu 1998) The research questions of the paper are: 1) What is the system of demonstrative pronouns in current South Estonian; 2) How do demonstratives survive under the pressure of the much simpler system of standard language? The data come from two kinds of sources: for new literary South Estonian mainly from the newspaper "Uma Leht" and for spoken variety from TV-series "Tagamхtsa", which is created and performed by native speakers of South Estonian of different age. Preliminary results show that all three demonstratives have survived but their usage is not the same as in previous varieties. References: Laanest, Arvo 1982. Einfьhrung in die ostseefinnischen Sprachen. Hamburg, Buske. Pajusalu, Karl 2009. The reforming of the Southern Finnic language area. Jussi Ylikoski (Ed.). The Quasquicentennial of the Finno-Ugrian Society (95 - 107). Suomalais-Ugrilainen Seura. Pajusalu, Renate 1998. Eesti pronoomeneid II. Vхru sjoo, taa ja tuu. Keel ja Kirjandus. 159­172. 119
The choice of the predicate form of Estonian purpose clause Helen Plado, University of Tartu Language of presentation: English
The aim of the presentation is to introduce the choice of the form of the main verb of Estonian purpose clause. The presentation is based on the purpose sentences that are taken from the Tartu University's Corpus of Written Estonian, both newspaper and fiction texts from 1990ies are used. Altogether I found 1.455 purpose sentences. A purpose clause expresses the purpose of the situation or event delivered by the main clause. In Estonian, the main verb of the purpose clause can be either in conditional mood or dainfinitive form (sentences (1) and (2) respectively). Estonian Academic Grammar (EKG II: 310) claims that da-infinitive can be used only if the actor of the subordinated clause and the actor of the main clause coincide.
(1) Pidasime
pцialt,
et "Kalevi" mдng хigel
keep:IPF:1PL thumb:PRT that Kalev.GEN play
ajal
lдbi
saaks. (AJA1990)
right:ADESS
time:ADESS through get:COND 'We kept our fingers crossed that the game of BC Kalev would end on time'
(2) A lдks
vдlja, et teha vдike jalutuskдik jдrve
A go.IPF.3SG out ддrde. (ILU1990)
that do.INF small walk
lake.GEN
edge:ILL 'A went out in order to go for a short walk to the lake'
However, I demonstrate that sometimes, like in sentence (3), da-infinitive is used in purpose clause, although the actor of the subordinated clause (men) is not the same as the actor of the main clause (president).
(3) President [--] ei olnud taganud
oma meestele
president
NEG be:PRTC guarantee:PRTC own.GEN man:PL:ALL
head
ettevalmistust ja piisavat kaitset,
et
good:PRT preparation:PRT and enough:PRT protection:PRT that
mitte surma saada. (AJA1990)
NEG death:PRT get:INF 'President had not guaranteed his men good training and sufficient protection to avoid
getting killed'
The presentation shows in which cases the predicate in da-infinitive and conditional mood can be used and under which circumstances either of these is preferred. I argue that the main difference between a purpose clause with predicate in da-infinitive and a purpose clause with predicate in conditional mood is referential dependency rather than obligatory identity of the actor in the case of purpose clause with the da-infinitive predicate. Purpose clause with predicate in da-infinitive form is referentially dependent on the main clause, whereas purpose clause with predicate verb in conditional mood is not. I introduce the constituents of the main clause (in addition to the agent), that can control the omission of the subject of the subordinated clause.
120
References: EKG II = Erelt, Mati, Reet Kasik, Helle Metslang, Henno Rajandi, Kristiina Ross, Henn Saari, Kaja Tael, Silvi Vare 1993. Eesti keele grammatika II. Sьntaks. Lisa: Kiri. Tallinn: Eesti TA Keele ja Kirjanduse Instituut. 121
Infinitival constructions with an unmarked object in old Hungarian and Mari. A cross-linguistic study Pйter Pomozi, Eцtvцs Lorбnd University Fanni Karбcsony, Eцtvцs Lorбnd University Language of presentation: English
The paper focuses on infinitival constructions which contained an unmarked object (IUOconstructions) in old Hungarian. The data are collected from corpora of Hungarian folklore and old Hungarian texts. The IUO-construction are analyzed typologically comparing old Hungarian to the Mari language. The following final clause structure occurs often in Hungarian folklore texts (A nagy hegyi tolvaj. Magyar nйpballadбk 1971: 56):
(1) Ne men-j
a vбros-ba kirбlylбny
NEG go-IMP.2SG ART town-ILLAT princess [NOM] 'Do not go to the town to kidnap the princess!'
rabol-ni! kidnap-INF
Its exact typological equivalent can be found in Mari, even in modern Mari language (Meadow Mari, Galkin 1964):
(2) - ц

Carpenter-PL house [NOM] build.INF 'Carpenters go/are going to build a house'
-. go-3PL
Typological equivalent can be found in Chuvash as well, however, the functions of unmarked object are wider there than in Mari or Hungarian (Paasonen 1949 cit., retranscribed: Krueger 1961: 157, 203):
(3) Karta-na
lasa
tit-ma
kay-n.
herd-OBV
horse [NOM] catch-INF
go-PST.PTCP
'he went into the herd to catch a horse' (lit. into the herd a horse to catch went [man])
Common feature of (1), (2) and (3) is that the argument of the infinitive is an unmarked object in the nominative case. It is also common in these sentences that their predicates always express motion or existence and it is always an implicit final construction. This typological coincidence can lead to the idea of historical areal-typological relationship between Mari and Hungarian. In our study we analyze the typological features which may cause the similarity of Hungarian and Mari IUO-constructions. Our research method relies on the theoretical study of Aikhenvald (Aikhenvald 2006: 1­49), applying the results of the Uralic Typology Database gained so far (Havas 2011). We use data of more than twenty Eurasian languages for the analysis of typological features and possible implications.
References and sources: Aikhenvald, Alexandra I., 2006. Grammars in contact. A cross-linguistic perspective. In: Aikhenvald, A.I.­Dixon R.M.W. ed. Grammars in contact. A cross-linguistic perspective. typology. Oxford University Press, Oxford. A nagy hegyi tolvaj. Magyar nйpballadбk. Ed. Gyula Ortutay. Magyar Helikon, Budapest, 1971. Galkin,, Ivan Stepanovich (, ) 1964. . . I. -. Havas Ferenc, 2011. A magyar йs rokon nyelvei ъj fйnyben: az urбli tipolуgiai adatbбzis tervezete. Pozna, 2011. (conference paper)
122
Krueger, John Richard, 1961. Chuvash manual: introduction, grammar, reader and vocabulary. Uralic and Altaic series 7. Indiana University, Bloomington. Paasonen, Heikki 1949. Gebrдuche und Volksdichtung der Tschuwassen. In: Mйmoires de la Sociйtй Finnoougrienne, vol. 94. Karahka, E.­Rдsдnen, M. eds., Suomalais-Ugrilainen Seura, Helsinki. 123
Completing and interpreting: Functions of the third person pronoun hдn `he, she' in reference to a co-participant in spoken Finnish Katri Priiki, University of Turku Language of presentation: English This paper is based on my doctoral thesis in progress which deals with Finnish third-person and demonstrative pronouns referring to people. In casual spoken Finnish the four pronouns hдn `he, she', se, `it, (he, she)', tдmд `this' and tuo `that' may be used to refer to people, the most common and the most neutral of them being se `it, (he, she)'. Yet in standard Finnish the thirdperson pronoun is hдn 'he, she' when the referent is human, and se `it' is only used with nonhuman referents. The use of these pronouns is, however, dependent on several factors other than the formality of the language, such as the conversational context. In all spoken Finnish dialects hдn may function as a logophoric pronoun referring to a speaker of another earlier discussion which is being spoken of. Another equally widely spread function of hдn is to convey a shade of underestimation in some contexts, sometimes referring even to inanimate objects. In addition to this, there is evidence of a dialectal difference: in some regional varieties of Finnish hдn `he, she' is more neutral and has more functions than in others. As for the demonstratives tдmд `this' and tuo `that', their use is shown to be strongly linked to their conversational context. The aim of this paper is to examine the use of hдn `he, she' in a regional dialect of the province of Satakunta in south-western Finland. The research is mainly being conducted using quantitative methods from sociolinguistics, but it also includes a qualitative part, in which the methods of interactional linguistics are employed to analyze the use of hдn, se, tдmд and tuo in actual conversations. The preliminary main data consists of 19 voice-recorded multi-party conversations (approximately 24 hours). On the recordings 34 speakers produce 3536 singular third-person and demonstrative pronouns that refer to a person ­ 316 of these refer to a coparticipant. This presentation focuses on the use of hдn `he, she' as a reference to a coparticipant. A preliminary analysis of the data indicates that in comparison with se `it, (he, she)', tдmд `this', or tuo `that', the pronoun hдn `he, she' is more frequently used as a subject of speech act verbs. This is naturally linked to the logophoric function of hдn. A closer analysis of the conversational contexts of hдn with speech act verbs reveals that hдn is rarely used when merely repeating or summarizing what a co-participant said in the same conversation earlier. Instead when using hдn speakers present an interpretation of or complete an earlier speaker's turn and solve problems that arise in their interaction. 124
Other-initiated self-repair in everyday interaction in Vхro language Maike-Liis Rebane, University of Tartu Tiit Hennoste, University of Tartu Sulev Iva, University of Tartu Language of presentation: English The aim of our presentation is to give the very first overview of the design of polar questions used in the other-initiated self-repair (OISR) in everyday conversation in Vхro language (on Estonian OISR see Strandson 2006, Hennoste et al 2012). The polar questions could be formulated by the different question particles (QPRT) in Vхro language (kas, va, kas...va, vai, ah, jah, egaq, хgaq), and also inversion and declaratives are used. Vхro language is a regional language used in South-East Estonia. It is used in a spoken everyday interaction and it has also a Standard form. Traditionally it has been considered as one of the South-Estonian dialects which is different from North-Estonian lexically, morphophonologically, and syntactically (Iva, Pajusalu 2004). OISR is the variant of the repair where the problem-source is in the text of the current speaker (Schegloff et al 1977). The recipient initiates repair and the speaker provides the solution to the problem. The other-initiations form a continuum from the "weaker" to the "stronger" formats: open class initiation, specific-class question words, (partial) repetition with and without question word, and offering a candidate (Sidnell 2010: 118-133). We will concentrate to the two "stronger" variants of the repair initiation where the speaker uses polar questions to initiate the OISR. Our data come from the Vхro and Seto language corpus which contains 80 000 running words of transcribed spoken dialogues. We have analyzed 100 examples of repair initiations for this presentation. The method used is Interactional Linguistics.
(1) 1. L: 2. (...) 3. K: 4. (...) 5. -> L: 6. K:
ooda *kelle puul sa *kдvet. wait whom did you visit *`Sigridi puul. Sigrid *tuu um *tuu kellel (...) *kддvд=хх *latsх sддl *trennih va. this is this whose children are participating there in the training PRT *jah, tuu=m [ka] yes this is too
In example (1) two women are speaking. L asks from K whom she visited and gets answer (lines 1 and 3). In line 5 she initiates OISR to specify who is Sigrid by offering a candidate using question particle va (literally 'or').
(2) 1. (...) 2. A: 3. B:
((B gives money to A)) ma=ot'si *tagasi sul[lх.] I find money back for you [a] *mis *latsх. but how about children
125
4. 5. A: A: 6. 7. -> B: 8. 9. A:
(...) no *Inga om *`haigх *jдlq. PRT Inga is sick again. (...) ah *Inga. PRT Inga (...) ((B looks for money)) h=*jah. yes.
In example (2) two women A and B are speaking. B asks a question how are the children of A and A answers that Inga is sick again. B initiates an OISR by partial repetition with a question particle ah (line 7).
References: Hennoste, Tiit, Rддbis, Andriela, Laanesoo, Kirsi 2013. Kьsimused eestikeelses infodialoogis II. Kьsimused ja tegevused. ­ Keel ja Kirjandus 1, 7­29. Iva, Sulev; Pajusalu, Karl 2004. The Vхro Language: Historical Development and Present Situation. ­ Language Policy and Sociolinguistics I: "Regional Languages in the New Europe" International Scientific Conference; Rzeknes Augstskola, Latvija; 20­23 May 2004. Rezekne: Rezekne Augstskolas Izdevnieceba, 2004, 58­63. Schegloff, Emanuel A., Jefferson, Gail, Sacks, Harvey 1977. The preference for self-correction in the organization of repair in conversation. ­ Language 52, 361­382. Sidnell, Jack 2010. Conversation Analysis. An Introduction. Oxford: Wiley & Blackwell. Strandson, Krista 2006. Kхneleja reaktsioon vestluskaaslase parandusalgatusele. ­ Keel ja arvuti. Tartu ьlikooli ьldkeeleteaduse хppetooli toimetised 6. Toim M. Koit, R. Pajusalu, H. Хim. Tartu, 170-182.
126
Spatial opposition, visual accessibility and contrast ­ an experiment with Estonian demonstratives Maria Reile, University of Tartu Language of presentation: English This presentation focuses on Standard Estonian demonstratives in situational context and on factors that affect the choice of demonstratives. The use of demonstratives can be divided into exophoric (situational use) and endophoric (textual use) reference (Halliday & Hasan 1977). The research of Estonian demonstratives has so far focused on endophoric reference of demonstratives. According to Pajusalu (2006, 2009) there are two demonstrative pronouns possible in Standard Estonian: see (this) and too (that), but the use of distal too is not that common and it is used mostly in South-Estonian. So far for the authors' knowledge exophoric usage of Estonian demonstratives has received little or no research attention. The aim of the study is to clarify whether the exophoric use of distal too is consistent and which factors influence the choice of demonstrative pronouns. Based on example of experiments in Indo-European languages (Coventry et al 2008, Piwek et al 2008), experimental approach to study exophoric demonstratives was chosen, as it allows to create controlled environment and makes it easier to determine the affective factors associated with demonstrative selection. The choice of stimuli is based on earlier research (e.g. Diessel 1999, 2006, Jarbou 2010) and includes spatial opposition, visual accessibility and contrast. Structure of the experiment: two participants ­ the instructor and the builder ­ take part of the experiment with the objective to build three different sculptures out of Lego blocks on a basis of ready built models. Only the instructor is allowed to see the models and only the builder is allowed to take the blocks from the table. The experiment tests three hypotheses: choice between demonstratives is affected by a) the location of the referent; b) visual saliency of the referent and; c) the need for contrast to make a distinction between two same kind of referents at the same place. The experiment was conducted in three South-Estonian schools. There were overall 39 pairs of students taking part of this experiment, but only material from 32 pairs is analysed on the basis of various reasons. Preliminary results of the experiment show that spatial opposition plays a crucial role on the choice of demonstratives. Distal too is used to refer to blocks out of one's hands reach and proximal see to blocks in one's hands reach. In addition visual saliency of the referents affects the pronoun choice ­ participants tend to use proximal demonstrative pronoun see (even if the referent is distant), rather than pronoun too, and distal or proximal demonstrative proadverb (depending on the location of the referent) to indicate the location. However this kind of use was not consistent. For contrast, participants use proximal see and pronoun teine (other). The study provides evidence that experimental approach proves to be effective for studying spatial demonstratives in Finno-Ugric languages and it provides linguistic data for comparison of spatial demonstrative use between Finno-Ugric languages. References: Coventry, Kenny R., Bernice Valdйs, Alejandro Castillo, Pedro Guijarro-Fuentes 2008. Language within your reach: Near-far perceptual space and spatial demonstratives. ­ Cognition, Volume 108, Issue 3, pp. 889-895. Diessel, Holger 1999. Demonstratives. Form, function and grammaticalization. John Benjamins Publishing Company. Amsterdam/Philadelphia. Diessel, Holger 2006. Demonstratives, joint attention, and the emergence of grammar. Cognitive Linguistics, Volume 17, Issue 4, pp. 463­489. Halliday, M. A. K, Ruqaiya Hasan 1976. Cohesion in English. Longman, London. 127
Jarbou, Samir Omar 2010. Accessibility vs. physical proximity: An analysis of exophoric demonstrative practice in Spoken Jordanian Arabic. Journal of Pragmatics, Volume 42, Issue 11, pp. 3078­3097. Pajusalu, Renate 2006. Death of a demonstrative: person and time the case of Estonian too. Linguistica Uralica 4. Tallinn. Pajusalu, Renate 2009. Pronouns and reference in Estonian. Sprachtypologie und Universalienforschung, 62 (1/2), 122-139. Piwek Paul, Robbert-Jan Beun, Anita Cremers 2008. 'Proximal' and 'Distal' in language and cognition: evidence from deictic demonstratives in Dutch. Journal of Pragmatics, volume, 40, isuse 4, pp. 694-718. 128
Some equivalents of the Estonian subordinate clause conditional in Hungarian (Eesti kхrvallause konditsionaali ungari vastetest) Tiina Rььtmaa, Tallinn University, University of Tartu Language of presentation: Estonian The Estonian conditional is a complex mood with many different functions. The question whether some of its functions have a conjunctive character has been raised by some researchers (e.g. Metslang 1999; Metslang & Sepper 2010). In Hungarian linguistics there has been a long discussion over a possible conjunctive mood sharing its morphological marker with the imperative (Prileszky 1974; Pomozi 1991; Pusztay 1998). Recent sources treat the conjunctive imperative as a separate linguistic category (e.g. Й. Kiss 1999; MG 2000) or even replace imperative with conjunctive in the Hungarian mood system (e.g. Hegeds 2005; Groot 2010). This suggests the Estonian conditional and Hungarian imperative should have at least some overlaps. This is also supported by a couple of studies of the Estonian equivalents of Hungarian moods in the subordinate clause as well as of Hungarian equivalents of Estonian moods in subordinate clause. This paper concentrates on the Estonian subordinate clause conditional and aims to map the main equivalents of the Estonian subordinate clause conditional in Hungarian. The study is based on original Estonian texts and their Hungarian translations. Studying moods in the subordinate clause allows us to take into consideration those factors of the mood choice that take effect only in the subordinate clause, for example, the semantics of the main clause predicate (e.g. Prileszky 1974) or some complementizers (e.g. Kehayov 2011). The Hungarian equivalents of the Estonian subordinate clauses with a conditional predicate confirm that there are several different equivalents in Hungarian associated with the Estonian conditional in the subordinate clause. The most frequent equivalents of the Estonian subordinate clause conditional in Hungarian are conditional or imperative. The conditional appears mostly in conditional clauses, the imperative in clauses where the Estonian conditional has conjunctivelike functions. This suggests that one of the main functions of the Estonian conditional is to express conditionality as well as to act as a conjunctive. However, a conditional subordinate clause in Estonian is also a widespread means of expressing an indirect command. Moreover, in some subordinate clauses in Hungarian the imperative can be replaced by the conditional. This means that the Hungarian equivalent of the Estonian subordinate clause conditional doesn't always reveal its function in a given clause as the Hungarian subordinate conditional can also have conjunctive functions. In future studies more of the semantics of the whole sentence as well as the various functions of the Estonian conditional should be taken into consideration in order to reach more detailed conclusions. Be that as it may, this paper already demonstrates that the Hungarian equivalent of the Estonian subordinate clause conditional with conditional function is mostly conditional while the equivalent of the Estonian subordinate clause conditional with conjunctive functions is mostly imperative. The paper is part of a larger research that aims to map the mutual relations of the Estonian and Hungarian subordinate clause mood systems. References: Й. Kiss, Katalin 1999. Mondattan. ­ Й. Kiss, Katalin; Kiefer, Ferenc; Spitбr, Pйter: Ъj magyar nyelvtan. Budapest, Osiris Kiadу, 15 ­ 184. Metslang, Helle 1999. Is the Estonian and Finnish conditional actually a conditional? ­ Mati Erelt (ed.) Estonian Typological Studies III. Tartu: Tartu University, 97­128. Metslang, Helle; Sepper, Maria-Maren 2010. Mood in Estonian. ­ Bjцrn Rothstein, Rolf Thieroff (eds.). Mood in the Languages of Europe. Studies in Language Companion Series 120. Amsterdam: John Benjamins, 529 ­ 550. 129
MG 2000 = Balogh, Judit, Lea Haader, Borbбla Keszler, Nуra Kugler, Krisztina Laczkу, Klбra Lengyel 2000. Magyar grammatika. Budapest: Nemzeti Tankцnyvkiadу. Groot, Casper de 2010. Mood in Hungarian. ­ Bjцrn Rothstein, Rolf Thieroff (eds.) Mood in the Languages of Europe. Studies in Language Companion Series 120. Amsterdam: John Benjamins, 551­568. Hegeds, Rita 2005. Magyar nyelvtan. Formбk, funkciуk, цsszefьggйsek. Budapest: Tinta Kцnyvkiadу. Kehayov, Petar 2011. Semantic functions of complementizers in Finnic (Estonian, Finnish, Karelian). http://dgcss.hum.ku.dk/arrangementer/semantic_functions_of_complementizers_in_european_languages/program me_-_semantic_functions/Finnic_complementizer_ abstract_revised.pdf/ (2.01.2012) Pomozi, Pйter 1991. Nйhбny gondolat a magyar "konjuktнvusz"-rуl. ­ Dolgozatok a magyar mint idegen nyelv йs a hungarolуgia kцrйbl 26. Zoltбn Йder (ed.). Budapest, 3 ­ 13. Prileszky, Csilla 1974. A felszуlнtу mуdъ mellйkmondat nйhбny kйrdйsйrl. ­ Imre Samu, Istvбn Szathmбri, Lбszlу Szt (eds.). Jelentйstan йs stilisztika. A magyar nyelvйszek II. Nemzetkцzi kongresszusбnak eladбsai. NytudЙrt. 83. Budapest: Akadйmiai Kiadу, 473­475. Pusztay, Jбnos 1998. A magyar imperatнvuszi бllнtmбnyъ mellйkmondatok (IБM) megfelelйsei az йsztben. ­ Folia Estonica VI. Йszt-magyar цsszevetйs II. Szombathely: Berzsenyi Daniel Tanarkepzц Fцiskola, 19­34. 130
Expressing temporary state of being in Mari ­ or how to manage without an essive case Sirkka Saarinen, University of Turku Language of presentation: English
The Mari language has two literary standards. The majority of Maris speak the so-called Eastern literary language, which is created on the basis of central (= Meadow) and eastern dialects, whereas there are nowadays only ca. 20,000 western (= Hill) Mari speakers. In the following, examples marked with a W were taken from western Mari. In Mari sentences expressing proper inclusion, equation and attribution, non-verbal predicates or copula constructions in the present tense can be used. Non-verbal predicates are more frequent in the 3rd persons, whereas the copula is more often used in the 1st and 2nd persons. In other tenses and moods, the copula is obligatory. In such sentences, the predicate, be it an adjective or noun, is in the nominative case (1). In non-verbal main predications expressing temporary location or state of being, predicative nominals are most often inflected in the inessive case (2).
(1) Tudo kьtьzц.
he
shepherd
'He is a shepherd.'
(2) Tudo jal-ste kьtьz-stц.
he
village-INE shepherd-INE
'He is a shepherd in the village (i.e., he is working as a shepherd).'
Often, in sentences such as (2), some other verb (such as sogem 'stand', kostam 'go, wander' or ilem 'live') is preferred rather than the copula or the non-verbal predicate, i.e., the predicate also stresses the temporariness of the state of being. In such contexts, the predicate seems to lose its primary meaning and becomes semantically empty (3­4). In eastern literary Mari, dative (3b) is also possible.
(3a) Tudo jal-ste kьtьz-stц sog-a.
he
village-INE shepherd-INE stand-PRES.3SG
'He is a shepherd in the village.'
(3b) Tudo jal-ste kьtьz-lan sog-a.
he
village-INE shepherd-DAT stand-PRES.3SG
'He is a shepherd in the village.'
(4) W A-m=at
дsndдrк кnde, pop-st ­ k-st,
NEG.PRES-1SG=CLT remember anymore, priest-INE what-INE
manak-st kast-n?
monk-INE
go-2PST.3SG
'I don't even remember anymore, was he a priest, a monk or what.'
Nominative encodes the predicative adjective. The difference between a permanent and a temporary state is expressed by the predicate.
(5) Tudo pojan. he rich 'He is rich.'
131
(6) Tudo jal-ste pojan il-a.
he
village-INE rich live-PRES.3SG
'He is living in the village as (a) rich (man).'
The modal case (in some grammars called comparative) in Mari expresses likeness and usually encodes adverbials (e.g., omsa-m tьmr-la perkal-as [door-ACC drum-MOD bang-INF] 'to bang the door like [as if it were] a drum'). The sentence (6) becomes a simile-like expression (7a), if the modal case is used. The postposition gaj can also be used in place of the modal (7b).
(7a) Tudo jal-ste pojan-la (~ pojan je-la) il-a.
he
village-INE rich-MOD ( ~ rich man-MOD) live-PRES.3SG
'He is living in the village like (a) rich (man) (as if he were rich).'
(7b) Tudo jal-ste pojan gaj il-a.
he
village-INE rich
like live-PRES.3SG
'He is living in the village like (a) rich (man).'
Other questions, such as predicative complements, will also be discussed in this paper.
132
Directives in teacher's guide books (Direktiivit opettajan opaskirjoissa) Minna Sддskilahti, University of Oulu Language of presentation: Finnish The purpose of this paper is to shed light on the use of directives in Finnish teacher's guide books of mother tongue and literature (later MTL) education. Teacher's guide books are supplementary learning material that is designed to help teacher's work and to offer diverse exercises for students to motivate them. Several studies show that Finnish teachers use textbooks and their supplementary material widely in their work. (Tainio 2012: 13­15.) This indicates that teacher's guide books are important tools for teachers, which motivates my study. My data consists of MTL guide books that are designed for class teachers and subject teachers. Directives are constructions that are used to command, urge, ask, and advise the recipient to act in a desired way. In Finnish, the prototypical directive is imperative clause but various constructions may be interpreted as directives in context. (ISK 2004: 1560­1561.) As Honkanen (2012) shows, the morphosyntactic variation of directives is related to contextual and interactional factors in several ways. As mentioned above, in teacher's guide books, the teacher is given instructions on how to design his/her teaching and the usage of other learning material, like students' textbooks. This is probably the reason why directives are a relatively common feature in teacher's guide books. My aims are to describe the morphosyntactic variation of directives in teacher's guidebooks and to discuss what this variation reveals of the interactional functions of directives in the data. I assume this enables to describe the relationship of the writer and the reader, as well. My preliminary results show that commands and requests are avoided in teacher's guidebooks, whereas recommendations and promises are favoured, like:
Opettaja voi
vaihtelevasti ohjata
teacher
can-3sg variably
direct-inf1
"Teacher can variably direct pupils to do..."
oppilaita pupil-pl-par
tekemддn... do-inf3
References: Honkanen, S. 2012. Kielioppi ja tekstilaji. Direktiivin muotoilusta viraston ryhmдkirjeissд. Helsingin yliopisto: Helsinki. ISK = Hakulinen, A., Vilkuna, M., Korhonen, R., Koivisto, V., Heinonen, T. R., and Alho, I. 2004. Iso suomen kielioppi. SKS: Helsinki. Tainio, L. 2012. The Role of Textbooks in Finnish Mother Tongue and Literature Classrooms. In Hanna LehtiEklund, Anna Slotte-Lьttge, Beatrice Silйn, and Ria Heilд-Ylikallio (eds.), Skriftpraktiker hos barn och unga. Rapport nr 35/2012. Pedagogiska fakulteten vid Еbo Akademi: Vasa.
133
Verb-framing vs. satellite-framing clauses and Information Structure in Estonian Heete Sahkai, Institute of the Estonian Language Language of presentation: English
L. Talmy (1985, 2000) introduced the notion of complex event, and the division of languages into verb-framing languages, which express the core schema of a complex event in the main verb, and satellite-framing languages, which express it in a ,,satellite", e.g. a particle or preposition. More recently, this division has been applied to individual complex event types within a language, rather than to languages as a whole (Croft et al. 2010). However, even within the same event type, both expression types may be available in a language. This raises the question as to what determines the choice of one expression type over the other. The aim of this study is to examine this by comparing, in the Balanced Corpus of Estonian, the distribution of the change-of-state verbs lхhkuma 'break' and tдitma 'fill' with that of the equivalent satelliteframing expressions involving, respectively, the satellites katki 'broken' and tдis 'full'. In addition, I will compare clauses with the verb lхhkuma 'break' with clauses involving lхhkuma in combination with the (redundant) perfectivising satellite дra. Previous literature predicts that the two expression types will differ with respect to register (Talmy 2000), as well as aspect, polysemy, abstract vs. concrete meaning (e.g. Eslon 2004). The present study will attempt to establish whether their distribution could additionally be related to certain information-structural, prosodic and syntactic differences between clauses with a simplex predicate as opposed to clauses with a framing satellite (or a complex predicate). Namely, in a transitive main clause with a simplex predicate, a given or non-focal object occurs clause-finally and is deaccented, whereas in a similar clause with a framing satellite, it is the latter that occurs in the clause-final position and bears the main sentence accent. It could be hypothesised that avoidance of clause-final non-focal and deaccented arguments is a potential factor that motivates the satellite-framing expression type, given that the clause-final position is the preferred position for focus (Tael 1988) and for the primary sentence accent (Ladd 2008). To illustrate, the study aims at testing the hypothesis that the satellite-framing expression type may partly be motivated by an avoidance of sentences of the type illustrated e.g. in (1), where both the subject and the object are given information, and the focus accent must occur early, on the verb. This avoidance might cause a preference for sentences of the type illustrated in (2) (an attested example) and (3), where the focus position is filled by a satellite, which would bear the primary sentence accent, the object being deaccented or bearing a pre-nuclear accent.
(1) Pдrast lхhkus
Meinart selle
afterwards break.PST.3SG Meinart this.GEN.SG
(2) Pдrast lхhkus
Meinart selle
afterwards break.PST.3SG Meinart this.GEN.SG
(3) Pдrast
tegi
Meinart selle
afterwards make.PST.3SG Meinart this.GEN.SG 'Afterwards Meinart broke this barrel'
tхrre. barrel.GEN.SG tхrre barrel.GEN.SG tхrre barrel.GEN.SG
дra. PERF katki. broken
To test this hypothesis, the study will examine the information-structural properties of the verbframing and satellite-framing clauses in the corpus (primarily, whether the object is given or new information). In addition, a series of other properties of the two clause types will be compared: register (as reflected in the proportion in which they occur in the different sub-
134
corpora of the balanced corpus), aspect, polysemy, abstract vs. concrete meaning, figurative vs. literal meaning, complementation, word order. References: Croft, William A., Jуhanna Barрdal, Willem Hollmann, Violeta Sotirova and Chiaki Taoka (2010). Revising Talmy's typological classification of complex event constructions. In H. C. Boas (ed.), Contrastive Studies in Construction Grammar. Amsterdam/Philadelphia: John Benjamins, 201­236. Eslon, Pille (2004). Mхningatest korrelatsioonidest vene ja eesti verbisьsteemis. In M.-M. Sepper, J. Lepasaar (eds.). Toimiv keel II. Tцid rakenduslingvistika alalt. Tallinn: TPЬ Kirjastus, 103­122. Ladd, D. Robert (2008). Intonational phonology. Cambridge: Cambridge University Press. Tael, Kaja (1988). Sхnajдrjemallid eesti keeles (vхrrelduna soome keelega). Preprint KKI-53. Eesti NSV Teaduste Akadeemia Keele ja Kirjanduse Instituut. Talmy, Leonard (1985). Lexicalization patterns: semantic structure in lexical forms. In T. Shopen, Language typology and syntactic description, vol. III: Grammatical categories and the lexicon. Cambridge: Cambridge University Press, 57­149. Talmy, Leonard (2000). Toward a cognitive semantics, vol. 2: Typology and process in concept structuring. Cambridge, Mass.: MIT Press. 135
From doubt to assumption: The Finnish verb epдillд as an example of a constructional meaning change Jutta Salminen, University of Helsinki Language of presentation: English The presentation concerns constructional meaning change displayed in the history of the Finnish mental verb epдillд. According to diachronic data (from Early Modern Finnish to Modern Finnish), in certain constructions the interpretation of this verb has undergone a meaning shift from `to doubt' to `to suspect' and `to assume'. As an evidence of this change, the verb is polysemous in Modern Finnish; the synchronic observation reveals a strong dependence between the complement type and the meaning interpretation (cf. Hansen 2012: 241). The reinterpretation concerned has a peculiar feature, since the subject's epistemic stance towards the proposition of the verb complement changes completely. Applied to the example Epдilen, ettд hдn tulee `I assume/doubt that he will come', the arrival of the person in question may either be supposed or questioned. The variant translated with `to doubt' will be called negation-inclining (the speaker is inclined to think that the proposition of the complement probably doesn't hold true), and the variants `to suspect' and `to assume' go under the label affirmation-inclining, since the stance is the opposite. (Salminen 2012: 5­14; for the terms, see also Langacker 2008: 450.) The focus of the presentation will be on the constructions consisting of the verb epдillд and its clausal complements: (1) epдillд + ettд `that' clause, (2) epдillд + non-finite construction, and (3) epдillд + embedded question. In addition, (4) elliptical use of the verb (with no overt complement) will be examined. The analysis reveals that the two first constructions have undergone an identical change from a negation-inclining interpretation to an affirmationinclining one. However, with an embedded question and in elliptical use there seems to be no such change. The explanations to this difference may be sought from the inherent semantics of the embedded question complement and from the context-bound nature of the elliptical use of the verb. Since the above meaning change doesn't apply to the verb in all its uses, it is reasonable to conceive of the change as a construction specific phenomenon: a constructional meaning change (cf. Hansen 2012: 240, Hilpert forthc.). Thus there exists a significant semantic dynamics between the verb and its complement ­ a relation that has been studied from several viewpoints (see e.g. Goldberg 1995: 59­65; 1998; forthc. 18­19; Rappaport Hovav & Levin 1998: 127­ 130). Since the current focus is on the verb's meaning variation, the dynamics will be elaborated as follows: the complement type and textual collocations affect the verb's semantics in each context (cf. Rappaport Hovav & Levin 1998: 98­99). These textual collocations may extend even to genre-specific conventions (Hilpert 2012: 153­156). References: GOLDBERG, ADELE E. (forthcoming): Constructionist Approaches. In Trousdale, G. & Hoffmann (eds.), The Oxford Handbook of Construction Grammar. Oxford: Oxford University Press. ­­­­­­ 1998: The relationship between Verbs and Constructions. In Verspoor, Lee & Sweetser (eds.), Lexical and Syntactical Constructions and the Construction Meaning. pp. 383­398. Amsterdam: John Benjamins Publishing Company. ­­­­­­ 1995: Constructions. A Construction Grammar Approach to Argument Structure. Chicago: The University of Chicago Press. HANSEN, MAJ-BRITT MOSEGAARD 2012: A Pragmatic approach to historical semantics. In Allan, K. & Robinson, J. A. (eds.), Current methods in historical semantics, pp. 233­258. Topics in English Linguistics 73. Berlin: De Gruyter Mouton. 136
HILPERT, MARTIN (forthcoming): Corpus-based Approaches to Constructional Change. In Trousdale, G. & Hoffmann (eds.), The Oxford Handbook of Construction Grammar. Oxford: Oxford University Press. http://members.unine.ch/martin.hilpert/OHCG.pdf ­­­­­­ 2012: Diachronic collostructional analysis: How to use it and how to deal with confounding factors. In Allan, K. & Robinson, J. A. (eds.), Current methods in historical semantics, pp. 133­160. Topics in English Linguistics 73. Berlin: De Gruyter Mouton. LANGACKER, RONALD W. 2008: Cognitive Grammar ­ A Basic Introduction. Oxford: Oxford University Press. RAPPAPORT HOVAV, MALKA & LEVIN, BETH 1998: Building verb meanings. In Butt & Geuder (eds.), The Projection of Arguments. Lexical and Compositional Factors. CSLI Lecture Notes no. 83. Stanford: CLSI. SALMINEN, JUTTA 2012: Epдilen, ettд tдmд verbi kyseenalaistaa tдydennyksensд proposition. Epдillд-verbin merkityksen kehityksestд ja monitulkintaisuudesta. [I doubt/assume that this verb questions the proposition of its complement. On the development and ambiguity of the meaning of the verb epдillд] Master's thesis. Finnish language. University of Helsinki. 137
Intonation of different syntactic structures in relation to different semantic contexts Nele Salveste, University of Tartu Language of presentation: English In languages with free word order, the word order of a sentence depends among other factors (animacy, constituent heaviness) on its information structure (IS) (see for example Kiss, 1995), phenomenon called scrambling. The IS in the intoNation Languages like English or Dutch is transmitted mainly by pitch accenting the focal information (Ladd, 1996). There are some indications that scrambling might cause restructuring of the intonation phrase and changes in the accent pattern of the sentence, which is not necessarily bound to the IS (Antonyk-Yudina & Mykhaylyk, 2010; Fйry, 2005). That phenomenon might be a consequence of the strong connection between the word order and its pragmatics. Pragmatically motivated word order permutations might have their own independent intonational patterns. For example, the left-dislocated objects may attract pitch accents at the beginning of the sentence and compress the accents in the rest of the sentence despite their given status in a discourse. A production experiment was carried out to test the hypothesis. A dialogue was constructed so that the target-sentence like `Lena painted a boat' was coupled with a context that was either (1) an echo question `Somebody painted a boat?'; (2) the sentence with contradicting information `Mary painted a boat' or (3) the general question `What's up?' The task of the subject was to utter the target-sentence as a reply to one of these contexts. This experimental design was expected to elicit broad focus, narrow focus and contrast in different sentence constituents. The assumption was that if one of the constituents was in focus, then the others were given and were encoded so. The focus domain was subject (S), object (O), adverb (A) or the whole sentence. Crossing four different contexts (broad, narrow, contrast, given) and different focus domains resulted in four information-structural conditions for each sentence constituent: 1. 1. A-Broad, A-Narrow, A-Contrast, A-Given 2. 2. O-Broad, O-Narrow, O-Contrast, O-Given 3. 3. S-Broad, S-Narrow, S-Contrast, S-Given The word order of the sentences was varied so that A, O and S occurred once in the sentence-initial position: AVO, OVS, OVA, SVO. The purpose of the experiment was to determine whether the O in the beginning of the sentence behaves prosodically differently from the other type of constituents (especially from the S). In the light of previous research (Salveste, 2013; Fйry & Kьgler, 2008; Pappert et al., 2004), we expected a strong correlation between intonation and the IS and no effect of the word order permutation. The results showed, however, that if in other sentence constituents the occurrence of the pitch accent varies in relation to IS, then this variation is partly canceled in the sentenceinitial O. This result indicates that the occurrence of the pitch accents goes hand in hand with the word order permutations and that the blocked prosodic variation can indicate a strong association between syntax and pragmatics. References: Fйry, C. (2005). Laute und leise Prosodie. Text - Verstehen. Grammatik und darьber hinaus. IDS-Jahrbuch 41, 162- 181. Fйry, C. & Kьgler, F. (2008). Pitch accent scaling on given, new and focused constituents in German. Journal of Phonetics 36, 680-703. Kiss, Й. K. (1995). Introduction. Katalin Й. (Ed.). Discourse configurational languages. Oxford studies in comparative syntax. New York: Oxford University Press. 138
Ladd, R. (1996). Intonational phonology. Cambridge: Cambridge University Press. Pappert M. S., SchlieЯer, J., Pechmann, T., & Lingel, S., (2004). Zur prosodischen Differenzierung zwischen syntaktischen Strukturen. Linguistische Arbeitsberichte, 81, 223-241. Salveste, N. (to appear in 2013). Focus perception in Estonian: Is it governed by syntax or by prosody? Proceedings of Nordic Prosody XI. Vilkuna, M. (1998). Word order in European Uralic. A. Siewierska (Ed.), Constituent Order in the Languages of Europe (pp. 173­233). Berlin, New York: Mounton de Gruyter. Antonyuk-Yudina, S. & Mykhaylyk, R. (2010). Prosody of scrambling. MIT Working Papers in Linguistics (NESL 40 proceedings). Retrieved from http://ling.auf.net/lingbuzz 139
From a proper name to an appellative. Appellativization of given names in Finnish slang compounds (Proprista appellatiiviksi. Etunimien appellatiivistuminen suomen slangiyhdyssanoissa) Maria Sarhemaa, University of Jyvдskylд Language of presentation: Finnish Finnish slang contains a number of compounds which include a homonym of a given name either as a head or a modifier. These include, among others, the noun ovimikko `doorman' (lit. `door + given name Mikko] and the adjective marttaikдinen 'middle-aged woman' (lit. given name Martta + `aged]. In these expressions, the words mikko and martta cannot be understood as authentic proper names but instead they have gone through a process of appellativization, in other words, they are understood as appellatives. Mikko is a common head of such a compound and for example a doorman can be referred to also by compounds vagemikko, vaksimikko and vahtimikko (lit. three different variants of `guard' in the Helsinki slang + given name Mikko) in colloquial language. Baarimikko means 'bartender' (lit. `bar' + given name Mikko) and jodimikko `assistant of a doctor in the army' (lit. `iodine' + given name Mikko). All these compounds refer to a profession or an assignment, and theoretically speaking mikko could perhaps be replaced by another appellative, for example mies `man' in these structures. It is also possible to create more compounds with the head mikko by this same pattern. These structures can be seen as constructions which are partially schematic as their modifier noun varies whereas the head is lexicalized and can be selected from a restricted list of given names, for instance, [[noun] + [mikko]] or [[adjective] + [kalle]]. These patterns seem to be quite productive, and even novel expressions elaborating these schemas may be introduced. For instance, it is by no means impossible to form and understand novel expressions such as tallimikko (lit. `stable' + given name Mikko) which could mean 'stable hand', koulumikko (lit. `school' + given name Mikko),'school's maintenance man' or kierokalle (lit. `twisted' + given name Kalle) `schemer'. However, the context plays a crucial role in the interpretation of such novel expressions as the meaning of these compounds is only partially understood on the grounds of the construction and partially in the relation to the actual context where the expression is uttered. These kinds of compounds are also found in the slang of another Finno-Ugric language, Hungarian. For instance, lуlujza (lit. 'horse' + given name Lujza) and tonnalujza (lit. 'ton' + given name Lujza) mean 'big woman'. In my paper, I will give a detailed analysis of the appellativization of the given names in Finnish slang compounds with Hungarian slang compounds as a point of comparison. 140
Attribute-Noun Agreement in the Uralic Languages: Evidence from Udmurt and Baltic Finnic Languages Olga Schitz, Tomsk State University Language of presentation: English The problem of agreement of attributes and nouns in one of the Permic group of the Uralic languages - Udmurt, will be presented in the report. Following the European grammatical tradition which originates from the Latin language model, we define agreement as a formal coincidence of one part of a noun phrase with another, i.e. the grammatical markers of the dependent word are similar to those of the key word. As for the Udmurt language, such phenomenon exists, but it is evident only when an attribute has an intensifying and demonstrative suffix which forces both noun and attribute to accept identical morphological markers. We consider such adaptation as apposition, as an attribute changes its typical function and becomes a noun. On the other hand, it is typical of the Udmurt language not to show agreement of attributes with nouns in number and case (there is no gender in the Uralic languages). At the same time it is pointed out that the particle ­e, which is added to an attribute when a noun has a plural form with flexions -os/-jos, is optionally used and the same particle is also considered as a plural marker of an attribute. In our opinion, the specified affix expresses not the plurality, but the intensity of quality, that it serves to emphasize the characteristics of persons or objects to individuate them. Thus the agreement category in the Udmurt language is shown only in special cases whereas in related Baltic Finnic languages attributes always show agreement with nouns. We suppose that such difference can be primarily explained by the contact influence of the European languages on the Baltic Finnic languages, namely Russian and Germanic, in which agreement is obligatory. In turn the Udmurt language is in close contact with the Tatar language where no such phenomenon exists, so the original characteristics of the Udmurt language are being preserved. Besides, a long history of writing in the Baltic Finnic languages also contributes to the establishment of agreement tradition in these languages, while active writing in Udmurt started in the 1930s of the 20th century. 141
Acquisition of Spanish L2: Estonian referential devices and Spanish articles Maarja Sepp, University of Barcelona Language of presentation: English From a historical perspective we know that Spanish definite article el derives from the Latin demonstrative ille/illa/illud, and the source of the indefinite article un is the Latin numeral unus/una/unum. Although Estonian is traditionally described as a language without articles, it is evident that in Estonian, similarly to Spanish, the pronoun see can perform the function of the definite article and the numeral ьks that of the indefinite article. However, they cover only some of the Spanish article uses. This presentation gives an overview of a typological study that contrasts different uses of Spanish articles with the expression of (in)definiteness in Estonian, in order to identify Estonian referential devices that systematically correspond to specific uses of Spanish articles, as well as to ascertain the uses that do not have agreement between the two languages. I argue that the Spanish article uses that where first to grammaticalize, are more probable to systematically resemble certain Estonian referential devices than the uses that fixated late. For example, the Estonian numeral ьks can play the role of the first mention un that developed early in the evolutionary process of Spanish articles, whereas ьks is normally not used in the attributive and generic interpretations, such as Kдnguru on kukkurloom / El canguro es un marsupial `The kangaroo is a marsupial'. Interestingly, up to the XVI century the attributive and generic interpretations of the Spanish indefinite article un were typically manifested by bare nouns. The study applies Error Analysis of written Spanish L2 by Estonian learners and aims to establish a systematic approach to analyzing common difficulties that could be applied in teaching Spanish as a foreign language to Estonians. Bibliography: CHESTERMAN, Andrew (1991). On definiteness. A study with special reference to English and Finnish. Cambridge University Press, Cambridge. ESLON, Pille (2010). "Suundumustest eesti keele grammatiliste kддnete kasutuses. Elson, Pille, Katre Хim (Eds.) Korpusuuring ja meetodid (7 - 36). Tallinn: TLЬ EKKI. GARCНA MAYO, Marнa del Pilar, Roger Hawkins (eds.) (2009). Second Language Acquisition of Articles: Empirical Findings and Theoretical implications. Vol. 49. Amsterdam/ Philadelphia: John Benjamins Publishing. LEONETTI, Manuel (1999) "El artнculo". In: Gramбtica Descriptiva de la Lengua Espaсola. Real Academia Espaсola. Madrid, Espasa-Calpe LEONETTI, Manuel (2000) "The Asymmetries between the Definite Article and Demonstratives: a Procedural Account", Comunicaciуn en la VII International Pragmatics Conference, Budapest. PAJUSALU, Renate (2001) "Definite and indefinite determiners in Estonian." In: Pragmatics in 2000: 7th International Pragmatics Conference; Budapest, Hungary; July 09-14, 2000. (Eds.) Nemeth, E. Antwerpen: International Pragmatics Association, 458 - 469. PAJUSALU, Renate (1997) ,,Is There an article in (spoken) Estonian?" ­ Mati Erelt (Ed.). In: Typological Studies II. Publications of the Department of Estonian of the University of Tartu, 8. Tartu: Tartu University Press, 146­177. RAJANDI, Henno, Helle Metslang (1979). Mддramata ja mддratud objekt. Tallinn: Valgus. REAL ACADEMIA ESPAСOLA: Nueva gramбtica de la lengua espaсola. Manual. Madrid, Espasa Libros, 2010. 142
Different conceptualizations in expressing "X becomes Y" in Finnish Mari Siiroinen, University of Helsinki Language of presentation: English
My paper deals with two syntactic constructions in Finnish which both express roughly the meaning "X becomes Y". I focus on the instantiations of these constructions where the verb tulla `come, become' is used. The sentences below are examples of instantiations of the two constructions. Both sentences have the meaning `the porridge became thick'.
1. Puuro
tuli
porridge-NOM become-PST.3SG
NP-nom become AP-part
2. Puurosta
tuli
porridge-ELA become-PST.3SG
NP-ela become AP-part
sakeaa. thick-PART sakeaa. thick-PART
The syntactic constructions are (type 1) NP-nom V AP/NP-nom/part and (type 2) NP-ela V AP/NP-nom/part. The most obvious formal difference between these constructions is in the case of the first nominal constituent. The first constituent, whose refererent's change is at issue, is either in the nominative (type 1) or elative case (type 2), and the subject predicative, which expresses the new state or property, is either in the partitive or nominative case (more on this case variation see Huumo 2009). At a closer look, type 1 resembles closely the Finnish copulative construction and type 2 the existential construction. Both of these constructions are found in Finnish dialects. Of these two constructions, type 1 is special because it is very infrequent in modern standard Finnish and often regarded by native speakers as odd and ungrammatical. Still it has been one of the standard examples mentioned in the traditional reference grammars, and in old dialectal data this construction is actually quite frequent. In my presentation I will discuss the dialectal distribution of type 1 and compare it with the type 2. I also look at the similar constructions in Baltic-Finnic languages. It seems that there are cognates in Karelian and Estonian (Erelt 2005). I will also try to describe the different conceptualizations of intransitive change that these two constructions manifest. I adhere to the Principle of Contrast by e.g. Croft "If two grammatical structures occur in the same language to describe the `same' experience, they will differ in their conceptualization of that experience in accordance with the difference in the two structures." (Croft 2001: 111.) Oddly enough, in today's standard Finnish type 1 seems to have virtually vanished. I try to find out what might have caused its demise: has its conceptualization changed, and if so, for what reason?
References: Croft, William 2001: Radical Construction Grammar. Syntactic Theory in Typological Perspective. Oxford:Oxford University Press. Erelt, Mati 2005: Source-marking resultatives in Estonian. Linguistica Uralica, Vol. 41 Issue 1, 20-29. Huumo, Tuomas 2009: Fictive dynamicity, nominal aspect, and the Finnish copulative construction. Cognitive Linguistics Vol. 20 Issue 1, 43-70.
143
How does the context disambiguate a construction? The case of the affective Finnish abstract motion construction mennд V-mA-An [go V-INF-ILL] `do something unwished' (Miten konteksti disambiguoi konstruktion? Tapaustutkimus suomen kielen abstraktin liikkeen konstruktiosta mennд V-mA-An tehdд jotakin epдtoivottavaa') Jari Sivonen, Eцtvцs Lorбnd University Language of presentation: Finnish
The Finnish deictic motion verb mennд `go' has several semantic extensions which do not express motion but have been grammaticalized to convey other meanings. One of these is the abstract motion construction mennд V-mA-An [go V-INF-ILL]. This construction evokes an affective meaning, speaker's disapproving stance towards the described event, and it can be described roughly as "do something unwished". An example is given in (1).
(1) Erдs hullu
men-i
ampu-ma-an Lennon-in.
certain crazy.NOM go-PST.3SG shoot-INF-ILL Lennon-ACC `Certain crazy shot Lennon [though he should not have].'
However, the mennд V-mAAn phrase, as such, can also be used in its literal sense for referring to physical motion, e.g. situations where some entity actually moves to some place and after that carries out the activity expressed in the constructions infinitival verb (2). In these cases there is no implication concerning negative evaluation.
(2) Men-i-n katso-ma-an asunto-a. go-PST-1SG look-INF-ILL flat-PTV `I went to take a look at the flat.'
The difference between the actual motion reading (2) and the non-motional affective interpretation (1) lies in the context where the mennд V-mAAn phrase occurs. In other words, there are syntactic and semantic features which are typical for the mennд V-mAAn construction and therefore guide the interpretation of the whole sentence. These include, among others, different adverbial modifiers such as tyhmyyt-tд-дn [stupidity-PTV-3PX] `due to stupidity', a locative adverbial in the inessive (`in') or adessive (`at') case and a subject noun connected with negative meaning such as hullu `crazy' in (1). In my presentation, based on a large corpus data, I will give a detailed analysis of the contextual determinants which disambiguate the affective mennд V-mAAn construction from the motion readings. In this light, my presentation gives an overt contribution to the analysis of the grammar and context interaction.
144
Depicitive in Nganasan Sбndor Szeverйnyi, University of Szeged Language of presentation: English The primary aim of the presentation is to introduce the constructions which can semantically be classified as depictive, and to expose the morphosyntactic realizations of these depictive meanings in the Nganasan language. Since the research work is at somewhat preliminary stage, it is not possible to provide a full and exhaustive, comprehensive analysis and overview. Presently number of questions exceeds the answers. The purpose of the research has two goals: at large, the broadening of our syntactic knowledge about the Samoyedic languages, and more specifically the explosure of a partly semantic and partly morphosyntactic phenomenon mostly positively missing from the documentation of the lesser Uralic languages. Mainly written sources have been being used for the research (e.g. Valentin Gusev's unpublished collection, 2008 ­ and other sources), and a questionnaire used for collecting relevant data by Nganasan natives in the summer of 2008 in Ust-Avam. The first part of the prersentation is a summary of the most important questions of the relating terminology. The second is the introduction of the main typological features of the DSPredicate (e.g. Groot 2008, Himmelmann­Schultze-Berndt 2005, Schultze-Berndt­ Himmelmann 2004). This will be followed by the Nganasan data, systematized from morphosyntactic point of view. Finally, these data will compared with typologic results. The following Nganasan constructions might be considered among the ones capable of expressing depictive content: 1. construction with prolative/adverbial suffix (-mnU / -mnU) 2. construction with caritive suffix (-Kai) 3. lexical (e.g. onda"laku 'lassan, уvatosan') 4. construction with mant 'as' postposition 5. construction with comitative suffix (-sbt) 6.a. converb (infinitive) 6.b. converb (Noun, Numeral or Adjective + ia 'be' construction) 7. word order References: Groot, C. de (2008), Depictive Secondary Predication in Hungarian. In: Schroeder C. ­ G. Hentschel ­ W. Boeder (eds.), Secondary predicates in Eastern European languages and beyond. Oldenburg. 69­96. Gusev, V. (2008), Nganasan texts. ms. Himmelmann, N. P. ­ E. Schultze-Berndt (eds.) (2005), Secondary predication and adverbial modification. The typology of depictives. Oxford. Schultze-Berndt, E. ­ N. P. Himmelmann (2004), Depictive secondary predicates. Linguistic Typology 8/1: 59­131. 145
Deictic and anaphoric grounding by possessive affixes in Nganasan Rйka Zayzon, University of Hamburg Language of presentation: English In Nganasan, as in a number of Uralic languages, 2SG and 3SG possessive affixes are used in contexts that are not prototypical for possessive markers, a phenomenon that has been notified in the literature (cf. Tereshchenko 1979: 95) but was not investigated in further detail, as regarding the Nganasan data. The analysis will be based on Nganasan language data from a corpus covering different discourse genres and written as well as spoken language. The data are mainly driven from the unpublished electronic database compiled by Valentin Gusev and Maria Brykina and are completed by some newspaper texts, comprising about 7500 sentences. In the presentation proposed I adopt Nikolaevaґs (i.p.) classification with slight modifications for analysing Nganasan language data. In her analysis of the usages of possessive affixes in Uralic languages, Nikolaeva showed that the "definiteness" account of the non-possessive usages, common in the descriptive literature on Uralic, can only partly explain relevant data and suggests a classification based on a conception of possession as a cognitive category. Possessive affixes link the referent of a noun (or pronoun) to 1) a speech act interlocutor (1st and 2nd person) or 2) another referent, by deictic and anaphoric procedures, respectively. Both types of procedures are aimed to facilitate joint attention of the speech act interlocutors, and are comparable to those performed by demonstratives. (For a description of reaching joint attention cf. Diessel 2006.) The range of relationships between referent and "possessor" encoded by the possessive affixes encompasses different kinds of associations between referent and "possessor" (Nikolaeva, i.p., 13) which may be based ­ beyond the conventionalised possessive meanings (cf. Lyons 1999, 124ff) ­ on situational or emotional proximity, as well as mental accessibility due to either encyclopaedic knowledge or knowledge construed or activated in the discourse. In case of the deictic personal affixes of the 1st and 2nd person, the "possessor" is identifiable by deixis. In case of anaphoric reference with the 3rd person affixes, the "possessor" must be identifiable based on either the speech situation or from former discourse (cf. example 1). In certain cases where the possessive affix is attested the function of marking of definiteness, the linked entity may not necessarily be identifiable by anaphora resolution, but by situational or encyclopaedic knowledge (example 2). In Nganasan, possessive affixes may combine with non-nouns, such as pronouns (demonstratives as well as indefinites, e.g. tti-r : thatANAPH-2SG, maa-r : whatINTERR-2SG) and particles (e.g. tbt-r : also-2SG). The joint attention account applies even in these cases, although the referent of these expressions might not be identifiable. The non-prototypical usages of the possessive affixes are pragmatically conditioned and not grammaticalized. They represent discursive strategies aiming to facilitate joint attention of the interlocutors. This accounts for the fact that possessive affixes encoding "possessors" of different persons may be iterated in Nganasan. Even different deictic procedures pointing to one and the same referent may be combined in one utterance (example 3). Furthermore, even within one lexical item, different suffixes may combine. The deictic linking of the referent to the interlocutor is conventionalised, e.g. ma-tu a : tentGen.3SG top, ґhomeґ; see also example (example 4). The aim of the analysis is to describe the distribution of the presented usages of the possessive markers throughout different text and discourse types. As these are not grammaticalized, the variation in using the markers will also be investigated. 146
Examples:
Examples (1), (3) and (4) stem from the Text meu djamezi, collected and transcribed by Beбta
Wagner-Nagy,
accessible
on
the
Internet
under
http://www.univie.ac.at/negation/sprachen/downloads/nganasan/nk-96_mou_djamezi.pdf.
(1) hiimtk-nd-t
tahariaa tundia-ku satr-tu
nanu
night.falls-VNtemp-Lat now
fox-Dim
tu-u--gj
hursi--gj
polar.fox-Gen.3SG basuр-gt
withPP
come-Ep-Aor-3Du
come.back-Aor-3Du hunting-El
`As darkness fell, the little fox and the(/his) polar fox returned from hunting.'
(meu djamezi.166)
(2) Sьrь-рь donьo
snow-3SG melted.3SG
The snow has melted. (Tereshchenko 1979:95)
(3) enaca-a
silak
centr-hiaр
tnd-mt
huge-AUG.ACC bak-mtu
planed.board.ACC make-INFER.3SG that-ACC.2SG
kmia-s
ma ni-t
scraper-ACC.3SG take-INF
in.this.way stand-AOR.3SG
`It prepared the large planed board, it seems, it holds the scraper, so it stands.' (meu
djamezi.141)
(4a) ba-n
m-t
tn
dog-Gen.1Sg this-Gen.2SG you
`Oh, you dog!'
(meu djamezi.147)
(4b) t maa-r t-gьmь
that what-2SG that-EMPH
(Idiomatic:) What the hell is this?
References: Diessel, Holger (2006): Demonstratives, joint attention, and the emergence of grammar. Cognitive Linguistics 17-4 (2006), 463-489. Lyons, Christopher (1999): Definiteness. Cambridge: CUP. Nikolaeva, Irina (in press): Possessive affixes in the pragmatic structuring of the utterance: Evidence from Uralic. In: Bernard Comrie and Pirkko Suihkonen (eds.) International Symposium on Deictic Systems and Quantification. Izhevsk. Retrieved from http://ling.uni-konstanz.de/pages/home/nikolaeva/. Tereshchenko, Natalja M. (1979): Nganasanskij jazyk. Leningrad: Nauka.
147
The construal of aspect ­ The interplay of boundedness and transitivity in Hungarian Edit Takбcs Language of presentation: English The presentation will discuss the phenomenon of aspect from a functional, cognitive linguistic viewpoint (cf. Langacker 1987, 2008; Croft­Cruse 2004). Aspect is treated as a linguistic phenomenon which determines how an event is represented in discourse, and thus its construal is heavily influenced by the dynamicity of conceptualization, especially in a language like Hungarian where aspect is not a clear syntactic or morphological category (Kiefer 2006). In Hungarian the imperfective viewpoint is the default aspectual category because the bare form of verbs is primarily associated with the intermediary phase of the event and the perfective is typically derived from the imperfective form (syntactically and conceptually alike). Due to that, in Hungarian the boundedness of an event plays a crucial role in the construal of the perfective aspect (cf. Dahl 1985, Huumo 2009, 2010). Under the term boundedness I mean the profiling process which elaborates an internal boundary of the event. This conceptual method (by means of drawing attention to a natural boundary) designates the entire event structure of the verb and thus it contributes to the holistic representation of the event. Boundedness can be achieved in various ways. From the perspective of the primary figure of the event (prototypically the actor) an action can be bounded if a previously set goal is attained. This "actor-oriented" type can be exemplified by motion verbs where the reach or the passing of a place might signal a natural boundary of an event (1). (1) Pйter mozi-ba ment. Peter cinema+into went `Peter went to the cinema' In another case, if an entity is manipulated (or even comes into being) during the proceeding of an event, the total affectedness of that entity might also serve as a natural limit of the action and thus it can signal completion. The best examples of this "theme-oriented" construal of perfective aspect are the achievement and the accomplishment events (2). (2) Krisz йpнtett egy hбz-at. Chris built a house+acc `Chris built a house' It has to be noted that the above outlined characteristic conceptual paths do not constitute exclusive and discrete categories. They can rather be regarded as conceptual preferences which enable the perfective depiction of an event but nonetheless they have a remarkable effect on the choice of syntactic structures. In Hungarian, the use of the accusative case marker and directional (lative and ablative) suffixes stands in accordance with the semantic construal of the events. In my presentation transitivity is treated as a scalar phenomenon, the central element of which is exemplified by the transitive construction while the directional structures constitute peripheral members of the category. In this approach transitivity is a collective syntactic phenomenon which reflects the boundedness properties of the situation (cf. Hopper­Thompson 1980, Goldberg 2006). The presentation will concentrate on the quantitative differences in the conceptualization of the events and the interchangeability of suffixes on the basis of corpus linguistic data. 148
References: Croft, William ­ Cruse, D. Alan 2004. Cognitive Linguistics. Cambridge: Cambridge University Press. Dahl, Цsten 1985. Tense and Aspect Systems. Oxford: Backwell. Goldberg, Adele E. 2006. Constructions at Work: The Nature of Generalisation in Language. Oxford: Oxford University Press. Hopper, Paul J. ­ Sandra A. Thompson 1980. Transitivity in Grammar and Discourse. Language 56. 251­299. Huumo, Tuomas 2009. Fictive dynamicity, nominal aspect, and the Finnish copulative construction. Cognitive Linguistics 20­21. 43­70. Huumo, Tuomas 2010. Nominal aspect, quantity, and time: The case of the Finnish object. Journal of Linguistics 46. 83­125. Kiefer, Ferenc 2006. Aspektus йs akciуminsйg: Kьlцnцs tekintettel a magyar nyelvre [Aspect and Aktionsart: With Special Attention to Hungarian]. Budapest: Akadamйiai Kiadу. Langacker, Ronald W. 1987. Foundations of Cognitive Grammar: Vol. I. Theoretical Prerequisites. Stanford: Standford University Press. Langacker, Ronald W. 2008. Cognitive Grammar: A Basic Introduction. Oxford: Oxford University Press. 149
Deixis and the referential interpretation of conversational narratives Szilбrd Tбtrai, Eцtvцs Lorбnd University Krisztina Laczkу, Eцtvцs Lorбnd University Language of presentation: English Adopting Sinha's (1999) and Tomasello's (1999) views on intersubjectivity and joint reference, this paper offers a model for the interpretation of stories as referential scenes in narrative discourses (cf. Georgakopoulou 2011). It is proposed that narrative interpretation crucially builds on the context-dependent processing of the physical, social and mental worlds presented by the narrative, pertaining to spatio-temporal relations, interpersonal relations and the mental states of participants, respectively (see Tбtrai 2011, cf. Verschueren 1999). Set against the background of social cognitive linguistics (cf. Croft 2009), the paper focuses on the role the referential centre as a context-dependent vantage point which forms the basis for the situative grounding of the narrative's spatial, temporal, and interpersonal relations. The paper demonstrates the model's applicability by a pragmatic analysis of storytelling in discourse mediated by internet (e.g. chats, forums), in particular, in spontaneous written Hungarian texts, thereby offering a contribution to a general approach to the way spatial, temporal and social deictic expression work in conversational narratives. References: Croft, William 2009. Towards a social cognitive linguistics. In: Evans, Vyvyan ­ Poursel, Stephanie (eds.): New directions in cognitive linguistics. Amsterdam: John Benjamins. 395­420. Georgakopoulou, Alexandra (2011). Narrative. In: Jan Zienkowski ­ Цstman, Jan-Ola ­ Verschueren, Jef (eds.): Discoursive Pragmatics (Handbook of Pragmatics Highlights 8). Amsterdam/Philadelphia: John Benjamins. 190­ 207. Sanders, Josй ­ Spooren, Wilbert 1997. Perspective, subjectivity, and modality from a cognitive point of view. In: Liebert, Wolf-Andreas ­ Redeker, Gisela ­ Waugh, Linda (szerk.): Discourse and perspective in cognitive linguistics. Amsterdam, Philadelphia: John Benjamins. 85­112. Sinha, Chris 1999. Grounding, mapping and acts of meaning. In: Janssen, Theo ­ Redeker, Gisela (eds.): Cognitive Linguistics: Foundations, Scope and Methodology. Berlin, New York: Mouton de Gruyter. 223­255. Tбtrai, Szilбrd (2011). Bevezetйs a pragmatikбba. Funkcionбlis kognitнv megkцzelнtйs [Introduction to pragmatics. A functional cognitive approach]. Budapest: Tinta Kцnyvkiadу. Tomasello, Michael 1999. The cultural origins of human cognition. Cambridge MA, London: Harvard University Press. Verschueren, Jef 1999. Understanding pragmatics. London, New York, Sydney, Auckland: Arnold. 150
The augmentative grade of verbal comparative in Komi Triin Todesk, University of Tartu Language of presentation: English
The topic of this paper is a unique phenomenon found in the Permic branch of the Uralic languages, the augmentative grade of verbal comparison or (Cypanov 2005; ЦKK). This paper gives an overview of the formation of the augmentative grade and the main restrictions its usage has with different verb classes in the modern Komi language. The results are based on examples from literature and questionnaires filled by native speakers of Komi. The formation of the augmentative grade of verbs in Komi is quite similar to the comparison of the adjectives ­ the same comparative suffix - is used (in the example glossed as CMPR). Morphosyntactically, there are few restrictions on its use and the suffix can occur with finite verb forms in all tenses and persons in both affirmative (a) and negative, and also with infinitive.
1. Kцi ц-
-ц-, /---/ (Popov 111­112)
where people-NOM.3SG go-3SG-CMPR
'where people go more /---/'
From the lexical semantics point of view, so far it has been stated that the inherent aspect of a verb determines whether it can be given an augmentative grade or not. According to Cypanov verbs of movement, action, thought, state, and change of quality are eligible for comparison, while existential and momental verbs are not (2005: 249). He, however, provides no supporting data for such a generalisation, leaving a major aspect of this phenomenon unexplained. This paper elaborates the question of lexical aspect of the verbs that do or do not have the augmentative grade based on the inherent oppositions stative/dynamic, telic/atelic, and punctual/durative. In addition to the lexical aspect of a verb, this paper also takes into consideration the context the verb can be found in (i.e. does the context change the way the verb could be interpreted and does that affect the usage of the -suffix). Preliminary results show that telicity and duration play an important role. For example, inherently atelic verbs which appear in a telic situation or certain verbs which are inherently telic are unlikely to occur in the augmentative grade. Semelfactive verbs are also rare to appear with the -suffix, mainly due to the lack of their inherent duration.
References: Cypanov 2005 = , .. 2005. . . 244­254. Popov, Aleksei. 2008. Grezd. Syktyvkar. ЦKK = Цi : 2000 / [. . , . . , . . ... .]; ц . . ; , ц, ,
151
Temporal factors of verbal aspect in Hungarian Gбbor Tolcsvai-Nagy, Eцtvцs Lorбnd University, Constantine University Language of presentation: English The paper gives a cognitive linguistic overview of aspect in the Hungarian language, with special focus on the temporal factors. Aspect has no direct and overt grammatical designation by auxiliaries or suffixes in Hungarian. Instead, aspect is construed in the domains of both schema and instantiation: in the semantic structure of the lexical verb itself and in the semantic and syntactic structure of the clause, with its pragmatic context. In the construal of the aspectual content of the verb the `inherent' temporal structure (based on the conceived time of the process, scanned sequentially) and the event structure has their basic roles. The paper focuses on the temporal structure: the component states of the conceived temporal process correspond to the continuous on-going dynamic process of the updating of present, past, and future. Present as present moment is based on dynamic factors of timeconsciousness: primordial event and perceptual moment. Present moment continuously changes into past moments (or past), whereby expected future moments change into present moments. Thus, conceived time is conceptualized through the perspective of the conceptualizer. The schematized sequence of present moments forms the basic sequence of the temporal structure of verbs. From the viewpoint of the conceptualizer, the sequence of present moments with continuous dynamic updating results in imperfective verbal semantic structures, i.e. imperfective verbs, related to the specific event structures. From the viewpoint of the conceptualizer, the sequence of present moments with a specific temporal reference point (a `starting' or `endpoint') subsequent or prior to the process designated by the content verb results in perfective verbal semantic structures, i.e. perfective verbs, related to the specific event structures. The paper gives an analysis of the two types of temporal structures in its variability in Hungarian, asserting that the difference between perfective and imperfective verb is a matter of degree. Aspect in Hungarian is construed via diverse ways of attention focusing on the schematized sequence of present moments. In the default case, verbal prefix (preverb) transforms aspectually less marked imperfective verbs into perfective ones by metonymic attention shift and metaphoric extension. The aspectual instantiation of verbs in the clause has a dynamic nature. The aspectual content of the lexical verb is reinterpreted and elaborated at least partly by tense (epistemic grounding), subjectification, the elaboration of the schematic figures tr and lm by nominals O and S, temporal and locative modifiers, in relation to the semantic type of the verb and the clause construction. These factors have a decisive role in the current attention focusing within the temporal and the event structure of the verb, consequently, in the discourse. References: Brisard, Frank 2002. The English present. In: Brisard, Frank (ed.): Grounding. The Epistemic Footing of Deixis and Reference. Berlin, New York: Mouton de Gruyter. 251­297. Chafe, Wallace 1994. Discourse, Consciousness, and Time: The Flow and Displacement of Conscious Experience in Speaking and Writing. Chicago, London: The University of Chicago Press. Evans, Vyvyan 2003. The Structure of Time. Language, meaning and temporal cognition. Amsterdam, Philadelphia: John Benjamins. Grady, Joseph E. 1999. A Typology of Motivation for Conceptual Metaphor: Correlation vs. Resemblance. In: Gibbs, R. W. ­ Steen, G. (eds.). Metaphor in Cognitive Linguistics. Amsterdam, Philadelphia: John Benjamins. 79­100. Heidegger, Martin 1972. Sein und Zeit. Tьbingen: Max Niemeyer. 12. Auflage. 152
Husserl, Edmund 1966. Zur Phдnomenologie des Inneren ZeitbewuЯtseins. Husseliana X. The Hague: Martinus Nijhoff. Huumo, Tuomas 2009. Fivtive dynamicity, nominal aspect, and the Finnish copulative construction. Cognitive Linguistics 20 (1): 43­70. Kiefer Ferenc 2006. Aspektus йs akciуminsйg. [Aspect and Aktionsart] Budapest: Akadйmiai Kiadу. Langacker, Ronald W. 1987. Foundations of Cognitive Grammar. Volume I. Theoretical Prerequisites. Stanford, California: Stanford University Press. Langacker, Ronald W. 1991. Foundations of Cognitive Grammar. Volume II. Descriptive Application. Stanford, California: Stanford University Press. Langacker, Ronald W. 2008. Cognitive Grammar. A Basic Introduction. Oxford: Oxford University Press. Patard, Adeline ­ Brisard, Frank (eds.) 2011. Cognitive approaches to tense, aspect and epistemic modality. Amsterdam: John Benjamins. Pйter, Mihбly 2008. A magyar aspektusrуl ­ mбs aspektusbуl. [On Hungarian aspect ­ from another aspect] Magyar Nyelv 2008: 1­11. Takбcs Edit 2012. A progresszнv jelentйstartalom konstruбlбsбnak sajбtossбgai a magyarban. [The progressive aspect in Hungarian] Magyar Nyelvr 325­335. Talmy, Leonard 2000. Toward a Cognitive Semantics. Cambridge, MA: The MIT Press. Tolcsvai Nagy, Gбbor 2001. Conceptual metaphors and blends of "understanding" and "knowledge" in Hungarian. Acta Linguistica Hungarica. 48 (1­3). 79­100. 153
Functions of Erzya Translative in Nonverbal Predication Rigina Turunen, University of Helsinki Language of presentation: English In the Uralic languages it is typical to encode construction 'being as an N' with an NP that is encoded by case markers, often called Essive or Translative. Erzya belongs to those Uralic languages that do not display a grammatical opposition between dynamic and stative predicative adverbials, but the functions of both are fulfilled by Erzya Translative (-ks). Also many Erzya verbs have a valency pattern including Translative marked argument. In this paper the focus is on the stative construction 'being (as) an N', in which the nonverbal predicate is Translative. These constructions are compared to those Erzya construction that are otherwise built with same morphosyntactic elements, despite of the nonverbal predicate that is in Nominative It is noticed that the construction types in which the nonverbal predicate is in Translative often seems to bear only vaguely the semantic content 'being as an N'. The data presents various genres, including written standard Erzya, translations and recorded speech. The Translative encoded nonverbal predicate constructions form only a minor group in all the data presenting nonverbal predication in Erzya: among 5000 clauses there are 150 clauses in which the predicative nominal argument occurs in Translative. The occurrence of Translative case depends on the predication pattern, since nominal and adjectival predicates may be encoded in either Nominative or Translative case when predication is made using the u()ems 'be' copula. Copula constructions are not employed to refer to present tense in Erzya, which affects the frequency of Translative in nonverbal predication. Also lexical class of the nonverbal predicate affects the use of Translative: Translative occurs far more often if the predicate is a noun than if it is an adjective. The adjectival pronouns ija 'other' , iamo 'such' and kodamo 'what kind' are, however, relatively often encoded in Translative (10 clauses). Additional data of 500 clauses with Translative encoded NP's in verbal predicate constructions is also gathered but the results are so far preliminary and not presented here. Stassen (2001) summarises that in the Circum-Baltic languages the employment of the nominative indicates class membership which is in some way an essential and permanent feature of the subject. The temporary, contingent or non-essential property or class-membership are encoded with the translative and essive cases (Finnic languages) or instrumental (Lithuanian, Russian, Polish). In the Slavic languages of the area, however, time stability does not play a major role; rather the switch is conditioned by many formal, semantic and stylistic factors. The Mordvin languages are not included in Stassen's study of the nominative-oblique switch. On the basis of my study it seems that the functions of Erzya Translative are to encode temporary properties in nonverbal predication, but there are also other factors, that show similarity with the functions of Russian Instrumental in corresponding clauses. Furthermore, Erzya Translative has strong function as a marker of comparison/similarity. This function is more frequent and more obvious in verbal predicate constructions, but it appears also in nonverbal predication. References: Stassen, Leon 2001: Nonverbal predication in the Circum-Baltic Languages. In Dahl, Цsten & Maria Koptjevskaja- Tamm (eds.) Circum-Baltic Languages. Volume 2. Grammar and Typology. Studies in Language Companion Series. Amsterdam/Philadelphia: John Benjamins. 569­590. 154
Mixed dialects ­ mixed syntaxes. A study of openly used person forms in Border Karelian dialects (Murteet ja lauseopit tцrmдyskurssilla. Tutkimus rajakarjalaismurteiden avoimista persoonaviittauksista) Milla Uusitupa, University of Eastern Finland Language of presentation: Finnish The category of person in grammatical system has been a topic of concern both with language typologists and Finnish linguists during the last two decades (i.a. Siewierska 2004, Laitinen 2006, Helasvuo­Vilkuna 2008, Helasvuo 2008). The picture of function and the expression of person forms have diversified notably (e.g. Helasvuo­Laitinen 2006, Leino­Цstman 2008). This paper strives to still widen the perspective to spoken language use and language contacts by introducing a contact situation between the Eastern Finnish dialects and their closest cognate, Karelian language. The paper has two major aims. Firstly, it seeks to illustrate the contact situation between Finnish and Karelian. Until recently the interrelationship between these two cognates has remained almost unexplored. Little attention has been paid to syntactic issues in particular (however see Sarhimaa 1999). Secondly, this paper yields new insight about the category of person in Finnic languages. There are various means of open reference among different cognates, two of which come across in the contact of Finnish and Karelian. This paper focuses on the area called Border Karelia situated in the borderline of Finland and Russia where both Finnish and Karelian were originally spoken side by side. In Standard Finnish, the most common means of open reference is the 3rd person singular, the so-called zero person. It is also by far the most examined. Yet, in many other Finnic languages, in Karelian and Estonian for instance, the more frequent one is the openly used 2nd person singular. Interestingly, it has originally belonged to the Eastern Finnish dialects as well. The present paper reveals how the open use of the 2nd and the 3rd person singular differ syntactically and semantically. Examples from primary data recorded in the 1960s will show that the 2nd and the 3rd person singular are not interchangeable. The choice of one type of person form over the other has an influence on mood, tense and other grammatical categories as well. Diverse examples will also illustrate why the concept of open reference is so practical. The reference is, indeed, constructed in the interplay between the interviewer and the informant. References: HELASVUO, MARJA-LIISA 2008: Minд ja muut. Puhujaviitteisyys ja kontekstuaalinen tulkinta. ­ Virittдjд 112, 186­ 206. HELASVUO, MARJA-LIISA ­ LAITINEN, LEA 2006: Person in Finnish: paradigmatic and syntagmatic relations in interaction. ­ Marja-LiisaHelasvuo& Lyle Campbell (toim.), Grammar from the human perspective, p. 173­207. Amsterdam: John Benjamins. HELASVUO, MARJA-LIISA ­ VILKUNA, MARIA 2008: Impersonal is personal: Finnish perspective. ­ Transactions of the Philological Society Volume 106:2, 216­245. LAITINEN, LEA 2006: Zero person in Finnish. A grammatical resource for construing human reference. ­ MarjaLiisaHelasvuoja Lyle Campbell (toim.), Grammar from the human perspective: case, space and person in Finnish. CILT 277, 209­231. Amsterdam: John Benjamins. LEINO, PENTTI ­ ЦSTMAN, JAN-OLA 2008: Language change, variability, and functional load. Finnish genericity from a constructional point of view.­JaakkoLeino (toim.), Constructional Reorganization. Constructional Approaches to Language 5, 37­54. Amsterdam: John Benjamins. SARHIMAA, ANNELI 1999: Syntactic transfer, contact-induced change, and the evolution of bilingual mixed codes. Focus on Karelian-Russian language alternation. StudiaFennicaLinguistica 9. Helsinki: Finnish Literature Society. SIEWIERSKA, ANNA 2004: Person. Cambridge University Press. 155
Pick it up: Referential Devices in Estonian Child-Directed Speech Virve-Anneli Vihman, University of Tartu Kriste Lauk, University of Tartu Language of presentation: English Because so much First Language Acquisition research has focussed on English, there are great gaps in knowledge on acquisition of languages with flexible word order, elaborate case systems, and extensive use of ellipsis (Slobin et al. 2011). Languages in the Uralic family provide fertile ground for study; in this talk we focus on the speech heard by children acquiring Estonian. Charting Child-Directed Speech (CDS) is an important step in both acquisition research and spoken language research more generally. We examine the expression of core arguments in one Estonian mother's speech to a child (aged 2;0, from CHILDES, childes.psy.cmu.edu), and compare the data with adult-to-adult spoken language data (adult-directed spech, ADS, from the University of Tartu's Corpus of Spoken Estonian, www.cl.ut.ee/suuline/Korpus). The context of recorded caregiver-child interactions is often limited to the two interlocutors, and the subjects of conversation are typically bounded by the immediate surroundings (Soderstrom 2007). Considering the circumscribed context of interaction, reference to shared visual information and joint focus of attention, one might expect patterns of reference to reflect this high degree of shared information, and indicate high salience and accessibility of referents (Gundel et al. 1993). However, our analysis paints a different picture. The use of referring expressions in the caregiver data analysed here is quite distinctive, leading us to ask how children arrive at a more complete grasp of the referential system and its usage. Preferred Argument Structure (PAS, DuBois et al. 2003) may provide a useful key to how the system is pieced together. Unsurprisingly, one-on-one conversation with toddlers includes a greatly reduced number of declarative sentences (43% in our CDS sample vs. 62.5% in ADS), in favor of imperatives (25%) and interrogatives (24%). The person-marking on verbs differs in some categories as well, showing far greater use of 2sg and smaller use of 1sg marking, an effect of both the subject matter and grammatical choice. The most interesting differences, however, bear relevance to PAS and acquisition of the semantics, syntax and pragmatics of referential devices (cf Allen & Schroder 2003, Clancey 2009). The use of full noun phrases is considerably greater in CDS than ADS, complemented by lower pronominal usage: whereas the ADS employed pronouns nearly twice as often as full NPs, in the CDS data full NPs accounted for half of all overt arguments. Ellipsis was used most frequently in intransitive S arguments, not A arguments as predicted by PAS, but full NPs also occurred more frequently in intransitive constructions than elsewhere. Evidence for the differential O-marking emblematic of Finnic languages also patterns differently in CDS than in adult data, mostly as an effect of the higher proportion of imperatives, which alternate partitive/nominative objects (rather than the usual partitive/genitive distinction). This paper pulls together information on grammatical role, morphosyntactic form and case to exemplify how CDS is marked in comparison with ADS, and to determine which differences cannot be accounted for by situational context and explore what effect they may have on the child acquiring argument structure in Estonian. References: Allen, S.E.M., & Schroder, H. 2003. The role of preferred argument structure in early Inuktitut spontaneous speech data. In J. Du Bois, L. Kumpf & W. Ashby (eds.), Preferred Argument Structure: Grammar as architecture for function, 301-338. Amsterdam: John Benjamins. 156
Clancy, P. 2009. The acquisition of argument structure and transitivity in Korean: a discourse-functional approach. In C. Lee, G. Simpson, and Y. Kim (eds.) The Handbook of East Asian Psycholinguistics, vol. III Korean, pp. 34-49. Du Bois, J.W., Kumpf, L.E. & Ashby, W.J. (eds.) 2003. Preferred Argument Structure: Grammar as architecture for function. Amsterdam: John Benjamins. Gundel, J.K., Hedberg, N. & Zacharski, R. 1993. Cognitive Status and the Form of Referring Expressions in Discourse. Language 69:274-307. Slobin, D. I., Bowerman, M., Brown, P. Eisenbeiss, S., & Narasimhan, B. 2011. Putting things in places: Developmental consequences of linguistic typology. In J. Bohnemeyer & E. Pederson (eds.), Event representation in language and cognition, 134-165. Cambridge: CUP. Soderstrom, M. 2007. Beyond babytalk: Re-evaluating the nature and content of speech input to preverbal infants. Developmental Review, 27: 501-32. 157
Something verb-final in Finnish: The case of conditional clauses
Maria Vilkuna, Institute for the Languages of Finland Language of presentation: English
Finnish, like most Uralic languages, allows both VO (more generally, VX) and OV (XV) order. The dominating order is (S)VX, and (S)XV exists as a minority pattern mainly confined to certain specific constructions; it is never obligatory or predictable. Moreover, rather than having a specific discourse function of its own, XV is more of an accompanying phenomenon, resorted to when the main focus of the utterance falls outside the XV part of the sentence. One case in point is the affirmation pattern with the initial particle kyllд, which is particularly likely to trigger XV order, illustrated in (1).
(1) Kyllд minд siitд jotakin
PARTICLE I
about.it something
`I do know something about it.'
tiedдn. know
The present paper discusses one of the syntactic contexts of XV order, viz. subordinate clauses, building on the ideas of Lindйn (1959) more recently discussed by Vilkuna (to appear), both of whom investigate the phenomenon in traditional dialects. Temporal `when' clauses with XV order typically present states of affairs as backgrounded in the sense of indicating routine, stereotypical temporal settings, while VX ordered clauses have a larger and more open set of functions. In relative clauses, XV correlates with referent identification as opposed to description or just adding new information about the referent, which is usually done with VX. My presentation will focus on conditional clauses in dialects and colloquial Finnish. In the dialectal data, XV order often has to do with picking out a potential state of affairs and contrasting it with alternatives, as opposed to just introducing such a state of affairs. More generally, XV ordered conditional clauses are quite specialized, occurring in particular contexts with particular rhetorical import. In one recognizable subconstruction, the conditional clause contains the indefinite pronoun/quantifier `someone, something' and has the function of challenging a presupposition that might serve as a background assumption for the utterance. There is a sharp contrast between the XV ordered (2) and the VX ordered (3); both examples come from informal internet discourse.
(2) Jos minд jotakin inhoan niin epдvarmuutta.
if I
something loathe so uncertainty
'If there is something I loathe, it is uncertainty.'
(3) Jos minд inhoan jotakin asiaa
if I loathe some thing
niin en taatusti eksy asiasta keskusteleville sivuille.
'If I loathe some issue, I certainly won't find myself on pages discussing that issue.'
In all different types of XV clauses, "old information" is somehow involved: individual referents as well as points of time are typically identified on the basis of known information, and it is information that is somehow entertained that can be challenged with a conditional like (2). I assume that this property connects the type in (2) with the XV ordered pattern in (1). The presentation will describe the focus properties of these kinds of XV order.
158
References: Lindйn, Eeva 1959: Hypotaktisen sanajдrjestyksen tehtдvistд lounaismurteissa [On the role of hypotactic order in south-western dialects]. In Virittдjд 63, 252--266. Vilkuna, Maria to appear: More subordinate? Verb-final order and subordination Finnish Dialects. In Kalliokoski, Jyrki & Helena Sorva & Laura Visapдд (eds.), Contexts of Subordination. Cognitive, interactional and typological perspectives. John Benjamins. Vilkuna, Maria 1989: Free word order in Finnish. Its syntax and discourse functions. Helsinki: Suomalaisen Kirjallisuuden Seura. 159
Transitivity in Eastern Mansi: An information structural approach Susanna Virtanen, University of Debrecen, University of Helsinki Language of presentation: English
In my paper I will present the system of expressing transitivity in Eastern Mansi from the point of view of information structure. I rely on the notion of transitivity based on semantic-functional factors, and I will provide a comprehensive description of the expression of transitivity as a whole. My aim is to show, how three separate syntactic notions ­ DO coding, three participant constructions and the passive ­ are connected to each other by very simple information structural factors. Concerning information structure, I mainly use the terminology and notions by Lambrecht (1994). My other theoretical dimension is the notion of Differential Object Marking (DOM) (see Bossong 1985, Aissen 2003, Iemmolo 2010/2011). My research data is collected from the folklore collection gathered by Artturi Kannisto in the beginning of 1900s and published some decades later (see Kannisto 1951, 1955, 1956, 1958 and 1963), which can be considered as a vast and variable database representing different text genres. In the Eastern dialects of Mansi, DO's can be either morphologically unmarked (1), only verb marked (Direct Object Indexation) (2), verb marked and case marked (3) or verb marked and possessive marked (4).
(1) kom kooj ­шs man meet ­PST 'He met a man.'
(2) nee
at tддrшt ­iitш.
woman NEG let ­SG'The woman does not let (the man) go.'
(3) sos ­mш uus kцдlt ­шs ­tш. moose ­ACC again frighten ­PST ­SG(4) ддk ­шn
komшly woеxtl ­шs ­lшn!
uncle ­ SG2SG how
leave ­ PST ­SG<2SG
'How could you leave your uncle?'
In some exceptional cases case marking may appear without object indexation, or object indexation appears with an unmarked noun. Furthermore, the place of a DO can be occupied by a Patient or an R-argument. The same action can also be expressed with the passive (5) where the Patient or the R-argument occupies the subject position.
(5) pеsшng-kom ­nш keet ­w ­шs ­шm
bright-man ­LAT
send ­PASS ­PST ­SG
`I was sent by the bright man.'
The basis of this complex on variation is that certain syntactic functions are occupied by certain pragmatic functions, and morphological marking is dependent on the topicality level of the argument. In my paper I will present the principles of the topicality-based variation and especially how the same principles cover the whole system of expressing transitivity.
160
References: Aissen, Judith 2003: Differential Object Marking: Iconicity vs. Economy. Natural Languages & Linguistic Theory 21, 435-483. Bossong, Georg 1985: Differentielle Objektmarkierung in den neuiranischen Sprachen. Empirische Universalienforschung. Tьbingen: Gunter Narr Verlag. Iemmolo, Giorgio 2010/2011: Towards a typological study of Differential Object Marking and Differential Object indexation. Dissertation. University of Pavia. Kannisto, Artturi 1951: Wogulische Volksdichtung I. Texte Mythischen Inhalts. Bearbeitet und hrsg. von Matti Liimola. MSFOu 101. Helsinki. Kannisto, Artturi 1955: Wogulische Volksdichtung II. Kriegs- und Heldensagen. Bearbeitet und hrsg. von Matti Liimola. MSFOu 109. Helsinki. Kannisto, Artturi 1956: Wogulische Volksdichtung III. Mдrchen. Bearbeitet und hrsg. von Matti Liimola. MSFOu 111. Helsinki. Kannisto, Artturi 1958: Wogulische Volksdichtung IV. Bдrenlieder. Bearbeitet und hrsg. von Matti Liimola. MSFOu 114. Helsinki. Kannisto, Artturi 1959: Wogulische Volksdichtung V. Auffьhrungen beim Bдrenfest. Bearbeitet und hrsg. von Matti Liimola. MSFOu 116. Helsinki. Kannisto, Artturi 1963: Wogulische Volksdichtung VI. Schicksalslieder. Bearbeitet und hrsg. von Matti Liimola. MSFOu 134. Helsinki. Lambrecht, Knud 1994: Information Structure and sentence form. Topic, focus, and the mental representations of discourse referents. Cambridge Studies in Linguistics 71. Cambridge. 161
Exploring the construal of change in Finnish translative expressions
Eero Voutilainen, University of Helsinki Language of presentation: English In this paper, I will examine the construal of change in the use of Finnish translative (`to/for') case, e.g. Hдn tuli iloiseksi (3SG.hum become-PST happy-TRA 'He/she became happy'). As theoretical and methodological framework, I will use cognitive grammar (e.g. Langacker 2008), especially the concept of construal, i.e. the way that conceptual content is viewed linguistically (ibid. ch. 3). Change is an essential category in human conceptualization (Langacker 2008: 33). It is important to note, however, that the concept of change can be treated in several different ways. According to a strict definition, for example, a relevant change occurs, when there is a shift in a basic conceptual relationship (e.g. inclusion, separation, identicality, association, and contact; Langacker 1987: 230; Lakoff 1987: 420) (Nurminen 2011). At the other end of the cline, any alteration in a thing or a relationship can be acknowledged as change. In my paper, I will consider these different perspectives and the consequences they have on the grammatical analysis of change. The construal of change is a holistic phenomenon that cannot be reduced to a single grammatical element, such as the finite verb in the clause. Instead, it depends on the expression as a whole and potentially even on its broader textual or situational context. I will demonstrate this by a morpho-syntactic and semantic analysis of translative, where the construed change, or resultativity (see Boas 2003; Fong 2003; Pдlsi 2000), varies diversely depending on the verb and argument structure of the clause. A basic division can be made between expressions of actual change (1) and fictive or subjective change (2) (cf. Huumo 2007; Matsumoto 1996; Sweetser 1997).
(1) Hдnet on valittu
johtoryhmдn jдseneksi.
He-ACC choose-PASS-PRF board-GEN 'He has been chosen as a board member'
member-TRA
(2) Ammoniakkikдsittely on havaittu
hyцdylliseksi.
Ammonia.treatment
perceive-PASS-PRF useful-TRA
'Ammonia treatment has been found useful'
Within these two basic categories, several distinctive types of translative expressions may be discovered. The expressions of actual change can be further divided, for example, into constructions of alteration, transition, switch, and coming into existence. The expressions of fictive change, on the other hand, profile mental processes, such as perceiving, turning out, evaluating, defining, calling, and describing. Furthermore, different translative expressions are used to describe directional and dynamically construed relationships that are closely related to change, such as leaving and remaining, purpose, consequence, suitability, comparison, and various temporal relationships. Together with seven other adverbial cases, translative is traditionally described as a local case (see Huumo ­ Ojutkangas 2006). However, as well as expressions with essive ('as') case, most of the translative expressions can be considered local only in a metaphorical way. To overcome this descriptive inadequacy, I will present an alternative view on Finnish adverbial cases based primarily on the notions of stativity and dynamicity. My data is a corpus of 2120 translative clauses collected from the Finnish daily newspaper Helsingin Sanomat in 1981.
162
References: Boas, Hans 2003: A constructional approach to resultatives. CSLI Publications. Fong, Vivienne 2003: Resultatives and depictives in Finnish. Diane Nelson ­ Satu Manninen (eds.), Generative approaches to Finnic and Saami linguistics. The University of Chicago Press, 201­233. Huumo, Tuomas 2007: Force dynamics, fictive dynamicity and the Finnish verbs of `remaining'. Folia Linguistica 41, 7398. Huumo, Tuomas ­ Krista Ojutkangas 2006: An introduction to Finnish spatial relations. Local cases and adpositions. Marja-Liisa Helasvuo & Lyle Campbell (eds.) Grammar from the human perspective: Case, space and person in Finnish. John Benjamins, 11­20. Langacker, Ronald 1987: Foundations of cognitive grammar, vol. 1: theoretical prerequisites. Stanford University Press. Langacker, Ronald 2008: Cognitive grammar. A basic introduction. Oxford University Press. Lakoff, George 1987: Women, fire, and dangerous things. What categories reveal about the mind. University of Chicago Press. Matsumoto, Yo 1996: Subjective-change expressions in Japanese and their cognitive and linguistic bases. ­ Gilles Fauconnier ­ Eve Sweetser (eds.) Spaces, worlds, and grammar. University of Chicago Press, 124­156. Nurminen, Salla 2011: Aspektin ilmaiseminen suomen kielessд ­ lauseen ominaisuus. Sananjalka 53, 6183. Pдlsi, Marja 2000: Finnish resultative sentences. SKY Journal of Linguistics 13, 211­250. Sweetser, Eve 1997: Role and individual interpretations of change predicates. ­ Jan Nuyts and Eric Pederson (eds.) Language and conceptualization. Cambridge University Press, 116­136. 163
Intermediate stages in the grammaticalization of Northern Samoyedic objective affixes Melani Wratil, University of Dьsseldorf Language of presentation: English It is a matter of a still ongoing debate, what determines the choice between the subjective and the objective inflection in Northern Samoyedic. Uralists like e.g. Terescenko (1956, 1979), Ristinen (1973), Kьnnap (1999), Pusztay (2001), Kцrtvйly (2005) and Dalrymple & Nikolaeva (2011), tentatively agree with each other that, roughly speaking, the choice of the respective allomorphs is in some way connected with the pragmatic statuses of the participants the arguments of a clause refer to and with clausal transitivity. However, their considerations neither offer a definite explanation for these specific correlations nor do they account for the differences in use of the conjugation types between the various Northern Samoyedic languages. So, why is there at all a close relationship between the insertion of verbs inflected in the objective conjugation and the topicality of object referents? Why is it nevertheless impossible for verbs that bear an objective suffix to co-occur with object personal pronouns, which undoubtedly refer to highly topic-worthy entities? And finally, why is the insertion of lexical objects in Nganasan much more syntactically restricted than in the other Northern Samoyedic languages? In my talk I will address these questions. More precisely, I will follow Havas' (2004) and Coppock's & Wechsler's (2010) considerations about the emergence of the Hungarian conjugations and apply them, at least partially, to the development of the Samoyedic objective conjugation. I will argue that only the 3Ps objective suffixes have arisen from the incorporation of the corresponding anaphoric object pronouns. The other objective agreement markers that indicate the presence of a pronominal object are relics of the original general conjugation. They have been reanalyzed as suffixes that enclose the least marked object pronoun. Their objective dual and plural endings have later come about by additional affixation. As I will illustrate, the 3Ps object personal pronouns of Nganasan are still exclusively morphologically realized by the suffixes of objectively conjugated predicates. They only cooccur with lexical objects if they are bound by the latter as resumptive pronouns in a typical Clitic Left-Dislocation construction. In Enets and Nenets the objective suffixes retain features of 3Ps object personal pronouns but already exhibit essential properties of verbal markers. Accordingly, they neither allow the additional insertion of overt 3Ps direct object personal pronouns nor are they unconditionally compatible with direct objects that exhibit any other phifeature specification. The lexical objects they may co-occur with still obligatorily refer to thematic entities and are preferably topicalized by fronting. However, resumptive binding from a left-dislocated position is no longer required. I will show that the agreement affixes of the objective conjugation in Enets and ­ albeit to a lesser extent ­ in Nenets nowadays significantly support the identification of grammatical relations. Hence, in my talk I will elucidate that the morphosyntactic properties of the objective inflection and its individual use in the various Northern Samoyedic languages can be explained by the attainment of different intermediate stages on the grammaticalization path from a pronominal category to a verbal agreement marker. References: Coppock, E. & S. Wechsler (2010): Less-travelled paths from pronoun to agreement: the case of the Uralic objective conjugations. In M. Butt & T. Holloway King (eds.): Proceedings of the LFG10 Conference. Stanford, CSLI. Dalrymple, M. & I. Nikolaeva (2011): Objects and information structure. Cambridge, CUP. Havas, F. (2004): Objective Conjugation and Medialisation. Acta Linguistica Hungarica 51, 95-141. Kцrtvйly, E. (2005): Verb Conjugation in Tundra Nenets, PhD thesis, SzTE Finnugor Tanszek, Szeged. 164
Kьnnap, A. (1999): Enets. Mьnchen / Newcastle, Lincom. Pusztay, J. (2001): Determinaltsag a konjugugacioban: Magyar-jurak-szamojed цsszevetes. Az 1999-es Veszprem Alkalmazott Nyelveszeti Kongreszus konstriv szekciojanak elцadasai. Veszprem. Ristinen, E.K. (1973): Some remarks on the Function of the Subjective and Objective Conjugation in Samoyedic. JSFOu 72, 337-347. Terescenko, N. (1956): Materialy i issledovanija po jazyku nencev. Leningrad, Akademija nauk. Terescenko, N. (1979): Nganasanskij jazyk. Leningrad, Nauka. 165

Z Gécseg

File: topics-in-focus-position-in-hungarian.pdf
Title: book_of_abstractsGC2.pdf
Author: Z Gécseg
Author: Maarja-Liisa
Published: Mon May 6 17:43:58 2013
Pages: 165
File size: 1.66 Mb


Health fair planning guide, 60 pages, 0.43 Mb
Copyright © 2018 doc.uments.com