Linguistics Prague 2017
Thursday 27 April – Saturday 29 April 2017
Charles University, Faculty of Arts
Department of Linguistics
Institute of Czech Language and Theory of Communication
1
2
PLENARY TALKS
Compilation and exploitation of corpora of previously unresearched languages............................. 5
Ulrike Mosel
The influence of the visual modality on language structure: Insights from sign language and
gesture............................................................................................................................................................ 5
Pamela Perniss
The interaction of corpus compilation and lexicography in language documentation projects .....6
Ulrike Mosel
TUTORIAL WORKSHOPS
Learning how to ask – a condensed introduction to fieldmethods ......................................................6
Florian Siegl
Acceptability judgment task: its design and interpretation .................................................................. 7
Jakub Dotlačil
Phonetics phonology interface: what is phonological? .........................................................................8
Míša Hejná
Multimodal corpora: annotation practices, tools and analysis of the (multimodal) spoken and
sign language corpus data ..........................................................................................................................8
Jakub Jehlička
ORAL PRESENTATIONS
Variation in Lovari morphology .................................................................................................................9
Márton András Baló
Existential and locative sentences in Central and Southern Selkup ................................................... 11
Josefina Budzisch
Effects of plausibility on sentence comprehension ...............................................................................12
Jan Chromý
“The law shall prevent our national language from devastation”: Language ideologies in the
Czech Parliament ........................................................................................................................................13
Ondřej Dufek
The Role of Interactional Alignment in Communication Accommodation Theory ......................... 14
Mirjam Elisabeth Eiswirth
Clitic dA’s Relevance to Conversation Mechanisms ............................................................................ 15
Münevver Erdem-Akşehirli
An areal study of valency markers – preliminary results.................................................................... 16
Jerzy Gaszewski
Complementizer, conjunction, discourse connective: the categorial status of co in spontaneous
conversational Polish ................................................................................................................................ 18
Wojciech Guz
Investigation of temporal and spatial reference frames in Hungarian ............................................. 19
Veronika Harmati-Pap
3
The semantics of -ILE in Nyamwezi .........................................................................................................21
Ponsiano Sawaka Kanijo
Lexical hybrids in German dialects: Investigations on agreement forms ......................................... 23
Stephanie Leser-Cronau
Vowel (Dis)harmony in Czech .................................................................................................................24
Jiří Milička & Hana Kalábová
Pre- and postverbal auxiliaries in Jinghpaw ..........................................................................................24
André Müller
English as-parenthetical construction: A construction-based perspective ......................................26
Seulkee Park & Jong-Bok Kim
The pragmatics of ‘definition’: verbal interaction and social reality ................................................. 27
Svitlana Pereplotchykova
The Spoken Production of Gender-Neutral Nouns in German – some Preliminary Results ..........29
Korbinian Slavik, Johanna Cronenberg, Erato Balafa, Christoph Draxler
Subject-verb agreement in German speakers: Pupillometry as a measure of sentence processing..... 31
Assunta Suess
Aspectual prefixes in Palaung Rumai? ................................................................................................... 32
Rachel Weymuth
POSTER PRESENTATIONS
A parallel text analysis of agreement in Germanic ............................................................................... 33
Magnus Breder Birkenes & Stephanie Leser-Cronau
The tense system of Dolgan – A corpus-based analysis ......................................................................34
Chris Lasse Däbritz
Grammatical status of the pointing sign “THIS” in Polish Sign Language (PJM) – corpus-based
study.............................................................................................................................................................36
Joanna Filipczak & Monika Krawczyk
The diachronic transformation of {-mAK} into {-mA} in genitive-possessive agreement contexts .. 37
Duygu Göksu
Areal variation in case government of the German verb vergessen ‘to forget‘ in contrast with
Czech zapomínat/zapomenout ‘to forget‘ ..............................................................................................39
Agnes Kim & Sebastian Scharf
Relationship between executive function and sign language skills in bilingual deaf children .... 41
Justyna Kotowicz & Magda Schromová
Grammaticalisation and Lexicalisation of the Chinese verb 断 .........................................................42
Wenchao Li
Serial Verb Constructions in Egyptian Arabic .......................................................................................43
Adam Pospíšil
Light verbs & morphologic disambiguation in Czech: nominative-accusative syncretism ......... 44
Karolína Vyskočilová
Pronoun vs. zero-anaphora: the expression of direct object referents in Selkup ............................ 45
Hannah Wegener
4
Compilation and exploitation of corpora of previously unresearched languages
Ulrike Mosel
Christian Albrechts Universität, Kiel
27 April, Thursday 10.45
After a short introduction into corpus linguistics and a comparison of corpora of underresearched languages (ULDs) with those of major European languages, Ulrike Mosel shows
how the corpus of an Oceanic language can be exploited for two kinds of syntactic studies.
From a semasiological perspective she first investigates different registers with respect to
clause linkage, and then she presents an onomasiological case study of comparative
constructions. The presentation concludes with recommendations for the compilation of
corpora of ULDs and leads to a discussion of the impact of this kind of corpus linguistics on
linguistic typology.
The influence of the visual modality on language structure: Insights from sign
language and gesture
Pamela Perniss
College of arts and humanities, University of Brighton
28 April, Friday 9.15
The human ability to communicate and use language is instantiated not only in the vocal
modality but also in the visual modality, as exemplified in sign languages and (co-speech)
gestures. Sign languages, the natural languages of Deaf communities, use systematic and
conventionalized movements of the hands, face, and body for linguistic expression. Cospeech gestures, though non-linguistic, are produced in tight semantic and temporal
integration with speech and have been argued to constitute an integral part of language. In
both signing and gesturing, the use of the hands and body allows visually motivated
representations of objects and events with a high degree of iconicity, i.e. resemblance between
form and meaning. This talk explores the influence of the visual modality on language
structure through corpus-based, cross-linguistic analysis of different sign languages, on the
one hand, and sign and co-speech gesture, on the other hand. How much structural similarity
vs. diversity can we expect across sign languages in domains that rely on the visual-spatial
affordances of the visual modality? How can we characterize the affordances of the visual
modality with respect to its use as the sole modality as expression, as in sign, or together with
the vocal modality, as in gesture?
5
The interaction of corpus compilation and lexicography in language
documentation projects
Ulrike Mosel
Christian Albrechts Universität, Kiel
29 April, Saturday 9.30
This paper gives first a general introduction into corpus-based lexicography and then
discusses the following issues:
1. the complementation of a corpus of narrative and descriptive texts by elicited lexical data
2. the compilation of a corpus-based lexical database
3. the problem of meaning descriptions and translation equivalents
4. the selection of examples
5. making a dictionary on the basis of a lexical database.
Learning how to ask – a condensed introduction to fieldmethods
Florian Siegl
UiT - Norges arktiske universitet
<[email protected]>
Interviews and elicitation are among the central tools of linguistic informant work. In this
respect, fieldlinguistics is just another technique of data gathering. Although informant work
has a firm position in disciplines such as e.g., sociology or anthropology, data gathering via
interviews and elicitation is not regularly taught in linguistics, sociolinguistics being perhaps
the only clear exception. However, when starting to work on an unknown language, asking
questions becomes mandatory as this is the way to accumulate data and to learn the basics of
the language (given that both the consultant and the linguist share another language). In order
to achieve one’s goal, the fieldworker needs to learn how to ask and what to ask.
This fieldwork tutorial will teach the premises of informant work typical for the early stage
of fieldwork. In this period, a researcher needs to rely on direct interaction with a consultant
almost exclusively. The language of this fieldwork tutorial with whom we will try to become
acquainted with is Sakha (also known as Yakut), a Turkic language of North-Eastern Siberia.
The working sessions will be held in English and when required Russian will be used as an
additional means of communication. Due to restriction of time, the sections will be centered
around several structured blocks a) recording settings b) initial work on phonetics/phonology
via lexical work c) non-verbal and intransitive predication and d) transitive predication.
Initially, the attendees get a short introduction in the beginning of blocks b)-d) how to
structure elicitation tasks and how such work could be done. After this, the instructor will take
a back seat in order to let the attendees try to gather data themselves. Additional guidance will
be offered if this becomes necessary. After this, a short demonstration of how to transliterate
6
narratives with a native speaker will follow in order to show how the next phase of consultant
work looks like.1
Students who are interested in attending this course should have passed the introduction to
linguistics and have active command of English in order to participate in the elicitation task;
ideally, a deeper background in morphology and syntax would be suitable, but this
requirement is optional. If you want to participate as an observer without engaging in
elicitation yourself, please contact the local organizers to see whether there is space.
Potential background reading:
Beer, Bettina (Hg.) 2003. Methoden und Techniken der Feldforschung. Ethnologische
Paperbacks. Berlin: Dietrich Reimer Verlag.
Biggs, Charles L. 1986. Learning How to Ask. A Sociolinguistics Appraisal of the Role of the
Interview in Social Science Research. Studies in the Social and Cultural Foundations of
Language 1. Cambridge: CUP.
Chelliah, Shobhana & Reuse, Willem J. de 2011. Handbook of Descriptive Linguistic
Fieldwork. Heidelberg, London, New York: Springer.
Gippert, Jost & Himmelmann, Nikolaus P. & Mosel, Ulrike (eds.) 2006. Essentials of
Language Documentation. Berlin, New York: Mouton deGruyter
Russel, Bernard H. 2006. Research Methods in Anthropology – Qualitative and Quantitative
Approaches. 4th edition. New York, Oxford: Altamira Press. online at
http://www.antropocaos.com.ar/Russel-Research-Method-in-Anthropology.pdf
Sakel, Janette & Everett, Daniel L. 2012. Linguistic Fieldwork: A Student Guide. Cambridge
Textbooks in Linguistics. Cambridge: CUP.
Tieberger, Nick (ed.) 2011. The Oxford Handbook of Linguistic Fieldwork. Oxford: OUP.
Acceptability judgment task: its design and interpretation
Jakub Dotlačil
University of Groningen
<[email protected]>
The acceptability judgment task belongs to the most common methodologies used to gather
data in linguistics. The task can be used to assess the status of a linguistic construction or the
availability of an interpretation, among other things.
In the workshop, we will study the task in detail: its use in linguistics, its application, its
analysis. We will try a hands-on approach, creating and analyzing one acceptability
experiment during the workshop. Issues that might trip you up when designing an
acceptability experiment will be also discussed.
1
In a later phase of fieldwork when the researcher has achieved more competence, data gathering via recording
of spontaneous speech, participant observation or interviews in the vernacular will take over. Still, elicitation
will continue to be relevant, even if its role will gradually decrease.
7
Phonetics phonology interface: what is phonological?
Míša Hejná
Newcastle University
<[email protected]>
The advent of more precise acoustic and articulatory techniques in the study of speech
production has stimulated debates surrounding the question which is fundamental within the
phonological domain of language: how do we establish whether a certain phenomenon is of a
phonological nature? The existence of uncertain or "difficult" phenomena in this regard has
stimulated interest in the phonetics-phonology interface.
The first half of the workshop will be based on the discussions of selected readings. These
readings will represent the main (extreme) approaches to the relationship between phonetics
and phonology (modular, quasi-modular, fully phonological non-modular, and fully phonetic
non-modular). The second half of the workshop will consist of discussions of
methodologically oriented exercises.
These will revolve around specific phenomena and how to determine whether their status
is phonetic or also phonological.
The readings used in the first half of the workshop will feed into this second half, and it is
essential that participants come prepared. Participants are further encouraged to email me
about phenomena that are of particular interest to them.
Multimodal corpora: annotation practices, tools and analysis of the (multimodal)
spoken and sign language corpus data
Jakub Jehlička
Charles University
<[email protected]>
The interactive workshop will focus on convergences between the study of co-speech gestures
in spoken language and the corpus-based sign language linguistics.
In the first half of the session, a general overview of multimodal corpus-based research
methods will be given, including both spoken and sign language corpora. Examples of
multimodal data annotation using ELAN annotation software and analyses of both spoken and
sign language data will be presented.
The second half of the workshop is intended as a collaborative data session – participants
are invited to bring their own multimodal (corpus) material for brief data demonstrations
(regardless of the current stage of the research) followed by discussion.
8
Variation in Lovari morphology
Márton András Baló
Research Institute for Linguistics, Hungarian Academy of Sciences, Budapest
<[email protected]>
Relying on newly collected data, I will attempt to explain two particular instances of variation
in the morphology of Lovari, a dialect of Romani belonging to the Vlax dialect group, in an
analogical framework, relying only on surface forms and their relationships, using the notion
of schemata (Blevins & Blevins 2009, Goldberg 1995, Booij 2010) and finding analogical
sources that can explain the variation. This holistic approach (Kálmán 2007) is also in line
with usage-based theories (e.g. Bybee 2010) and recent experimental research in phonetics,
speech perception and speech production (Port 2007, Port 2010, Pisoni 1997).
Romani is peculiar among the languages spoken in Europe in that it is used extensively in
online communication (Leggio 2011), but it lacks a written standard. Vlax dialects are
generally said to be the best documented, but even Boretzky (2003), a comprehensive work
on Vlax Romani can only draw on a small amount of data which are either old or consist of
fairy tales or contain pre-written texts. Therefore, I will use fresh data (both questionnairebased and spontaneous) I collected during my recent fieldwork.
The two phenomena to be discussed appear as weak points in the nominal morphology of
Lovari. In order to clarify what a weak point is, 2 we will use the idea that the regularities on a
particular level of linguistic description can be expressed in terms of schemata (Booij 2010).
Although closely related, schemata represent a more general notion than constructions. While
the latter denote a pairing of form and meaning (Goldberg 1995, Jackendoff 2008), the
former, in case of morphological schemata, contains phonological, syntactic and semantic
information.
One of the two weak points in Lovari under scrutiny, where the surface forms (surface
similarities and differences) and analogical effects might play a role in producing and
maintaining variation is the masculine oblique base, where there are two possible sets of
oblique markers: -es-/-en- (singular/plural), so the oblique bases of a word like šēró ‘head’ are
šērés- and šērén-, and -os-/-on-, so the oblique bases of the word fṓro ‘town’ are fōrós- and
fōrón-. These could be illustrated by the schema for the -es-/-en- form in Figure 1. (The
schema for the -os-/-on- form would look very similar.)
Figure 1
The three kinds of linguistic information included here are the phonological form ω, the
syntactic information S (that it is an affix), and the semantic information. The symbol ↔
stands for correspondence. However, in order to be able to show the variation, I suggest
instead a circular representation, where every kind of information is connected bidirectionally
2
A weak point is also similar to the notion of an unstable point, that is, a point in a paradigm ‘where more than
one conflicting analogical requirement applies with approximately equal strength’ (Rebrus & Törkenczy 2011:
139).
9
to the other two through correspondences, as there is a relationship between the semantic and
the phonological information as well. With the help of this improvement, variation can be
represented as the combination of two schemata, as shown in Figure 2, and the notion of a
weak point in morphology can be expressed in terms of schemata: it is a schema where at
least one of the correspondences is not mutually unambiguous and where, therefore, variation
may emerge.
Figure 2
The two schemata weaken each other and this results in intra-speaker variation, where
there are two possible oblique forms for the same word, like čokanés-/čokanós for the word
čókano ‘hammer’.
The other weak point can be found in the feminine class. The singular oblique marker is
invariably -a-: šej ‘girl’ ~ šejá-, žuv ‘louse’ ~ žuvá-, but the plural oblique marker varies: it
can be -an-, for example the plural oblique base of šej ‘girl’ is šeján-, and it can be -en-, see
for example žuv ‘louse’, whose plural oblique base is žuvén-. The two competing patterns
result in variation again: both tjīrén- and tjīrán- exist as oblique bases of the word tjīrí ‘ant’,
which can also be represented by way of schemata.
References:
Blevins, James P. and Juliette Blevins (eds.). 2009. Analogy in Grammar: Form and Acquisition. Oxford:
Oxford University Press.
Booij, Geert. 2010. Construction Morphology. Oxford: Oxford University Press.
Boretzky, Norbert. 2003. Die Vlach-Dialekte des Romani. Wiesbaden: Harrassowitz.
Bybee, Joan. 2010. Language, Usage and Cognition. Cambridge: Cambridge University Press.
Goldberg, Adele E. 2006. Constructions at Work. Oxford: Oxford University Press.
Jackendoff, Ray. 2008. Construction after construction and its theoretical challenges. Language 84. 8–28.
Kálmán, László. 2007. Holisztikus szemlélet a nyelvészetben. [Online] Szabad Változók. Available:
http://www.szv.hu/cikkek/holisztikus-szemlelet-a-nyelveszetben [Accessed 10th November 2016].
Leggio, Daniele Viktor. 2011. “The Romani Internet: Language Codification and Identity Formation.”
Pesentation given at the Annual Meeting of the Gypsy Lore Society, 1st-3rd September 2011, Graz,
Austria.
Pisoni, David B. 1997. Some Thoughts on "Normalization" in Speech Perception. In K. Johnson and J. W.
Mullennix (eds.), Talker Variability in Speech Processing. San Diego: Academic Press. 9–32.
Port, Robert. 2007. How are words stored in memory? Beyond phones and phonemes. New Ideas in Psychology
25. 143–170.
— 2010. Rich memory and distributed phonology. Language Sciences 32. 43–55.
Rebrus, Péter & Miklós Törkenczy (2011). Paradigmatic variation in Hungarian. In Tibor Laczkó & Catherine F.
Ringen (eds.), Approaches to Hungarian, Volume 12: Papers from the 2009 Debrecen Conference.
Amsterdam: John Benjamins. 135–162.
10
Existential and locative sentences in Central and Southern Selkup
Josefina Budzisch
Hamburg, Germany
<josefina.budzisch@uni-hamburg>
This study is a corpus-based analysis of the Central and Southern Selkup dialects regarding
the usage and differences of existential and locative sentences. Selkup belongs to the
Samoyedic branch of the Uralic language family. Most Selkup speakers speak a variant of the
Northern Selkup dialects; Central and Southern Selkup are highly endangered, as there are
only around 5 speakers. The corpus consists of 4,362 sentences with 26,389 tokens, being an
amalgamation of published field work data of various researchers dating from 1879 to 2015,
which is now being consistently annotated.
Existential sentences state the existence of something and are usually seen as sentences
which consist of three possible elements: the theme which is mostly the subject, a verb and
possibly a locative phrase. This type of sentence is often discussed in relation to locative
sentences as they contain the same basic elements (e.g. Lyons 1969, Freeze 1992).
In Selkup in general, no special verb exists to mark existence in affirmative sentences; the
copula verb equ ‘to be’ is used in existential as well as locative sentences. It is stated for
Northern Selkup that existential and locative sentences therefore only differ in word order
which also seems to hold true for the Central and Southern variants: while in existential
sentences (example 1) the general order is (Locative) – Theme – Copula, the locative meaning
is marked by the order Theme – Locative – Copula (example 2).
(1)
Central Selkup, Vasjugan dialect
Natʼe-ɣɨt
tudo-t
ukkɨr
haj-he
e-ja-dɨt.
there-LOC.ADV
crucian-PL
one
eye-INS
be-AOR-3PL
‘There are one eyed crucians.’ (ChDN_1983_Nikita_flk_005)
(2)
Southern Selkup, Ket dialect
Paja
maːt-qɨn ä-ku-s
iː-n-d-opti.
old.woman house-LOC be-ITER-PST.3SG son-GEN-3SGPOSS-withPP
‘The old woman was in the house with her son.’ (AGS_1968_FairytaleSnake_flk_004)
In affirmative existential sentences also the occurrence of zero copula can be detected.
Negated sentences behave differently regarding the verb – čaŋkɨgu ‘to be absent’ or the
Russian loan netu have to be used in negated existential sentences, while in locative sentences
the standard negation of ‘to be’ applies.
Finally, this topic is relevant in terms of definiteness: it is stated that in existential
sentences the theme cannot be definite, while in its locative opponent it usually is (Milsark
1977). In my study I will look further into that phenomenon for Central and Southern Selkup
and analyze the definiteness effect in these variants.
List of abbreviations
AOR aorist
ITER iterative
PP
postposition
References
Freeze, Ray 1992. Existentials and other locatives. Language 68, 553–595.
11
Lyons, John 1969. A note on possessive, existential and locative sentences. Foundation of Language 3, 390–396.
Milsark, Gary 1977. Towards an explanation of certain peculiarities of the existential construction in English.
Linguistic Analysis 3, 1–29.
Wagner-Nagy, Beáta 2016. Existentials, Possessives and Definiteness in Samoyedic Languages. In: Fischer,
Susann – Tanja Kupisch – Ester Ringe (eds.): Definiteness Effects: Bilingual, Typological and Diachronic
Variation, Cambridge: Cambridge Scholars Publishing, 213–243.
Effects of plausibility on sentence comprehension
Jan Chromý
Charles University, Prague
<[email protected]>
The paper will examine the good-enough processing hypothesis (Ferreira 2003) which states
that readers do have a tendency for a shallow and superficial representations.
Two experiments will be presented. The first deals with plausibility in simple constructions
like (1a) and (1b).
(1)
a.
b.
Agresivní žralok sežral bezbranného plavčíka, který plaval u pobřeží.
(‘An aggressive shark ate a helpless lifeguard who swam along the coast.’)
Agresivní plavčík sežral bezbranného žraloka, který plaval u pobřeží.
(‘An aggressive lifeguard ate a helpless shark who swam along the coast.’)
Self-paced reading method was used and after each sentence (12 experimental items, 102
fillers), the participants (N=122) were asked a yes-no comprehension question. For
experimental items, they were asked for the sentence object (i.e. Zemřel plavčík? ‘Did the
lifeguard die?’). The linear-mixed effects model showed a significant effect of implausibility.
RTs on the object and subsequent words were significantly higher in implausible than in
plausible sentences, i.e. in sentences like (1b). Also, questions on implausible sentences
yielded less correct answers (86.86% vs. 96.48%).
The second experiment dealt with plausibility in garden-path sentences like (2a) and (2b).
(2)
a.
b.
Ostraha uklidňovala opilce a fanynku na stadionu sledovali střídající hráči.
‘The security calmed down the drunks and the fan at the stadium was watched by
the substitute players.’
Ostraha uklidňovala opilce a kameru na stadionu sledovali střídající hráči.
‘The security calmed down the drunks and the camera at the stadium was watched
by the substitute players.’
Also, two control sentences were used where the garden-path was not possible (the second
clause was changed to a SVO structure, e.g. fanynka…sledovala…hráče ‘a
fan…watched…the players’). There were 16 experimental items and 102 fillers and selfpaced reading paradigm was used. Every sentence was followed by a comprehension
question. After experimental items, the questions were aimed on the possible garden-path
structures (e.g. Uklidňovala ostraha fanynku? ‘Did the security calm down the fan?’). The
linear-mixed effects model showed the possibility of garden-path slows down reading on the
second-clause verb and the next word. Answers on the comprehension questions for plausible
condition were correct only in 66.79% if the garden-path was possible and in 83.93% if the
garden path was impossible.
12
These findings support the good-enough processing hypothesis. Readers seem to have a
tendency for an incorrect reading which corresponds with their real-world knowledge. Also,
their garden-path analyses tend to persist even after sentence reanalysis.
References:
Ferreira, Fernanda. (2003): The misinterpretation of noncanonical sentences. Cognitive Psychology 47. 164–203.
“The law shall prevent our national language from devastation”:
Language ideologies in the Czech Parliament
Ondřej Dufek
Institute of the Czech Language, Prague
<[email protected]>
Many countries have adopted language laws for various reasons. In the Czech Republic, there
were a few attempts to establish Czech as the state language, too. It never got through but the
parliamentary debates were highly interesting from the ideological, especially nationalist
point of view.
The aim of the study is to examine language ideologies employed when there is argument
whether a language law should be introduced in Czech Chamber of Deputies. In more general
words, it intends to find out how is language conceptualized when being discussed by the
politicians. In particular, it focuses on the metaphorical concepts used in the political speeches
such as language as a treasure, beautiful entity, living organism etc. Special attention is paid
to the national aspect of argumentation, both covert and overt.
The corpus collected for this purpose consists of ca. 30,000 words and contains the debates
on language law proposals held in the Czech Chamber of Deputies from 1995 until 20017.
The paper combines the approach of corpus linguistics allowing to deal with larger sets of
data and employ highly controlled procedures (inspired by the Lancaster approach (e.g. Baker
2006)) with a rather traditional CDA approach benefiting from detailed and interpretative
analyses of texts in context (using mainly the tools of the discourse historical approach (e.g.
Reisigl – Wodak 2009)). It starts with keywords analysis compared with thematic concentration
and continues with closer concordance analysis of the identified terms with regards to relevant
context. The utterances are assigned to particular speakers and their ideological positions. The
study also pays attention to discursive approach to language policy (e.g. Barakos 2016).
The analysis shows a strong accent on the national aspect of Czech language, which is in
compliance with previous surveys (Bermel 2007; another perspective on language ideologies
in Czech context cf. Nekvapil & Sherman 2013). Anyway, what is of particular interest is
how stable and widely spread they are. The analysis shows that it almost does not matter to
what political orientation particular MP adheres to, be it left or right, liberal or conservative –
while it is likely that he or she may hold the simplest purist and nationalist stance towards
language, the civic principle embodied in the constitution is almost not heard.
References:
Baker, Paul. 2006. Using Corpora in Discourse Analysis. London / New York: Continuum.
Barakos, Elisabeth. 2016. Language Policy and Critical Discourse Studies: Toward a Combined Approach. In E.
Barakos, J. W. Unger (eds.), Discursive Approaches to Language Policy. London: Palgrave Macmillan, 23–49.
Bermel, Neil. 2007. Linguistic authority, language ideology, and metaphor: the Czech orthography wars. Berlin:
Mouton de Gruyter.
Nekvapil, Jiří & Sherman, Tamah. 2013. Language ideologies and linguistic practices: The case of multinational
companies in Central Europe. In E. Barát – P. Studer – J. Nekvapil (eds.), Ideological Conceptualizations
of Language. Discourses of Linguistic Diversity. Frankfurt am Main: Peter Lang, 85–117.
13
Reisigl, Martin & Wodak, Ruth. 2009. The discourse-historical approach (DHA). In R. Wodak – M. Meyer
(eds.), Methods for critical discourse analysis. London: Sage, 87–121.
The Role of Interactional Alignment in Communication Accommodation Theory
Mirjam Elisabeth Eiswirth
University of Edinburgh, Edinburgh
<[email protected]>
Communication Accommodation Theory (Giles 1973; H. Giles, J. Coupland, & N. Coupland
1991; Giles, Taylor, & Bourhis 1973) has been dominated by quantitative analyses of
convergence on a number of linguistic variables averaged out across a conversation, based on
speakers’ relationship, personality traits, and other social factors (Soliz & Giles 2014).
Interaction happens, however, on a turn-by-turn basis and the averages used in quantitative
studies are rooted in interactional contexts that are not normally included in these analyses.
Qualitative research has shown that phonetic and interactional alignment in assessment
sequences or backchannels go hand in hand (Gorisch, Wells, & Brown, 2012; Ogden, 2006).
Quantitative CAT studies thus far have rarely considered the turn-by-turn nature of interaction
as a contributor to accommodation, despite early calls for an integration of qualitative and
quantitative methods (H. Giles, N. Coupland, & J. Coupland, 1991). One notable mixed
methods paper is Nilsson (2015), but more work remains to be done.
The present paper presents an analysis of the relationship between interactional alignment
and overall rates of convergence on vowel duration and rate of speech in a subset of the “One
speaker two dialects” Bidialectal corpus (http://thevariationist.com/). The ubiquity of those
two features allows continuous modeling and visualization of dynamic changes throughout
the interaction for both speakers. It adds to our knowledge of the relationship between
linguistic accommodation and interactional alignment through an in-depth analysis of a small
corpus of interviews involving the same set of speakers.
The subset of the corpus consists of seven conversations: three participants from Buckie
are interviewed first by a local interviewer and second by one from the South of England. The
seventh conversation is between the two interviewers. All participants are women between the
ages of 30 and 40, and each interview lasts about an hour. All interviews have been
transcribed in ELAN and phonetically aligned with FAVE (Rosenfelder et al. 2011).
Applying both quantitative and qualitative methods, I show that moments in the interaction
where participants’ vowel durations are very similar are often extended assessment sequences
in which the speakers agree with each other. On the other hand, moments in which speakers’
vowel durations diverge widely are often interactional situations in which one speaker takes
the role of the narrator and the other provides backchannels to signal involvement and
perform listenership – they are interactionally aligned, but perform different roles. In a next
step, I will move to a quantitative analysis of similarity of average vowel duration between
turns, hypothesising that turns in which the speakers are interactionally aligned will be more
similar than turns in which there is interactional misalignment.
These results have implications for the CAT paradigm: given that variable realisations
change dynamically throughout an interaction and that turn-by-turn alignment is related to
interactional activities, CAT needs to incorporate this qualitative perspective into the analysis of
accommodation.
References
Giles, Howard. 1973. Accent Mobility: A Model and some Data. Anthropological Linguistics, 15(2). 87-105.
Giles, Howard, Coupland, Justine, & Coupland, Nikolas. 1991. Contexts of Accommodation: Cambridge
University Press.
14
Giles, Howard, Coupland, Nikolas, & Coupland, Justine. 1991. Accommodation theory: Communication,
context, and consequence. In Howard Giles, Nikolas Coupland, & Justine Coupland (Eds.), Contexts of
Accommodation (pp. 1-68): Cambridge University Press.
Giles, Howard, Taylor, D. M., & Bourhis, Richard. 1973. Towards a Theory of Interpersonal Accommodation
through Language: Some Canadian Data. Language in Society, 2(2). 177-192.
Gorisch, J., Wells, B., & Brown, G. J. 2012. Pitch Contour Matching and Interactional Alignment across Turns:
An Acoustic Investigation. Language and Speech, 55(1). 57- 76.
Nilsson, Jenny. 2015. Dialect accommodation in interaction: Explaining dialect change and stability. Language
& Communication, 41. 6-16.
Ogden, Richard. 2006. Phonetics and social action in agreements and disagreements. Journal of Pragmatics,
38(10). 1752-1775.
Rosenfelder, Ingrid; Fruehwald, Joe; Evanini, Keelan and Jiahong Yuan. 2011. FAVE (Forced Alignment and
Vowel Extraction) Program Suite. http://fave.ling.upenn.edu.
Soliz, Jordan & Giles, Howard. 2014. Relational and Identity Processes in Communication: A Contextual and
Meta-Analytical Review of Communication Accommodation Theory. In E. L. Cohen (Ed.),
Communication Yearbook, Vol. 38. 107-143. New York: Routledge.
Clitic dA’s Relevance to Conversation Mechanisms
Münevver Erdem-Akşehirli
Bogazici University, Istanbul
<[email protected]>
dA is generally described as a multifunctional clitic in Turkish grammar. Göksel and Kerslake
(2005) define it as a conjunction and discourse connective with additive, adversative, topicshifting and enumerating functions. However, from a conversation analytic perspective, such
markers may be of totally different significance. This paper explores whether various dA
segments in Turkish can be identified with conversation mechanisms of speaker positioning,
turn-taking and preference structures in spontaneous talks-in-interaction.
The data used in the paper were taken from six different recordings each of which lasts
nearly an hour.
In terms of positioning, dispreferred responses are delayed by inter-turn or turn-initial
delays (Sidnell 2010). Adversative dA signals a turning of the discourse to an
opposite/different direction from what has been previously said. Thus, dA in this sense
frequently follows “turn-initial delays” as observed in the data below:
(1)
1. Müesser : temiz(.)lemeye yardıma gelen kadından memnun kaldın mı:
cleaning
help
come woman happy
were Q
“Were you happy with the cleaner?”
2. →Sevgi : hh. vallaha kaldım da evi
çok değiştirmiş
well I was adv. dA house a lot she.changed
“Well, I was but she changed the design a lot.”
3. Müesser : @@@@[@@@@]
4. Sevgi :
[@@@@]
5. Müesser : [ne de(h)mek o(h)]
what mean it
“ What do you mean?”
6. Sevgi : [eve geldim acaba] dedim benim evim mi
house I.came whether I.said my house Q
“I came home and asked myself if it was my flat.”
15
In this piece of conversation, two close friends are talking about the cleaning lady Müesser
found for Sevgi. In line 1 the question whether Sevgi was happy with the lady normally seeks
a positive feedback as a “preferred response”. Sevgi, trying to initiate a dispreferred response,
“not really” in this context, uses two “turn-initial delay” devices: “hh.” indicating breathing
and valla kaldım da… “well, I was but…”. I propose that generally surfacing in dispreferred
responses, adversative dA has a direct relevance to preference structure in conversation.
The two other conclusions of this paper are that additive dA corresponding to “too”
observed in contexts like ben de “me too” clearly indicates alignment as speaker positioning
and topic shifting dA corresponding to “as to” is used as a device of self-selecting in turntaking either by the current or next speaker.
References:
Bilmes, J. 1988. The concept of preference in conversation analysis. Language in society, 17(02), 161-181.
Göksel, A. and A. Sumru Özsoy. 2003. "dA: a focus/topic associated clitic in Turkish." Lingua 113.11 : 1431167.
Göksel, A. and Celia Kerslake. 2004.Turkish: A Comprehensive Grammar. Routledge.
Kornfilt, J. 1996. On some copular clitics in Turkish. ZAS Papers in Linguistics, 6, 96-114.
Schegloff, E. A. 1990. On the organization of sequences as a source of “coherence” in talk-in-interaction.
Conversational organization and its development, 38, 51-77.
Schegloff, E. A. 2007. Sequence organization in interaction: Volume 1: A primer in conversation analysis (Vol.
1). Cambridge University Press.
Sidnell, J. Conversation analysis: An introduction. Vol. 45. John Wiley & Sons, 2011.
Wooffitt, R. 2005. Conversation analysis and discourse analysis: A comparative and critical introduction. Sage.
An areal study of valency markers – preliminary results
Jerzy Gaszewski
University of Wrocław, Wrocław
<[email protected]>
Geographically close languages share structural features, even without close genetic relationship. The focus of my research is Central and Eastern Europe. The languages of the region are
well described individually, but comparative studies with an areal perspective are infrequent
and usually narrow in scope (e.g. Maćkiewicz 2004, Newerkla 2007). On the other hand,
studies dealing with the whole of Europe (i.e. the purported European linguistic area) are
much more common (e.g. van der Auwera 1998, Haspelmath 2001, Feuillet 1998, Heine and
Kuteva 2006). Yet, they tend to give more attention to the languages of Western Europe.
The paper presents the results of a pilot study on six languages of the region (Czech,
Polish, German, Ukrainian, Hungarian and Romanian). The investigated constructions are
valency patterns of verbs. Valency structures involving lexically-governed oblique markers
are idiom-like and we might expect them to exhibit more areal than genetic similarity, like
idiomatic constructions (Aikhenvald 2006).
The study relies on a questionnaire with simple sentences that native speaker informants
translate into their languages. From each sentence an underlying valency pattern is extracted
as shown by points a and b in the following examples.
(1)
a.
b.
Ich warte auf meinen Papa.
“Iʼm waiting for my dad”
warten > auf + Acc.
(German)
16
(2)
a.
b.
(3)
a.
b.
Čekám na mého otce.
“Iʼm waiting for my dad”
čekat > na + Acc.
(Czech)
(Az) apukámat várom.
“Iʼm waiting for my dad”
vár > Acc.
(Hungarian)
The grammatical markers in the valency patterns (cases and adpositions) flag their
respective phrases as particular participants of the situation denoted by the verb. Each valency
position of each verbal meaning is treated as a separate semantic microrole (cf. Hartmann et
al. 2014). The objective of the study is to assess the degree of formal parallelism in the
distribution of the markers between the languages (co-expression of microroles).
Thus, on the basis of (1-3) we can establish the equivalence of German auf + Acc., Czech
na + Acc. and Hungarian Acc.3 in the context of verbs meaning “to wait”. In the study I seek
to establish how regular such correspondences are, how often they repeat in the context of
other verbs, which markers are involved in the recurrent equivalences and which languages
show the greatest correlation.
Let me stress that my interest in the paper is solely the formal correspondence found in the
data i.e. which marker in one language corresponds to which marker in another. It may be the
case that these markers are good general equivalents in terms of meaning as German auf +
Acc. and Czech na + Acc., but this is not a prerequisite here. Yet, semantic equivalence
deserves an analysis of its own. Note that Hungarian Accusative is not a ready equivalent of
the other two markers outside of the context shown in (3). Another example is provided by the
markers in (4) none of which could be claimed to be another's general equivalent.
(4)
a.
danken > für + Acc.
b.
poděkovat > za + Acc.
c.
(meg)köszön > Acc.
“to thank for something”
(German)
(Czech)
(Hungarian)
The study's database comprises some 600 datapoints (valency patterns in individual
languages). The data is analysed with the use of statistical methods like clustering on the basis
of computed pair-wise similarity ratios between languages. The data is acquired from native
speaker informants.
References:
Aikhenvald, Alexandra Y. 2006. Grammars in Contact: A Cross-Linguistic Perspective. In Alexandra Y. Aikhenvald
& R. M. W. Dixon (eds.) Grammars in Contact. A Cross-Linguistic Typology. Oxford: OUP. 1-66.
Feuillet, Jack (ed.) 1998. Actance et Valence dans les Langues de l’Europe. Berlin, New York: Mouton de
Gruyter.
Hartmann, Iren, Martin Haspelmath & Michael Cysouw. 2014. Identifying semantic role clusters and alignment
types via microrole coexpression tendencies. Studies in Language 38 (3), 463-484.
Haspelmath, Martin. 2001. The European linguistic area: Standard Average European. In Martin Haspelmath,
Ekkehard König, Wulf Oesterreicher & Wolfgang Raible (eds.) Language Typology and Language
Universals. Berlin, New York: Walter de Gruyter. Vol. II, 1492-1510.
Heine, Bernd & Tania Kuteva. 2006. The Changing Languages of Europe. Oxford: OUP.
3
The Hungarian Accusative is of course not an oblique marker (see above), it is included because it surfaced as
an equivalent of oblique markers in other languages. In fact, Hungarian verb vár has an alternative pattern with
the Sublative case. Such variation of markers is present in the data and it complicates the analysis.
17
Maćkiewicz, Jolanta. 2004. Czy istnieje środkowoeuropejska wspólnota językowa? In Andrzej Kątny (ed.)
Kontakty językowe w Europie środkowej. Olecko: Wszechnica Mazurska, 7-14.
Newerkla, Stefan Michael. 2007. Kontaktareale in Mitteleuropa. In In Andrzej Kątny (ed.) Słowiańskoniesłowiańskie kontakty językowe. Slawisch-Nichtslawische Sprachkontakte. Olecko: Wszechnica
Mazurska, 29-38.
van der Auwera, Johan (ed.) 1998. Adverbial Constructions in the Languages of Europe. Berlin, New York:
Mouton de Gruyter.
Complementizer, conjunction, discourse connective: the categorial status of co in
spontaneous conversational Polish
Wojciech Guz
The John Paul II Catholic University of Lublin
<[email protected]>
The paper discusses a range of uses of the uninflected relativizer co. To complement previous
accounts, which concentrated on its straightforward relativizing function (e.g. Mykowiecka
2001, Szczegielniak 2006, Hladnik 2015), the study draws on corpus data to examine noncanonical relatives and other co clauses that depart from the traditional perception of co’s
relativizing function. The data from the Spokes corpus of conversational Polish (Pęzik 2015)
indicate that co is polyfunctional.
First, there is the regular relativizing use with the gapped structure typical of relative
clauses, as in (1):
(1) gdzie jest mój klej co kupiłam [gap]?
where is my glue CO bought:1SG
“Where is my glue that I bought?”
Second, one notes a non-canonical relativizing use whereby co conjoins a head NP with a
clause that has its own complete set of argument NPs (the gap typically found in relatives is
absent), as in (2). Such gapless relative structure is reminiscent of noun complement clauses
(cf. the idea that he doesn’t send her the alimony).
(2) o
tym terminie co nie przysyła jej tych alimentów
about this deadline CO not sends
her these alimony:PL
“about this deadline that he doesn’t send her the alimony”
Third, co may be a general conjunction, time or place conjunction, or a discourse-linking
element rather than a relativizer/complementizer, as in (3) and (4), where the semantic
contribution and categorial status of co is ambiguous.
(3)
pamiętasz
z matematyki opowiadałeś
co ci
chciała
remember:2SG in maths
told:2SG
CO
you:DAT wanted:3SG
w
pierwszym yyy w
pierwszej klasie
na koniec
postawić piątkę
in
first
in
first
class
at
end
give
five
“Remember? In your maths class, you were telling me, that/how/when she
wanted at the end of the first class to give you a five”
18
(4)
A: widziałeś ten ostatni
wypadek co
się
stał?
saw:2SG this last
accident CO
REFL happened
B: y co ciężarówka wjechała?
CO
lorry
went in
“Did you see this last accident that happened?”
“Where/when/the one in which the lorry crashed?”
Fourth, there may be no head NP for the co clause to refer back to, so that a relative clause
interpretation is untenable, as in (5).
(5)
przed Borowcem
co ten dom
jest pobudowany
before Borowiec:INSTR CO this house
is
built
“Next to the Borowiec place, where this house has been built”
The analysis offers insight into more functions of co than previously reported on the basis of
introspection. I argue that there is a cline of uses of co, which has been grammaticalized to serve
these other purposes besides the basic relativizing function. Such multifunctionality of relativizers
/complementizers has been noted in the literature cross-linguistically (Boye & Kehayov 2016).
References
Boye, Kasper & Kehayov, Petar (eds.). 2016. Complementizer semantics in European languages [Empirical
Approaches to Language Typology 57]. Berlin/Boston: Mouton de Gruyter.
Hladnik, Marko. 2015. Mind the gap. Resumption in Slavic relative clauses. Ph.D. dissertation, LOT Dissertation
Series, 390. Utrecht University.
Mykowiecka, Agnieszka. 2001. Polish relatives with the marker co. In Przepiórkowski, Adam and Piotr Bański,
(eds.), Generative linguistics in Poland. Proceedings of the GLiP-2 conference,149–57. Warszawa: Instytut
Podstaw Informatyki PAN.
Pęzik, Piotr. 2015. Spokes – a search and exploration service for conversational corpus data. In Odijk, Jan (ed.),
Selected papers from the CLARIN 2014 conference, 99–109. Linköping: Linköping University Electronic
Press. [Spokes available on-line, URL: http://spokes.clarin-pl.eu; accessed December 15, 2016]
Szczegielniak, Adam. 2006. Two types of resumptive pronouns in Polish relative clauses. In Pica, Pierre, Johan
Rooryck, Jeroen van Craenenbroeck (eds.), Linguistic Variation Yearbook 5, 165–185. Amsterdam: John
Benjamins.
Investigation of temporal and spatial reference frames in Hungarian
Veronika Harmati-Pap
Pázmány Péter Catholic University, Budapest / RIL-HAS Budapest
<[email protected]>
Languages can be distinguished by the reference-frames they prefer to express temporal and
spatial relations. In this study I investigated: a) which reference-frames are preferred in
Hungarian? b) Does the meaning of the verb have an influence on choosing a referenceframe? c) Can we find any correspondence between the usage of temporal and spatial
reference-frames? (Boroditsky 2009 vs. Bender et al. 2012)
A frame of reference (FoR) is a coordinate system required to describe the relation
between objects or events from a given perspective. (Levinson 2003). There are three main
types: absolute FoR, where the relation is defined by an external structure (e.g. cardinal
directions); intrinsic FoR, where the relation is defined by the internal structure of the objects;
relative FoR, where the relation is defined by the observer’s point of view – it has
translational and reflectional subtypes:
19
120 Hungarian adult native speakers participated in this research. I collected the data with
a questionnaire which included temporal (a) and spatial (b) tasks using the term előre
“forward” in the stimuli. There were 12 test sentences in the temporal part and 14 sentencepicture pairs in the spatial part. Half of the participants got the spatial task first, half of them
got the temporal part first to test the hypothesized primacy of space. (Boroditsky 2009).
a)
b)
The opening time of the shop has been [moved]
forward 2 hours from 8 a.m. When will it open? (write
down your answer)
[Move] the front piece two positions forward. (draw it
in its new position in the picture)
Six verbs alternated in both of the tasks: rak (‘put’), tesz (‘put’), helyez (‘place’) and hoz
(~‘bring’), tol (‘push’), visz (~’take /away/’).
The results are the following:
20
In the spatial part the translational relative FoR, in the temporal part the intrinsic FoR was
preferred: I could not find any correspondence between the two, which supports Bender’s
(2012) claim that the spatial concept do not have primacy. Contrary my expectations the
meaning of the verbs did not have a crucial effect on the usage of the reference-frames: only
the temporal “tol” (‘push’) showed a significant difference from the others.
Main References:
Bender A. & Rothe-Wulf A. & Hüther L. & Beller S. 2012. Moving forward in space and time: How strong is
the conceptual link between spatial and temporal frames of reference? Frontiers in Psychology, 3. 486.
http://journal.frontiersin.org/article/10.3389/fpsyg.2012.00486/full (2016.03.22.)
Bender, A. & Beller, S. & Bennardo, G. 2010. Temporal frames of reference: conceptual analysis and empirical
evidence from German, English, Mandarin Chinese, and Tongan. Journal of Cognition and Culture. 10.
283–307.
Boroditsky, L. 2000. Metaphoric structuring: understanding time through spatial metaphors. Cognition 75, 1–28.
Boroditsky, L. 2009. How does our language shape the way we think?. In Brockman, M (ed.): What’s Next?Dispatches on the Future Science, Vintage books original, 116-119.
Levinson, S. C. 2003. Space in Language and Cognition. Cambridge: Cambridge University Press.
Núñez, R. E. & Sweetser, E. (2006). With the future behind them: convergent evidence from Aymara language
and gesture in the crosslinguistic comparison of spatial construals of time. Cognitive Science. 30, 401–450
The semantics of -ILE in Nyamwezi
Ponsiano Sawaka Kanijo
University of Gothenburg, Sweden
<[email protected]>
This study investigates different types of meanings encoded by the perfect marker -ILE in
Nyamwezi, a Bantu language spoken in Tabora, Tanzania. The study is motivated by three
observations: (a) in previous work, it was noticed that -ILE in Nyamwezi selects few verbs,
but it was not possible to identify the patterns (Maganga & Schadeberg 1992), (b) not much
has been done on describing the semantics of Nyamwezi verbs, which appears to be the root
of variable temporal interpretations of -ILE, and (c) although the interaction of -ILE with
different types of verbs has been studied in other Bantu languages (e.g. Botne 2010, Botne
and Kershner 2000, Crane 2012, 2013), the general tendency has been to associate this marker
with inchoative verbs (or change-of-state) which give a resultative reading (i.e., an ongoing,
previously-entered state). The resultative reading is also restricted to inchoative verbs (e.g. sha ‘grind’, -ßola ‘rotten’ and -ßɩɩ́nzá ‘break’) in Nyamwezi, as illustrated in (1).
1. a) ga-ø-sh-iílé
(#gataálí ga-ø-sh-iílé ‘It (the flour) is still being ground’)
cl.6-null-grind-ILE
‘It (the flour) is ground (because it was ground in the past)’
b) ga-βol-ilé
(#gataálí ga-βol-ilé ‘They (mangoes) are still rotten’)
cl.6-be rotten-ILE
‘They (mangoes) are rotten (because they got rotten in the past)’
Based on fieldwork data (for a bigger project) gathered in August 2016 in Tabora,
Tanzania, I argue that this picture is incomplete. There are inchoative verbs in Nyamwezi
which do not neatly indicate a cause-result relationship between the prior eventuality and
current state like the one illustrated in (1) above. This includes “perception” verbs (e.g., chilwa ‘hate’, -saaya ‘be sad’, -mana ‘know’, etc.), posture verbs (e.g. -laála ‘sleep/lie’, -ɩmá
21
‘stand’, -ikala ‘sit’, etc.), and those verbs which denote mental processes and physical
conditions (e.g. -sala ‘be insane’, -dota ‘be wet’, -gina ‘be fat’, -saata ‘be sick’, etc.). These
verbs give a continuative reading, i.e., the result state in these verbs is equivalent to the
permanent state of the event having occurred (see e.g. (2)). Since in this reading the prior
eventuality and the state of the event having occurred are not different, thus the event
structures referred by these verbs can be modified by persistive form ataálɩ́ ‘still (being)’ while
those in (1) cannot.
2. a) a-saay-ílé
3SG-be sad-ILE
‘S/he is (still) sad’
(ataálí a-ø-saay-ílé ‘S/he is still sad’)
b) a-ø-laál-ílé
(ataálí a-ø-laál-ílé ´S/he is still sleeping’)
3SG-null-sleep-ILE
‘S/he has fallen asleep and s/he is still sleeping’
c) a-ø-sal-ílé
(ataálí a-ø-sal-ílé S/he is still insane’)
3SG-null-ILE
‘S/he is (still) insane’
It is less common for non-inchoative verbs to be inflected by -ile in Nyamwezi (also in many
other Bantu languages), as these verbs have no result state. However, motion verbs (e.g. -ja
‘go’, -peela ‘run’, and -toónga ‘go in front’) can also be inflected by this form to give a
continuative reading, which is translated in English using progressive sentences (3).
3. a) a-ø-peel-ilé
(a-taali a-ø-peel-ilé ‘s/he is still running’)
3SG-null-run-ILE
‘S/he is running (S/he is in the state of running)’
b) a-ø-βa-tóóng-ílé
(a-taali a-ø-βa-tóóng-ílé ‘S/he is still walking ahead of them’)
3SG-null-ahead-ILE
‘S/he is walking ahead of them’
Motion verbs by virtual of their lexical semantics encode movement, so the continuative
reading in these verbs exists as the result of the situation described by the verb itself.
I argue that -ile, as a grammatical category, indicates two different types of
relationships between the prior eventuality and the current state/action. In (1), a causal
relationship exists between the prior eventuality and the current state, whereas on a
continuative reading, the prior eventuality entails the current state in (2) or current action in
(3). The common core of these usages remains an open question.
List of abbreviations: 3SG – Third person singular; cl.6 – noun class marker 6
References
Botne, R. (2010). Perfectives, perfect and pasts, oh my!: On the semantics of -ILE in Bantu. Africana
Linguistica(16), 31-63.
Botne, R., & Kershner, T. L. (2000). Time, tense, and the perfect in Zulu. Afrika und Übersee(83), 161-180.
Crane, T. M. (2012). -ile and the pragmatic pathways of the resultative in Bantu Botatwe. Africana Linguistica,
18, 41-96.
Crane, T. M. (2013). Resultatives, progressives, statives, and relevance: The temporal pragmatics of the-ite
suffix in Totela. Lingua(133), 164-188.
Maganga, C., & Schadeberg, T. (1992). Kinyamwezi: grammar, texts, vocabulary. Köln: Rüdiger Köppe.
22
Lexical hybrids in German dialects: Investigations on agreement forms
Stephanie Leser-Cronau
Philipps-Universität, Marburg
[email protected]
The term lexical hybrid (Corbett 2006: 213ff.) designates nouns with inconsistent agreement
patterns concerning number or gender. The current presentation focuses on lexical hybrids
with an internal conflict between grammatical gender and biological sex. One famous
example is the German lexeme Mädchen ‘girl’, which is biologically female but
grammatically neutral. Agreement targets of Mädchen may either show agreement with
grammatical gender (neutral forms) or biological sex (female forms). The former is called
formal agreement, the latter semantic agreement. This is shown by the following example:
(1)
DasN
Mädchen hat seinN/ihrF Essen
nicht
TheN
Girl
has itsN/herF food
not
hat keinen
Hunger.
has no
hunger.
“The girl has not touched her food. She is not hungry.”
angerührt. EsN/SieF
touched. ItN/SheF
In the last years, several studies have shown that different factors influence the choice of
formal or semantic agreement, e.g. the distance between controller and target, the part of
speech, or semantic factors (cf. Fleischer 2012, Thurmair 2006, Robinson 2010). All these
studies have in common that they analyze written standard language. In contrast, the current
presentation uses oral data collected from dialect speakers. For this purpose, data from two
different sources are analyzed.
On the one hand, data from a corpus investigation is used. Basis for this investigation are
three corpora of the “Database of Spoken German” from the “Institute of German Language”
(IDS). The included corpora contain more than 2500 recordings of dialect speakers from the
Federal Republic of Germany, the German Democratic Republic and former eastern areas
(Silesia, Bohemia etc.). This corpus investigation shows data highly close to natural speech.
In addition, the project “Syntax of Hessian dialects” (SyHD) collected written and oral data
from so called NORMs and NORFs (non-mobile, old rural males/females). The written
surveys were answered by more than 900 participants and include several multiple-choice
tasks concerning agreement forms of lexical hybrids. In the oral interviews with 141 persons a
video and pictures were used to elicit different agreement forms. This data is extremely
comparable and can thus complement the corpus investigation.
This large amount of data from different sources is used to test as to whether German
dialects behave differently from the standard language concerning agreement patterns. The
diversity of methods help to develop a detailed picture of variation in agreement of lexical
hybrids. First results show that the above mentioned factors, such as the part of speech, play a
fundamental role, but for the first time it can be shown that the diatopic variation is an
important influencing factor on the choice of semantic or formal agreement. Maps of the
results show that there is an isoglosse between the western and the eastern parts of Germany:
semantic agreement is the dominant form in the west, formal agreement in the east.
References:
Corbett, Greville G. 2006. Agreement: Cambridge University Press.
Fleischer, Jürg. 2012. Grammatische und semantische Kongruenz in der Geschichte des Deutschen. eine
diachrone Studie zu den Kongruenzformen von ahd. wīb, nhd. Weib. Beiträge zur Geschichte der
deutschen Sprache und Literatur 134. 163–203.
23
Robinson, Orrin W. 2010. Grimm Language. Grammar, Gender and Genuineness in the Fairy Tales.
Amsterdam, Philadelphia: Benjamins. (=Linguistic Approaches to Literature).
Thurmair, Maria. 2006. Das Model und ihr Prinz. Kongruenz und Texteinbettung bei Genus-Sexus-Divergenz.
Deutsche Sprache 34: 191–220.
Vowel (Dis)harmony in Czech
Jiří Milička & Hana Kalábová
Charles University, Prague
<[email protected]>, <[email protected]>
In some languages, vowel harmony is an integral part of the language system (cf. Vargo 1980
for main issues and further literature). E.g. in Finnish, the vowel harmony is such a strong
principle that it can be utilized as a cue for the detection of word onsets in unsegmented text
(Suomi – Mcqueen – Cutler 1996).
Czech is traditionally considered to be a language system that lacks vowel harmony, but we
should not assume that there are no tendencies in the vocalic structure of Czech words at all.
The occurrence of a vowel is definitely not independent of vowels in its neighbourhood,
for example the “ú–o” pair (as in “způsobem”) is overrepresented (the frequency is about 1.6
times higher than in the random model), while the “é–é” pair (as in “švédské”) is
underrepresented.
The paper presents the data that were acquired from the SYN 2010 corpus and compares
them with the data from Hungarian (HNC), Turkish and German corpora. The data show
unexpected and inspirational patterns and the interpretation we propose is only one of many
possible interpretations.
References:
Český národní korpus – SYN2010. Ústav Českého národního korpusu FF UK, Praha 2010.
<http://www.korpus.cz>.
Hungarian National Corpus (HNC). Hungarian Academy of Sciences. Budapest 2017.
<http://corpus.nytud.hu/mnsz/index_eng.html>
Suomi, Kari, Mcqueen, James M., Cutler, Anne 1997. Vowel Harmony and Speech Segmentation in Finnish.
Journal of Memory and Language 36/3. 422–444.
Vargo, Robert M. (ed.) 1980. Issues in vowel harmony: proceedings of the CUNY linguistics conference on
Vowel Harmony. Amsterdam: JOHN BENJAMINS.
Pre- and postverbal auxiliaries in Jinghpaw
André Müller
University of Zurich, Zurich
<[email protected]>
The languages of East and Southeast Asia are known for their productive strategies of verb
serialization and marking of grammatical categories such as aspect, negation, causative,
applicatives, etc. by means of verbal juxtaposition (cf. Matisoff 1973, Lord 1993, Aikhenvald
& Dixon 2006, Anderson 2006). In most cases, these versatile verbs or auxiliaries are
restricted in their placement either before or after the main verb. Yet Jinghpaw is an exception
from this pattern as it allows a subset of auxiliaries to be flexible according to their position,
thus being able to appear either before or after the lexical verb. This study seeks to determine
which underlying factors – sociolinguistic, syntactic, pragmatic, lexical, or other – determine
the actual placement of these auxiliaries in relation to the main verb.
24
Jinghpaw is a Tibeto-Burman language spoken mainly in the Kachin State of northern
Myanmar and adjacent regions in the Yunnan province of China. Jinghpaw auxiliaries
(sometimes called “versatile verbs” or “secondary verbs”) are a subclass of verbs, derived
from lexical verbs, that assume related grammatical functions when combined with lexical
verbs (Matisoff 1974, Dai 2012, Kurabe 2016). Most of them have a fixed position, either
preceding or following the main verb, yet certain verbs, such as lù ‘get’ or chye ‘know’ can
appear on either side without changing the meaning, apparently in free variation, as
examplified below:
(1)
(2)
Shi gàw sumpyi chye dùm
3SG TOP flute
know play
‘S/he can play the flute.’
Shi gàw sumpyi dùm
3SG TOP flute
play
‘S/he can play the flute.’
ai.
DECL
chye ai.
know DECL
Pre- and postverbal auxiliaries have been reported for other languages, such as Dutch,
Russian, Lahu (Matisoff 1973), Purépecha, Gurindji (both Anderson 2006), but the
phenomenon is still rare in the languages of the world. Insight into the causal factors will
further our understanding of language variation in this aspect of grammar, as well as in this
part of the world.
Corpus data from spoken and written Jinghpaw from both China and Myanmar, in part
collected in the field, indicate an AUX V order preference for most flexible-order auxiliaries.
Preliminary analyses already indicate that both internal factors like the complexity of the
serial verb construction, and external factors such as the language variety and period of text,
do play a role for determining the order. The present study aims to explain their distribution
and function in the corpus.
References:
Aikhenvald, Aleksandra Y. & R.M.W. Dixon. 2006. Serial Verb Constructions: A Cross-linguistic Typology.
Oxford/New York: Oxford University Press.
Anderson, Gregory D.S. 2006. Auxiliary Verb Constructions. Oxford/New York: Oxford University Press.
Dai Qingxia [戴庆厦]. 2012. Jǐngpōyǔ cānkǎo yǔfǎ (Reference Grammar of the Jingpo Language)
[景颇语参考语法]. Beijing: Chinese Academy of Social Sciences Press.
Kurabe Keita. 2016. A Grammar of Jinghpaw, from Northern Burma (PhD Dissertation). Tokyo: Tokyo
University of Foreign Studies.
Lord, Carol. 1993. Historical Change in Serial Verb Constructions. Amsterdam/Philadelphia: John Benjamins
Publishing.
Matisoff, James A. 1973. The Grammar of Lahu. Berkeley: University of California Press.
Matisoff, James A. 1974. Verb Concatenation in Kachin. Linguistics of the Tibeto-Burman Area 1. 186–206.
25
English as-parenthetical construction: A construction-based perspective
Seulkee Park & Jong-Bok Kim
Kyung Hee University, Seoul
<[email protected]>, <[email protected]>
The as-parenthetical clause (APC) illustrated by the COCA (Corpus of Contemporary
American English) example (1a) displays a significant resemblance to the nonrestrictive
relative clause (NRC) given in (1b), but has unique internal and external distributional
properties distinct from the NRC:
(1)
a.
b.
They have both been married before, as we all know. (COCA:2005SPOKCNNEvent)
It’s hard to get a movie made, which we all know. (COCA: 1999 SPOK CNN King)
In this paper, we claim that the interaction between the lexicon (in particular, the lexical
properties of as) and constructional constraints are responsible for the APC’s distinctive
grammatical properties. As noted by Huddleston and Pullum (2002), Potts (2002), LeeGoldman (2012), among others, the APC and the NRC behave much alike. They both have
the function words as and which in the initial position, and contain an obligatory syntactic gap
(*as/which we all know it) which observes the island constraints as seen in *Nina quickly
bought two durians, exactly as we met a chef who did __ . Except for these commonalities, the
two constructions are quite different in many respects. The first difference comes from the
property of the gap in the internal syntax. The gap in the APC is not nominal, but verbal as
seen from the following contrast (Potts 2002):
(2)
a.
My hearing is quite acute, as I am sure you are aware (*of). (COCA: 2000
FIC Bk:Excalibur)
b.
My hearing is quite acute, which I am sure you are aware *(of).
Given that the adjectival predicate aware requires a sentential complement or the PP[of], the
contrast here shows that the gap in (2a) is a CP while the one in (2b) is an NP. The two
constructions are also quite different in terms of external syntax. The NRC is restricted to the
postmodifier position, but the APC behaves just like sentential adverbs, appearing in sentence
initial, medial, or final. In terms of meaning, the gap in the APC must be supplied by a sister
constituent and does not fall within the scope of negation (cf. Huddleston and Pullum 2002):
(3)
a.
b.
When a mayor tells you he opposes charter schools out of concern for the well
being of students currently in failing schools, you know that he is pulling your leg,
as does everyone in his audience. (COCA: 2013 MAG AmSpect)
The president’s proposal was not a new initiative, as he implied. (COCA: 2004
ACAD Hemisphere)
In (3a), the antecedent of the gap in the APC cannot be the remote embedded clause, but must
be the adjacent clause. In (3b), the negation cannot scope over the APC, conveying that he did
not imply.
In this paper, we claim that such similarities and differences have to do with tight
interactions between the lexicon and constructional constraints. In particular, we suggest that
the functor as in the APC selects as its argument an incomplete sentence with a verbal gap
and the construction evokes a conventional implicature rather than an at-issue meaning (Potts
26
2002). The present analysis supports the view of Construction Grammar in which languagespecific generalizations across constructions are captured via inheritance net- works,
reflecting commonalities or differences among constructions.
Selected References:
Huddleston, Rodney & Geoffrey K. Pullum. 2002. The cambridge grammar of the English language.
Cambridge, UK: Cambridge University Press.
Lee-Goldman, Russell. 2012. Supplemental relative clauses: Internal and external syntax. Journal of Linguistics
28(3). 573-608.
Potts, Christopher. 2002. The syntax and semantics of as-parentheticals. Natural Language and Linguistic
Theory 20(3). 623-689.
The pragmatics of ‘definition’: verbal interaction and social reality
Svitlana Pereplotchykova
Taras Shevchenko National University of Kyiv
<[email protected]>
In their interactions with others, people try to be cooperative if they want to be understood
by their interlocutor(s). Spontaneous speech presupposes a good command of the language in
use, but the overall meaning of certain words, their recognizibility (visibility), is closely
connected with the associations the relevant notions have acquired within the language codes
of particular social groups.
Since 2013 the television show Hollywood Game Night has run successfully on NBC
(USA). In each episode two teams consisting of four players compete for a money prize by
playing various games. Some of the games, where the host or the team members have to elicit
a certain word/expression from other players by giving them clues, a sort of definition,
depend crucially on particular verbalizations that are closely connected with the players’
current ‘reality’. Those words and expressions usually are embedded into some thematic field
or they are words beginning with a particular letter, or they are words consisting of four letters
etc. The explanations have to be short but are intended to provide the players with the most
salient, recognizable features of the object in question. According to the rules of the show
neither translation equivalents, nor cognate words can be used. And also there is a time limit
of 90 seconds (each game for each team). This factor is crucial for our research as the time
pressure makes players think quickly, thus providing the clues which most spontaneously
appear in the mind of the clue-giver. Those clues, hypothetically, are the most salient features
of the real-world object in question and are therefore supposed to be those most closely
associated with it subconsciously.
In so far as this game show has been adapted, and is currently shown in other countries and
in other languages, an attempt is made to compare the different associations of the words in
question, and the different strategies speakers employ to construct a pragmatically cooperative
communicative framework. The following localised versions of the show are analysed:
Celebrity Game Night (Mega Channel, Greece), Добрый вечер на Интере (Good evening,
on the Channel Inter, Ukraine), Hvězdna Párty (Stellar Party, on TV Nova, Czech Republic).
The concern of the present paper is to trace the reasons for both positive and negative
outcomes of the word-based games by means of semantic analysis of the clues provided and
the words/expressions elicited, and through conversation analysis of participants’ minidialogues, taking into consideration the communicative and social characteristics of the
interlocutors. The paper is therefore intended as a specific contribution to the development of
social and cognitive semiotics and pragmatics.
27
The results of the analysis show that the successful eliciting of a pre-determined
word/expression mainly depends on players’ assumptions of likely mutual knowledge and
reference to the most salient feature of an object in question. Sometimes a player names the
predetermined word/expression at once, sometimes two or three salient features are needed,
and there are also times when the word/expression remains unrecognized.
For example, the clues provided for ‘Grinch’ in my sample of games are ‘monster’,
‘Christmas’, “Jim Carrey played”, where ‘monster’ concerns his look, and ‘Christmas’ is the
time of the year when he exists. But the most important feature in this case is “Jim Carrey
played”, because this is what connects the general clues ‘monster’ and ‘Christmas’ to the
specific answer: the corresponding film with this actor is very famous, especially for those
aged 40-50, and the explanation works for people belonging to various social groups. On the
other hand ‘Mary Poppins’ can be efficiently represented with the clue
‘supercalifragilisticexpialidocious’, but chiefly when addressing English-speaking people of a
certain age. This nonsense word was coined by the writers of one of the best known songs of
American cinematography and was for a long time one of the most famous long words in
English. For younger speakers, and in languages other than English, the value of the clue will
vary depending on whether the story of Mary Poppins is still popular and on whether the film
is still shown, dubbed or subtitled. We may compare a very famous Soviet film made in 1983,
where the heroine is surely described, and instantly recognised by Russian-speakers born in
the 1970s, as “Samo sovershenstvo” [perfection itself, pluperfection].
In so far as this show is a team game, the players also typically try to offer clues connected
with the occupation or appearance of their co-players, thus connecting the personal ‘reality’ of
the player they are addressing with their own. For example, when the word ‘glasses’ has to be
elicited from the person wearing glasses, the most secure/salient feature in this situation is
‘you are wearing them right now’. Similarly, the color in the predetermined expression ‘Red
Square’ is presented as ‘it’s not black, ... I’m wearing it now’ (the player was in red shirt).
Sometimes such clues are not enough, however, and other salient features are referred to
verbally and non-verbally. For example, the word ‘song’ within the thematic field ‘nursery
school’:
A: I’m writing...
B: On the school-desk?
A: My job... what I am doing
B: Lyrics writer.
Sometimes the most salient feature of the object in question may be identified through its
opposite, e.g. the compass point of ‘North Pole’ is represented as ‘not South’. But sometimes
the most salient such feature, though potentially valid for more than one language, does not
secure a successful outcome. For example, in the Greek version of the show the word ‘melon’
was represented as ‘yellow’, ‘water-melon’ (a non-cognate word) and ‘opposite’. But in the
American show the attempt to give a clue ‘melon’ for the predetermined ‘water-melon’ was
not successful, as in English it is a cognate word.
Overall, the salience of different aspects of a word’s meaning is determined by its
frequency in participants’ mental lexicon and by the strength of the shared associations
engendered by its real-world familiarity, conventionality and prototypicality (Giora, 2003:10).
But these associations are not necessarily universal, and there are also important languagespecific and culture-specific constraints that have to be considered. The results of this
research comprise a preliminary overview of how speakers of different languages instinctively
process and understand 'familiar' words and phrases, and of how they interact to maximise the
efficiency of their communication.
28
References:
Giora, Rachel. 2003. On Our Mind: Salience, Context, and Figurative Language. Oxford: Oxford University
Press.
Kecskes, Istvan. 2013. Intercultural pragmatics. Oxford: Oxford University Press.
The Spoken Production of Gender-Neutral Nouns in German – some Preliminary
Results
Korbinian Slavik, Johanna Cronenberg, Erato Balafa, Christoph Draxler
Institute of Phonetics and Speech Processing
Ludwig Maximilian University, Munich, Germany
<[email protected]>, <[email protected]>,
<[email protected]>, <[email protected]>
In our current MA seminar on Speech Databases we are investigating the pronunciation of
gendered expressions. In written German, there exist a number of conventions to denote the
inclusion of all genders: a capital ‘I’ inside the word, e.g. FreundInnen (friends), either an
underscore or an asterisk preceding the ‘i’, e.g. Freund_innen or Freund*innen, or, to avoid
such novel orthographic forms, the explicit use of the female and the male nouns, e.g.
Freundinnen und Freunde. The use of such forms is a topic of public debates – see e.g.
(Theissl 2017).
‘I’, ‘_i’, and ‘*I’ do not adhere to standard orthography, and currently there does not exist
a widely accepted standard for their spoken production. We hypothesize that these forms are
produced with a stop before the /I/ vowel in the gender-neutral suffix to express the difference
to a non-gender-neutral form.
For this pilot study, we have recorded 17 speakers (10 f, 7 m) in three age groups (pupils:
16-18, students: 20-32, seniors: 80+) using the SpeechRecorder software (Draxler/Jänsch
2004). Each speaker read 25 sentences, 12 with gender-neutral forms, 13 with both male and
female forms (or vice versa).
These recordings were transcribed orthographically in a crowd-sourcing manner using a
simple web-based transcription editor. This resulted in a total of 450 orthographic
transcriptions.
We then performed a linguistic analysis to determine the use of syntactic variants in the
spoken production of gender-neutral forms. We found four variants: (ST) no change, i.e.
speakers produced the one form given in the prompt, (EE) elliptic expansion, i.e. female and
male noun without a junctor, (VE) a complete expansion, i.e. female and male form with a
junctor in between, and (MA) male form. Note that variant ii) is not grammatical according to
German standard syntax. The table shows the count for each variant type in the 207 read
sentences with only one gender-neutral form (the remaining 18 sentences contained more than
one gender-neutral form):
Type
ST
EE
VE
MA
Description
no change
elliptic expansion
complete expansion
male form
other
Count
147
33
14
9
4
29
Percent
71.0%
15.9%
6.8%
4.3%
1.9%
In the transcriptions, mispronunciations and word fragments as a result of reading errors
are marked. The percentage of such errors is almost twice as high in the items with the
gender-neutral forms: 23.5% of the prompt texts with gender-neutral forms were read with
errors, compared to only 12.8% for the texts with standard orthography. A possible
explanation is that the cognitive workload needed to parse the gender-neutral forms increases
to such a degree that speech production is affected.
A segmental duration analysis requires a phonemic segmentation and labelling of the
speech signal. Using the orthographic transcriptions, the audio files were segmented
automatically using the WebMAUS (Kisler/Schiel/Sloetjes 2012) service. As output format,
we chose the TextGrid file format, which was then converted to tab-separated text files for
import into a relational database system.
This database system can be accessed from the statistics package R for a detailed analysis.
In our analysis we focused on the absolute duration of the /I/ vowel in the target words.
The following figure shows the results for Freundinnen vs. Freund_innen and Lehrerinnen
vs. Lehrer*innen (durations are given in sample points; 44100 sample points are one second).
There are two interesting observations to be made: both a) the median and b) the variation
of the duration of /I/ are clearly much higher in the gender-neutral form than in the regular
form. a) implies that speakers mark gender-neutral forms in their speech production, and b)
indicates that speakers are not consistent in how to mark these forms. Note that the effect is
quite marked: in the boxplot for Lehrerinnen, the duration variation boxes do not overlap at
all. For Freundinnen, these boxes overlap – an explanation for this may be the nasal context
of the vowels in the suffix, which leads to marked co-articulatory effects.
Note that this segmental analysis is preliminary, and the amount of data available is
limited. For most words, we have only 8 or 9 tokens respectively. The segmentation was
performed fully automatically, and although the quality of the automatic segmentation is
good, it is does not reach that of human transcribers. We did not group our results by speaker
age or dialectal background because this would have yielded too small sample sets.
The results of the pilot study presented here show that novel orthographic forms may lead
to linguistic change in a language. Although the sample is small, effects can be observed and
measured, both in the linguistic and the phonetic domain.
References
Draxler, Chr.; Jänsch, K. (2004) SpeechRecorder – a Universal Platform Independent Multi-Channel Audio
Recording Software. Proc. of LREC, pp. 559-562, Lisbon.
Kisler, T.; Schiel, F.; Sloetjes, H. (2012) Signal Processing via Web Services: the use case WebMAUS. Proc. of
Digital Humanities 2012, pp. 30-34, Hamburg
Theissl, B. (2017) Auf der Suche nach einer Sprache, die nicht diskriminiert.
http://derstandard.at/2000051773350/Auf-der-Suche-nach-einer-Sprache-die-nicht-diskriminiert (last
checked 22.02.2017)
30
Subject-verb agreement in German speakers: Pupillometry as a measure of
sentence processing
Assunta Suess
University of Potsdam
<[email protected]>
In this study, we tested whether pupillometry is a suitable tool to study processing of subjectverb agreement in 17 German healthy adults. Previous research has shown that the pupil
dilates not only because of changes in illumination or emotions like fear, but also relates to
higher cognitive processing like working memory load (Kahneman & Beatty 1966).
Pupillometry has been used to investigate the processing of linguistic phenomena in various
domains. In sentence processing, pupil size increased according to structural complexity
(Schluroff 1982; Just & Carpenter 1993). Amongst other, increased pupil dilation was also
observed in participants' processing of ambiguous sentences (Engelhardt et al. 2010) and in
the detection of mispronunciation of words (Fritzsche & Höhle 2015).
The aim of the current study was to use pupillometry to investigate adults processing of
syntactic agreement between subject and verb in a sentence. Subject-verb agreement, like the
-s in “she works”, is found in a large number of languages. Certain syntactic information – so
called features of the subject, like person and number – are expressed on the finite verb, typically by an affix. Our hypothesis was, that the processing of subject-verb agreement would be
reflected in participants' pupil sizes, with increase in the case of an agreement violation.
Pupillometry was conducted as a single-picture eye-tracking experiment. Participants were
seated in front of a screen, presenting a picture of some action, after 1 second then heard a
related sentence that either had congruent subject-verb agreement or a violation of subjectverb-agreement with respect to number features. For each participant, 40 sentences were
presented, half of which were ungrammatical, with no fillers in between. During this
procedure, participants’ pupils were tracked via an eye-tracker. The data was then baselined,
collapsed over participants and analysed using R and linear mixed models.
The results revealed that pupil dilation correlates with the congruency of the agreement in
the sentences. Given violated agreement, after verb offset pupil dilation went on longer,
resulting in a larger pupil diameter in comparison to sentences with congruent agreement. We
propose that the higher increase of pupil dilation after agreement violations in our experiment
reflects the higher effort that is needed to process these syntactic violations. This is in line
with other studies, showing that ungrammatical sentences caused higher processing load (e.g.
Meyer et al. 2000; Vos et al. 2001), and suggests pupillometry as a promising choice for
research on syntactic processing in different populations.
References:
Engelhardt, Paul; Ferreira, Fernanda., & Patsenko, Elena G. 2010. Pupillometry reveals processing load during
spoken language comprehension. Quarterly Journal of Experimental Psychology, 63(4), 639-645.
Fritzsche, Tom & Höhle, Barbara. 2015. Phonological and lexical mismatch detection in 30-month-olds and
adults measured by pupillometry. In The Scottish Consortium for ICPhS 2015 (Ed.), Proceedings of the
18th International Congress of Phonetic Sciences. Glasgow, UK: University of Glasgow. ISBN 978-085261-941-4.
Kahneman, Daniel, & Beatty, Joy. 1966. Pupil diameter and load on memory. Science, 154(3756), 1583-1585.
Just, Marcel A. & Carpenter, Patricia A. 1993. The intensity dimension of thought: Pupillometric indices of
sentence processing. Canadian Journal of Experimental Psychology, 47(2), 310-339.
Meyer, Martin, Friederici, Angela D., & von Cramon, D. Yves. 2000. Neurocognition of auditory sentence
comprehension: event related fMRI reveals sensitivity to syntactic violations and task demands. Cognitive
Brain Research, 9(1), 19-33.
31
Schluroff, Michael. 1982. Pupil responses to grammatical complexity of sentences. Brain and language, 17(1),
133-145.
Vos, Sandra H., et al. 2001. Working memory constraints on syntactic processing: An electrophysiological
investigation. Psychophysiology, 38(01), 41-63.
Aspectual prefixes in Palaung Rumai?
Rachel Weymuth
University of Zurich, Zurich
<[email protected]>
Rumai (Austroasiatic; Myanmar, Southwest China) is a mostly isolating language, but there
are some prefixes that might have had or still have a grammatical or derivational meaning,
two of which, namely ʔɯN- and ʔə-, are the focus of this paper. The prefix ʔɯN-, whereby the
“N” is a nasal that adapts to the place of articulation of the following consonant, is often
spoken simply as /ʔə-/ which makes it sometimes difficult to distinguish between ʔɯN- and
ʔə-. The following examples show the two prefixes:
cʌ̂ m tɔ̂n
ʔɯn-hɲyê tɔ̂n
ʔɯn-hɲyê kɛ
hâw
COND reach every ?-morning every ?-morning 3PL go
tì
kɛ tʰan.hmeɛ ʔɯn-dâŋ tʰan.hmeɛ ʔɯn-jâ tʰan.hmeɛ
plant 3PL how.much ?-big
how.much ?-high
how.much
‘Every day, early in the morning, they went to look at the paddy
(and to look) how big, how high, how good it was then.’
(1) kʰʌ
tɤh lʌ̆ hɲyɐ̂w
look to paddy
ʔɯm-pen
?-good
that they had planted,
(2) ʔăn ʔə-jɔh
leɛp kʰêɛŋ ʔêɛm kʰôm
hlêɛm
3SG PFV?-fall enter LOC water reservoir deep
‘He had fallen into a deep drinking water reservoir.’
Rumai is hardly described. So far published work by foreign scholars contains few topics
like short lexicons (Milne 1931 (appendix: Palaung and Palê dialects), Newlands 1994), and a
study about tones in a Rumai dialect in Yunnan, China (Dai and Liu 1996). There are also two
dictionaries published by the Ta’ang Literature and Culture Committee (Unknown 2012; Ya
Sein 2011), the latter of them containing an English translation. The texts for the corpus of the
present author have been collected by five fieldwork trips to Myanmar since 2013. They
include elicited sentences from questionnaires about topics like negation and causative
constructions, an interview and two picture stories, all recorded, and written texts from a
journal, a book with short texts and a phrasebook. All the recorded texts and parts of the
written texts are glossed and annotated (altogether about 14,000 words).
The current findings in this data show that ʔɯN- attaches to both nominal and verbal
elements and seems to indicate a state or fact that is present at the time of the event, whereas
ʔə- only attaches to verbs and might have a perfective meaning.
The supposed meanings of the two prefixes imply an aspectual function for both. Prefixes
with TAM functions are very rare in Austroasiatic (Jenny et al. 2015) and also generally in the
languages of Southeast Asia. If there are really prefixes with such functions in Rumai, a
further question will be, how they have developed. This paper aims to explain the actual
distributions and the real meanings of the two prefixes.
During the last fieldwork trip to Myanmar at the beginning of 2017, one focus was to
answer the research questions through targeted interviews and questionnaires, among them
the one from Dahl (1985) for aspect. The results shall be presented at the conference in
Prague.
32
References
Dahl, Östen. 1985. Tense and Aspect Systems. Oxford: Basil Blackwell Ltd.
Dai, Qingxia &Liu, Yan. 1996. Analysis of the tones in the Guangka subdialect of De’ang. Mon Khmer Studies
27. 91-108.
Jenny, Mathias; Weber, Tobias & Weymuth, Rachel. 2015. The Austroasiatic Languages: A Typological
Overview. In: Jenny, Mathias & Sidwell, Paul (eds.), The Handbook of Austroasiatic Languages, 13-143.
Leiden, Boston: Brill.
Newlands, Elizabeth. 1994. Rumai wordlist for linguistic analysis, Pedaen village Namkham. Unpublished ms.
http://www.sealang.net/monkhmer/database/ Newlands, Elizabeth 1994.
U Sudhamma. 2015. Byat gawy khaeng bae (Clever stay in text). Mandalay: U Sudhamma.
Unknown. 2012. Pap-om-ngae Bran-Ta’ang (Sam Long, Rumay, Rucing). (Dictionary Burmese-Palaung (Shwe,
Rumai, Ruching)). Namhsan: Ta’ang Literature and Culture Committee.
Unknown. 2014. Pap kaseh tamay (Phrasebook Rumai-Shwe-Burmese-English). Lashio: Ta'ang Students and
Youth Union (TSYU).
Yai Sein. 2011. Ta’ang (Rumai)-English-Burmese Dictionary. Namhkam: Ta’ang Literature and Culture
Committee.
A parallel text analysis of agreement in Germanic
Magnus Breder Birkenes & Stephanie Leser-Cronau
University of Marburg, Marburg
<[email protected]>, <[email protected]>
Parallel texts, i.e. the same text in various translations, have recently been discovered as a
valuable data source in typological research (see e.g. Cysouw & Wälchli 2007, Mayer &
Cysouw 2014). In this paper we try to extend the advantages of parallel text analysis to
diachronic research. In a comparison of a passage of the New Testament (Luke 2;1–2;20), we
will explore diachronic developments and typological differences of agreement in Germanic.
A quantitative analysis provides interesting results with respect to what we call the
“pervasiveness” of agreement: How many, and which, instances of agreement can be found in
the selected passage? All relations for which we find agreement in at least one of the
morphological categories gender, number, or person were annotated for all parallel texts (i.e.,
relations between a controller and a target, in the terms of e.g. Corbett 2006).
Our first results (based on ca. 1200 agreement relations in nine bible versions) show that in
Continental West Germanic, the number of agreement relations is stable or even increases in
the history of German. Modern Dutch displays a similar profile as German, while the number
of agreement relations declines markedly in Afrikaans. In Mainland Scandinavian, it is
decreasing substantially in modern Norwegian. This development is mostly due to the loss of
agreement on the verb. In the history of German, predicative agreement decreases as well, but
this is not due to loss of morphological differences on the verb, but rather to the loss of
inflection in predicative adjectives and participles, which, interestingly, is preserved to some
degree in Mainland Scandinavian. In Afrikaans, no predicative agreement at all takes place.
On the other hand, West and North Germanic concur on one development: We find an
increasing use of attributive agreement morphology, mostly due to the grammaticalization of
the article, and in the case of German, due to more frequent use of nominal constructions.
The analysis of more than 20 versions, representing different languages, language stages
and dialects, will allow to establish a thorough typological profile of agreement and its
differing developments in Germanic.
33
References:
Corbett, Greville G. 2006: Agreement. Cambridge: Cambridge University Press.
Cysouw, Michael & Bernhard Wälchli (eds.) 2007. Focus on parallel texts. Sprachtypologie und
Universalienforschung (Thematic Issue) 60.2.
Mayer, Thomas & Michael Cysouw (2014): Creating a massively parallel Bible corpus. Proceedings of LREC
2014: 3158–3162.
The tense system of Dolgan – A corpus-based analysis
Chris Lasse Däbritz
University of Hamburg, Hamburg
<[email protected]>
Dolgan is spoken by approx. 1,000 people (VPN 2010) on the Taimyr Peninsula in Northern
Siberia, Russia. It belongs to the northeastern branch of the Turkic languages, its closest
relative being Yakut (Sakha) (Boeschoten 1998: 11f.). The dialectal division of Dolgan is still
a desideratum, however, various sources imply that there are two main dialects, Upper Dolgan
and Lower Dolgan (Stachowski 1998: 126).
The three existing grammatical descriptions (Ubryatova 1985, Li 2011, Artemyev 2013)
differ in length and scope (Ubryatova 1985 only describes the (sub-)dialect of Norilsk,
Li 2011 and Artemyev 2013 seem to describe Dolgan as a whole) and are at several points
quite contradictive. As for example, the tense system is described by Ubryatova (1985: 167ff.)
and Artemyev (2013: 201ff.) as follows:
Suffix/Form
-IAktI
-IAk
-V / -Vː ~ -r
-Ø
-TI
-BIT
Ubryatova 1985
--future tense
present tense
--past tense (just or not yet
completed action)
past tense (completed action)
present participle + pluperfect
auxiliary with -TI
past participle + --auxiliary with -TI
Artemyev 2013
future tense 2 (remote)
future tense 1
--present tense
past tense, form 1 (just or not
yet completed action)
past tense, form 2 (just
completed action)
past tense (resultative)
past tense (not completed
action)
pluperfect
Thus, both the underlying structure of the tense system of Dolgan and the particular
functions of each form seem to be quite unclear. Moreover, the given grammatical descriptions
are at least partly more tentative than well-grounded and it is not always possible to reconstruct
to which data they refer to. Most likely, it is not possible to describe the given system of
“tenses” in purely temporal categories. Having in mind that the interaction of tense and aspect
in other Turkic languages is a well-known fact (cf. Johanson 1971, Buder 1989 i.a.), it seems
quite likely that neither in Dolgan tense and aspect can be separated from each other. “Tense”
shall hereafter be understood as a grammatical category which expresses the location of an
event in time (Comrie 1995: 1244). “Aspect” in turn is to be understood as a grammatical
category which expresses the internal temporal structure of an event and its relation to a certain
period of time in view (Comrie 1995: 1244).
34
The future tense(s) in Dolgan show only the temporal notion of futureness (i.e., an event
taking place after the speech time), the future tense with the suffix -IAK being the most used
form, the future tense with the suffix -IAktI being used only sporadically in relatively clear-cut
imperative-like contexts (Artemyev 2013: 205f.).
The present tense most often points towards a currently ongoing event or towards a general
statement, irrespective of any aspectuality. However, in combination with past tenses it can
also be used conveying the imperfectivity, i.e. an aspectual notion, of an action in the past:
(1)
u͡ol-a
bar-bɨt-a
biːr uskaːn
olor-o-r
boy-3SG go-PST2-3SG
one hare
sit-EP-3SG
“The boy went and there was a hare sitting.”
(BaR_1931_OldManHaresPolarFoxes_flk, 5)
The several past tenses all point to an event that at least partially occurs in the past (i.e.
before the speech time). Their exact functions, however, are complex and cannot be merely
described within the category of tense, but rather taking into account categories like aspect
and evidentiality:
(2)
a.
b.
bɨlɨr
biːr dʼaktar-daːk
e-ti-m
long.ago one wife-ADJZ
be-PST1-1SG
‘Long ago I had a wife.’
ol
dʼaktar-bɨ-ttan ogdu͡oba kaːl-bɨt-ɨ-m
that wife-1SG-ABL widower stay-PST2-EP-1SG
‘I remained as a widower of that wife.’
(MiAI_1964_OldPeasantOldWoman_flk, 81-82)
The aim of my talk is, hence, to provide a corpus-based analysis of the tense and aspect
system of Dolgan, special interest will be drawn to the system of past tenses. The data for the
analysis will be taken from the Dolgan corpus of the research project “INEL” (Grammatical
Descriptions, Corpora and Language Technology for Indigenous Northern Eurasian
Languages) which is located at the University of Hamburg. The corpus is in progress and
consists of 3371 sentences, 6246 unique words (types) and 18644 words at all (tokens), by
now. It is made up of folklore (tales and legends) and narrative texts. The origin of the texts is
sometimes quite unclear, but the vast majority of those seem to be from Upper Dolgan areas.
References:
Artemyev, N.M. 2013. Dolganskij jazyk. Čast’ 2. Morfologija. Sankt-Peterburg: Almaz-Graf.
Boeschoten, Henrik. 1998. The Speakers of Turkic Languages. In: Johanson, Lars; Csató, Éva. (eds.). The Turkic
Languages, 1-15. London; New York: Routledge.
Buder, Anja. 1989. Aspekto-temporale Kategorien im Jakutischen. Turcologica 5. Wiesbaden: Harrassowitz.
Comrie, Bernard. 1995. Tense and Aspect. In: Jacobs, Joachim i.a. (ed.). Syntax. Bd. 2. Handbücher zur Sprachund Kommunikationswissenschaft 9, 1244-1251. Berlin: De Gruyter.
Johanson, Lars. 1971. Aspekt im Türkischen. Acta Universitatis Upsaliensis. Studia Turcica Upsaliensia 1.
Stockholm: Almqvist & Wiksell.
Li, Yong-Song. 2011. A Study of Dolgan. Altaic Language Series 5. Seoul: Seoul National University Press.
Stachowski, Marek. 1998. An example of Nganasan-Dolgan linguistic contact. In: Turkic Languages 2, 126-129.
VPN 2010 = Vserossijskij perepis’ naselenija 2010. Tom 4. Nacional’nyj sostav i vladenie jazykami. Online at:
http://www.gks.ru/free_doc/new_site/perepis2010/croc/perepis_ itogi1612.htm, 24.11.2016.
Ubryatova, E.I. 1985. Jazyk noril’skix dolgan. Novosibirsk: Nauka.
35
Grammatical status of the pointing sign “THIS” in Polish Sign Language (PJM) –
corpus-based study
Joanna Filipczak & Monika Krawczyk
University of Warsaw, Warsaw
<[email protected]>, <[email protected]>
This study focuses on specific use of the sign THIS in Polish Sign Language (polski język
migowy, PJM). PJM is a natural, visual-spatial language used by the Deaf community in
Poland. Our study is based on data extracted from the corpus of PJM that is currently being
compiled at the University of Warsaw by the Section for Sign Linguistics.
While we direct our focus at the pointing sign THIS (TO/WŁAŚNIE), we analyze its role
in the syntax as well as in the information structure of PJM. There is an ongoing debate on the
status of pointing signs in sign languages (as pronouns, symbolic gestures or gesture-like,
non-linguistic elements) (Meier & Lillo-Martin 2000, McBurney 2002, Cormier, Schembri, &
Woll 2013, Hodge & Johnston 2014, Johnston 2016). We are not aware of any other research
on this subject for PJM.
Firstly, we examine a pointing sign paying special attention to the use of THIS in
comparison to the role of pointings in PJM texts and its relation to the expression ‘exactly
this’ (właśnie to) in spoken Polish (1). Secondly, we focus on its influence on the semantics of
the sentence and its role in the information structure as a particle/responsive element (2),
element signaling contrastive focus (3) and a topic marker (4):
(1)
YES BASIC THIS
“Yes, basic [level], exactly this!”
(2)
RIGTH THIS PT:ME NOT UNDERSTAND CHOOSE WHAT?
“Rigth, exactly. I didn’t understand what I should choose.”
(3)
POSSIBLE JOIN SEVENTEEN THIS GOOD GOOD PROBLEM NOT-HAVE PROBLEM
“Yes, you can join [me] on the seventeenth. Yes, good, there’s no problem!”
(4)
WAS THIS DIVORCE LATER AFTER IMMEDIATELY TO HOLLAND ABROAD WAS
“Immediately after the divorce [he moved] abroad to Holland.”
Based on the analysis of the preliminary data we note a tendency of the changing shift in
the grammatical status of “THIS” as a pointing sign. It plays different functions on a
discourse level - more often than any other pointing sign in PJM corpus: THIS is rarely used
to point to the objects located in a signing space and does not take the accent (in form of a
strong manual articulation or nonmanual). We argue that THIS can be presented as an
example of slow grammaticalization of gesture-like element towards partly-lexicalized sign
with special discourse functions in PJM.
List of abbreviations: PT1 – pointing sign; PJM – Polish Sign Language.
References:
Cormier, K., J. Fenlon, R. Rentelis & A. Schembri. 2011. Lexical frequency in British Sign Language
conversation: A corpus-based approach. In Peter K. Austin, Oliver Bond, Lutz Marten & David Nathan
(eds.), Proceedings of the Conference on Language Documentation and Linguistic Theory 3. London:
SOAS.
36
Hodge, G. & T. Johnston. 2014. Points, Depictions, Gestures and Enactment: Partly Lexical and Non-Lexical
Signs as Core Elements of Single Clause-Like Units in Auslan (Australian Sign Language), Australian
Journal of Linguistics, 34:2, 262-291.
Johnston, T. 2013. Formational and functional characteristics of pointing signs in a corpus of Auslan (Australian
sign language): are the data sufficient to posit a grammatical class of’pronouns’ in Auslan?’. Corpus
Linguistics and Linguistic Theory 9: 109–159.
Morford, J. & J. MacFarlane. 2003. Frequency characteristics of American Sign Language. Sign Language
Studies 3, 213-225.
Sandler, W. & D. Lillo-Martin. 2006. Sign Language and Linguistic Universals. New York: Cambridge
University Press.
The diachronic transformation of {-mAK} into {-mA} in genitive-possessive
agreement contexts
Duygu Göksu
Boğaziçi University, İstanbul
<[email protected]>
In Modern Turkish (MT: early 20C-present), in embedded infinitival clauses there are two
subordinators {-mAK} and {-mA}. The verbs are argued to select for the structure formed by
{-mA} in a subject-verb agreement context and the one formed by {-mAK} in PRO-control
constructions (Kural, 1993)(1a-b).
(1)
a. Beni [PROi araba kullan-mak] isti-yor-um.
I
car
use -mAK want-IMPERF-1.SG
‘I want to use a car.’
b. Ayşe [ben-im
araba kullan-ma]-m-ı
isti-yor-Ø.
Ayse I -GEN car use -mA -POSS.1SG-ACC want-IMPERF-3.SG
‘Ayse wants me to use the car.”
Today, generally [-mAjI/-mAjA] surface in accusative and dative case marked PROcontrol subordinations while the elder speakers might use [-mAɣI/-mAɣA] in these contexts
(2a-b).
(2)
a. Beni [PROi araba kullan-mak/-ma]-(y)ı sev-iyor-um. (-maɣı > -majı)
I
car
use -mAK/-mA-ACC like-IMPERF-1.SG
‘I like using a car.’
b. Beni [PROi araba kullan-mak/-ma]-(y)a bayıl-ıyor-um. (-maɣa > -maja)
I
car
use -mAK/-mA-DAT love-IMPERF-1.SG
‘I love using a car.’
This paper argues that what on the surface looks like a change in the phonological shape of
a single morpheme (mAɣ>mAj) is actually one of the recent steps of a grammaticalization
process. The two markers have been recorded to coexist in Old Anatolian (OAT) and Ottoman
Turkish (OT). The data analyzed are taken from a dictionary of affixes that consists of
excerpts from literary texts dating back to 13C-19C (Aksoy & Dilçin, 1996). The dictionary is
organized as a list of examples comparing the MT forms of affixes with their older versions.
The data seem to support the following claims:
{-mAK} used to be compatible with both a pro (3) and a PRO (4) subject and bear POSS
in agreement (AGR) with genitive subject (if overt).
37
(3)
Eyit-miş-ler-dir
kim İsa-nın doğ-mağ-ı
Efsertis
say-EVID-3.PL-MOD that İsa-GEN be born-mAK-POSS.3.SG Efsertis
melikliğ-in-in
kırk iki yıl geç-tiğ-in-den
sonra i-di-
kingdom-CM-GEN fourty two year past-NOM-POSS.3.SG after be-PST-3.SG
‘It is told that the birth of Jesus was 42 years after the Efsertis Kingdom.’
(14C)
(4)
proi Var-ır-dı-
ay-da
bir gez PROi gör-meğ-in-e
come-AOR-PST-3.SG month-LOC one time
see-mAK-POSS.3.SG-DAT
‘(He) used to come once a month to see (her).’
(14C)
The MT forms of the examples above are as follows:
(5)
… İsa-nın
doğ-ma-sı …
Jesus-GEN
be.born-mA-POSS.3.SG
‘The birth of Jesus’
(6)
PRO gör-me-ye
gel-ir-di-
see-mA-DAT come-AOR-PST-3.SG
‘(He used to come to see (her).’
Two changes are observed here:
i. {-mAK} used to be the subordinator in pro-AGR and PRO-AGR contexts,
ii. MT’s PRO-no AGR structures used to be PRO-AGR.
Comparison of the data from OAT, OT and MT for other subjects are summarized here:
OAT and OT Forms: (14-17C)
MT Forms:
i)Verb+mAɣ+Um/+Iŋ/+I(n)
Verb+mA+m/+n/+sI(n)
(mAK+POSS.1.SG/2.SG/3.SG)
(mA+POSS.1.SG)
ii)Verb+mAɣ+In+I/+A
Verb+ma+jI/+jA
(mAK+POSS.3.SG+ACC/DAT)
(mA+ACC/DAT)
The co-existence of (7a) and (7b) in MT nowadays provides further evidence for these
hypotheses.
(7)
a. PROArb kitap oku-mak zor.
Book read-mAK difficult.
b. Kitap PROArb oku-mA-sI
zor. (preferred by elders)
Book
read-mA-POSS.3.SG difficult
‘It is difficult to read a book.’
The paper concludes that this process should not be analyzed as a purely phonological
change of {-mAK} resulting in {-mA} contrary to what has been argued in the literature
(Banguoglu, 1990), but as a diachronic change in syntactic compatibility of a morpheme.
References:
Aksoy, Ö. A. & Dilçin, D. (1996). Tarama Sözlüğü VII: Ekler. TDK yayınları: Ankara.
38
Banguoğlu, T. (1990). Türkçenin Grameri, TDK yayınları: Ankara.
Bayraktar, N. (2004). Türkçede Fiilimsiler, TDK yayınları: Ankara.
Boyd, C. (1842). The Turkish Interpreter or New Grammar of The Turkish Language, Smith Elder: London.
Ediskun, H. (1985). Türk Dilbilgisi: Ses Bilgisi, Biçim Bilgisi, Cümle bilgisi, Remzi Kitabevi.
Gülsevin, G. (1997). Eski Anadolu Türkçesinde Ekler, TDK Yayınları: Ankara.
Kanar, M. (2011). Eski Anadolu Türkçesi Sözlüğü, Say Yayınları: Ankara.
Korkmaz, Z. (1994). Türkçede Eklerin Kullanılış Şekilleri ve Ek Kalıplaşması Olayları, TDK yayınları: Ankara.
Korkmaz, Z. (2009). Türkiye Türkçesi Grameri Şekil Bilgisi, TDK yayınları: Ankara.
Kural, M. (1993). V-to(I-to)-C in Turkish. UCLA: Occasional Papers in Linguistics, 11, 17-55.
Raposo, E. (1987). Case Theory and Infl-to-Comp: The Inflected Infinitive in European Portuguese. Linguistic
Inquiry, 18(1), 85-109.
Yılmaz, Y. (2012). {-mAk} ve {-mA} eklerinin hal ekleriyle münasebeti, Turkish Studies, 7(2), 1267-1280.
Areal variation in case government of the German verb vergessen ‘to forget‘
in contrast with Czech zapomínat/zapomenout ‘to forget‘4
Agnes Kim & Sebastian Scharf
University of Vienna & Humboldt University, Berlin
<[email protected]>, <[email protected]>
For a row of syntactical constructions typical for (Eastern) Austrian German an explanation
based on the intense historical language contact with Czech has been proposed (see e. g. Zeman
2003, Newerkla 2007). Still, the contact explanations have been controversial (see e. g. Zeman
2003, Ernst 2008). Even though they are plausible within modern contact linguistic theoretical
frameworks (see Hickey 2010) they have not been investigated empirically yet. Our paper
proposes and outlines a corpus linguistic approach to close this gap5.
The construction in focus is the verb vergessen ‘to forget’, which may either govern a plain
accusative argument (vergessen + NP[acc.]) or a prepositional argument (vergessen auf +
NP[acc.]). The latter is supposed to be both diatopically and diastratically restricted to certain
German registers. It seems to occur in some dialects in the Federal Republic, but it is not
considered part of the according standard variety, in which only the construction vergessen +
NP[acc.] is accepted. In Austrian German however, the construction vergessen auf + NP[acc.] is
widely used in colloquial varieties and is even codified within the standard variety.
Contrastively, in Czech standard language the equivalent zapomínat (ipf) / zapomenout (pf.)
‘to forget’ occurs with both the plain accusative (zapomínat/zapomenout + NP[acc.]) and the
prepositive argument (zapomínat/zapomenout na + NP[acc.]). One of the main arguments in
favor the contact explanation for this phenomenon in Austrian German is the semantic
correspondence of the German preposition auf ‘on’ and the Czech na ‘on’ (see Blahak 2015:
487‒589, Bednarský 2002).
In our corpus linguistic approach, we first of all confirm the areal distribution of the
construction vergessen auf + NP[acc.] within the German language area. In a second step, we
argue that the prepositive object construction is restricted to certain semantics of the verb ‘to
forget’ in both Austrian German and Czech: In contexts
4
The presented research has been conducted in the context of the CENTRAL-Kolleg ‘Empirical perspectives on
area-typological aspects of language contact and change’. This research workshop for students lead by young
researchers in fall 2016 was a collaboration of the Humboldt University in Berlin, the Charles University and the
University of Vienna funded by the DAAD in the context of the CENTRAL network. Also, it is supported by the
SFB “German in Austria: Variation – Contact – Perception” (FWF 060).
5
The empirical approach is mainly based on written language data from the German Referenzkorpus (DeReKo,
URL: http://www1.idsmannheim.de/kl/projekte/korpora/, 18. 11. 2016) and the Český národní korpus (URL:
https://www.korpus.cz/, 18. 11. 2016).
39
In Austrian German the prepositional argument exclusively occurs with the meaning ‘to
not think about something in the context of another action’ (1a). In our Czech sample the
corresponding construction zapomínat/zapomenout na + NP[acc.] is even preferred over
zapomínat/zapomenout + NP[acc.] when it comes to expressing the same meaning (1b).
(1)
a.
Oft sind es auch die Eltern, die auf die Sicherheit vergessen. (DeReKO, NÖN)
“Often it is the parents, who forget about the safety.”
b.
[…] jeho zamrkání jí dává spočinout v přítomnosti a na moment zapomenout na
všechny touhy a vlastní zatracený neklid. (ČNK, SYN2015, Páté roční období)
“It’s blinking lets her rest in the present and forget about all desires and her own
stupid anxiety.”
On the other hand, both languages use the accusative argument construction to express the
meaning ‘to not be able to recall/remember something in a context, in which this information
is needed’ (2) and other meanings.
(2)
a.
Ich vergaß seinen Namen, was keine große Sache ist, […]. (DeReKo, Die Presse)
“I forgot his name, which is not a big deal […].”
b.
Na ni budu mít milion otázek a většinu z toho zapomenu. (ČNK, SYN2015,
Mluvíme o sexu)
“I will have a million questions for her and will [certainly] forget most of them”
The accordance in the distribution of the accusative and prepositive argument construction
leads us to the assumption that we are dealing with a phenomenon of language contact, i.e.
borrowing of pattern. Additionally, we give diachronic perspectives on the development of
the constructions in focus.
References:
Adelung, Johann Christoph. 1794–1806. Grammatisch-kritisches Wörterbuch der Hochdeutschen Mundart mit
beständiger Vergleichung der übrigen Mundarten, besonders aber der oberdeutschen. Zweyte, vermehrte
und verbesserte Ausgabe. 2. Nachdruck. Hildesheim: Olms. (URL: http://woerterbuchnetz.de/, 18. 11.
2016).
Ammon, Ulrich & Bickel, Hans & Lenz, Alexandra Nicole. 2016. Variantenwörterbuch des Deutschen. Die
Standardsprache in Österreich, der Schweiz und Deutschland sowie in Liechtenstein, Luxemburg,
Ostbelgien und Südtirol. 2., völlig neu beab. u. erw. Aufl. Berlin, New York: De Gruyter.
Bednarský, Petr. 2002. Deutsche und tschechische Präpositionen kontrastiv am Beispiel von an, auf und na.
Münster: Waxmann.
Blahak, Boris. 2015. Franz Kafkas Literatursprache. Deutsch im Kontext des Prager Multilingualismus. Köln,
Weimar, Wien: Böhlau.
Grimm, Jakob & Grimm, Wilhelm. 1854‒1961. Deutsches Wörterbuch. 16 Bde. in 32 Teilbänden. Leipzig:
Hirzel. (URL: http://woerterbuchnetz.de/, 18. 11. 2016).
Eichhoff, Jürgen. 1993. Wortatlas der deutschen Umganssprache. München: Sauer.
Ernst, Peter. 2008. Die Tschechen in Wien und ihr Einfluss auf das Wienerische. Eine kritische
Bestandsaufnahme. In Nekula, Marek & Bauer, Verena & Greule, Albrecht. (ed.), Deutsch in
multilingualen Stadtzentren Mittel- und Osteuropas. Um die Jahrhundertwende vom 19. zum 20.
Jahrhundert, 99–107. Wien: Praesens.
Hickey, Raymond (ed.). 2010. The handbook of language contact. Malden, Mass.: Wiley-Blackwell.
Newerkla, Stefan Michael. 2007. Areály jazykového kontaktu ve střední Evropě a německo-český mikroareál ve
východním Rakousku. Slovo a slovesnost 68. 271–286.
Wiemer, Björn & Wälchli, Bernhard. 2012. Contact-induced grammatical change: diverse phenomena, diverse
perspectives. In Wiemer, Björn & Wälchli, Bernhard & Hansen, Björn (ed.), Grammatical Replication and
Borrowability in Language Contact, 3–64. Berlin: De Gruyter.
Zeman, Dalibor. 2003. Das österreichische Deutsch und die österreichisch-tschechischen Sprachbeziehungen.
Ein kulturhistorischer und sprachlicher Abriß. Wien: Dissertation.
40
Relationship between executive function and sign language skills in bilingual deaf
children
Justyna Kotowicz & Magda Schromová
Department of Special Education Pedagogical University in Cracow, Poland & Section for
Sign Linguistics University of Warsaw, Poland
<[email protected]>, <[email protected]>
Although it is not fully understood, the developmental relationship between executive
function and spoken language skills are reported in studies investigating hearing children
(e.g., Müller, Jacques, Brocki, & Zelazo, 2009). Sign language and its correlation with
higher cognitive functions of deaf children still need to be explored (Hauser, Lukomski, &
Hillman, 2008; Oberg, & Lukomski, 2011). This poster presents a study of linguistic and
cognitive functioning of bilingual deaf school-aged children to determine the relationship
between executive function and sign language skills, as those factors are crucial foundations
for learning.
Two groups of bilingual deaf children participated in the study: deaf children of Deaf
parents without cochlear implant (CI) (N=20, age: M=9.11 years SD=2.00) and deaf children
of hearing parents with CI (N=19, age: M=10.8 years SD=1.9). Their bilingualism included:
Polish Sign Language (PJM) and Polish as those two languages are independent. All
children attended special schools for deaf children (none of the school has
bilingual/bicultural curriculum but all children could sign with their peers and with some
teachers in schools).
The four components of executive function were analyzed with the following assessment
tools: cognitive flexibility-Wisconsin Card Sorting Task; inhibitory control (interference
suppression)-Simon task; inhibitory control (response inhibition)-Go/ no go task; working
memory-Corsi block; and planning-Tower of London. All tasks were presented on the
computer. In the Wisconsin Card Sorting Task children were asked to sort the cards into
different and changing categories (cognitive flexibility was defined as the capacity to switch
from one to another category). The goal of the Simon task was to press the key button to
appearing butterflies (left Shift key for blue butterfly and right Shift key for red butterfly).
The interference suppression was measured as the ability to react when the stimulus occurred
on the opposite side of the response (e.g. pressing the right button when blue butterfly was on
the right side of the screen). The Go/ No go task required to react to one stimuli (boat) and to
inhibit the reaction to another one (fountain), which enabled to measure the ability to inhibit
the prepotent response. In the Corsi block task children were asked to repeat a sequence of lit
up blocks in backward order (as a measurement of working memory span). The aim of Tower
of London was to put beads on pegs on one board according to the model from the second
board (with limit of movements and rule that allowed just to take one bead at a time). This
task required to plan the sequence of movements. The pilot version of the PJM adaptation of
the BSL Receptive Skills Test (RST) was used to determine proficiency in PJM. The nonverbal intelligence measured by Raven’s Progressive Matrices was controlled.
For the deaf children of Deaf parents, the Pearson correlation coefficient showed a
correlation between sign language raw scores and summary EF coefficient (Spearman’s rank
correlation: Rho= .64, p< .01); further analysis presented the correlation between PJM RST
and two inhibitory control tasks: Simon task (interference suppression) (r= .61, p< .01) and
Go/ no-go task (response inhibition) (r= .46, p< .05). Additionally, there was a positive trend
between PJM skills and working memory (Rho= .40, p=0,08) in this group. In deaf children
of hearing parents with CI, although there were no correlation between sign language scores
41
and summary EF coefficient (Rho= .19, p> .05), there was a significant correlation between
PJM skills (PJM-RST) and working memory (r= .61, p< .01).
A large scale study of deaf children (Botting et al. 2015) showed that the language skills
play a crucial role for EF performance rather than vice versa. Based on this findings we have
determined in the regression analysis that sign language skills were a significant predictor of
EF (β= .62, t=3.32 p< .05). Sign language predicted 38% of variance of summary EF
coefficient (R2= .38).
The findings of this study suggest that sign language can play role in executive function
not only in deaf children of deaf parents who use sign language as first language, but also in
deaf children of hearing parents with CI with late and impoverished access to sign language.
Two components of higher cognitive functions: inhibitory control (interference suppression
and response inhibition) and working memory seem to be related to sign language.
The poster presents first steps in research on executive function in deaf children using
Polish Sign Language. In the future, we need to enlarge our sample in order to generalize with
more probability the research results on executive function in deaf children.
References:
Botting, N., Jones, A., Marshall, C., Denmark, T., Atkinson, J. & Morgan, G. (2016). Non-verbal executive
function is mediated by language: A study of deaf and hearing children. Child Development, 1-12.
Hauser, P.C., Lukomski, J., & Hillman, T. (2008). Development of Deaf and Hard-of-Hearing Students’
executive function, [In:] M. Marschark & P. C. Hauser (eds.), Deaf Cognition, Oxford: Oxford University
Press.
Müller, U., Jacques, S., Brocki, K., & Zelazo, P.D. (2009). The executive functions of language in preschool
children. In A. Winsler, C. Fernyhough & I. Montero (Eds.), Private speech, executive functioning, and the
development of verbal self-regulation. New York: Cambridge University Press.
Oberg, E., & Lukomski, J. (2011). Executive functioning and the impact of hearing loss: performance-based
measures and the Behavior Rating Inventory of Executive Functioning (BRIEF), Child Psychology 17:
521-545.
Grammaticalisation and Lexicalisation of the Chinese verb 断
Wenchao Li
Zhejiang University, Hangzhou-city
<[email protected]>
This study explores the evolution of the Chinese verb 断 (duàn ‘break’), exploring how it has
transformed from being a substantive transitive verb to the various forms it fulfils in
contemporary Chinese: i.e. a resultative complement (切断 qiē-duàn ‘cut-broken’), a noun
(断头台 duàn-tóu-tái ‘cut-head-platform (scaffold)’), an adjective (断枝 duàn-zhī ‘brokenbranch’) and an adverb (断定duàn-dìng ‘judge’). It aims to explicate: (i) the pathways of
transition of Chinese change-of-state verbs; (ii) whether the transformation is unidirectional or
reversible. The data for Old Chinese, Middle Chinese and Pre-Modern Chinese are
predominantly drawn from the corpus: Chinese Text Project (http://ctext.org/).
The findings reveal that duàn ‘break’ was initially transitive (three tokens of transitive use
in the Odes (1046BC–771BC), eight tokens in the Book of Change (1046BC–771BC)). Duàn
derived its intransitive function during the Spring and Autumn period (772BC–476BC). The
two functions co-existed for the rest of the pre-Qin period. Duàn’s transition into a resultative
complement began during the Eastern Han Dynasty. The noun function is developed in the
Warring State period. Duàn’s adjective and adverb functions began in Middle Chinese.
Although some transitions took place during overlapping periods, they developed via two
separate paths: the process by which duàn developed into a resultative complement, an
42
adverb, and an adjective is a case of grammaticalisation (Heine 1993; Lehmann 1995; Heine
and Kuteva 2007); the process by which duàn developed into a noun is a case of lexicalisation
(Kuryłowicz 1965; Hopper and Traugott 1993; Brinton and Traugott 2005). Essentially, it is
the intransitive function that is subject to grammaticalisation and it is the transitive function
that is subject to lexicalisation. This confirms the idea that grammaticalisation and
lexicalisation can occur to a lexeme at the same time, but the two processes do not intertwine.
References:
Brinton, Laurel and Elizabeth Traugott. 2005. Lexicalization and Language Change: Research Surveys in
Linguistics, Cambridge: Cambridge University Press.
Heine, Bernd. 1993. Auxiliaries: Cognitive Forces and Grammaticalization. Oxford: Oxford University Press.
Heine, Bernd and Tania Kuteva. 2007. The Genesis of Grammar: A Reconstruction. Oxford: Oxford University
Press.
Hopper, Paul and Traugott, Elizabeth Closs. 1993[2003]. Grammaticalization. Cambridge: Cambridge
University Press.
Kuryłowicz, Jeryz. 1965. The evolution of grammatical categories. Diogenes 51: 55-71. Reprint: Kuryłowicz, J.
1975, Esquisses linguisique II. München: W: Fink.
Lehmann, Christian. 1995[1982]. Thoughts on Grammaticalization. Munich: LINCOM EUROPA (originally
published as Thoughts on Grammaticalization: A Programmatic Sketch, Vol. 1. University of Cologne:
Arbeiten des Kölner Universalienprojekts 49).
Serial Verb Constructions in Egyptian Arabic
Adam Pospíšil
Charles University, Prague
<[email protected]>
The concept of serial verb constructions (SVC) has been used in descriptive work on various
languages of the world. Usually, it would refer to constructions in which two or more verbal
forms are concatenated without any overt linking element, like in the colloquial English “Go
eat your dinner!”. However, as Martin Haspelmath points out in Haspelmath (2015), a
unifying cross-linguistic definition of an SVC has never been provided. This holds despite the
profound treatise of different typical characteristics of SVCs provided in the volume by
Aikhenwald & Dixon (2006). To make the notion of SVC more typologically useful,
Haspelmath presents the following definition of an SVC as a comparative concept: “A serial
verb construction is a monoclausal construction consisting of multiple independent verbs with
no element linking them and with no predicate-argument relation between the verbs.”
(Haspelmath 2015).
The aim of this paper is to find out what this comparative concept can yield in the case of
Egyptian Arabic. Like in other colloquial varieties of Arabic, verbal constructions like (1) that
are prone to be classified as SVCs are abundant. By applying the above definition to Egyptian
Arabic data I tried to gain a deeper understanding of the properties of the observed
constructions, and thus first to arrive at a language-specific notion of SVC for Egyptian Arabic
and then to see how it matches Haspelmath´s comparative concept. Moreover, such an
enterprise can be of wider importance as there are good reasons to believe along with Hussein
(1990) that a very similar type of SVC can be identified across most colloquial varieties Arabic.
The data used is taken mostly from written comments and conversations on social
networks, however some spoken data which I gathered is also drawn upon.
It turns out that most of the constructions in question satisfy the proposed definition of
SVC and in turn, the definition offers convenient grounds for excluding constructions which
at first sight behave like SVCs morphosyntacticaly, but would make the notion of SVC too
43
wide and maybe less interesting comparatively. This is illustrated by example (3) in which the
verbal form yiktibu is indeed a complement of nisi ‘to forget’, thus failing to satisfy the
condition of ‘no predicate-argument relation between the verbs’. Haspelmath´s proposal also
proves useful in not insisting on the condition that all the verb forms share the same TAM values,
as shown in (2).
(1)
Ma-staḥmal-it-ši l-ʔacda
wi-qām-it
raqaṣ-it.
She didn´t bear the-sitting and-stand_up.PST-3SG.F
dance.PST-3SG.FEM
‘She couldn´t bear sitting and stood up dancing.’ (“and stood up danced”)
(2)
Fiḍil-Ø
yi-liff
fi-l-mudarrag.
remain.PST-3SG.M 3SG.M-walk_around.SUBJ in-the-lecture_hall
‘He kept walking around in the lecture hall.’ (“He remained walks”)
(3)
Nisi-Ø
yi-ktib-u.
Forget.PST-3SG.M
3SG.M-walk_around.SUBJ-3.SG.OBJ
‘He forgot to write it.’ (“He forgot writes it.”)
References:
Haspelmath, Martin. 2016. The serial verb construction: Comparative concept and cross-linguistic
generalizations. Language and Linguistics 17(3). 291–319.
Aikhenvald, A.Y, and Dixon, R.M.W. (eds.). 2006. Serial Verb Constructions: a cross-linguistic typology.
Oxford: Oxford University Press.
Hussein, Lutfi. 1990. Serial Verbs in Colloquial Arabic. In Brian D. Joseph & Zwicky, Arnold. (ed.), When
Verbs Collide: Papers from the Ohio State Mini-Conference on Serial Verbs (Columbus, Ohio, May 26-27,
1990). Working Papers in Linguistics 39. 340–354.
Light verbs & morphologic disambiguation in Czech: nominative-accusative
syncretism
Karolína Vyskočilová
Charles University, Prague
<[email protected]>
Morphological homonymy in Czech involves almost 60 % of nouns, the most spread case syncretism is nominative-accusative. This has an impact on many linguistics tasks, and one among
them is a morphological annotation of texts used for a corpus creation. In the case of Czech
corpus in the CNC project, both manual and automatic (statistical) disambiguation are used.
Nowadays, there are only a few handwritten rules which solve nominative-accusative
syncretism, and stochastic approach has to been used instead. With the Czech free word order,
it’s one of the problematic parts of morphological annotation.
Light verbs constructions are analytical predicates where the verb is semantically empty or
impoverished, and the noun (often deverbal) is showing both nominal and verbal properties
and the meaning. They work together almost as a phraseme, but with exceptions: meaning is
usually clearer, verbal part is variable, nominal part is possible to pronominalize.
Phraseme or fast structure of light verbs constructions means that they can be used for
determining if the noun is in accusative or nominative case. If there are two ambiguous
(nominative or accusative) nouns in one sentence, and we know that one of them is part of
light verb construction with the verb used in the given sentence, we can determine that the
noun will be in the case of the verb's complement (in this case accusative) with higher
probability of success than a statistical tagger.
44
This work in progress contribution will present rule-based morphological disambiguation
in the accusative/nominative case with light verb construction in Czech. The list of the light
verbs constructions used in this study is obtained from the previous studies, mainly from
Radimský.
References:
Butt, Miriam Jessica. 2010. The Light Verb Jungle: Still Hacking Away. Complex Predicates: Cross-Linguistic
Perspectives on Event Structure, 48–78.
Butt, Mitiam Jessica. 2003. The Light Verb Jungle. In Aygen, G., Bowern, C., Quinn (ed.), Harvard Working
Papers in Linguistics, 1–49.
Petkevič, Vladimír. 2014. Morfologická Homonymie v Současné Češtině. Praha: Nakladatelství Lidové noviny.
Petkevič, Vladimír. 2006. Reliable Morphological Disambiguation of Czech: Rule-Based Approach Is
Necessary. In Insight into the Slovak and Czech Corpus Linguistics, 26–44. Bratislava: Veda.
Radimský, Jan. 2010. Verbo-Nominální predikát s kategoriálním slovesem. České Budějovice: Jihočeská
univerzita v ČB.
Spoustová, Drahomíra, Jan Hajič, Jan Votrubec, Pavel Krbec, & Pavel Květoň. 2007. The Best of Two Worlds:
Cooperation of Statistical and Rule-Based Taggers for Czech. In Proceedings of the Workshop on BaltoSlavonic Natural Language Processing, 67–74. Praha.
Pronoun vs. zero-anaphora: the expression of direct object referents in Selkup
Hannah Wegener
University of Hamburg, Hamburg
<[email protected]>
Selkup is a Uralic language belonging to the Samoyedic subgroup and is spoken in NorthWest Siberia. Selkup can be divided into three major dialectal groups, i.e. Northern, Central,
and Southern Selkup. The Northern Selkup language variety forms the largest group of
Selkup speakers. Central and Southern Selkup are on the verge of extinction. Based on the
Russian census of 2010 the overall number of Selkup speakers amounts to 1.023.
Investigating the form of the direct object it can be noted that in Southern and Central
Selkup there are two ways of referring to the object referents other than by a noun, namely
pronouns and zero anaphora.
(1)
kū
katbo-nǯ-al
when get-FUT-2SG.O
adv v
mašem
aža
qonǯyr-ǯenǯ-al
1SG.ACC
pronp
NEG
see-FUT-2SG.O
v
ptcl
If you get it somewhere, you will not see me anymore.
Southern Selkup (Maksimova et al. 2010)
(2)
nʼadek
tab-ɨ-p
kwat-pa-t
girl
n
he-EP-ACC
pronp
kill-PST.REP-3SG.O
v
The girl killed her
Central Selkup (ChDN_1983_GirlAndIce_flk.030 6)
6
Examples (2) and (3) taken from the corpus of Southern and Central Selkup currently compiled within the
project Syntactic description of southern and central selkup dialects: a corpus-based analysis (WA 3153/3-1),
funded by the DFG.
45
(3)
tebɨ-lɣup
na
pöː-llaga-dɨn
šeːge-nd
čoče-mba-d
man-ADJZ-human.being
n
this
dem
stone-sing-3PL
n
thread-ILL
n
rank-PST.REP-3SG.O girl-LOC.AN
v
n
holʼ-ɣend
ödɨ-mba-d
neck-DAT/ALL.3SG
n
put.on-PST.REP-3SG.O
v
nade-nnan
The man strung the stones on a thread and put it around the girl’s neck.
Central Selkup (MNS_1984_DaughterOfEarth_flk.017)
Example (1) exibits pronominal reference and zero anaphora within the same sentence for
Southern Selkup. Pronominal reference is indicated by the 1sg.acc refering back to the
speaker. Zero-anaphora can be detected within the 2SG suffix of the verb katbo-nǯ-al (get).
Sentences (2) and (3) exemplify a similar situation for Central Selkup. Back reference takes
place via personal pronoun (2) or zero-anaphora (3), again visible in the verbform ödɨ-mba-d
(to put on). With two possibilities at hand the question arises which properties form the basis
for object realization with recurring referents in Selkup.
Gundel et al. (1993) observe a correlation between cognitive status of the referent and
phonological form of the referring expression. The forms referring to referents highly
restricted in cognitive status, e.g. activated, in focus, tend to be of little phonetic form, i.e.
pronouns, clitics and zero forms (Gundel et.al 1993: 285). For personal pronouns Rohde –
Kehler (2014) claim topicality to be an influencing factor for pronoun production. This,
however, accounts for personal pronouns referring to subject referents (Rohde – Kehler
2014:20).
In line with previous research on referring expressions it is argued that for Southern and
Central Selkup direct object realization referential as well as information structural properties
of the referent are crucial. The corpus examined contains data on Southern and Central Selkup
narrative data. The total size amounts to 5.376 sentences. The corpus was analyzed for
transitive sentences containing direct objects which were referred to in immediately following
sentences. Instances in which a personal pronoun is chosen instead of zero anaphora show a
greater need for distinguishing the object referent from any other referent. Thus referents
expressed by zero-anaphora are highly referential and do not need to be further distinguished.
References:
Bajdak, A. – Maksimova, N. – Fedotova, N. 2010. Sel’kupskie teksty. In Filchenko (ed.): Annotirovannye
fol’klornye teksty obsko-enisejskogo jazykovogo areala [Annotated folklore prose texts of Ob-Yenisei
language area]. Tomsk: Veter. 133-184.
Gundel, J. – Hedberg, N. – Zacharski, R. 1993. Cognitive Status and the Form of Referring Expressions in
Discourse. Language 69 (2). 274-307.
Rohde, H. – Kehler, A. 2014. Grammatical and information-structural influences on pronoun production.
Language, Cognition, and Neuroscience, Special Issue on Production of Referring Expressions: Models
and Empirical Data. 912-927.
46
47
48