ChapterPDF Available

Salience and definiteness in Turkish children's narratives

Authors:
- 192 -
Salience and definiteness in Turkish children’s narratives
Duygu Özge and Özlem Yeter – Middle East Technical University
duyguo@metu.edu.tr and ozlemyeter1@gmail.com
1 Introduction
Von Heusinger has posed definite descriptions as dynamic expressions that (i) pick the most
salient entity as their referents and (ii) change the salience hierarchies of the introduced
referents and their supersets (von Heusinger 1997; 2003; 2007). With this proposal, he
departs from the assumption of the uniqueness attributed to definite expressions in Russel’s
(1905) Theory of Descriptions, and from the assumption that an expression can either have
context dependence or salience changing potential, but not both (e.g., Kamp 1981; Heim
1982; Groenendijk & Stokhof 1991). In this altered dynamic semantics approach, definite
descriptions have double dynamics (von Heusinger 2003). First, they do not gain their
uniqueness condition via the lexical semantics of the definite article, but dynamically, via the
function it has in a discourse, so they are context-dependent. Second, they do have the power
to alter the context and update the salience structure of the discourse.
Among others, two of the discourse functions of definite nouns that will be relevant
for our purposes are situational salience, where the referent is the most salient entity that is
accessible to both interlocuters, as in the glacier in (1a), and the anaphoric relation where the
referent is introduced with an indefinite article (a glacier) in its first mention and is referred
with a definite noun in its re-mention (the glacier), as in (1b) (von Heusinger 1997). These
examples clearly illustrate that the definite expressions tend to refer to the most salient entity
and that they gain their meaning by their function in the discourse.
(1)
a. Oh my God! The glacier is very steep and almost impossible to climb.
b. There is a glacier in our town. It is very steep and almost impossible to climb. Even the
most advanced climbers need continuous belaying and extreme care and it has caused many
injuries in the past. The ice waterfall route in the Skyler area is much better though. Despite
the dangers, some climbers insist on going for the glacier.
c. The glacier is very steep and almost impossible to climb. Even the most advanced
climbers need continuous belaying and extreme care and it has caused many injuries in the
past. Despite this some climbers insist on going for the route.
Also, the example in (1b) is a good case where another definite noun (the ice waterfall route)
shifts the salience from one entity (the glacier) to another (the waterfall). Furthermore, the re-
mentioning of the first definite entity (the glacier) after this other entity (the waterfall) again
shifts the salience back to the glacier as the most salient noun in this discourse, which
demonstrates the discourse changing potential of the definite referents. Finally, in (1c), the
glacier is re-mentioned with another relevant noun that includes the glacier as its member the
route, an example of salience spreading where the definite expression changes the salience of
their supersets as well (von Heusinger 2003).
In this study, we analyze the narratives of Turkish-speaking children focusing on their
referential choices for discourse referents and their function, with a special focus on the
functions of definite descriptions in children’s referential chains.
- 193 -
2 Method
Participants
We tested 21 primary school monolingual Turkish children (12 Females) (Mage = 9.56, SD =
0.61). Tested as a control group for another study, these children come from a relatively
lower SES.
Materials
We used the Multilingual Assessment Instrument for Narratives (Gagarina et al. 2012, 2019).
It consists of 4 stories, each two having the same number of story characters: Cat Story/Dog
Story and Baby Birds/Baby Goats. We analysed the first stories told by children (i.e., cat
story and dog story). These stories contain 3 characters (i.e., cat, butterfly and the child in the
“Cat Story” and dog, mouse and child in the “Dog Story”). Reference to the setting, an
initiating event, goal, attempt and outcome in an organized manner constitute the story
structure.
Procedure
Story pictures were printed out in vivid colours, cut and stuck next to each other as suggested
in the manual. Each story was printed 3 times and was put in separate envelopes. Later, the
child was asked to choose among the envelopes, although they all contained the same story.
This was made in order to make the child believe that the experimenter does not know what
story was going to be told. This was crucial to prevent shared knowledge effect between the
child and the experimenter (Gagarina et al. 2012). The narration was audio-recorded and
transcribed by 2 Turkish native speakers and was checked by the second author.
Coding
Only the animate story characters were coded. Ambiguous and incomprehensible sentences
were excluded (N = 3). 372 utterances were coded in total (ungrammatical sentences causing
ambiguity and utterances containing no animate reference have not been coded, N = 18). 9
children told the cat story while 12 told the dog story. Introduction refers to the first mention
of the story characters. Maintenance is when the story character that is being mentioned is
referred to also in the previous utterance. We coded the story characters that re-appeared after
being interrupted by the mention of another story character as re-mention.
3 Results and Discussion
While the number of indefinite expressions was greater than the number of definite entities
when introducing the referent, the number of definite expressions was greater when
maintaining and re-mentioning the already introduced entity [X2(1, N=372)=208.23,
p<.00001] (Table 1).
Table 1: Function of definite and indefinite expressions in children's narratives
# of total occurrences
# of definite occurrences
# of indefinite occurrences
Introduction
61
21
39 (4 generic)
Re-mention
104
102
1
Maintenance
207
205
2
Total
372
328
42
- 194 -
Despite this pattern, the number of using bare definite nouns while introducing a subject
referent was still at a remarkable rate (%43). This is an example of using the definites for
situational salience (von Heusinger 2003). This is in line with previous studies in Turkish
children’s narratives (Küntay 2002; Aksu-Koç & Nicolopoulou 2015). However, different
from previous studies, our participants did not have a shared visual narrative context with
their interlocutors but they were still treating the subject entities as the situationally the
obvious and shared character. This may be due to limited Theory of Mind skills, so further
studies should address this possible correlation. We also see that the number of subject
mentions was also greater than the number of object mentions [X2(1, N=285)=10.58, p=.005]
(Table 2). This also concurs with previous studies underlying the subject-bias in referential
chains (for a summary, Schumacher & von Heusinger 2019).
Table 2: The number of subject and object roles depending on the function of the entity
Character function
# of total subjects
# of total objects
Introduction (N = 61)
43
14
Re-mention (N = 104)
86
4
Maintenance (N = 207)
128
21
Total (N = 372)
257
39
We then examined the percentage of definite nouns and their morphosyntactic realizations in
different word orders within subject and object referents separately. The pattern for the
definiteness (i.e., greater indefinites for introduction, greater definites for maintenance and
re-mention) and the subject-bias persisted in this analysis. Analyzing the word order for
subject referents, we found that the number of SV utterances was greater than that of SOV
when introducing a referent while we observed just the opposite pattern for character re-
mention [X2(1, N=236)=111.64, p<.00001]. Furthermore, OV and V orders were greater for
maintaining the reference while there was no difference between SOV and SV orders. Hence,
the subject is more likely to be introduced with the verb without any other referents to
establish the salience of this initial entity and once this is done, an additional referent (i.e.,
object) is introduced in an SOV order. For object referents, the word order did not differ by
the function of the referent [X2(1, N=42)=5.99, p=.42]. This may be due to the smaller
number of object mentions in total. We then looked at the morphosyntactic realization of the
referents in subject mentions as the majority of the cases. A greater number of indefinite
referents were introduced by an indefinite article, as in (2a) (Aksu-Koç & Nicolopoulou
2015), whereas all of the definite subject referents were introduced by a bare noun in SOV
and SV order while they were all introduced by a null pronoun in V order (2b). All referents
were definite during the re-mention and maintenance, and they were mentioned by a null
pronoun in OV and V and by a bare noun, demonstrative noun and pronoun in SOV and SV.
Finally, there were also some examples of definite expressions referring the supersets of the
main entity, as in (2b), a la von Heusinger’s (2003) observation of salience spreading.
(2) a. Bir tane kedi varmış. (There was a cat)
O sırada bir tane çocuk gelmiş. (Meanwhile a boy came)
b. Şimdi bir tane fare varmış. Böyle ağacın deliğinden girmiş. (SV order)
(Now there is a mouse. (The mouse) passed through the hole in the tree.)
c. Bir tane köpek varmış burada. Fareyi kovalıyormuş. Yakalamış. Fareyi kovalarken
içine girmiş. Kaçmaya çalışıyormuş. Ondan sonra hayvan kafasını vurmuş.
- 195 -
(There is a dog here. (It: dog) is chasing the mouse. (It: dog) caught (it: mouse). As
(it: dog) was chasing the mouse, (it: dog) got into it. (It: mouse) was trying to run
away. Then the animal (i.e., dog) hit his head.)
In conclusion, the analysis of Turkish 9-year-olds’ referential chains in their narratives
indicates that the subject is the most salient entity and although the characters are generally
introduced by an indefinite noun, they are oddly and frequently introduced by a bare definite
noun, which is supposed to mark the situationally salient entity. Children are adultlike in
using more definite expressions for character maintenance and re-mention. The
morphological realization of these nouns (i.e., bare nouns, (null)pronouns, demonstratives)
interacted with the word order/information structure and the discourse function of these
entities. Although infrequently observed, children’s use of definite expressions for salience
spreading is also adultlike at age 9. Therefore, for children just like adults, definite
expressions are dynamic expressions marking the salience and updating the salience
hierarchies of the referents and their supersets (von Heusinger 1997). The only developing
feature at this age seems to be situational salience that may be related to other socio-cognitive
abilities, which we intend to investigate in future studies.
References
Aksu-Koç, Ayhan & Angeliki Nicolopoulou. 2015. Character reference in young children’s
narratives: A crosslinguistic comparison of English, Greek, and Turkish. Lingua 155. 62–
84.
Gagarina, Natalia, Daleen Klop, Sari Kunnari, Koula Tantele, Taina Välimaa, Ingrida
Balčiūnienė, Ute Bohnacker & Joel Walters. 2012. MAIN: Multilingual Assessment
Instrument for Narratives. ZAS Papers in Linguistics 56. 1–140.
Gagarina, Natalia, Daleen Klop, Sari Kunnari, Koula Tantele, Taina Välimaa, Ute Bohnacker
& Joel Walters. 2019. MAIN: Multilingual Assessment Instrument for Narratives
Revised. ZAS Papers in Linguistics 63. 1–36.
Groenendijk, Jeroen & Martin Stokhof. 1991. Dynamic predicate logic. Linguistics and
Philosophy 14. 39–100.
Heim, Irene. 1982. The semantics of definite and indefinite noun phrases. PhD dissertation,
University of Massachusetts Amherst. Published in 1988 by Garland.
von Heusinger, Klaus. 1997. Salienz und Referenz. Der Epsilonoperator in der Semantik der
Nominalphrase und anaphorischer Pronomen. Berlin: Akademie Verlag.
von Heusinger, Klaus. 2003. The double dynamics of definite descriptions. In Jaroslav
Peregrin (ed.), Meaning: The dynamic turn, 149–168. Leiden: Brill.
von Heusinger, Klaus. 2007. Anaphors in Text: Cognitive, formal and applied approaches to
anaphoric reference. In Monika Schwarz-Friesel, Manfred Consten & Mareile Knees
(eds.), Anaphors in Text: Cognitive, Formal and Applied Approaches to Anaphoric
Reference, 123–144. Amsterdam: John Benjamins.
Kamp, Hans. 1981. A theory of truth and semantic representation. In Jeroen Groenendijk,
Theo Janssen & Martin Stokhof (eds.), Formal methods in the study of language, 277–
322. Amsterdam: Mathematical Centre.
Küntay, Aylin C. 2002. Development of the expression of indefiniteness: Presenting new
referents in Turkish picture-series stories. Discourse Processes 33(1). 77–101.
Schumacher, Petra B. & Klaus von Heusinger. 2019. Introduction to prominence in
discourse. Journal of Pragmatics 154. 18–21.
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
The Multilingual Assessment Instrument for Narratives (MAIN) is part of LITMUS (Language Impairment Testing in Multilingual Settings). LITMUS is a battery of tests that have been developed in connection with the COST Action IS0804 Language Impairment in a Multilingual Society: Linguistic Patterns and the Road to Assessment (2009-2013).
Article
Full-text available
This study investigates how native Turkish-speaking participants of different ages produce new referents in narrative discourse about a 6-frame picture series. Turkish does not obligatorily encode the distinction between indefinite and definite reference with a formal article system. The expression of indefiniteness is instead achieved through a conglomeration of devices, including an optional indefinite numeral, case-ending variation, and word order. The main motivation of this study was to specify the means and the extent of indicating the nondefinite status of newly introduced story participants by Turkish narrators of different ages. The results indicate that Turkish children, similar to young speakers of other languages, do not exhibit a tendency to mark the indefinite status of referents until around 7 years of age. The centrality and animacy of the story characters constrain the introductory referential strategies of speakers. The implications of these findings are discussed in a cross-linguistic developmental framework.
Article
The Multilingual Assessment Instrument for Narratives (MAIN) was designed in order to assess narrative skills in children who acquire one or more languages from birth or from early age. MAIN is suitable for children from 3 to 10 years and evaluates both comprehension and production of narratives. Its design allows for the assessment of several languages in the same child, as well as for different elicitation modes: Model Story, Retelling, and Telling. MAIN contains four parallel stories, each with a carefully designed six-picture sequence. The stories are controlled for cognitive and linguistic complexity, parallelism in macrostructure and microstructure, as well as for cultural appropriateness and robustness. The instrument has been developed on the basis of extensive piloting with more than 550 monolingual and bilingual children aged 3 to 10, for 15 different languages and language combinations. Even though MAIN has not been norm-referenced yet, its standardized procedures can be used for evaluation, intervention and research purposes. MAIN is currently available in the following languages: English, Afrikaans, Albanian, Basque, Bulgarian, Croatian, Cypriot Greek, Danish, Dutch, Estonian, Finnish, French, German, Greek, Hebrew, Icelandic, Italian, Lithuanian, Norwegian, Polish, Russian, Spanish, Standard Arabic, Swedish, Turkish, Vietnamese, and Welsh.
Article
This Special Issue on prominence in discourse originated from collaborative research at the University of Cologne within the research center Prominence in Language that considers prominence an organizational principle of language that operates at all levels of linguistic representation. Here, we focus on reflections of prominence at the level of discourse structure. The contributions that make up this Special Issue address prominence-related research questions for different discourse notions, including referents, time points, propositions, coherence relations and inferences.
Anaphors in Text: Cognitive, formal and applied approaches to anaphoric reference
  • Klaus Von Heusinger
von Heusinger, Klaus. 2007. Anaphors in Text: Cognitive, formal and applied approaches to anaphoric reference. In Monika Schwarz-Friesel, Manfred Consten & Mareile Knees (eds.), Anaphors in Text: Cognitive, Formal and Applied Approaches to Anaphoric Reference, 123-144. Amsterdam: John Benjamins.