Children’s language development and quality are matters of public concern that linguists, educators, and psychologists strive to address. There is increasing discussion of how the language of adults—and especially children—has noticeably deteriorated over the past decade, is acquired more slowly, and involves greater difficulty in developing linguistic and communication skills. The abundance and diversity of factors that may influence language development encourage the search for new research approaches and the integration of linguistic, sociolinguistic, and psycholinguistic research methods.
Interest in the language development of Lithuanian children emerged before World War II, when researchers began observing and recording children’s spontaneous speech (Kalnius 1993), but this activity soon came to an end. Later, with the establishment and development of the Lithuanian speech therapy school, considerable attention was devoted to pedagogical aspects of children’s language development, the diagnosis and remediation of language disorders, and children’s linguistic readiness for school (Glebuvienė 1977; Federavičienė 1978; Giedrienė 1986; Garšvienė, Ivoškuvienė 1993; Ivoškuvienė et al. 1997; Mazolevskienė 2000, 2003; Grigaitė 2004, etc.). In recent decades, naturalistic longitudinal observation has often been used to study Lithuanian children’s language (Savickienė 1999, 2003; Wójcik 2000; Balčiūnienė 2009; Kamandulytė 2010). Experimental studies are increasingly common (most language tests fall into this category), and the age range of the children studied has been broadened (Dabašinskienė 2010; Ruzaitė, Dabašinskienė 2010; Kamandulytė-Merfeldienė et al. 2010); however, the semi-experimental method—also called special-
212
KALBOS KULTŪRA | 86
Page 2
...of the task—the method is only just beginning to be applied (Savickienė et al. 2009; Balčiūnienė 2013) to the Lithuanian language. One of the classic semi-experimental methods is narrative analysis. This method was developed in the United States in the 1970s and 1980s and soon—because it is particularly flexible and easily adapted to different participant groups and research objectives—became popular worldwide. At present, narrative methodology is used in comparative language studies (Berman, Slobin 1994), in assessing children’s linguistic readiness for school (Hudson, Shapiro 1991), in checking or refining a diagnosis of a particular mental illness (Garrad 2005), and in assessing language interference in bilingualism (multilingualism) (Gülzov, Gagarina 2007; Gagarina et al. 2012). In Lithuania, as already mentioned, narrative methodology is only beginning to be applied. Diagnostic and corrective tools for language development are being developed and tested (Balčiūnienė, Povilaitienė 2012; Gagarina et al. 2012), attempts are being made to establish the natural development of narrative skills (Balčiūnienė 2013), and studies of narrative skills in bilingual Lithuanian children have begun (Balčiūnienė et al. 2011; Balčiūnienė, Kalninytė 2013; Gagarina et al. 2013). This article reviews the narrative characteristics of preschool-age children, identified using the narrative analysis (RiPA) methodology (Balčiūnienė 2013). The study group was selected because the preschool age (i.e., the final year before school) is particularly important with respect to linguistic maturity and linguistic readiness for school. It is hoped that the results will broaden knowledge, at least to some extent, about the language of preschool-age children and will be useful not only to researchers but also to practitioners whose work is directly related to the diagnosis and remediation of language development and to preparing children linguistically for school.
2. RESEARCH MATERIAL AND METHODOLOGY
2.1. Collection of research material
In studying narrative skills in Lithuanian preschool-age children, the researchers chose to have the children create a story about an imaginary event based on a sequence of story pictures. This choice was motivated by arguments presented in international scholarly and methodological literature: (a) story generation reveals the quality of a child’s narrative skills better than retelling does
I. BALČIŪNIENĖ, A. KALNINYTĖ. Priešmokyklinio amžiaus vaikų rišliojo pasakojimo ypatybės
213
Page 3
(Hedberg, Westby 1993, cited in Hughes et al. 1997), and (b) a sequence of pictures, compared with individual pictures and films, best helps children understand and create a story’s plot and structure (Hughes et al. 1997). Taking into account the participants’ age and linguistic and cultural experience, the researchers selected picture sequences depicting characters familiar to children and plots close to Lithuanian culture. This took into account that the characteristics of children’s narratives are directly related to their cultural experience and environment (Gutierrez, Quinn 1993); therefore, when selecting visual materials, it is important to assess whether they reflect the culture of the child’s environment.
The sample (24 typically developing monolingual Lithuanian children attending kindergarten and educated according to the Pre-primary Education Curriculum, see www.smm.lt) was tested twice—at the beginning and at the end of the academic year. During the first testing session, the children created a story based on the story-picture sequence The Fox Story (see Fig. 1).
1 PAV. Lapės istorijos siužetinių paveikslėlių seka (Gülzov, Gagarina 2007)
During the second testing session, the children created a story based on the story-picture sequence The Cat Story (see Fig. 2).
214
KALBOS KULTŪRA | 86
Page 4
2 PAV. Katės istorijos siužetinių paveikslėlių seka (Hickmann 2003)
Both picture sequences consist of a set of six black-and-white pictures. During the study, all the pictures are shown (laid out on the table in front of the child in a single horizontal row from left to right), and the child is told that they depict a fairy tale. The child is then asked to look carefully at the pictures (with no time limit). After the child has looked at the pictures, the following is said: And now I would like you to tell me this fairy tale. The story begins here… (pointing to the first picture). While the participant is telling the story, the researcher does not interrupt, help, or prompt them, but only encourages them to continue speaking by responding with words such as yes, mm-hmm, aha, good and so on, along with appropriate facial expressions and intonation. If the child is reluctant to speak or asks for advice, the researcher does not answer directly, but asks: What do you think? What does it seem like to you? What else could you say? This is intended to minimize influence on the plot and structure of the story, and on the choice of particular lexemes, grammatical categories, and syntactic constructions. Each child is tested individually, and their story is recorded
I. BALČIŪNIENĖ, A. KALNINYTĖ. Priešmokyklinio amžiaus vaikų rišliojo pasakojimo ypatybės
215
Page 5
recorded on a dictaphone. All recorded narratives (48 in total) were transcribed and coded for automatic linguistic analysis using the CHILDES (English Child Language Data Exchange System, MacWhinney 2010) program. These transcripts were used to compile a morphologically annotated corpus of coherent narratives (17,861 words in total) and prepare for the analysis of coherent narratives.
2.2. Analysis of Coherent Narratives
Among the various methods for analyzing coherent narratives, two of the most popular and versatile can be identified; they allow the quality of a coherent narrative to be assessed quickly and objectively. These are the analysis of the language of coherent narratives (also referred to as microstructure in English-language scholarly literature) and of their structure (macrostructure, respectively).
Analysis of the language of coherent narratives involves examining certain linguistic features that best reveal the quality of the narrative language. Drawing on the experience of researchers in other countries (Loban 1976; Hughes et al. 1997; Justice et al. 2006; Mäkinen, Kunnari 2009; Soodla 2011), three key indicators of the linguistic features of coherent narratives were selected for this study: a) overall productivity, b) lexical diversity, and c) syntactic complexity.
The overall productivity of coherent narratives was determined by assessing two indicators: the mean length of utterance (hereafter MLU) index and the lexical and grammatical form diversity (hereafter LGFD) index. MLU is one of the key indicators of grammatical acquisition and is particularly often required in studies of early language development. This criterion is also used to analyze coherent narratives, as it helps to objectively assess the quantitative features of the narrative under analysis. The LGFD index can be described as the ratio of all words used to the number of their distinct grammatical forms. Like MLU, this indicator is often used in studies of early grammatical development, but it is also recommended for the quantitative analysis of coherent narrative language. In the study of coherent narratives produced by preschool-age children, the LGFD index was calculated for a) all words, b) nouns, c) verbs, and d) adjectives.
Lexical diversity in coherent narratives was examined by analyzing how the main characters in the narrative were referred to using nominal phrases. For the study, all character references used in the children’s narratives were classified as semantically neutral1 (e.g., dog, bird), hypernyms (e.g., animal), and
1 The description “semantically neutral” is used provisionally as a working term, in the absence, for now, of a more suitable one.
216
KALBOS KULTŪRA | 86
Page 6
hyponyms (e.g., bulldog, magpie), after which their percentage distribution was calculated.
The syntactic complexity of coherent narratives was assessed using the communication unit analysis method (Loban 1976). In accordance with the requirements of this method, all narratives were divided into communication units2; their number was then calculated and their structure examined.
Analysis of the structure of coherent narratives involves examining textual cohesion and structural balance. Drawing on the experience of researchers in other countries (Baltaxe, Angiola 1996; Benson 1997; Hughes et al. 1997; Bliss et al. 1998; Batoréo, Costa 2000; Mäkinen, Kunnari 2009; Soodla 2011), four key indicators of textual cohesion and structural quality were selected for this study: a) the balance of the structural parts of a coherent narrative, b) the expression of the main idea, c) the amount of information provided, and d) textual cohesion. When examining the balance of the structural parts of a narrative, it is checked whether the coherent narrative contains all the required structural parts, and whether these parts are appropriately organized and clearly expressed. To determine whether the main idea is conveyed, it is checked whether the plot of the coherent narrative corresponds to the plot depicted in the pictures. In evaluating the amount of information provided, consideration is given to whether the child provides all the plot information and additional information. Textual cohesion is assessed on the basis of the expression of reference and the links between communication units in coherent narratives.
3. RESULTS
3.1. Language features of coherent narrative
Overall productivity
Following quantitative analysis, it emerged that the sentences in preschool-age children’s Fox story consist of ~7–8 words, those in the Cat story of ~9–10 words, while the overall typical MLU index ranges from 4,700 to 15,000. Atypical MLU index cases were recorded in the study group—one extremely low (2,909)
2W. Loban described a communication unit as an independent sentence (or part of one) with all its modifiable elements (1976: 9). In traditional Lithuanian grammar, a communication unit would correspond to (a) a simple sentence; (b) a complex sentence with a subordinate clause; or (c) a clause in a compound sentence. When segmenting a mixed or asyndetic sentence, the boundary of a communication unit is determined by the type of syntactic relation, i.e., coordination or subordination.
I. BALČIŪNIENĖ, A. KALNINYTĖ. Priešmokyklinio amžiaus vaikų rišliojo pasakojimo ypatybės
217
Page 7
and one extremely high (31,000) MLU index value; that is, one narrative consisted of sentences of ~3 words, while the other consisted of sentences of ~31 words. An extremely low MLU index indicates that the coherent narrative contains too few long constructions (relative to the age group) and is limited to short sentences, e.g.:
(1) The fox took the fish. The bird bit her. The bird took the fish from the fox.
An extremely high MLU index suggests insufficient competence in segmenting a coherent narrative, e.g.:
(2) The dog saw her, that the cat was already climbing the tree, pulled at her, bit her tail and tugged, and at that time a bird flew in with a worm, to the children, and then the cat ran down from the branch and the dog began to chase it, and then the bird fed the birds.
The narrative features captured in the audio recording (intonation, pauses) reveal that the child does not try to divide the text into separate segments, but instead tries to convey as much information as possible in a single sentence.
The results of the quantitative analysis show that preschool-age children’s narratives have an overall LGFI index mean of 0.596 (Fox story)–0.727 (Cat story). The narratives contain ~66 words in total³ (~38 distinct forms), yielding an LGFI index of ~0.661. To assess more precisely the diversity of vocabulary and grammatical forms in coherent narratives, it is important to consider the semantically significant words in the narrative; therefore, this study examined the diversity of nouns, verbs, and adjectives in vocabulary and grammatical forms. Quantitative calculations showed that preschool-age children’s narratives contain ~16 nouns (~9 distinct forms), and the LGFI index is ~0.614 (0.491 in the Fox story; 0.737 in the Cat story). Four extremely low noun LGFI index values (0.273, 0.286, 0.333, and 0.353) were recorded (the mean typical noun LGFI index is =0.641), indicating a limited noun vocabulary and little diversity in grammatical forms, e.g.:
(3) Attacks the fish. bird. Wants to eat it. And wanted to eat that fish. And he ate that fish, and the fox came. The fox was unfed, hungry. Then wanted to catch him,
3When counting words, self-corrections (e.g., Here comes the fox... oh, the wolf), cut-off words (e.g., There were fish... fish bones), and filled pauses (e.g., And asks that mm give him the fish) are excluded. Morphological compounds (e.g., somehow, whatever) are counted as one word.
218
KALBOS KULTŪRA | 86
Page 8
well, and he wanted to eat it. Then he let the fish go, and then he would have eaten the fox. Then the fox ran off with it, but was already trying to attack. Then the fox wanted to run away. Because he was attacking her. Then he took it, then the fox wanted to swallow it, to catch it, and caught it.
An exceptionally high (1.000) noun LGFI index means that there are no repeated noun forms at all in the narrative; in other words, the child’s noun vocabulary is more developed, e.g.:
(4) Once a bird watched everything so that no bad animals would come. Then it flew off to bring food to its little ones, and a cat wanted to steal the chicks and eat them. Then she began waiting for the nest to fall. Then she thought she could climb onto the branch herself and eat them. A dog saw this, came over at once with its tongue hanging out, and bit her on the tail. Then it started pulling; when it had pulled her down, it chased the cat until she returned home.
On the other hand, an exceptionally high (1.000) noun LGFI index may also indicate a particularly poor and short narrative, e.g.:
(5) A cat climbed up a tree. A pigeon flew over. And it chased the cat away by its tail.
In this case, the narrative consists of only three sentences, which contain just five different nouns in lexical and/or grammatical forms altogether. The verb LGFI index is higher than the noun index. Narratives by preschool-aged children contain ~16 verbs (~13 distinct forms), and the LGFI index is ~0.892. Two exceptionally low verb LGFI index values (0.600) were recorded (the mean typical verb LGFI index is = 0.863). In general, the verb could be described as the part of speech with the highest LGFI index; even when an LGFI index below the typical value is recorded, it cannot be said that the narrative lacks verb vocabulary, e.g.:
(6) Once a crow laid eggs, and chicks hatched from them. She flew off to look for food, while at the same time a cat noticed the nest that had been built. Well, she watched for a long time and saw the eggs. She began to climb up; the dog saw her and did not want her to eat the little birds, so it grabbed her by the tail, knocked the cat down, and then the mother returned with a worm, saw the little birds, and the dog chased the cat away.
In this narrative, the verb saw is repeated three times, which considerably lowers the verb LGFI index.
The highest verb LGFI index (= 1.000) most often indicates that the narrative is characterized by a particularly well-developed verb vocabulary, e.g.:
I. BALČIŪNIENĖ, A. KALNINYTĖ. Priešmokyklinio amžiaus vaikų rišliojo pasakojimo ypatybės
219
Page 9
(7) There lived a bird who had chicks of her own and loved them very much. One day, she wanted to fly away to bring them food. A cat noticed all this and thought of climbing that tree and snatching the chicks. A dog was lurking around the corner; he knew the bird as his best friend and thought he must not let the cat climb into that nest. He grabbed her by the tail, caught her, and dragged her out of the nest. The cat ran off, and the dog chased after her, while the bird happily returned to her chicks with food.
This narrative is characterized not only by verb synonymy (grabbed – caught), but also by two-, three-, and even four-part verb constructions (thought of climbing, must not let [her] climb, wanted to fly away to bring food).
Quite different results emerge from examining the adjective LGFI indicators: this part of speech is almost never used in coherent narratives by preschool-age children. Of all 48 narratives, adjectives appeared in only six; each was a different lexeme, so their LGFI index was =1.000, e.g.:
(8) A big bird had three little ones. [...] Then the cat looks. [It] saw the nest. [It] wants to eat them, the little birds. [...] And then the mother is happy that her chicks are safe.
Considering that the adjective is generally one of the most difficult parts of speech to acquire and is used least frequently in speech, it could be hypothesized that the use of adjectives in coherent narratives by preschool-age children is a sign of especially rapid language development.
Lexical diversity
In the study, all references to narrative characters using nominal phrases were classified as semantically neutral, hyperonyms, or hyponyms. After counting the words in each group (the number of occurrences), it became clear that semantically neutral character labels predominate in all narratives, while the ratio of hyperonyms, semantically neutral labels, and hyponyms depends on the picture sequence. In the Fox story, semantically neutral labels (bird, female bird) account for 77 percent, hyponyms (eagle, seagull) for 22 percent, and hyperonyms (animal) for just 1 percent of all character labels. In the Cat story, semantically neutral labels account for as much as 80 percent, hyponyms for 4 percent, and hyperonyms for 16 percent of all character labels. The ratio of hyperonyms, semantically neutral labels, and hyponyms also depends on which character is being named. When naming the predator in the Fox story, children choose only semantically neutral labels fox, male fox, wolf, whereas the labels for the bird are more varied—from hyperonyms bird, bir-
220
KALBOS KULTŪRA | 86
Page 10
d⟧ to names of various bird species. When naming the fish, the semantically neutral label fish, skeleton, bone is chosen most often; more specific labels fish skeleton, fish bones are used less frequently, and the hyperonym food occurs once. In creating the Cat story, the bird is most often referred to by the semantically neutral labels bird and female bird; a few hyperonyms (mother) and hyponyms (crow, swan, hen, pigeon) are also used. The bird’s offspring are most often referred to by semantically neutral labels chicks, little birds, tiny birds and by the hyperonyms little ones, children, as well as their diminutive forms. The dog and cat are referred to only by semantically neutral labels (dog, cat) and their diminutive forms; the sole exception is the hyperonym beast, which occurs once.
Syntactic complexity
In examining the syntactic complexity of coherent narratives, all narratives were divided into communication units; the number of these communication units was then calculated and their structure examined. As mentioned above, a communication unit may correspond to (a) a simple sentence; (b) a subordinate clause in a complex sentence; or (c) a clause in a compound sentence, e.g.:
(9) 1 The little bird wanted to bring food to its babies,
2 and the cat wanted to eat them.
3 And then a dog ran up, // when she climbed into the tree, and pulled her by the tail,
4 and she fell and started running,
5 and then the dog chased her.
6 And the little bird brought food to its babies,
7 and the cat did not eat them.
Užsienio lingvistų (Hughes ir kt. 1997: 41) teigimu, toks pasakojimo segmentavimas yra patogesnis nei skaidymas į frazes ar sakinius, nes leidžia objektyviau įvertinti pasakojimo sintaksinių konstrukcijų sudėtingumą. Pavyzdyje pateiktą pasakojimą sudaro 7 komunikaciniai vienetai: 1-asis atitinka sudėtinio sujungiamojo sakinio dėmenį; 2-asis – sudėtinio sujungiamojo sakinio dėmenį; 3-iasis – sudėtinio mišriojo sakinio pagrindinį ir priklausomąjį dėmenis; 4-asis – sudėtinio mišriojo sakinio pagrindinį dėmenį; 5-asis – sudėtinio mišriojo sakinio pagrindinį dėmenį; 6-asis – sudėtinio sujungiamojo sakinio dėmenį; 7-asis – sudėtinio sujungiamojo sakinio dėmenį. Pagrindiniai kriterijai, kuriais remiantis sakinys ar jo dalis yra prilyginami komunikaciniam vienetui, yra a) atskiras gramatinis centras (veiksnys ar tarinys gali būti išreikšti arba numanomi iš konteksto) ir b) sujungiamasis sintaksinis ryšys. Kadangi pateiktame pasakojime yra vienas prijungia-
I. BALČIŪNIENĖ, A. KALNINYTĖ. Priešmokyklinio amžiaus vaikų rišliojo pasakojimo ypatybės
221
Page 11
In the case of a coordinate clause (a subordinate clause together with a main clause, forming one communicative unit), the narrative is calculated to consist of 7 communicative units and 8 grammatical centers, and the ratio of communicative units to grammatical centers is = 1.142.
The research results show that children’s narratives consist of ~7.7–11.3 communicative units, each containing one or more grammatical centers. The ratio of the number of communicative units to grammatical centers (hereafter, CU/GC) is = ~1.043 (Fox Story)–1.125 (Cat Story), while the individual CU/GC index ranges from 1.000 to 1.714. Nearly half (23 of 48) of the narratives produced by preschool-age children still have a one-clause communicative-unit structure; that is, the communicative unit in these narratives comprises only one grammatical center, for example:
(10)
1 The bird wanted to take the food.
2 Then he took it.
3 The wolf came and wanted to eat that bird.
4 Then that wolf wanted to catch him.
5 He tried to catch it but didn’t.
6 Then the bird dropped the fish,
7 and the fox caught the fish.
8 Then the fox ran away from the bird,
9 and the bird flew away.
10 Then a bird flies over and takes the fish,
11 and then the wolf chases the bird.
12 He wants to take the fish.
Comparing these results with those of earlier studies that revealed tendencies in the natural development of coherent narration, it can be concluded that the syntactic constructions in preschool-age children’s coherent narratives are at a transitional stage—from simple and compound coordinate sentences toward complex sentences with subordinate clauses.
3.2. Features of Coherent Narrative Structure
Balance of Structural Components
Depending on the research methodology and visual materials selected, up to 8 obligatory structural components of a coherent narrative can be distinguished: 1) introduction to the situation and/or characters; 2) introduction to the problem; 3) description of the main character’s state (intentions); 4) description of the main character’s plan of action; 5) account of the character’s attempt to solve the problem;
222
KALBOS KULTŪRA | 86
Page 12
6) statement of the consequences of the character’s actions; 7) final assessment of whether the problem was solved or remained unsolved; 8) ending (a sentence or phrase indicating that the narrative is complete) (Hughes et al. 1997: 118–119). Some of these components may partially overlap or adjoin one another; therefore, depending on the research objective and the characteristics of the participant sample, the structure of a coherent narrative may be modified, or only certain components may be examined. Given the participants’ age, this study adopted one of the simpler ways of examining coherent narrative structure: assessing whether the narrative contains three required structural elements: 1) beginning (introduction to the situation and/or characters and/or description of the problem); 2) middle (description of the main character’s state (intentions), description of the main character’s plan of action, or account of the character’s attempt to solve the problem, and statement of the consequences of the character’s actions); and 3) ending (final assessment of whether the problem was solved or a phrase indicating that the narrative is complete). If all three structural elements are present in the narrative, its structure is considered to be a formed three-part structure. If one component is missing (usually the beginning or ending), but there are indications of it, the narrative structure is considered to be an emerging three-part structure. If two components are missing (usually the beginning and ending), the narrative structure is considered to be unformed.
Analysis of narrative structure showed that 36 of the 48 narratives had an emerging three-part structure. Most of these narratives have a beginning (e.g., the situation is described, the characters are introduced, or a conventional fairy-tale opening is used (Once upon a time..., One day... Long, long ago...), and so on) and a middle section, but no explicit ending, for example:
(11) BEGINNING: A bird was flying and wanted a fish. MIDDLE: It took it; the wolf wanted it. And it stood up; the bird had already eaten that fish, and the wolf wanted to eat that bird. The bird dropped the fish and flew away, and the fish fell for him. Then the wolf took it, the bird flew away, and the wolf ran off. Then the bird took it back, and the wolf wanted to eat the bird. ENDING: –
The structure of ten narratives was assessed as unformed; that is, the child describes what they see in the pictures but does not seek logical cause-and-effect relationships, and the narrative has neither a beginning nor an ending, for example:
(12) That cat ran over and is chasing the bird. The bird isn’t here anymore; only the cat is standing there. The dogs are dragging the cat, and the bird is flying back. The cat is running away from the dog, and the bird—to help the pasta.
I. BALČIŪNIENĖ, A. KALNINYTĖ. Priešmokyklinio amžiaus vaikų rišliojo pasakojimo ypatybės
223
Page 13
The structure of only two narratives could be described as fully formed and tripartite, e.g.:
(13) BEGINNING: Once upon a time, a little bird lived with her babies. DEVELOPMENT: And then the little bird flew out of the nest, and the cat thought of a way to catch the chicks. Then she thought of a plan and climbed up, but a dog was nearby. Then the dog pulled the cat, and she—bang—fell down. And the dog attacked the cat, so then the cat ran away. ENDING: And that was that.
Lyginant su ankstesnių tyrimų rezultatais (Balčiūnienė 2013), galima pastebėti tam tikrų rišliojo pasakojimo struktūros raidos požymių: pirmiausia pradeda formuotis pasakojimo pradžios užuomazgos (kartą..., vieną dieną...), vėliau išmokstama informuoti klausytoją apie pasakojimo pabaigą (viskas, pabaiga), dar vėliau pradedama išplėtoti pasakojimo pabaigą ir galiausiai išmokstama išplėtoti pasakojimo pradžią ir sukurti priešistorę. Atsižvelgiant į šiuos polinkius, struktūrinių dalių subalansuotumą galima laikyti vienu iš svarbių rišliojo pasakojimo struktūros rodiklių.
Expressing the Main Idea
One component of the analysis of coherent narrative structure is the expression of the main plot idea; that is, the assessment of whether the narrative plot corresponds to the sequence of pictures. The main idea is most often related to the logical sequence of the protagonist’s goal–attempts to achieve the goal–outcome. If the sequence of pictures includes several equally important characters, all pursuing their own goals, the assessment considers how many plotlines (i.e., character goal–attempts to achieve the goal–outcome sequences) are present in the narrative.
Both the Fox story and the Cat story have three main characters pursuing their goals: in the Fox story: 1) a bird steals a fish; 2) a fox tricks the bird into giving up the fish; 3) the bird gets the fish back. In the Cat story: 1) a bird brings food to her young; 2) a cat tries to catch the chicks; 3) a dog chases the cat away. For the study, all narratives were divided into three groups. The first group comprises narratives in which the goals of all characters, their attempts to achieve them, and the consequences are expressed, e.g.:
(13) One beautiful summer day, (1a) the crow’s chicks hatched. And she flew away to find food for them. As she passed by, (2a) the cat saw the nest and the crow’s chicks sitting in it. She began climbing the tree, but (3a) the dog saw her climbing the tree and (3b) grabbed her by the tail, because he felt very sorry for the little birds. So (2b, 3b)
224
KALBOS KULTŪRA | 86
Page 14
the cat jumped down from the tree and began to run, while the dog started chasing her, and (1b) the crow happily brought food to her babies.
The main idea in the narratives in this group is fully expressed. Some characters’ goals may be stated indirectly (for example, in this case, it is not said why the cat climbed the tree, but her goal is explained by the motive for the dog’s behavior (he felt very sorry for the little birds)); one sentence may reveal the behavior of several characters and the outcome of their actions (for example, here the sentence the cat jumped down from the tree and began to run, while the dog started chasing her shows that the cat did not achieve her goal, whereas the dog did). However, the most important criterion is whether the goals, actions, and consequences for all characters can be identified in the narrative.
The second group comprises narratives in which the goals, actions, and consequences are expressed for only one or two characters, e.g.:
(14) A crow was sitting in the nest with her chicks. And then she flew away; (1a) a cat came. She saw that the nest was empty and climbed the tree. (1b, 2a) Then a dog came and pulled her down from the tree. (2b) He chased her away from the tree and left the chicks alone so the cat wouldn’t take them.
Here, only the goals and actions of the cat and dog, and the consequences of those actions, are described; the bird’s goal of feeding her young is not mentioned. The main idea in narratives of this kind is expressed only partially.
The third group comprises narratives in which the goals of none of the characters, their attempts to achieve them, or the consequences are stated, e.g.:
(15) Once there was a duck in her nest, and they could not reach her. A seagull climbed up with its claws. A cat tried to climb the tree. Then a dog came. The cat was almost holding on, and the dog tried to pull her down. And she ran away.
The main idea in the narratives in this group is not expressed at all; that is, the narrative does not mention any character’s motives or goals, etc., but merely comments on what can be seen in the pictures.
After calculating the proportions of these narrative groups, it emerged that among preschool-aged children, the main idea is fully expressed in 50 percent of narratives, partially expressed in 35 percent, and not expressed at all in 15 percent. Compared with the results of earlier studies, it can be concluded that the ability to express the main idea of a narrative is directly related to the participants’ age and language development, and could therefore be an important indicator of the development of coherent narrative structure.
I. BALČIŪNIENĖ, A. KALNINIYTĖ. Priešmokyklinio amžiaus vaikų rišliojo pasakojimo ypatybės
225
Page 15
Amount of information provided
When creating a narrative based on a sequence of pictures, participants may provide different amounts of information: they may tell only what they see in the pictures (this information may be complete or partial), or add their own interpretations (provide additional information). In conducting the study, all narratives were divided into four groups:
1) narratives that provide all the plot information and a) provide or b) do not provide additional information;
2) narratives that do not provide all the plot information and a) provide or b) do not provide additional information.
Analysis of the coherent narratives showed that most of them (62%) fall into group 2b. These narratives describe only what is depicted in the pictures, without offering any interpretations or information about what is not depicted but could be inferred or surmised from the context, e.g.:
(16) Mother bird has to fly off to look for food. Mother had already flown off to look for food, and the cat appeared. The cat looked at those little birds and climbed the tree. The dog saw her, bit her by the tail, and pulled her. And the mother bird had flown back with a little worm in her beak. And the dog started chasing the cat.
Some narratives (14%) that provide all the plot information also include new details or interpretations, e.g.:
(17) Probably someone had lunch outside here and finished eating the fish, so a thrush flew over, stole it, and carried it away [...].
(18) There once lived a bird who had her own little birds and loved them very much. [...]
Some narratives (18%) were also found to provide incomplete plot information. Most of these lack a particular element of the plot (most often, the goal of one of the characters, the effort to achieve that goal, and the outcome of that effort are not mentioned); this partly coincides with only partial disclosure of the main point of the narrative, e.g.:
(19) The fox wants to take the fish. The fox sees the fish too. The fox wants to take the fish. The fox took the fish. The bird bit her. The bird took the fish from the fox.
(20) The bird had little ones. Then the cat saw them, and the bird flew away. The chicks were left, and the cat wanted to take those little birds. But the dog stopped her and pulled off her tail, then chased the cat. And afterward the bird flew back, and the bird stayed.
226
KALBOS KULTŪRA | 86
Page 16
It can be argued that younger children are not always able to comprehend and reveal all the plot sequences (and therefore certain gaps remain in their narratives), but these skills improve increasingly over time.
Particular attention should be paid to cases in which not all the plot information is provided in the narrative, but additional information and the child’s own interpretations are inserted, e.g.:
(21) The crow went to get the fish and took it. Then she, then... There was grass on the table. There was a tree over there. And a little squirrel lived in that tree. There was a cup with a knife and fork on the table over there. The crow flew onto the tree, and the fox came, and that fox wanted to take the fish, but couldn’t reach it because it couldn’t fly and the tree branches were very high. Because the crow’s beak took the fish. The one where the squirrel lived.
In the preschool-age group, such narratives (they make up 4 percent of all narratives) may indicate a child’s inability to comprehend the plot: the child focuses on the first picture and creates a backstory for the narrative situation, but is unable to reveal the narrative’s plotlines.
Textual cohesion
In examining the textual cohesion of connected narratives, two features were analyzed—reference expression and the relations between sentence constituents. In exploring reference expression, drawing on the experience of foreign linguists (Nicolopoulu et al. 2011), the first three mentions of each character in every narrative were first examined, and the type of these mentions was identified: a) naming with a noun; b) omitted reference (marked 0-REF in the examples below); c) naming with a demonstrative pronoun–noun combination; d) naming with a personal pronoun. All connected narratives were then divided into 3 groups according to how the referent was named. The first group consists of narratives in which referents are always named with nouns, e.g.:
(22) Here a bird is flying for the fish. The wolf came for the fish, and the bird took the fish. The wolf stood up and wanted to grab the bird. The bird dropped the fish, and the wolf turned around. And only bones were left of the fish. The wolf ran away with the fish. The crow grabbed the fish, and the wolf chased it.
The second group consists of narratives in which referents are named with nouns or the reference is omitted, e.g.:
(23) There was a fish on the table. And the bird took [the fish:0-REF]. The fox came. And saw that the bird had taken the fish. The fox wanted to jump, but couldn’t reach it. And
I. BALČIŪNIENĖ, A. KALNINYTĖ. Priešmokyklinio amžiaus vaikų rišliojo pasakojimo ypatybės
227
Page 17
The b i r d dropped it, the f o x dropped it. And the f o x ran away with the f i s h. And the b i r d chased it. And then the b i r d took it from the f o x [fish:0-REF].
The third group comprises narratives in which referents are designated by nouns, personal pronouns, and/or combinations of demonstrative pronouns and nouns, while some designations are omitted, for example:
(24) The f i s h is attacking the b i r d; it wants to eat her. And it wanted to eat t h a t f i s h. And h e ate t h a t f i s h, and the f o x came. The f o x had not eaten, was hungry. Then [fo–x:0-REF] wanted to catch him, and, well, h e wanted to eat her. Then he let the f i s h go, and then the f o x would have eaten it. Then the f o x ran off with it, but then tried to attack. Then the f o x wanted to run away. Because it attacked h e r. Then h e took it, then the f o x wanted to swallow it, to catch t h a t, and caught it.
Narratives in the first group account for 8 percent of the preschool-age study group (none were recorded in the other age groups—Balčiūnienė 2013), narratives in the second group account for 33 percent, and those in the third group for 58 percent of all narratives. It can be concluded that younger children tend to designate referents with nouns; later, they learn to omit designations in certain positions (thus making the text more cohesive); and, finally, they move to a mixed method of designating referents using nouns (usually when mentioning a character for the first time), combinations of demonstrative pronouns and nouns, personal pronouns, or omission of the designation. Nevertheless, it should be noted that appropriately designating referents is one of the most difficult tasks in producing a coherent narrative at preschool age. Even when the mixed (third-group) method of designating referents is used, children do not always manage to use a personal pronoun appropriately and/or to omit a designation in the appropriate position, for example:
(25) Once a bird gave birth to three babies and flew away to fetch food, and a cat came because it wanted to eat the little birds. The d o g s a w h e r, b e c a u s e t h e c a t c l i m b e d i n t o t h e t r e e, and pulled at her, bit her tail [...].
Narratives were also found in which a personal pronoun is used first, but the speaker later doubts this choice and corrects it—the referent is clarified with a noun, for example:
(26) [...] and the dog came. And h e r, t h e c a t, he grabbed by the tail.
(27) [...] Then the cat looks. It saw the nest. It wants to eat t h e m, t h e l i t t l e b i r d s [...].
228
KALBOS KULTŪRA | 86
Page 18
Such cases are not considered errors—on the contrary, self-corrections and clarifications show that the child understands the rules governing anaphoric pronoun use and strives to apply them in coherent narration without disrupting textual cohesion. Thus, taking into account the development of referent expression and the difficulties children encounter, this criterion can be regarded as one of the important indicators of the development of coherent narrative structure.
The second criterion of textual cohesion is the relations between sentences and/or their clauses. Studies by foreign linguists (Reilly et al. 2011) show that the structure and relations of sentences and/or clauses in coherent narratives develop through four stages. At the earliest stage, a narrative consists of clauses (sentences) that are not connected to one another and describe the objects depicted in pictures, e.g.:
(28) Here is a dog. Here is a cat.
Later, the next stage is reached, when a narrative consists of clauses/sentences that are not connected to one another and describe the actions performed by the characters, e.g.:
(29) The fox took the fish. The bird bit her. The bird took the fish away from the fox.
Later still, children learn to connect sentences (clauses) through linear relations, e.g.:
(30) The fox came, and the crow was sitting on a tree branch at that time, and the fox kind of stood on a stick and tried to move the branch. It moved it a little, and the fish fell down there. Then the fox caught it and started running, while the crow started chasing her and got the fish, and then the fox tried to chase her again, but the crow didn’t catch up with her.
Finally, a stage is reached at which the sentences (clauses) of a coherent narrative begin to be connected by causal (purpose) relations, e.g.:
(31) Then the cat pulled at the nest, but could not pull it down because the dog was pulling her by the tail [...].
The study drew on this classification and calculated the frequency index for each of these types of relations. It emerged that linear relations between sentences and/or their clauses predominate in preschool age—their frequency index is approximately 0.656. Sentences and/or their clauses are connected much less often not at all (the characters’ actions are merely described)—the frequency index for this type is approximately 0.321. The least frequent relation between sentences and/or their clauses is causal (consequential), with a frequency index of approximately 0.023. In almost all coherent narratives, not just one but several ways of connecting sentences (clauses) are used; only the proportions of these methods differ. However, it should
I. BALČIŪNIENĖ, A. KALNINIYTĖ. Priešmokyklinio amžiaus vaikų rišliojo pasakojimo ypatybės
229
Page 19
Particular attention should be paid to cases in which sentences (clauses) in a coherent narrative are not linked at all, that is, the interrelations between sentences (clauses) have reached only the first stage of development. This indicator means that the narrative contains neither linear nor causal (goal-oriented) relations; this may be a sign of atypical development of narrative structure.
CONCLUSIONS
In summarizing the characteristics of coherent narratives typical of preschool age, the abundance of deviations of individual indicators from the established mean should be emphasized. Such deviations were recorded most frequently at both the linguistic and structural levels of coherent narratives, compared with other age groups (Balčiūnienė 2013). This may be related to individual differences in linguistic and cognitive development, the child’s experience of listening to and following stories, and family traditions and attitudes concerning reading or telling stories to children and incorporating them into everyday communication4.
1. Coherent narratives produced by preschool-age children have the following characteristics: the mean utterance length is ~8–9 words, but the individual score ranges from 2.909 to 31.000, which means that some narratives contain utterances of only ~3 words (an extremely low score), whereas others contain utterances of more than ~30 words (an extremely high score). Nevertheless, despite these atypical cases, it can be stated that the mean utterance length in most children’s coherent narratives does not deviate far from the typical mean MLU (8.295); this figure could therefore be considered a reference point when assessing the mean utterance length in coherent narratives produced by preschool-age children.
2. Narratives based on a sequence of six plot pictures consist of ~66 words (excluding repairs and incomplete words), of which ~38 are distinct forms. Thus, the lexical and grammatical form diversity (LGFI) index is ~0.661; in other words, each narrative word is repeated-
4 According to the children who participated in the study, at home their parents and older siblings often read children’s books to them (“about a little goat,” “a little bear,” “a princess,” “Thumbelina,” “Little Red Riding Hood,” “Bambi,” “Puss in Boots,” “Harry Potter”), and at kindergarten teachers sometimes read books before the midday nap. Some children said that they read books themselves, but such answers should be treated with caution (some children who gave this answer said that they did not know the letters; moreover, conversations with some children revealed that a book was considered read once all of its pictures had been looked through). Some children mentioned that they liked not only books but also children’s magazines such as “Barbie,” “Donald Duck,” “Flint,” and others.
230
KALBOS KULTŪRA | 86
Page 20
ed ~1.7 times. Narratives show a wide range of individual LGFI scores—from 0.422 to 0.800—and atypical cases (with extremely low scores) are far removed from the typical LGFI score. After additionally calculating the LGFI score for content words, it can be stated that narratives produced by preschool-age children contain ~16 nouns (~9 distinct forms, LGFI ~0.614) and ~16 verbs (~13 distinct forms, LGFI ~0.892). It should be noted that both an extremely high and an extremely low noun LGFI score in this age group may indicate poor narrative quality (a limited noun vocabulary or an underdeveloped plot). By contrast, even an extremely low verb LGFI score does not indicate a substantial lack of verb vocabulary. The infrequent use of adjectives can be considered a distinctive feature of coherent narratives produced by preschool-age children (compared with other age groups; Balčiūnienė 2013). Given that adjectives are generally among the most difficult parts of speech to acquire and are used least frequently in speech, it may be assumed that even the use of a single adjective in a coherent narrative by a preschool-age child may be a sign of faster-than-typical language development.
3. The vocabulary is characterized by an abundance of semantically neutral nouns used to name the narrative characters (bird, female bird, fox, cat). Semantically neutral character names account for as much as 79 percent, hyponyms (magpie, crow) for 13 percent, and hypernyms (animal, beast) for 8 percent of all character references (instances of use). This age group also included names that were not entirely appropriate to the narrative plot (thrush, seagull), but this may be related to the child’s cognitive rather than linguistic development.
4. Syntax is characterized by an index of the ratio between the number of communication units and the number of grammatical centers—~1.084. A large proportion of the narratives (23 of 48) consist of one-member communication units (independent clauses with dependent elements), that is, the communication unit in these narratives contains only one grammatical center.
5. Overall, the narrative structure can be considered to be developing into a three-part structure; this is characteristic of 36 of the 48 narratives examined. Narratives most often have an opening and a body, but no explicit ending. As many as 10 narratives do not meet the requirements for coherent narrative structure at all. The structure of only 2 narratives in this age group could be described as a fully developed three-part structure, that is, the narrative has a clearly expressed beginning, body, and ending.
6. Difficulties in conveying the main idea of narratives become apparent. Only half of the narratives have a fully conveyed main idea,
I. BALČIŪNIENĖ, A. KALNINYTĖ. Priešmokyklinio amžiaus vaikų rišliojo pasakojimo ypatybės
231
Page 21
for others (35 percent), it was partially revealed or (15 percent) not revealed at all, i.e., half of the children studied did not understand all the narrative plotlines and/or were unable to express them appropriately. When creating a coherent narrative, a large proportion of the children (62 percent) provided all the necessary plot information but did not offer any interpretations or information about what was not depicted but could be inferred or surmised from the context. Several narratives (14 percent) not only revealed all the plot information but were also supplemented with new details or interpretations; however, as many as 18 percent of the narratives lacked some plot information. This suggests that preschool-aged children are not yet always able to comprehend and reveal all the plot elements, and therefore certain gaps remain in their narratives. In this age group, more cases were found than in other groups (Balčiūnienė 2013) in which the narrative did not include all the plot information but did incorporate additional information and the children’s own interpretations.
7. A wide variety of referential expression is characteristic. Some narratives (4 out of 48) use only the simplest means of referential expression—naming characters with noun phrases; in others (16 out of 48), in addition to using noun phrases, the ability to omit reference to a character where it is unnecessary was observed; and more than half (28 out of 48) of the coherent narratives demonstrate the ability to use all possible means of referential expression: noun phrases, omitted reference to a character, demonstrative pronouns, and personal pronouns.
An examination of the cohesion of sentences and/or their clauses shows that the simplest way of creating a coherent narrative—merely describing the objects depicted in the story pictures—is no longer characteristic of preschool-aged children. Approximately one third of the sentences or clauses in coherent narratives are still not connected, but they describe not only objects but also actions, and thus imply plot relations. Approximately two thirds of narrative sentences or clauses are characterized by a linear relationship, i.e., they are connected by the conjunctions and, or. Connecting sentences and their clauses through causal (purpose) relations is particularly rare; however, given the difficulty of comprehending and expressing this relation, it should not be considered a major shortcoming in narrative cohesion. On the other hand, the expressed causal (purpose) relation in a preschool-aged child’s narrative could indicate more rapid development of coherent narrative skills.
Although limited in scope, the study revealed essential tendencies in coherent narratives produced by preschool-aged children; it is therefore hoped that the research results reviewed in this article will be useful in conducting diagnostic assessments of language development disorders and/or drawing up individualized language intervention plans.
232
KALBOS KULTŪRA | 86
Page 22
LITERATŪRA
B a l č i ū n i e n ė I. 2009: Pokalbio struktūros analizė kalbos įsisavinimo požiūriu. Daktaro disertacija, Kaunas: Vytauto Didžiojo universitetas.
B a l č i ū n i e n ė I. 2013: Rišliojo pasakojimo analizės metodas kalbos raidos tyrimui. Podoktorantūros tyrimo „Rišliojo pasakojimo analizės metodas kalbos raidos tyrimui“ mokslinė ataskaita, Vilnius: Lietuvos mokslo taryba.
B a l č i ū n i e n ė I., K a l n i n y t ė A. 2013: Narrative production in Lithuanian TD Mono- and Bilingual 5–6-year Olds. Pranešimas skaitytas tarptautinėje mokslinėje konferencijoje Child Language Seminar-2013, 2013 m. birželio 24–25 d. Mančesteris: Mančesterio universitetas.
B a l č i ū n i e n ė I., K r i v i c k a i t ė E., K a l n i n y t ė A. 2011: Lietuvių kilmės vaikų, gyvenančių ne Lietuvoje, kalba: nuostatos, įtakos, tendencijos. Pranešimas skaitytas tarpdalykinėje mokslinėje konferencijoje Pasaulio lietuvybės idėja: šiuolaikinės migracijos patirtys, 2011 m. spalio 20 d. Kaunas: Vytauto Didžiojo universitetas.
B a l č i ū n i e n ė I., P o v i l a i t i e n ė G. 2012: Rišliosios kalbos ugdymas ikimokykliniame amžiuje: pasakojimo pagal paveikslėlių seką metodas. Pranešimas skaitytas Prof. Artūro Uozuolo 48-ojoje tarptautinėje konferencijoje Kalbos gramatinė ir leksinė sistema, jos funkcionalumas ir normatyvumas, 2012 m. kovo 15–16 d. Ryga: Latvijos universitetas.
B a l t a x e A. M., D ’ A n g i o l a N. 1996: Referencing skills in children with autism and specific language impairment. – European Journal of Disorders of Communication 31, 245–258.
B a t o r é o H. J., C o s t a A. 2000: Reference mechanisms in children’s oral and written narratives at the age of ten. – Psycholinguistics on the Treshold of the year, 265–270.
B e n s o n M. S. 1997: Psychological causation and goal-based episodes: low-income children’s emerging narrative skills. – Early Childhood Research Quarterly 12, 439–457.
B e r m a n R. A., S l o b i n D. I. 1994: Relating Events in Narrative. A Crosslinguistic Developmental Study. Psychology Press, Hillsdale, NJ: Erlbaum.
B l i s s L. S., M a c C a b e A., M i r a n d a A. E. 1998: Narrative assessment profile: discourse analysis for school-age children. – Journal of Communication Disorders 31, 347–363.
D a b a š i n s k i e n ė I. 2010: The comprehension of derivational morphemes in early childhood: an experimental study for Lithuanian. – Estonian Papers in Applied Linguistics 6, 43–51.
F e d e r a v i č i e n ė E. 1978: Ikimokyklinio amžiaus vaikų rišliosios kalbos ugdymas, Vilnius.
G a g a r i n a N., K l o p D., K u n n a r i S., T a n t e l e K., V ä l i m a a T., B a l č i ū n i e n ė I., B o h n a c k e r U., W a l t e r s J. 2012: MAIN: Multilingual Assessment Instrument for Narratives. ZASPiL 56 – December 2012, Berlin: ZAS.
I. BALČIŪNIENĖ, A. KALNINYTĖ. Priešmokyklinio amžiaus vaikų rišliojo pasakojimo ypatybės
233
Page 23
Gagarina N., Klop D., Kunnari S., Tantele K., Välimaa T., Balčiūnienė I., Bohnacker U., Walters J. 2013: Production and perception of narrative texts across languages and cultures: MAIN. Pranešimas skaitytas tarptautinėje mokslinėje konferencijoje Child Language Impairment in Multilingual Contexts, 2013 m. gegužės 27–29 d. Krokuva: Krokuvos Jogailaičių universitetas.
Garrard P. 2005: The effects of very early Alzheimer’s disease on the characteristics of writing by a renowned author. – Brain 128 (2), 250–260.
Garšvienė A., Ivoškuvienė R. 1993: Logopedija, Kaunas: Šviesa.
Giedrienė R. 1986: Pradinukų kalbėjimo sutrikimai, Kaunas: Šviesa.
Glebuvienė V. 1977: Vaikų ankstyvosios garsinės kalbos raida: metodinė medžiaga, Vilnius: Pedagogikos mokslinio tyrimo institutas.
Grigaitė B. 2004: Lietuvos vaikų brandumas mokyklai: tyrimas ir problemos, Vilnius: VPU.
Gülzow I., Gagarina N. 2007: Noun phrases, pronouns and anaphoric reference in young children narratives. – ZAS Papers in Linguistics 48, 203–223.
Gutierrez-Clellen V., Quinn R. 1993: Assessing narratives of children from diverse Cultural/Linguistic groups. – Language, Speech, and Hearing Services in Schools 24, 2–9.
Hickmann M. 2003: Children’s discourse: person, space and time across languages. Cambridge: Cambridge University Press.
Hudson J. A., Shapiro L. B. 1991: From knowing to telling: The development of children’s scripts, stories and personal narratives. – Developing narrative structure (eds. A. McCabe and C. Peterson), Hillsdale NJ: Erlbaum, 89–136.
Hughes D., McGillivray L., Schmidek M. 1997: Guide to Narrative Language. Procedures for Assessment, Austin: PRO-ED, Inc.
Ivoškuvienė R., Mamonienė Z., Pečiulienė O., Suveizdienė V. 1997. Ikimokyklinio amžiaus vaikų žymiai ir vidutiniškai neišsivysčiusios kalbos ugdymas, Šiauliai.
Justice L. E., Bowles R. P., Kaderavek J. N., Ukrainetz T. A., Eisenberg S. L., Gillam R. B. 2006: The index of narrative microstructure: a clinical tool for analyzing school-age children’s narrative performances. – American Journal of Speech-Language Pathology 15, 177–191.
Kalnius A. 1939: Vaikas ir kalba. Perspausdinta: Gimtoji kalba, 1997 (1), 16–23.
Kamandulytė L. 2010: Lietuvių kalbos būdvardžio įsisavinimas: leksinės ir morfosintaksinės ypatybės. Daktaro disertacija, Kaunas: Vytauto Didžiojo universitetas.
234
KALBOS KULTŪRA | 86
Page 24
Kamandulytė-Merfeldienė L., Fürst B., Dressler W. U., Sauerland U. 2010: On the acquisition of adjective gradation in Lithuanian and German. – Darbai ir dienos 54, 267–276.
Loban W. 1976: Language Development: Kindergarten through grade twelve, National Council of Teachers of English.
Mäkinen L., Kunnari S. 2009: Narrative as an assessment method in children with SLI and AS/HFA. Pranešimas skaitytas COST veiklos IS0804 darbo grupės WG2 Narratives Meeting narių susitikime, 2009 m. lapkričio 16–18 d. Amsterdamas: Amsterdamo universitetas.
Mazolevskienė A. 2000: Dvikalbystė Lietuvos šeimose. – Pedagogika 42, 52–55.
Mazolevskienė A. 2003: Dvikalbių vaikų lietuvių kalbos mokėjimo ypatumai. – Mano vaikai. Priešmokyklinis vaiko ugdymas (Sud. O. Monkevičienė), Kaunas: Šviesa, 76–96.
Nicolopoulou A., Aksu-Koc A., Küntay A., Andersen Ch., Chang Ch., Ignacio M., Slancová D. 2011: A cross-linguistic study of character reference in young children’s narratives. Pranešimas skaitytas XII-ajame tarptautiniame kongrese 12th International Congress for the Study of Child Language, 2011 m. liepos 19–23 d. Monrealis: Monrealio Kvebeko universitetas.
Reilly J. S., ir kt. 2011: Emerging narratives: A cross-linguistic study. Pranešimas skaitytas XII-ajame tarptautiniame kongrese 12th International Congress for the Study of Child Language, 2011 m. liepos 19–23 d. Monrealis: Monrealio Kvebeko universitetas.
Ruzaitė J., Dabašinskienė I. 2010: Specific language impairment and diagnosis: adaptation of a diagnostic test for Lithuanian. – Darbai ir dienos 54, 277–300.
Savickienė I. 1999: Lietuvio vaiko daiktavardžio morfologija. Daktaro disertacija, Kaunas: Vytauto Didžiojo universitetas.
Savickienė I. 2003: The Acquisition of Lithuanian Noun Morphology, Wien: Verlag der Österreichischen Akademie der Wissenschaften.
Savickienė I., Kempe V., Brooks P. 2009: Acquisition of Gender Agreement in Lithuanian: Exploring the Effect of Diminutive Usage in an Elicited Production Task. – Journal of Child Language 36, 477–494.
Soodla P. 2011: Picture-Elicited Narratives of Estonian Children at the Kindergarten-School Transition as a Measure of Language Competence. PhD Dissertation, Tartu: Tartu University Press.
Wójcik P. 2000: The Acquisition of Lithuanian Verb Morphology: a Case Study, Kraków: Quartis.
Gauta 2013 10 06
Priimta 2013 12 20
I. BALČIŪNIENĖ, A. KALNINYTĖ. Priešmokyklinio amžiaus vaikų rišliojo pasakojimo ypatybės
235
Page 25
THE MAIN CHARACTERISTICS OF NARRATIVE IN PRESCHOOLERS
Summary
The article presents the study of monologue (coherent) language of preschool-age children establishing the main micro- and macro-structural indications of Lithuanian preschool age children’s narratives. The study was based on the method of story generation according to a picture sequence: the children participating in the study were asked to tell a story according to a sequence of six related pictures. 24 children of 5–6 years old attending preschool education groups participated in the study; the sample was balanced in respect of social properties; all children were monolingual; they were not diagnosed with any speech-related or other developmental disorders. Every child was individually tested twice; hence, 48 narratives in total were collected. The narratives were recorded by a voice recorder, transcribed and morphologically annotated for an automatic analysis of narrative language and structure using CHILDES software. The analysis of the data collected identified the following individual and general properties: general productivity (including MLU and TTR), lexical diversity (namely, NP references), syntactic complexity (namely, CU structure), narrative structure, quantity of information and cohesion (namely, referential and linkage development) of the narrative. The results obtained enable us to compare individual results, to make general conclusions and to identify the general tendencies of narrative development.
