Lexical Cohesion in the Narrative Produced by Javanese School-Age Children in Pati

Abstract

This chapter presents a study which discussed the narrations in Bahasa Indonesia by children in Pati, Central Java, who still use Javanese at home. Our focus is on lexical cohesive devices, based on Halliday and Hasan’s (Cohesion in English. Longman, 1976) theory. The participants in this study are school-age children (N = 51) aged 6–9 years old, classified into three groups based on their grades at school. This study uses semi-structured elicitation technique with silent film The Pear Story (developed by Chafe in The pear film, 1975) as the main instrument. Instruction is in Indonesian. The analysis is supported by our observation at school and parental questionnaires on language use. Descriptive statistics analysis is conducted to observe the tendency of the use of the lexical cohesive device. It is found that children use Indonesian- and Javanese cohesive devices in their story and use code-mixing, as well. The result shows that the use of repetition cohesive device in Indonesian is being very dominant, compared to other cohesive devices. The most dominant ones in all grades are the repetitions of nouns and verbs.

Introduction

In Pati, Central Java, people still use both the regional language, Javanese, and the national language, Bahasa Indonesia, in daily activities; the regional language for informal speech and Bahasa Indonesia for formal communication. In the educational domain, the use of Indonesian, especially at school, is compulsory. Being competent in Bahasa Indonesia, the formal variety of Indonesian is one of the requirements in this domain. However, code-switching and code-mixing of Indonesian and Javanese are often used in daily conversations. Based on our preliminary observation in Pati, these phenomena are not uncommon, as have also been found in several studies (e.g., Dewi, 2006; Sudono, 2011), and certainly affect children’s language use.

When children enroll in school, the use of Indonesian increases as it is used as the main medium of instruction. In early elementary school, children—especially those whose first language is Javanese—experience a transition to using Indonesian. Some materials of certain subjects at school are given in Bahasa Indonesia, but in order to understand, children still need explanations in their first language. In such a situation, teachers need to code-switch or code-mix. The need for code-switching and code-mixing from Indonesian to Javanese to explain difficult materials shows that children’s understanding of Indonesian has not yet been perfected. The transition commonly ranges from children’s first enrollment to approximately the third grade.

Concurrent use of Javanese and Indonesian evokes a diglossic situation for children. They tend to use Javanese with persons closest to them, such as family or friends, and Indonesian to other people or in formal situations, such as at school. The diglossic situation also happens at school, where both, in this case Javanese and Indonesian, are used. While the former is mostly used outside the classroom, the latter is used inside the classroom.

At school, narration is used to improve children’s speaking skills. They are encouraged to speak in Bahasa Indonesia. In this activity, children learn how to tell a story in Bahasa Indonesia in a well-composed, coherent, and cohesive structure. Based on our preliminary observation, these children, especially those who are in the first, second, and third grades, are starting to gain more exposure to Indonesian while they are still using Javanese. The question to be addressed is: How do these children create a cohesive story in Bahasa Indonesia?

This chapter discusses the use of cohesive devices in these children, focusing on the lexical cohesive devices, as proposed by Halliday and Hasan (1976). Our participants in this study were school-age children, who were in the first, second, and third grades of an elementary school in Pati, Central Java. The main data of this study was children’s narrations which were elicited from the silent film The Pear Story (developed by Chafe in 1975). Observation by researchers and a questionnaire completed by parents also supported the study.

This chapter is part of the first author’s thesis, completed under the second author’s supervision. In this chapter, we briefly discuss the theoretical framework and research method. With data obtained from questionnaires, we also explain the use of Indonesian and Javanese and the result from analysis of the usage of lexical cohesive devices. The conclusion also suggests several findings that could be discussed in further research.

Narratives: By Children and the Cohesive Devices

This section discusses some reviews on the theoretical background of this study. First, we present a brief discussion on children’s language development, especially in narrative. We also discuss briefly the language of bilingual children, as a comparison to the present chapter. As we examine children’s use of cohesive devices, we discuss briefly Halliday and Hasan’s theory on cohesion.

Children and Narrative Ability

Children’s narrative production emerges at an early age and the pace of development grows in the preschool- and early school-age (Justice et al., 2006; Mäkinen et al., 2013). This ability continues to develop into production of long, coherent, and cohesive narration during the school-age period (Berman, 2009; Hoff, 2009).

Willenberg (2016) who studied the development of narrative production with mixed-race preschool and third-grade children in South Africa found that vocabulary, macrostructure elements in narration, and discourse features, improved with age. Children’s early experience with at-home reading positively correlates with the macrostructure of third-grade children’s narratives. The study also found no correlation between the education level and language use of mothers who were non-native speakers of English. Additionally, Mäkinen et al. (2013) studied the development of narrative production in Finnish children, aged 4–8. These researchers found a rapid increase in production capability and utterance of events in children aged 4–5. Five to six-year-old children showed a rapid increase in their production of reference cohesion capability.

This study deals with children who are exposed to more than one language. Hoff (2009, p. 298) stated that children’s exposure to two languages can happen in diverse situations. For instance, children are not necessarily equally exposed to both languages. Hoff (2009, p. 304) also showed evidence of bilingual children’s overlapping use of lexical forms because they have two lexical systems for expressing the same concept. In Hatay, Turkey, Coşkun (2011) studied fifth-graders’ use of cohesive devices with immigrant children from Uzbekistan, who spoke Uzbek and Turkish. Coşkun found no significant difference between the use of cohesive devices by immigrant and nonimmigrant children in Turkey. Besides Coşkun (2011), Kim (2011) also studied bilingual (Korean–English) and monolingual children’s use of discourse markers. Similar to Coşkun’s study (2011), Kim’s study showed no difference between bilingual and monolingual children. Keith and Nicoladis (2013) studied 7- to 10-year-old bilingual (French–English) and monolingual (English-speaking) children’s lexical categories. Compared with monolingual children, bilingual children tended to make more mistakes in word-meaning categorization. One language’s influence on the other was also observable in bilingual children. Squires et al. (2014) studied story retelling by bilingual Spanish–English, typically developing (TD), and primary language impaired (PLI) children. They found that at school, TD children made greater growth in microstructure in their first language, while PLI children did not. This study shows the growth patterns of macrostructure (hierarchical structure) and microstructure (linguistic) elements in the narratives of bilingual children, both with a primary language impairment and without. Squires et al. also pointed out that the influence of schooling played an important role in first language development, especially for the bilingual TD children.

In Indonesia, Purwo and Sukamto (2016) studied Javanese and Indonesian first to sixth graders’ language competence in Gunung Kidul, Yogyakarta. Through data elicitation using a silent film, results showed that children in lower grades were more eloquent telling a story in Javanese than in Indonesian. In addition, through data elicitation using images, Indonesian competence increased in the lower to higher grades, while Javanese competence showed no development. In this study, we also observed Javanese–Indonesian children’s narrative production in the school setting. However, our focus was on the use of lexical cohesive devices.

Cohesion and Cohesive Devices

Indicators of a good narration include coherence and cohesion. Coherence, in narrative, is the connection that is brought about by something outside the text, whereas cohesion is the connection between its elements (Renkema, 2004, p. 49). According to Peterson and McCabe (1991, p. 29), cohesion is the relationship between sentences. Messages and meaning are communicated effectively as it ties together and organizes structure (Horton-Ikard, 2009, p. 394).

A cohesive narration is manifested by use of appropriate cohesive devices. According to Halliday and Hasan (1976), there are two kinds of cohesive devices related to grammatical and lexical aspects. Grammatical cohesive devices include reference, substitution, conjunction, and ellipsis. Lexical cohesive devices include repetition, synonymy, hyponymy and hypernymy, meronymy, antonymy, and collocation.

Since the focus of this chapter is the lexical cohesive devices, a more detailed description is presented in the following.

• Repetition is a cohesive device which refers to the recurrence of a certain element of language. It can be lexical or phrasal repetition and is usually nominal or verbal.

• Synonymy is characterized by the repetition of the same meaning, such as field and area.

• Hyponymy and hypernymy are characterized by the recurrence of a form with subordinate and superordinate meanings, such as fruit and apple.

• Meronymy is characterized by the recurrence of a form where the meaning refers to a part of a whole, such as wall and house.

• Antonymy is characterized by the recurrence of a form with contrasting meaning, such as happy and sad.

• Collocation is manifested by words that usually appear together in the same context, such as sand, beach, and coconut tree.

Method

The main data of this study are children’s narrative productions, elicited from a silent film. We also observed children’s social environment and activities, as well as their language use inside and outside the school environment. Interaction between others was also our main concern. Even though we had observed children’s language use at school, we needed supporting information on language use at home, to explain some of the findings that might occur. Therefore, we also constructed a parental questionnaire.

Before we collected the data, we conducted a pilot test on five children in grades one through three, in Pati. We performed a trial procedure by asking several children, who were not involved in this research, to watch the film and retell the story. As this study’s focus is Indonesian cohesive devices, we used Indonesian as the instructional language. In the pilot study, we observed that these children mixed Indonesian and Javanese. Nevertheless, they could follow and tell the whole story, despite differences—which were basically due to the influence of local knowledge—in lexical choices when they referred to certain objects. For example, they used jambu “guava” to refer to the fruit instead of using pir “pear.”

Instruments

The main instrument of this study was the silent film The Pear Story (downloaded from http://pearstories.org.), which was designed by Wallace Chafe in 1975 at the University of California, Berkeley, to elicit language samples around the world. The film is about a farmer who picks some pears, his stolen pear, and the boy who steals it. It was developed for narrative elicitation from speakers of different languages around the world and used by many researchers to examine numerous aspects of linguistic competence, including narration. Since the film contains no dialog, informants can tell the story without any influence from the characters’ speech. Based on this consideration, we chose it as the instrument for narrative elicitation. As we also observed in this study, children’s lexical choices in narration originated from what they saw, understood, and used daily.

Apart from the silent film, parental questionnaires were used to support this research. The questionnaire was adapted from the second author’s research (see Kushartanti, 2014) and completed by selected children’s parents (a description of the selection process is presented in the following subsection). The questionnaire contained questions about the children’s demographic data (date and place of birth) and the parents’ sociocultural information. Based on our preliminary observation that Javanese as well as Bahasa Indonesia is used in daily conversation, we assumed that the children’s linguistic input would be in both languages. However, it was important to use the most dominant language as the input language. Therefore, the questionnaire has needed to examine the extent to which the parents used both languages at home, at school, and in public areas. Besides collecting information on children’s demographic data, the questionnaires also collected information on the patterns of language use of both children and the people around them.

Selection of Participants

Before the data collection, we went into the children’s classes and sat in and made ourselves known to the children. Once they became familiar with us in the classroom, we showed them the equipment (i.e., a laptop) that we were going to use and several films (which were not used for the research). We watched these films together and talked about them afterward. We then selected several active children with good communication skills as participants. Furthermore, the researchers also mingled with the children when they were playing during the breaks.

Data Collection Procedure

Children’s speech data were obtained from individual interviews, conducted in a small room provided by the school. Prior to each interview, the researcher enquired about the child’s daily activities to put her/him at ease. We sat together and watched the complete film, The Pear Story. Following that, the film was replayed, and the interviewer asked the child to tell the story while watching it. The child’s speech was then recorded. The interviewer responded to the child’s story only with back-channel behavior (see Renkema, 2004, p. 165), such as hmm, iya “yes,” wah “wow,” o, gitu “I see,” so that he/she did not feel awkward or self-conscious. These prompts encouraged the child to talk. At times, the interviewer repeated some parts of the child’s speech, as an acknowledgment or confirmation. An example of such conversation is the conversational segment below, which was taken after the interviewer played the silent film (INT = interviewer; CHI = child).

(1)
[…]
INT	: Udah ya?	“so”
	…Itu:.. tadi ceritanya tentang apa?	“what is the film about?”
	Sekarang Kak Rizma puterin lagi,	“Now, I will replay it”
	nanti kamu cerita ya.	“and you will tell me the story”
	Sambil cerita ya.	“tell me the story”
	Bisa?	“can you?”
CHI	: (nodded)
(INT replayed the film, and the CHI started to tell the story and pointing some objects)
CHI	: (pointing)Orang yang.. memetik apel.	“there is the man who picks apples”
INT	: Mmm	“mmm”
	memetik apel.	“picks apples”
CHI	: Memetik apel yang banyak.	“he is picking so many apples”
INT	: Mmm iya.	“mmm, yes”
CHI	: Keranjang..nya ada:… tiga	“he has three baskets”
INT	: Mmm.	“mmm”
	He’e:m.	“yes”
CHI	:Itu,	“there…”
	kok ad..a #dua.. diambil orang.	“it seems…two...(it is) taken by someone”
INT	: Mmm.	mmm
[...]

There were 56 children who were interviewed (20 from the first grade, 19 from the second grade, and 17 from the third grade). However, five recordings could not be used, either because the sound was not properly recorded or the child remained silent. Therefore, the total number of recordings for the purposes of the data source was 51 (17 from each grade).

The parental questionnaires were distributed at school and completed by parents at home, with a 100% return rate (N = 51). However, one questionnaire was not entirely completed, so we excluded it, leaving 50 questionnaires for the analysis.

Data Processing

Recordings were transcribed as orthographic transcription, adapted from Du Bois (2006). From the transcription, we identified and classified the cohesive devices and conducted a quantitative analysis to examine the use of Bahasa Indonesia in the children’s speech. Nevertheless, the use of Javanese cohesive devices and whether there was some mixing of both languages were also analyzed. The demographic data from the parental questionnaires were analyzed quantitatively, as well.

An Overview of the Participants

From the questionnaires (N = 50), we obtained demographic information, such as place of birth, birth order, parents’ place of origin, and parents’ occupation. Forty-one children were born in the Pati district; others were born in the Central Javanese-speaking areas such as Grobogan, Kudus, Wonosobo, and Kendal.

We found that both parents of these children came from Central Javanese-speaking areas, and all of them came from Javanese ethnic groups. This information showed us that no child came from an inter-ethnic marriage (which sometimes indicates that the child uses Indonesian as the language of the family, instead of either language of the parents).

Overview of the Language Use

In the following section, we discuss the results from the analysis of the children’s language use, based on data from the parental questionnaires. This analysis examines how the children learned and used both Javanese and Indonesian, as it could potentially influence the narrative production. To describe the children’s linguistic profile, we analyzed the information in the parental questionnaires. In this section, we present comparisons between Javanese and Indonesian in terms of how the children use the languages in their surroundings (home, school, and community) and in terms of how other people use the language when speaking to the children. We present figures to show the comparisons. Figure 1 shows the mean scores of language use by individual children in those environments, whereas Fig. 2 shows mean scores of languages used by people around the children—father, mother, siblings, paternal grandparents, maternal grandparents, and other close relatives, based on the children’s grade level.

Fig. 1

Use of Indonesian and Javanese by children at home, school, and public space/community (1 = never; 4 = always)

Fig. 2

Languages used by others while interacting with children (1 = never; 4 = always)

Figure 1 shows that children used both Indonesian and Javanese. However, we can see that the use of these languages was influenced by the setting. At home and in public spaces, children tended to use Javanese more frequently than Indonesian. A paired samples t-test revealed significant differences between the use of Javanese and Indonesian at home (p = 0.00) and in the public space (p = 0.00), where Javanese tended to dominate. At school, children used both Indonesian and Javanese. The figure shows that the use of both Indonesian and Javanese is more or less equal. We found that there was no significant difference between the use of Javanese and Indonesian (p = 0.351) at school.

Figure 2 shows the language used by others when interacting with children.

Figure 2 shows that both Javanese and Indonesian were used by fathers, mothers, siblings, paternal grandmothers and fathers, maternal grandmothers and fathers, and others (relatives and new acquaintances) to talk with the children. In other words, these children have both Javanese and Indonesian input from their extended family and others and are already familiar with both languages. The figure shows that extended family members (grandparents) and others tended to use Javanese more frequently than Indonesian. Nevertheless, the difference is not significant. There is also a reason that these people used both languages. One reason for using Indonesian, as we observed, was to avoid Javanese stratified styles, namely ngoko (the lowest style), ngoko alus (the low style), krama alus (the higher style), and krama inggil (the highest style). Inappropriate use of style can be a problem when speaking to certain people from a different social status; for example, it is considered impolite for a younger speaker to speak ngoko instead of krama alus to an older person. We observed that Indonesian is used at home to avoid Javanese styles or registers. The finding was in line with Poedjosudarmo (as cited in Cohn & Ravindranath, 2014, pp. 140–141) and Gunarwan (2006). Therefore, it is understandable that Indonesian plays an important role in their social life.

Lexical Cohesive Devices in Children’s Narrative Speech

From the children’s speech data, we found repetition, hyponymy, hypernymy, synonymy, and antonymy, in both Indonesian and Javanese and also in a mix of both languages. Table 1 presents the cohesive devices’ mean ratios. The formula of the ratio was as follows:

$$ \frac{{{\text{Total}}\;{\text{score}}\;{\text{of}}\;{\text{cohesion}}\;{\text{per}}\;{\text{individual}}}}{{{\text{Total}}\;{\text{of}}\;{\text{all}}\;{\text{cohesion}}\;{\text{per}}\;{\text{individual}}}} $$

Table 1 Mean and standard deviation of ratio of lexical cohesive devices

Table 1 shows that repetition dominated the children’s speech. The first graders (M = 0.902, SD = 0.095), second graders (M = 0.890; SD = 0.093), and third graders (M = 0.891; SD = 0.088) showed no significant differences in use and distribution; however, equitable distribution occurred only in the use of Indonesian repetition. Table 1 also indicates that Indonesian cohesive devices were more frequently used than Javanese cohesive devices, and the third-grade children no longer mixed Javanese and Indonesian cohesive devices. We also found that the use of other cohesive devices was limited. Moreover, our analysis showed that the mean scores (M) of these devices were less than the score of the standard deviation (SD), indicating that there was a large individual difference.

The findings on Indonesian dominance in the data can be attributed to the fact that it was the language of instruction in the data collection. In this case, these children were already capable of using the language; at least they understood the instructions and tried to accommodate the interviewer’s speech. As they were still developing the ability to use Indonesian, some were still influenced by their first language, Javanese. Nevertheless, we also found that competence in using Indonesian increased as the children grew older, as indicated in the oldest children’s data.

We also analyzed each cohesive device’s mean scores based on the category, at the lexical, phrasal, and clausal level. The scores presented in Table 2 below are the ratio scores. Not all types of cohesive devices, as proposed by Halliday and Hasan (1976), were found in Javanese or in a mix of Indonesian and Javanese. Besides that, only repetition of nouns and verbs was frequently used. We also found that each grade had a different tendency in using cohesive devices, and the children did not use all of them. Table 2 shows a more detailed representation of the lexical cohesive devices, based on the two languages.

Table 2 Use of lexical cohesive devices in Indonesian and Javanese

From Table 2, we can see that the most dominant lexical cohesive device in all grades was nominal and verbal repetition. This could be because the research instrument was a film, and the children told the story while watching it. Moreover, the use of repetitions indicated that it worked well in maintaining the characters, which is in line with Colozzo and Whitely’s (2014) findings. In the current study, the children mentioned the “objects” they saw while watching the film, several of which are always visible in the film, for instance, the farmer, the boy who steals, and the pear. In the final scene of the film, three other boys help the thief. Furthermore, each scene connects to another through these objects, so the children repeatedly mentioned the visible objects, which resulted in frequent repetition. Table 3 shows the mean ratio scores of the cohesive device of repetition in the data.

Table 3 Mean ratio of categories and types of lexical cohesive devices

All the children used nominal repetition to mention objects reappearing in each scene, for example, the fruit and the thief. The oldest children tended to use verbal repetition, nominal phrase repetition, and verbal phrase repetition. Meanwhile, repetition of other categories was used only by certain participants.

No child mentioned “pear,” which was the name of the fruit that was one of the objects in the film, since they were unfamiliar with that fruit. Instead, they tended to call it jambu “guava,” mangga “mango,” apel “apple,” or alpukat “avocado.” Here, we see that culture has influenced the children’s perception on the object. It is understandable that these Pati children did not mention “pear,” as they were not familiar with it. Instead, they mentioned things they knew from their own culture. Nevertheless, they were able to mention the object as “fruit.”

As previously mentioned, the boy who steals the pear is always visible in the film. The children mentioned him, not only in nominal repetition, but also in nominal phrases. The first graders (M = 0.668; SD = 0.152) and the second graders (M = 0.603; SD = 0.238) tended to use nominal repetition; however, their references were not clear because they overlapped with the other main character, the farmer. For example, the children mentioned orang “person,” seseorang “a person,” and anak “child.” The third graders used noun phrases (M = 0.633; SD = 0.122); therefore, their references were clearer, for example, tiga anak “three children” or tiga orang “three people” for the boys who help the thief.

Besides repetition, the children used other cohesive devices, such as hyponymy, hypernymy, synonymy, and antonymy. A few Javanese interferences were also found in the use of other lexical cohesive devices, namely hyponymy and synonymy, in Javanese and a mix of Indonesian and Javanese. Only the first and second graders used such cohesive devices; the third graders used only Indonesian cohesive devices. The findings suggest that the first and second graders were still influenced by Javanese, which was still used in the classroom. Nevertheless, given that the use of Javanese cohesive devices was not as frequent as the Indonesian ones, it implies that the use of the latter at school has had an impact on the children’s speech. At school, the children are trained to speak individually using Indonesian, in certain contexts, such as when they explain content on a certain subject. The higher the grade, the greater the children’s exposure to Indonesian. Because of that stimulus, third graders can speak Indonesian without any intervention from Javanese. Studies have shown that school greatly affects children’s language development (Squires et al., 2014). As the school environment has a different language requirement, compared with other environments, the teaching and learning process tends to create an environment that encourages interaction. How teachers speak and interact with their colleagues also affects children’s language. Moreover, the findings in this study also confirm that as the children in this study grow older, their ability to use the second language—Indonesian—develops further.

We found that if there was any interference, especially in their use of lexical cohesive devices, the children were not able to find the specific word in Indonesian with which to label a certain event, for example, the occurrence of hyponymy jatuh (Ind. for “fall down”) and manting (Jv. for “fall down”). Manting refers to a two-wheeled vehicle that is shaky or unbalanced and then crashes and flips over. Semantically, manting is a subordinate of tiba (Jv. “fall down”) which has no equivalent in Indonesian. Another finding was that children tended to use the same word for different events. The word mengambil “take” sometimes implies mencuri “steal” or memetik “pick.” The word usually occurs in hyponymy. This means that children use the word mengambil “take” instead of mencuri “steal” or memetik “pick.” We need to consider the context (in this case, the film) to determine whether mengambil is the hyponym of mencuri or of memetik.

Conclusions and Recommendations

This study has shown that the children in Pati were able to produce cohesive narratives by using lexical cohesive devices, such as repetition, hyponymy, hypernymy, synonymy, and antonymy. Compared to other cohesive devices, repetition, especially repetition of nouns, was most frequently used by the children. The findings show that the children used this kind of cohesive device to maintain the narration, which was also in line with Colozzo and Whitely’s (2014) findings. We also found that the children’s narration was influenced by the instrument, the silent film The Pear Story, especially when they were focusing on the objects and the actions they saw.

The first and second graders also used cohesive devices in Javanese and a mix of Indonesian–Javanese, but the third graders did not. We observed that the school was influential in exposing them to Indonesian and not to Javanese; therefore, the use of Javanese decreased as the children reached higher grades. The questionnaire responses regarding the dominant language used at home, at school, and with other people also supported this research. Nevertheless, our findings show that both Indonesian and Javanese have their own functions, respectively, and that the children had almost equal Indonesian and Javanese input from the people around them, especially their extended families.

At the early stage of elementary school, children are still in a transition period between using Javanese as the dominant language and Indonesian as the language of education. But at school they have to speak mainly in Indonesian. By the third grade, the participants in this study were already accustomed to speaking Indonesian. In the language used by the oldest participants, therefore, we found neither Javanese cohesive devices, nor mixed Javanese and Indonesian cohesive devices.

We also found other tendencies in the use of lexical cohesive devices. The first is that the children used cohesive devices as tools for topic maintenance (see also Colozzo & Whitely, 2014). The children used repetition while thinking and staying on topic, especially for finding words and correcting their language. They also sometimes used hyponymy and synonymy to find more specific words and correct their language. In some cases, the children used lexical cohesive devices to refer to something vague, for instance, they used a generic word that is a hyponym of certain word with a different meaning. In this study, the film’s context played an important role, and therefore, further study on grammatical cohesive devices to examine the use of references in children’s narrative is very important.

This research was conducted in Pati, a Javanese region where people still speak the regional language. The findings of this research paint a small picture of children’s language use in this region. Nevertheless, this study has some limitations. First, it was conducted at only one school, and the number of participants was also limited. Moreover, the focus of this study was only on the use of lexical cohesive devices in which linguistic input at home and school was accounted for. We did not examine other factors, such as age nor gender differences, nor the socioeconomic status, which might add further explanation to the findings. As Javanese still has a large number of speakers with Indonesian being their second language, research with more participants is still needed to have a comprehensive illustration of how Javanese-speaking children use and learn their first and second languages, especially in narration.