Just Like Conversation?: Speech and Thought Presentation in Indonesian Adolescent Fiction

Abstract

This chapter examines a remark by an observer of Indonesian adolescent literature that the language in Indonesian Teenlit novels resembles “spoken language,” implying that it is conversational in style. Drawing on approaches to speech and thought presentation initially applied to the study of English language texts, the chapter shows that direct speech and free direct speech are the main techniques employed by Indonesian authors to represent the speech and thought of adolescent characters. It is argued that the observer’s remark is not an indication that the language in Teenlit is indeed like conversational language. Rather, it represents an intuitive judgment that attests to the authors’ effective use of the techniques.

Introduction

Toward the end of the 1990s, a new genre of popular fiction entered the Indonesian publishing market and drew the attention of adolescent readers, educators and literary critics alike. “Teenlit,” as the genre became known, was introduced to the Indonesian audience through the translation of American adolescent novels at a time when there was little on offer in terms of popular fiction for adolescent readers beyond translation of Japanese comics and Indonesian folk tales. As a popular fiction genre, Teenlit was welcomed by young readers but, perhaps predictably, treated with disdain by critics and educators. In the opinion of the latter, novels in this genre lack educational values. They objected to the fact that the stories were mainly focused on the lifestyles of wealthy Jakartan teens, and the writing style was, in their view, overly colloquial. However,  not everyone agreed. Some argued, for instance, that the writing style in Teenlit approximated the speech style of young urban Indonesians and was precisely the reason why young readers were drawn to the new genre. Gunawan (2006), for example, described the style as “basically spoken language that is written,” implying that what was written was a faithful representation of informal conversation. Though Gunawan’s statement is an oversimplification of the way language is employed in fiction, it nonetheless had an intuitive appeal for those who read it, as evident by the fact the statement was recycled numerous times in various blogs and chat forums, both by readers who appreciated the genre and critics who objected to its existence. In his blog, Gunawan gave several examples of colloquial language in Teenlit novels but did not say whether he based his claim on a systematic study or a cursory observation. Even if the remark was simplistic, similar  remarks by others that the language was ‘speech like’ tells us something about how stylistic choices are recognized. Bucholtz (2015, p. 29) defines style as “a way of doing things,” of “engaging in culturally significant activities and practices of any kind using a range of stylistic features in both established and innovative ways” (2015, p. 32). What stylistic choices can be identified in Teenlit that led observers such as Gunawan to conclude that the language is speech like or conversational?

In a study of style in contemporary Indonesian youth fiction (novels and comics), Djenar and Ewing (2015) point out  that one of the characteristics of contemporary Indonesian popular fiction produced around the 1990s is the blurring of the boundary between the language of narration, which is traditionally rendered in standard Indonesian, and that of dialogue, which may incorporate colloquial terms. In Teenlit, colloquial expressions are commonly employed to represent the speech and thought of teen characters. Djenar and Ewing suggest that colloquialism or standardness is a matter of degree, that is, the extent to which and the way in which elements from standard and colloquial varieties of Indonesian are deployed in youth fiction varies across individual works  and genres.

Although Djenar and Ewing’s study has shown that authors of contemporary Indonesian popular fiction have moved away from the norm that assigns standard and colloquial varieties to different narratorial functions, the question of whether the writing style in this genre, and more specifically in Teenlit, can be characterized as conversational has not been explored. This study addresses this question by drawing on the literature on speech and thought presentation (Fludernik, 1993; Leech & Short, 2007; Semino & Short, 2004; Short, 2014). By describing the different types of speech and thought presentation and showing how they are deployed in Teenlit, I hope to demonstrate that the techniques for presenting speech and thought employed by Teenlit authors tend to be biased toward the adolescent characters’ points of view. The application of these techniques in the novels creates an impression that the characters are talking to the reader with little intervention from the narrator. This factor coupled with the stereotype of adolescent characters as heavy users of colloquial language led observers such as Gunawan to equate the writing style in Teenlit with colloquial speech.

The analysis presented here is based on data from two novels: Still, written by Esti Kinasih, and Dilan: dia adalah Dilanku tahun 1991 (hereafter, Dilan), by Pidi Baiq. It might be objected that this dataset is too small to draw generalizations about Teenlit as a genre. To allay this concern, I have selected novels that are different from each other in several respects. Still is written by a Jakarta-based female author and narrated in the third person. The novel was published in 2007, a time when public attention on Teenlit as a new genre peaked, as indicated by the many  public debates around the question of whether reading Teenlit novels had educational merit for adolescents, and the number of publishers willing to promote the novels in the genre. Like Still, Dilan is a love story narrated in the first person. Written by a male author based in Bandung, Dilan was published in 2015 when public debates about Teenlit—including objections to the use of colloquial language in adolescent literature—had largely subsided. The difference in the temporal and sociocultural contexts between the two novels and the ethnolinguistic background of the authors ensures the analysis is not biased toward one writing style. That being said, it is acknowledged that given the small data size, a quantitative analysis would likely yield unreliable results. The quantitative findings presented here should therefore be considered preliminary and a prompt for further research.

Types of Speech and Thought Presentation

Fiction authors draw on a range of techniques to present the speech and thought of their characters. These techniques have been the subject of many studies in stylistics and literary studies. The analysis of speech and thought in this paper is based on the theory proposed by Leech and Short (2007),¹ extended in Semino and Short (2004), and further refined in Short (2012). Leech and Short (2007) distinguish between the presentation of speech and that of thought by drawing two parallel sets of categories.

Figure 1 shows the categories from Leech and Short (2007, p. 276), which include only speech and thought presentations (with some adaptation²). These categories form a cline, from the type of speech or thought presentation in which the narrator has the strongest degree of control of what is reported, to the type of presentation in which the narrator has the least degree of control. These are discussed in further detail in the following sections, drawing on Teenlit examples.

Fig. 1

Categories of speech and thought presentation (adapted from Leech and Short (2007, p. 276))

To the two sets of categories above, Semino and Short (2004) added an additional set, namely the presentation of writing, referring to the “reporting” of speech, thought or writing that is anterior to the one presented (2004, p. 30) and defined as  “any linguistic structures used to introduce a stretch of speech, thought or writing presentation.” The category “report of writing” accounts for instances in which what is reported is not the speech or thought of a character but some writing, such as a quotation from some written sources (e.g., written documents or reports). The example below from Semino and Short (2004, p. 102), taken from an autobiography, illustrates this category.

(1)	I know he suspected that I ate the wrong food for while I was convalescent in the country he wrote to me frequently: I still have his letters. “Be sure to eat the right food,” he says repeatedly. (Muriel Spark, Curriculum Vitae, p. 204; bold letters in Semino and Short (2004)).

This category is also relevant for the Indonesian data, as illustrated in the examples from Dilan shown later in the chapter.

Speech Presentation

In terms of speech presentation, the two main categories, direct speech (DS) and indirect speech (IS), are related in that both are concerned with the presentation (or “report”) of what someone said, the main difference being that in DS, the report is supposedly faithful to the original, while in IS, the reporter conveys what was said using the reporter’s own words (Leech & Short, 2007, p. 255). Short et al. (2002) stress that the notion of “faithfulness” and the related notion of “verbatim” do not simply refer to an exact reproduction of original words and should be understood differently for different text types. For example, in conversation, the words someone reports to someone else may be able to be traced back to their original source, while in fiction, this is not the case. Fictional speech, presented as DS, is essentially produced for the first time as the reader encounters it in reading, and in that regard, the DS has no antecedent: “…the words of a character in DS are not being reported at all, but being witnessed within the world of fiction as they are being produced for the first time” (Short et al., 2002, p. 335; italics in original). This is different from a report by a character of another character’s speech, which conveys words uttered by a character in a fictional past (to which we, as readers, are privy). The reader knows the source of the words and interprets the report accordingly with an assured degree of certainty about its faithfulness. This is because, by the time the report is presented, the reader has already witnessed the event in which the antecedent is produced: “In fiction which reports events that happened in a fictional past, we usually assume that the ‘reports’ are real within the fiction; therefore, we use the canonical assumptions associated with various categories of discourse presentation scales analogically, and with effectively guaranteed faithful results” (Short et al., 2002, p. 336).

The structural difference between DS and IS in the Indonesian data is illustrated in Excerpts (1) and (2) respectively. In (1), Aldo’s speech is presented in between quotation marks and followed by the reporting frame kata Aldo “said Aldo,” indicating that the speech is conveyed through the voice of the narrator. In excerpt (2), the information is conveyed through the words of the narrator without any graphological marking separating  the character’s speech and that of the narrator, and the person deixis is anchored accordingly. Notice the difference between the first-person pronoun gua (1sg) used by Aldo in the quoted speech to refer to himself in (1) and the third person pronoun dia (3sg), which has the co-reference ‘Aldo’ in (2).

(2)	“Oke deh. Gua nggak pingin tau,” kata Aldo (Kinasih, 2007, p. 28).
	‘“Okay, whatever. I don’t want to know,” said Aldo.’

(3)	Aldo bilang oke. Dia nggak pingin tau.
	‘Aldo said okay. He didn’t want to know.’ (my adaptation of [2])

When a character’s words are reported in IS, linguistic elements that are adjusted may include subjectively-anchored elements, such as discourse markers. As one of the functions of discourse markers is to index the speaker’s stance toward the addressee and what is said, if the character’s words are reported by another character, the discourse marker is adjusted so that it indexes the reporter’s stance. Alternatively, the discourse marker may be omitted altogether in IS. In the DS example in (1), Aldo uses the discourse marker deh to index his indifference toward what his friend Rei said. In the previous discourse (not shown), Rei had asked Aldo if he could use his bedroom, promising to leave the room in its original condition when he finished. Rei gave no reason for his request, leaving Aldo puzzled. Being a supportive friend, Aldo granted Rei’s request without asking questions. The discourse marker deh, which follows the expression of agreement oke “okay,” shown above in (2), indexes his indifference toward Rei’s situation. The following clause, Gua nggak pingin tau “I’m not interested,” constructs this stance more explicitly. Given deh forms part of Aldo’s stance, it would be nonsensical if it were included in the narrator’s IS in (3). However, a different discourse marker that indexes the narrator’s stance toward what is said could be used, as shown in (4).

(4)	Aldo bilang oke dan dia nggak pingin tau lho.
	‘Aldo said okay and he didn’t want to know, you see.’ (my adaptation of [2])

Free direct speech (FDS) is similar to DS in being a faithful report of what someone said but differs from DS in that it is rendered in a freer form. DS may include both or either one of the following formal features: a reporting frame and quotation marks. Correspondingly, in the freer form, either the reporting frame and the quotation marks are both absent or only one of these is included. When the formal indicator of the narrator’s presence is not there, the reader is given an impression that the character speaks directly to them (Leech & Short, 2007, p. 258). An example of FDS that includes quotation marks but does not include the reporting frame is given in (5). Three friends—Langen, Fani and Febi—are talking. Langen has just told Fani and Febi that her boyfriend, Rei, abruptly ended their relationship. The first two speaking turns, in (5a) and (5b), are rendered in DS, with a reporting frame included, therefore we know that the narrator is present. In the reporting frame, the narrator explicitly informs the reader that it is Fani’s and Langen’s words that are reported in (5a) and (5b), respectively. However, in (5c), the speech is rendered in FDS with quotation marks but without a reporting frame, thus it is unclear whose speech is presented. Leech and Short (2007, pp. 258–259) note that a long stretch of FDS can create confusion in the reader’s mind but they add that this may be precisely the effect some authors want to produce. In (5c-f), confusion may arise because the reader is presented with a multiparty fictional interaction, but the reporting frame specifying the name of the speaking character is missing, and the content of what is said does not provide a clue as to which character is speaking. The problem may be less complex if this interaction is dyadic.

Let us examine the interaction more closely. As noted, in speech turns (5a) and (5b) the narrator presents Fani’s and Langen’s words, respectively. Given that in turn (5b) we are presented with Langen’s words, we can surmise that in turn (5c), it must be either Fani’s or Febi’s words that are presented. However, in the absence of a reporting frame, we cannot be certain which of these two characters is speaking. In (5d) we are presented with Langen’s speech again, and we know this by inferring from what is said and from the previous turn, in which Langen is addressed by name (notice the name “La” in (5c), thus projecting that she would be speaking next. The subsequent two turns, (5e) and (5f), are similarly rendered in FDS, and, as with (5c), there are no clues to help us ascertain who is speaking, and in which turn. The excerpt ends in (5g) with a DS showing Langen’s speech again, as indicated by the mention of her name in the reporting frame. (The characters’ names are underlined for convenience).³

(5)
	a.	“Gila!” Fani geleng-geleng kepala. “Ini ya, yang lo maksud waktu itu? Kayaknya ada yang aneh sama Rei.”
	b.	“Iya.” Langen mengangguk
	c.	“Tapi udah deh, La. Nggak usah diperpanjang. Kalau maunya Rei gitu, ya udah.”
	d.	“Siapa juga yang mau memperpanjang?”
	e.	“Nggak apa-apa, kan?”
	f.	“Lupain aja!”
	g.	“Oke.” Langen menghela napas, lalu mengangguk lemah. (Kinasih, 2007, p. 39)

	a.	‘“(It’s) crazy!” Fani shook her head. “Is this what you said before? That it looks like there’s something strange happening to Rei.”’
	b.	‘“Yeah.” Langen nodded’
	c.	‘“But just forget it, La . Why drag it on. If that’s what Rei wants, let it be.”’
	d.	‘“Who’s dragging it on?”’
	e.	‘“It’s okay, right?”’
	f.	‘“Don’t worry about it.”’
	g.	‘“Okay.” Langen drew a deep breath, then (she) nodded weakly.’

If we consider the context created by these framed and frameless turns—by which are meant turns that contain and do not contain a reporting frame to indicate who is speaking, respectively—it is interesting to observe that frameless turns are used here to construct a context of grievance-sharing among intimates. The first two turns in extract (5) establish the cause of the grievance (turn [5a]) and the identity of the character who experiences the grievance (turn [5b]), respectively, whereas the remainder of the dialogue, except the last turn, provides a context through which intimacy among the characters is built. In this regard, individuating the characters through use of a reporting frame and specifying their names in each turn seem less of a concern for the author than highlighting how intimacy is fostered through the display of empathy for the grieved party. It is also instructive that the final turn includes a reporting frame, which clarifies that it is Langen’s voice that the narrator is presenting. This turn marks the closing of grievance-sharing and indicates that the aggrieved person has achieved a resolution.

The alternation between frameless (FDS) and framed (DS) turns in this example is particularly effective in conveying in-group intimacy. The absence of a reporting frame suggests to the reader that hearing what someone says is more important than being informed about who says it. At the same time, for grievance-sharing to be understood as intimacy building, the reader must know who consoles whom, and the reporting frames used in the first two turns are devices deployed by the author to achieve this rhetorical goal.

A variation of FDS, in which a reporting frame is provided but quotation marks are not, is shown in (6). The clause, tidak terjadi apa-apa ‘nothing happened’ in the emboldened sentence is separated with a comma from the reporting frame Langen mengatakan “Langen said” to indicate that the clause represents Langen’s speech. Without the comma, the sentence would read like a straightforward IS, as can be seen in the adapted version of the example in (7). Because Indonesian verbs do not inflect for tense, the comma in (6) is the only indicator of FDS. If we insert the complementizer bahwa ‘that’ after the verb mengatakan ‘say,’ as shown in (8), we can be more certain that the sentence is an IS.

(6)	Raditya juga cemas melihat kondisi Langen yang kini terlihat murung dan agak pendiam. Tapi tidak ada yang bisa dilakukannya karena Langen mengatakan, tidak terjadi apa-apa (Kinasih, 2007, pp. 39–40).
	‘Raditya is also worried about Langen who is looking miserable and prefers to keep to herself. But there is nothing he could do because Langen said, nothing happened.’

(7)	Tapi tidak ada yang bisa dilakukannya karena Langen mengatakan tidak terjadi apa-apa.
	‘But there is nothing he could do because Langen said nothing happened.’

(8)	Tapi tidak ada yang bisa dilakukannya karena Langen mengatakan bahwa tidak terjadi apa-apa.
	‘But there is nothing he could do because Langen said that nothing happened.’

In (9), the FDS consists of the reporting frame dia bertanya ‘he asked’ and ada apa ‘what’s the matter?’ The use of a question mark indicates that it is the voice of the father that the narrator is conveying. Without this question mark, the sentence could be interpreted as a straightforward IS, as shown in the adapted version in (10). (Note that in the English translation, the past form of “be” has to be moved to the end of the sentence in order to turn the question into a statement; no such grammatical adjustment is required for Indonesian.)

(9)	Sepertinya, Ayah melihat aku gelisah, dia bertanya ada apa? (Baiq, 2015, p. 156).
	‘It seems, Dad saw me feeling anxious, he asked what’s the matter?’

(10)	Sepertinya, Ayah melihat aku gelisah, dia bertanya ada apa.
	‘It seems, Dad saw me feeling anxious, he asked what the matter was.’

Unlike turns (5c)-(5f) in example (5), in which the voice of the characters are given between quotation marks, the use of a reporting frame in (9), as we also saw in (6), indicates a greater degree of narrator interference. Nevertheless, unlike DS, the absence of quotation marks in these FDS examples results in a partial fusion between the narrator’s voice and that of the character.

Free indirect speech (FIS) is the freer form of IS (Leech & Short, 2007, p. 260). Like IS, a reporting frame may be included in FIS, but unlike IS, the choice of any grammatical marking indicating time and person (tense in English) in the quoted speech reflects the perspective of the character. In example (11) the author describes the situation from the narrator’s point of view, except the temporal phrase waktu ini “at this time,” which references the time of Langen’s experience and reflects the character’s point of view. In a straightforward IS, a temporal phrase deictically anchored to the narrator’s point of view would be used, as shown in (12).

(11)	Tapi tidak ada yang bisa dilakukannya karena Langen mengatakan tidak terjadi apa-apa waktu ini.
	“But there is nothing he could do because Langen said that nothing had happened at this time.”

(12)	Tapi tidak ada yang bisa dilakukannya karena Langen mengatakan tidak terjadi apa-apa waktu itu.
	“But there is nothing he could do because Langen said nothing had happened at that time.”

Another example of FIS is taken from Dilan and shown in (13). The story is narrated in the first person from the point of view of the female protagonist, Milea. Milea is telling the reader that Yugo, a male friend of hers who is keen on dating her, asked her what had happened between her and Dilan, her boyfriend. The temporal phrase tadi malam ‘last night’ in (13), is deictically anchored to the time when Milea and Dilan were talking—that is, the night before Milea talked to Yugo—and not  the time Milea is narrating the event to the reader. This produces an interesting effect of drawing the reader to the time Milea and Dilan were talking, thus highlighting Milea’s point of view with regard to her relationship with Dilan. Notice that if we adjust the temporal phrase to malam sebelumnya ‘the night before,’ reflecting the narrative time (i.e., the time Milea is narrating the story to us, the reader), it creates a distancing effect; it is as if Milea is merely reporting to us that Yugo asked what had happened between her and Dilan rather than taking us to the time she and Dilan were together.

(13)	Yugo nanya lagi soal kejadian tadi malam antara aku dan Dilan (Baiq, 2015, p. 170).
	‘Yugo asked again about what happened last night between me and Dilan.’

(14)	Yugo nanya lagi soal kejadian malam sebelumnya antara aku dan Dilan.
	‘Yugo asked again about what happened the night before between me and Dilan.’

The different temporal points in this example are schematized in Fig. 2. Time₁ is the time of the event (Event_e) being talked about , Time₂ is the time Milea talked to Yugo, and Time₃ is the time that Event_e is being reported to the reader (narrative time). The arrows are used below to indicate “appropriate temporal phrase, to refer to an Event_e at narrative time.”

Fig. 2

Temporal reference for example (11)

In the Narrative Presentation of Speech Act (NPSA), the narrator informs us that a speech act has been performed by a character but does not provide details of what was said. In this type of speech presentation, the narrator is “apparently in total control” of the report (Leech & Short, 2007, p. 260). An example is given in (15). The narrator is telling us that Rei’s heart berbisik mengingatkan “whispered a warning” but does not specify what words were used in the warning (in this example, the action of whispering is to be understood metaphorically).

(15)	Tubuh Rei terguncang seketika. Membentur tembok dan bersandar di sana tanpa tenaga. Kemudian tubuh itu meluruh lunglai. Bahkan hatinya yang tadi sudah berbisik mengingatkan, kini ikut diam (Kinasih, 2007, p. 38).
	‘Rei’s body shook suddenly. (It) hit the wall and lay there weakly. That body then crumbled, having lost all its strength. Even his heart that previously whispered a warning now grows silent.’

Similarly, in (16), the reader is informed that Raditya asked his younger sister to admit that the only reason she is looking low-spirited is that she has just argued with her boyfriend, but we don’t know what Raditya actually said to his sister.

(16)	Akhirnya setelah sekali lagi meminta ketegasan bahwa memang benar-benar tidak ada yang terjadi selain pertengkaran, Raditya menghibur satu-satunya adik perempuannya itu dengan cara seorang kakak yang baru terima gaji bulanan (Kinasih, 2007, p. 40).
	‘Finally, after once again requesting a confirmation that nothing else had happened except an argument, Raditya consoled his younger sister by doing what an older brother, who has just received his monthly salary, does.’

Thought Presentation

Direct thought (DT) is similar to DS in that it may include both or either one of the following formal features: a reporting frame and quotation marks. In example (17) from Dilan, the DT (shown in bold) is preceded by two sentences describing how the female protagonist Milea is feeling. These sentences are not grammatically constructed as a reporting frame but function similarly to a reporting frame in specifying whose thought is being presented.

(17)	Aku rindu Dilan. Aku hanya merasa begitu kosong.
	“Dilan … kamu di mana?” (Baiq, 2015, p. 139)
	‘I miss Dilan. I just feel so empty.’
	‘“Dilan … where are you?”’

Later in the same chapter, Milea’s thoughts are presented in Free Direct Thought (FDT). As with FDS, both the reporting frame and quotation marks, or either one of these features, may be absent. Example (18) includes neither feature but the personal pronoun aku “1sg” explicitly indicates that we are dealing with Milea’s thoughts and are viewing the situation from her perspective.

(18)	Apakah aku harus nelepon Dilan? Atau, menunggu Dilan nelepon? Kalau nunggu Dilan yang nelepon, entah kapan dia akan (Baiq, 2015, p. 158).
	‘Do I have to ring Dilan? Or, wait for Dilan to ring? If (I) wait for Dilan to ring, who knows when he’ll (do it).’

Example (19) includes a reporting frame, and the example looks like indirect thought (IT). However, upon closer inspection, it is in fact ambiguous between IT and FDT, and the ambiguity is shown through the English translation in (19) and (20).

(19)	Jadi, aku berpikir biar aku saja yang nelepon duluan tapi entah mengapa mendadak aku urungkan niatku, bersamaan dengan muncul perasaan gak enak karena semalam sudah marah-marah ke Dilan (Baiq, 2015, pp. 158–160).
	‘So, I thought I should be the one who rings first but for some reason, I suddenly changed my mind, as I felt a little guilty for being cross with Dilan last night.’

(20)	Jadi, aku berpikir biar aku saja yang nelepon duluan tapi entah mengapa mendadak aku urungkan niatku, bersamaan dengan muncul perasaan gak enak karena semalam sudah marah-marah ke Dilan (Baiq, 2015, pp. 158–160).
	‘So, I thought let me be the one who rings first but for some reason I suddenly changed my mind, as I felt a little guilty for being cross with Dilan last night.’

In free indirect thought (FIT), the narrator’s presentation of a character’s thought is woven into the character’s presentation of their own thought, similar to FIS. In both cases, it is often difficult to ascertain whether we are seeing the situation from the narrator’s or the character’s viewpoint; this scenario becomes complicated by the fact that in first person narration, the narrator is also a character. Leech and Short (2007, p. 272) point out that there is an interesting difference between FIT and FIS. While FIS distances us from the speaking character, FIT produces a sense of vividness and immediacy, drawing us closer to the character’s thoughts. This is illustrated in the Indonesian example shown below in (21).

The example begins with evaluative statements. In the previous discourse (not shown here), Bima is described as feeling dispirited as his desperate search for a past girlfriend, whom he had wronged and to whom he wants to apologize, is not yielding any result. Based on this information, we can interpret the first three evaluative statements (in bold) in the example as presentations of Bima’s thoughts. This character-based point of view shifts in the following (indented) sentence, as indicated by the shift to the mention of a third party, “Bima,” which indicates that the situation is now viewed from the narrator’s perspective. This perspective is sustained in the following paragraph where the reader is given a dramatic description of Bima slumping as he broke into tears. Another perspective shift occurs as we read the last two expressions describing Bima’s feelings (Mengecil. Kerdil. ‘Shriking. Small.’). The absence of overt indicators of Bima’s self-reference, in this example, renders his thoughts less readily distinguishable from the narrator’s thoughts; nevertheless, due to the use of FIT, the reader is provided with a closer insight into the character’s thought.

(21)	Menangis bukanlah cengeng. Menangis meredakan sakit, meskipun tidak mengubah keadaan. Meskipun sama sekali tidak menebus kesalahan.
	Dan… Bima menangis!
	Terisak. Air matanya turun. Tubuhnya terhuyung mundur membentur Jeep. Perlahan, dia jatuh terduduk. Masih terisak, dia lalu tertunduk. Meringkuk. Mengecil. Kerdil (Kinasih, 2007, p. 135)
	‘Crying is not silly. Crying reduces pain, though (it) doesn’t change things. Though it doesn’t erase your mistakes at all.’
	‘And… Bima cried!’
	‘(He) sobbed. His tears rolled (down his cheeks). His body staggered backwards hitting the Jeep. Slowly, he sat down. Still sobbing, he (sat with) his head down. Crouching. Shrinking. Small.’

The final category in the types of thought presentation is the Narrative Presentation of Thought Act (NPTA). Similar to the NPSA, in NPTA, we, as the reader, can access a character’s thoughts through the narrator’s description. The reader is thus positioned some distance away from the content of the thought. Nevertheless, as Leech and Short (2007, p. 271) remind us, even NPTA allows the reader to access a character’s points of view, though less directly than if they are presented through other types of thought presentation: “A writer who decides to let us know the thoughts of a character at all, even by the mere use of thought act reporting, is inviting us to see things from that character’s point of view.” An example of NPTA from the Indonesian data is  given in (22). In this example, the narrator reports on Rei’s intention to scream as loudly as he can, but that his desire to not create commotion stops him from doing so. This is preceded by two sentences describing the physical manifestation of the character’s feeling of frustration. The NPTA and the narrative description thus work in concert to illustrate the character’s state of mind, as viewed through the narrator’s perspective.

(22)	Rei mengatupkan kedua rahangnya kuat-kuat. Kesepuluh jarinya mengepal. Ia ingin berteriak keras-keras, tapi sadar, akan menyebabkan kegemparan di tetangga sekitar (Kinasih, 2007, p. 35).
	‘Rei clenched his jaws. His ten fingers clenched into a fist. He wants to scream, but (he) realizes (that it) would greatly disturb the neighbours.’

The aforementioned examples illustrate the application of Leech and Short’s speech and thought categories to the Indonesian data. A note was made regarding the ambiguity of example (19). This example may be categorized into more than one type depending on interpretation and how it is translated into English (see Semino & Short, 2004, pp. 197–198 for a discussion on ambiguities in their English language corpus). This ambiguity is not explored further here but is an interesting issue to investigate in more depth in a future study. In the following section, I consider the extent to which the different categories of speech, thought and writing presentations are represented in the data.

The Distribution of Speech and Thought Presentations in the Data

In a study of speech and thought presentation based on three text types of English writing—fiction, news report, and (auto)biography—Semino and Short (2004) found that DS and FDS are the most frequently occurring of all speech and thought presentation categories. Moreover, the number of occurrences of these two categories combined is higher in fiction than in news reporting or (auto)biography. In speech presentation categories alone, (F)DS constitutes 26% compared with 2.9% of IS and FIS combined. (F)DS occurrences are also more frequent in “popular” fiction compared with “serious” fiction (60 and 40%, respectively; see Semino and Short, 2004, p. 67). Short et al., (2002, p. 342) attribute this finding to several factors, one being that (free) direct speech “helps to produce the effects of immediacy, drama, and involvement that are particularly important in the telling of fictional stories.” DS and FDS also provide the reader with an impression of direct access to the characters’ words and fulfill our conventional expectations that reported/quoted speech is a faithful “reproduction” of the original, though in fiction the speech has no independent antecedent outside the fictional world itself (Short et al., 2002, p. 343).

Assuming direct speech categories in Indonesian popular fiction shares similar formal features as those in English fiction (e.g., reporting frame, graphological marking in DS), one would expect that the occurrences of DS and FDS in Teenlit are also relatively high compared to non-popular fiction. At this preliminary stage, my aim is modest, namely to determine the degree of likeness to speech, or conversationality, in Teenlit by examining the occurrences of DS and FDS compared to  the other categories of speech/thought presentation. As such, we are not concerned with comparing “popular” with “serious” fiction. By focusing on the degree of similarity between the language in Teenlit and informal conversation, one could then argue that Gunawan’s claim can be understood relative to the degree to which the reader is afforded access to the speech and thought of the characters. I suggest that adolescent interaction in the novels tends to be presented in ways that create a sense of immediacy, drama, and involvement, and DS and FDS help to produce this. The many colloquial expressions used by the authors in DS and FDS is what has created an impression that the language in Teenlit is similar to adolescent speech.

The analysis presented here is based on approximately the first 3000 words of each novel, and includes narration and dialogue. The data are coded manually to determine the number of occurrences for each of the speech/thought/writing presentation categories. To determine the distribution of the different categories, graphological sentences (Short, 2014) are counted and categorized. DS and DT presentations are distinguished between those that are followed or preceded by a reporting frame (coded Narrative Presentation of Speech [NPS]) and those that are not, to obtain an indication of the narrator’s control of speech and thought. Following Semino and Short (2004), I code reporting frames separately from the reported material (i.e., the quoted speech). Separating the reporting frame from the speech allows us to group reporting frames and clauses that function similarly to reporting frames but are not grammatically constructed in the same way. For example, the emboldened materials in (23) and (24) are not, strictly speaking, reporting frames but they function similarly to reporting frames in introducing the speech that follows (the underlined material).

(23)	Febi berkhotbah tentang batas yang jelas antara cowok dan cewek (Kinasih, 2007, p. 41).
	‘Febi preached about the clear distinction between boys and girls.’

(24)	Alasannya, berselingkuh sudah menjadi kodrat cowok (Kinasih, 2007, p. 41).
	‘Her reason is, having an affair is men’s nature.’

In these examples, the frame-like materials are categorized as NPS, and the following material as IS. For indirect thought presentation, the corresponding categories, Narrative Presentation of Thought (NPT) and IT, are used.⁴

In addition to NPS and NPT, the category, Narration of Internal States (NI), is used to account for instances in which the narrator “reports a character’s cognitive and emotional experiences without presenting any specific thoughts” (Semino & Short, 2004, p. 46). As we saw earlier, in example (17), reproduced below as (25) for convenience, the narrator, who is also the main character, is describing her internal state; the propositional content is given following the NI, in DT. The emboldened two sentences are therefore coded as NI, while the following sentence is coded as DT. Although NI, by virtue of its being the narrator’s voice, affords us less immediate access to the character’s thought compared with DT, it helps convey the character’s inner state. Compare this, for example, with the “pure” narration in (26), which tells us about the character’s actions but not their cognitive state.

(25)	Aku rindu Dilan. Aku hanya merasa begitu kosong.
	“Dilan … kamu di mana?” (Baiq, 2015, p. 139).
	‘I miss Dilan. I just feel so empty.’
	‘“Dilan … where are you?”’

(26)	Diraihnya setumpuk buku dari atas meja, lalu sekuat tenaga dilemparnya ke dinding (Kinasih, 2007, p. 20).
	‘He grabbed a pile of books on the table, and with all his might threw them to the wall.’

The distribution of the different categories in the data is shown in Table 1. In the categories related to the narrator’s presentation of the characters’ speech and thought, the following categories are added for Dilan: Narrator’s Presentation of Writing (NPW) and Direct Writing (DW). NPW accounts for the narrator’s citation of a character’s thought rendered in writing, while DW is the writing cited, and unaccompanied by the narrator’s framing. In Dilan, the narrator (Milea), who is also one of the main characters, cites the writing she found in her textbook (written by Dilan, her boyfriend).

Table 1 Occurrences of speech and thought presentation in the data (highlighted boxes are the “norm” for speech presentation and thought presentation, respectively)

Table 1 shows that DS and FDS have the highest percentages of occurrence in the data, and higher than the percentages for narration (N). If we consider that the data are taken from the first 3000 words of each novel, and the beginning part of a popular novel generally contains background material pertaining to the spatio-temporal setting of the story and preliminary introduction to the main character(s), the high percentages of DS and FDS suggest that these types of speech presentations occur early in both novels. In fact, the first instance of DS occurs as the 20th and 24th sentences in Still and Dilan, respectively.

DS and FDS sentences in the data occur within speech turns that stretch over several turns, the longest consisting of 15 turns and two NRS and was found in Dilan, as shown in (27). In this stretch of DS/FDS turns, the reader is given access to a talk between Dilan and Milea, the protagonists in the novel. These turns do not move the story temporally but serve the affective function of indexing intimacy; they invite the reader to listen in and empathise with the two characters as they build intimacy through talk.

(27)	Longest stretch of DS & FDS (from Baiq, 2015, pp. 20–21).
1	“Si Dadang, kau tau si Dadang gak?” tanya Dilan ‘“Dadang, do you know Dadang?” asked Dilan’
2	“Dadang mana?” ‘“Which Dadang?”’
3	“Gak tau, ya?” ‘“So you don’t know?”’
4	“Enggak,” kataku ‘“No, I don’t” I said’
5	“Kok, sama, ya? Aku juga gak tau.” ‘“Wow that’s like me. I don’t know (him) either.”’
6	“Ih!” ‘“Honestly!”’
7	“Kalau si Guntur?” ‘“What about Guntur?”’
8	“Kamu gak tau juga?” kutanya balik ‘“You don’t know (him) either?” I asked in return’
9	“Itu teman sekelasmu.” ‘“That’s your classmate.”’
10	“Oh, iya. Kenapa dia?” ‘“Oh, that’s right. What about him?”’
11	“Dia itu diam, bukan karena baik.” ‘“He’s quiet, not because he’s a good person.”’
12	“Karena apa?” ‘“Because of what?”’
13	“Karena, gak berani. Karena, takut. Gak siap dimarahin.” ‘“Because he’s spineless. Because he’s scared. Doesn’t want to be grounded.”’
14	Aku diam ‘I didn’t respond’
15	“Harusnya, dia juga dimarah karena penakut. Dunia butuh orang pemberani. Yes?” ‘“He should be grounded for being a scaredy-cat. The world needs brave people. Yes?”’
16	Aku diam ‘I didn’t respond’
17	“Kamu pikir bandel itu gampang? Susah. Harus tanggung jawab sama yang dia udah perbuat,” kata Dilan lagi ‘“You think being a rebel is easy? It’s hard. You have to take responsibility for what you’ve done,” said Dilan further’

Table 2 shows that DS and FDS occur in 71 speech turns in Still and in 114 speech turns in Dilan. In both novels, the number of speech turns that do not have a reporting frame (FDS) is 1.5 times higher than those that do (DS). This suggests that most speech turns in the data are presented without the narrator’s intervention. This frameless speech presentation technique creates vividness and a greater sense of immediacy compared with other techniques in which the narrator’s presence is more visible.

Table 2 Distribution of DS and FDS in the data

The quantitative results for DS and FDS in this study support the findings by Semino and Short (2004) that these speech presentation types are the most common in fiction. Indeed, among the categories for speech presentations, Leech and Short (2007, p. 276) identify DS as the “norm,” pointing out that it “represents speech in the form in which it is directly manifested to a listener,” while the indirect modes (IS and FIS) provide “the substance” of what someone said without commitment to her/his words.

In terms of thought presentation, Leech and Short point out that IT is the norm. If we look at the Indonesian data in Table 1above, IT and FIT are indeed more common than their direct mode counterparts, DT and FDT. Unlike speech, someone’s thought is not directly perceptible to others and so cannot be faithfully reproduced in the way speech can (“faithful” in the sense of Short et al., 2002). It seems then, that the norms for speech and thought presentation in English texts, which are DS and FDS for speech presentation and IT and FIT for thought presentation, respectively, also seem to apply to the two novels considered here.

That DS and FDS are the most common types of speech presentation in the Indonesian data may be taken as empirical evidence supporting Gunawan’s claim that Teenlit style is “speech-like.” If we consider that these speech presentation categories are also norms for speech presentation in English fiction, particularly in popular fiction, we could then say that Indonesian Teenlit authors are not unique in preferring those techniques.

Conclusion

This study has considered the claim by F. X. Gunawan that the language in Teenlit is “basically spoken language.” Drawing on the categories of speech and thought presentation discussed by Leech and Short (2007) and Semino and Short (2004), I have shown that in the two novels examined, DS and FDS are the most frequently used techniques for presenting the speech of adolescent characters. The high degree of occurrences of these types of speech presentation in the data may have led observers like Gunawan to conclude that the writing style in Teenlit is conversational. Conversationality in fiction, as Toolan (2010, p. 327) points out, may be produced by means of various techniques, including repetition. This chapter has focused on one of those techniques, namely the presentation of speech and thought.

Scholars of narratology (see Nielsen, 2004, 2018; Nielsen et al., 2015) have strongly argued that fictional discourse is invented discourse and that it is a mistake to confuse real-world narrative situations with fictional narrative. Gunawan’s remark that the language in Teenlit is similar to language used in real-life conversation confuses fictional interaction with real-life interaction. The fact that the conversation among the characters in Teenlit has captured the attention of readers and critics alike can be understood as less of an evidence of its close resemblance to real-life conversation and more of an attestation to the authors’ relative success in using techniques of speech and thought presentation to create vividness and a sense of immediacy in fictional adolescent interaction.