Textual construction of the hearing-impaired students’ recount texts: A case of students with special needs in writing to mean

Being cohesive in writing for the hearing-impaired students is very crucial, serving as an alternative means of expressing ideas without using sign language in communicating with people in general. This study aims to analyze the textual structure of the hearing-impaired students’ recount texts to instantiate how they structure language in communication. This study focused on investigating the choice of theme and thematic progression patterns as an instantiation of making meanings among the hearing-impaired students as a minority group of students with special needs. The data were 36 recount texts of high and low achiever students collected from three educational levels: elementary, junior high, and senior high schools. Following a qualitative approach to data collection and analysis, this study relied on Systemic Functional Linguistics as the main conceptual framework for investigating types of theme and thematic progression patterns, and how these two elements showed the features of writing to mean among the hearing-impaired students. In terms of theme choices, the findings indicate that low achiever hearing-impaired students employed more marked themes with a lack of ability in writing to mean. In terms of thematic progression, theme reiteration pattern was the most frequently employed pattern, indicating a strong intention to the topic. The study informs that the textual description of the hearing-impaired students’ writings could help the teachers understand what and how to teach to improve their writing to mean.


INTRODUCTION
Writing is a fundamental skill to learn as a means of expressing meaning. For this reason, writing skill is important to acquire including by people with special needs such as students with hearing loss and hard of hearing (Aram et al., 2015;Sugaya et al., 2019). It is reported that over 5 % of the world population has problems in their hearing organs resulting in hearing disabilities (World Health Organization, 2020). Since many people with hearing loss or other hearing predicaments have limitations in spoken communication, they need to learn to write in order to communicate with the hearing people, who mostly do not understand sign language.
Hearing loss or impairment, however, has an impact on language development and speech, Many studies have been conducted to search for best practices in providing support for hearingimpaired students to write as a potential means of communication at the level of pedagogical practices (e.g., Karasu, 2014;Sirois et al., 2008;Strassman & Schirmer, 2012) and planning (e.g., Sharifi et al., 2016;Sharifi et al., 2017;Xu, 2018). Those studies have indicated urgent steps to take to support hearing-impaired students' writing development, which offers the potentials for their mental and cognitive development and their better quality of life despite some challenges (Sugaya et al., 2019). This study attempted to add to the existing studies on investigating current practices of writing among hearing-impaired students as the basis for developing pedagogical plans and practices in the teaching of writing. Some studies have indicated the distinguished linguistic features of hearing-impaired students in terms of their cohesion in writing and grammatical features in general (e.g., Crossley, 2020;Mang'oka et al., 2019;Samadian & Mohseni, 2019). While those studies offer some structural bases for supporting hearing-impaired students, this study specifically focused on analyzing the interrelation of words, phrases, and clauses configured in the concept of theme in recount texts to instantiate the extent to which they manage to write more meaningful texts by following the framework of organizational text analysis (e.g., Butt et al., 2000;Danes, 1974;Eggins, 2004;Gunawan & Aziza, 2017;Zhang & Lee, 2019). In the analysis of textual realization in the texts, this study paid attention to attention to the construction of theme, thematic progression, and the semantic interweaving of the two elements to build the cohesion of the texts.

Theme system in a text
In the field of Systemic Functional Linguistics (SFL), theme system is the term to explicate the realization of the textual metafunction, which has an important role in constructing the flow of a discourse (Halliday, 1994). According to Halliday, theme is an element that comes first in the clause, representing the point of departure of a message or the starting point for the message that informs what the clause is going to be about (Halliday, 1994). In theme system, a clause also consists of rheme as the rest element of the clause, which usually contains new information. Rheme explains what the clause is about, that is, rheme adds new information to what theme has been introduced before (Halliday, 1994). From Halliday's Systemic Functional Linguistics, the interaction of theme and rheme governs how the information in a text develops, including the construction of information in a clause, how people structure information in a clause. Those matter to text development. To contribute to the development of a text, there are three types of theme that can be identified according to the three strands of meaning in SFL, which are ideational (topical), interpersonal, and textual (Eggins, 2004;Halliday, 1994;Halliday & Matthiessen, 2004) (see Table 1).

Table 1
Types of Theme (Halliday, 1994)  The interconnectedness of the theme types, according to Halliday (1994), is seen from the process of meaning-maki realization in which if a conjunctive, modal or relative adjunct or conjunction is present in a clause, they will make an interpersonal theme (modal adjunct, vocative) or textual theme (all the others), while topical theme, the most important one, comprises only one experiential element and ends the thematic segment of a clause. In addition, the theme of a clause extends to and includes the topical theme. Thus, the elements which come before the topical theme are also thematic, but the elements which come after the topical theme are not thematic (Butt et al., 2000;Eggins, 2004;Knapp & Watkins, 2005).
A cohesive text is seen from how the information in the clauses unfolds. As they unfold, they connect to the themes and rhemes of preceding clauses in various ways, picking up or repeating the important concepts and developing them further; these connections form patterns of thematic progression (Danes, 1974). Eggins (2004) defines thematic progression as the flow of information between successive themes and rhemes in a text. Eggins (2004) identifies that there are three patterns of thematic progression, which are reiteration pattern, zig-zag pattern or simple linear progression, and multiple theme pattern. The reiteration pattern occurs when the theme of the first clause is picked up and repeated as the theme of the following clause, signaling that each clause will have something to say about the theme (Paltridge, 2006). The zig-zag pattern is when an element introduced in the rheme in clause 1 is picked up to be the theme of clause 2 (Eggins, 2004). Multiple theme pattern occurs when a rheme includes a number of different pieces of information, and then each information may be taken up as the theme in a number of subsequent clauses (Paltridge, 2006).

METHOD
This study followed a qualitative method involving textual analysis to describe, categorize, and interpret the data by exploring theme and theme progressions in the hearing-impaired students' recount texts. This study deployed the Systemic Functional Linguistic framework (see Halliday, 1994), focusing on the textual metafunction to acquire a holistic description in a particular situation or circumstance (e.g., Fraenkel et al., 2012;Gunawan & Aziza, 2017;Zhang & Lee, 2019). This study was conducted in a special school for hearing-impaired students in one major city in Indonesia. The accessibility of the school in terms of location and permission from the principals enabled us to conduct this study. Despite the granted permission from the principals, we provided informed consent for the students and the teacher. The teacher assisted the students in understanding the information written in the informed consent and told the students that they had the right to withdraw from participating in our study at any time. The schools also showed available resources that met the purpose of this study, that is, to understand the meaning-making process of the texts written by the hearing-impaired students. The school also provided education for all educational levels, from kindergarten to senior high schools, making the data collection possible for comparing recount texts written by the hearing-impaired students from elementary to senior high s school in the same site. This site led to our judgment, in the beginning, to believe that the site provided all the needed data for this study (Fraenkel et al., 2012). In addition to being differentiated based on the educational level, the students were also distinguished based on their performance in writing, categorized into low achievers and high achievers. The decision to categorize low and high achiever students was made by grading their writing. The teacher provided the data of the low and high achiever students.
The data were collected from 36 recount texts written in the Indonesian language by the hearing-impaired students at three educational levels: elementary school (ES), junior high school (JHS), and senior high school (SHS). The recount texts were also distinguished by whether the writers were categorized into high achiever or low achiever students. There were 12 recount texts from elementary school students (three from low achiever 4th grader students, three from high achiever 4th grader students, three from low achiever 5th grader students, and three from high achiever 5th grader students), 12 recount texts from junior high school students (three from low achiever 7th grader students, three from high achiever 7th grader students, three from low achiever 8th grader students, and three from high achiever 8th grader students), and the rest 12 recount texts from senior high school students (six from low achiever 10th grader students and six from high achiever 10th grader students).
Each student was asked to write a recount text based on a series of pictures given. The teacher in each class guided the writing activities by frequently using sign language to help them understand the instruction. The picture series was adopted from the book "Aku Bisa Bahasa Indonesia" by Nurhayati (2009). There were nine related pictures for the participants to arrange into a story about a family holiday to a beach. The study used picture series as a research instrument since the hearing-impaired students were believed to be visual learners (Oyewuni, 2008) and have the advantage of visualspatial ability (Marschark et al., 2013).
There were several steps of data analysis. The first step was reading the recount texts closely. Then, the recount texts were broken down into clauses. The clauses were then placed into a table for an initial analysis. The fourth step was determining the theme and rheme following the analytical framework proposed by Halliday (1994). The sixth step was classifying the thematic progression pattern by using thematic progression theory proposed by Eggins (2004), which could be investigated in terms of reiteration, zig-zag, or multiple themes. The last step was discussing how the choice of theme and thematic progression patterns support the cohesion of the texts.

FINDINGS AND DISCUSSION Types of themes in the hearing-impaired students' recount texts
From three types of theme, namely topical theme, interpersonal theme, and textual theme, the analysis shows that all three types of theme were used by the hearing-impaired students in writing their recount texts. However, the interpersonal theme can only be found in senior high school students' recount texts (see Table 2). The table above shows that unmarked topical theme is the most frequent type of theme that occurred in the hearing-impaired students' recount texts (73.13%), followed by marked topical theme (17.52%), then textual theme which appears 38 times (8.88%), and the interpersonal theme which only appears five times (0.47%). The findings show the commonality of themes written in the genre of recount texts.

Topical themes in the hearing-impaired students' recount texts
The finding shows that topical theme is the most frequently used theme in the hearing-impaired students' recount texts. This phenomenon is not a surprise since a topical theme is an element that refers to participants, circumstances, or processes (Arunsirot, 2013), which are usually contained in a clause. A topical theme plays a prominent role in the development of a clause since it occupies a transitivity role (Eggins, 2004).
The analysis shows that an unmarked topical theme was the most dominant theme found in the hearing-impaired students' recount texts, indicating an intention to propose the topic under discussion. As seen in Table 3, the tendency in the use of unmarked topical theme in students' recount texts indicates that high achiever students used more unmarked topical themes than low achiever students. The grammatical awareness of high achiever students in this case related to their understanding of how to make sentences in a simple way, that is, involving common ways of beginning sentences with a topic (Arunsirot, 2013), which then contributes to the cohesion of the text. Most of the unmarked topical themes in both high achiever and low achiever elementary school students' recount texts were realized by mentioning family members. The following are examples of clauses taken from low achiever student's recount texts at the elementary school level. Extract 1. An instance of clauses written by an elementary school high achiever student Ayah, ibu, kakak dan adik "Dad, mom, older brother, and younger brother" akan membawa baju renang "will bring their swimsuits" Unmarked Top. Theme Rheme

Extract 2. An instance of clauses written by an elementary school low achiever student
Ayah mamah adik kakak anak ayah "Dad Mom younger brother older brother dad" naik mobil "go by car" Unmarked Top. Theme Rheme In junior high school, the unmarked topical themes in high achiever students were mostly realized as keluarga "family." It could be an indication that the high achiever students had already obtained more vocabulary than low achiever students. Instead of mentioning family members one by one, they used the word keluarga "family" to make it more effective. However, the low achiever students realized their unmarked topical theme through mentioning family members as what elementary school students did in communicating. This phenomenon indicates that their mastery of vocabulary was less than high achiever students. The examples below are the realization of unmarked topical themes higher and lower achiever students' recount texts, respectively, at the junior high school level.
Extract 3. Unmarked topical theme written by a higher achiever junior high school student Keluarga "family" mau berangkat ke pangandaran naik mobil jarak jauh "is going to Pangandaran by car a long way" Unmarked Top. Theme Rheme Extract 4. Unmarked topical theme written by a lower achiever junior high school student Bapak dan ibu, adik "Dad and Mom, younger brother" pergi akan mobil ke panti "go will by car to the house" Unmarked Top. Theme Rheme In senior high school, similar to junior high school students, the high achiever students realized their unmarked topical theme as keluarga "family." However, the unmarked topical themes in low achiever students' recount texts were mostly realized as orang "person," signaling that the students wrote a random topic of the texts by referring to random people as orang "people." Below are the examples showing the realization of the unmarked topical theme in higher and lower achiever students' recount texts, respectively, in senior high school level.
Extract 5. Unmarked topical theme written by a higher achiever senior high school student Keluarga saya "myfamily" mau pergi ke pangandaran untuk bermain bersama-sama di pantai "want to go to Pangandaran to play together on the beach" Unmarked Top. Theme Rheme Extract 6. Unmarked topical theme written by a lower achiever student Orang "a person" mau pergi ke pantai "wants to go to the beach" Unmarked Top. Theme Rheme Regarding the marked topical theme, Wang (2007) believes that the use of a more marked topical theme is a typicality of an academic text since it indicates the writer's ability to produce more dynamic text. However, in this study, the hearingimpaired students used marked topical theme due to their lack of ability to arrange words to make more grammatical sentences. As can be seen in Table 3, the low achiever students made more marked topical themes in writing recount texts than high achiever students. Lack of access to making meanings at the sentence level among the low achievers was the key to understanding this phenomenon. The following is an example of a marked topical theme written by higher achiever students at the elementary school level.
Extract 7. Marked topical theme written by a higher achiever student Di pantai "on the beach" mamah, adik, kakak, bapak main air "mom, younger brother, older brother, dad play water"

Marked Top. Theme Rheme
Higher achiever students made meanings more appropriately with marked themes. Here, the high achiever student, for example, used prepositional phrases as a marked topical theme. Marked topical themes were constructed by adverbial, prepositional phrase, or complement at the beginning of a sentence, which means that the use of prepositional phrase di pantai "on the beach" as instantiated in the clause above is common and acceptable. However, this pattern did not really unfold in texts written by the other groups of the hearing-impaired students in this study, such as the following written by a low achiever student at the elementary school level.
Extract 8. Marked topical theme written by a lower achiever student Sudah pergi "has gone" rumah mobil malam "house car night"

Marked Top. Theme Rheme
From the examples above, the low achiever students tended to use a predicator as a topical theme in a declarative clause. It signals that low achiever students had a problem in structuring a grammatically correct sentence, which, therefore, may be related to the argument that the vocabulary and the grammar of the hearing-impaired students are inevitably low (Cannon & Kirby, 2013;Takahashi et al., 2016;Zamani et al., 2018). The use of marked topical themes may attract someone's attention for their distinguished intention in sentences. In this case, this argument may be true, especially in low achiever students' recount texts. People normally do not expect to find a clause arranged in this structure because about 86.5% of languages in the world possess the system of either SVO (Subject Verb Object) or SOV (Subject Object Verb) (Tomlin, 1986). The Indonesian language follows SVO pattern, which is similar to English. In the clause above, low achiever students used VOS pattern, which is extremely rare; only 9% of languages use this pattern (Tomlin, 1986).
When marked topical themes are used, there are at least two purposes (see Butt et al., 2000). It may be either to draw readers' attention to a particular group or phrase or to build coherent text to make it easy to follow. In some cases, it is true that the use of marked topical themes by high achiever students is to emphasize some words in a clause, such as di pantai "on the beach" in the clause above. However, low achiever students did not achieve those purposes by using marked topical theme. They made an effort to emphasize some words in their recount texts but ended up confusing readers as they emphasized words without paying attention to the sentence structure.
In short, the analysis shows that marked topical themes in the hearing-impaired students' texts were not generally designed to emphasize the focus but were constructed due to their inability to arrange more grammatical sentences. The higher their level of education, the more capable they control the text. For example, the senior high school hearing-impaired students were able to use marked theme appropriately. On the other hand, the dominant use of unmarked themes, which were common in recount texts, were interrupted by choice of general and specific participants such as family (in high achiever students' texts) and person (in low achiever students' texts) to address the participants involved. Inappropriate choices of words were due to their lack of vocabulary knowledge.

Textual theme in the hearing-impaired students' recount texts
Textual theme is another feature of how cohesion is built within the hearing-impaired students' recount texts. This study indicates that textual theme is the second type of theme which occurs the most in the students' recount texts. A textual theme functions to connect a clause to its neighboring clause. It can be in the form of continuatives, conjunctions, and conjunctive adjuncts (Eggins, 2004). As can be seen in Table 4, there were 37 textual themes found in students' recount texts. Conjunction was the most frequently used textual theme in the hearing-impaired students' recount texts (54.06%), followed by conjunctive adjunct (43.24%), while continuative was the least frequently used theme (2.70%). This finding implies that the students had little knowledge about writing a good text due to the typical use of conjunctions as a clause chaining strategy in spontaneous spoken language for written language (Muddhi & Hussein, 2014;Schleppegrell, 1996), for example, in clauses written by a high achiever student of senior high school level.
The use of tapi "but" to link two clauses was not used as adversative conjunction to express a contrastive relation. Meanwhile, Extract 10 shows the use of conjunctive adjunct in a senior high school student's recount text. Extract 11 shows a clause using continuative, which was quite rare in general communication since the number of continuative is not more than conjunctions and conjunctive adjuncts (Muddhi & Hussein, 2014 Extract 11. Continuative in a senior high school student's recount text. Oh "Oh" baru melihat "Have just seen" pantai Bali indah "Beautiful Bali beach" Continuative Topical Rheme Textual "Oh," a continuative, here signaled a new move in the text, and it is usually at the beginning of the clause (Knapp & Watkins, 2005). The occurrences of textual themes in most of senior high school students' recount texts indicate that the students were aware of the important role of textual theme in creating a cohesive work in relating the clause to its context. Such an example shows a rather meaningful use of textual theme.
However, some textual themes were used in their recount texts inappropriately and unclearly. It can be found in the clauses: Terus keluarga sedang duduk sambil melihat di depan laut. Sehingga keluarga sedang melihat sama anaknya dan anaknya main pasir "then family was sitting while looking at the front beach, so that the family is watching their children playing with sand." The use of conjunctive adjunct terus "then" was to signal a new move. However, in the second clause, the use of sehingga "so that" was inappropriate since there was no causal relation in the clauses. The use of dan "and" in the sentence was also inappropriate and ineffective to mean a causal relation with an additive function. To make it more cohesive, the conjunction dan "and" could be changed into relative pronoun yang "who" in English to introduce a new part of a clause that further describes people, as previously mentioned.
Similar to the findings in terms of the topical theme, despite inappropriateness, the hearingimpaired students of the higher level of education tended to use more textual themes; they showed awareness of how meanings should be semantically woven. However, the lack of exposure to written language and disability to engage in spoken language was the main reason for having no modeling of using language more appropriately.

Interpersonal theme in the hearing-impaired students' recount texts
The analysis shows that interpersonal theme was the least employed theme in the hearing-impaired students' recount texts by only five times of occurrences. An interpersonal theme deals with the interpersonal meaning of a clause. Interpersonal theme is used to indicate the kind of interaction between speakers or the positions which they are taking (Butt et al., 2000;Matthiessen, 1995). Interpersonal themes could be in the form of modal adjuncts, vocatives, finite, or wh-elements (Eggins, 2004). There were five interpersonal themes found in the students' recount texts (see Table 2 In the example above, taken from a high achiever student's recount text in senior high school level, kenapa "why" was used as interpersonal theme, which functions to set up the mood of a clause (see Eggins, 2004). Another example was taken from a high achiever student's recount text in senior high school level.
Extract 13. Interpersonal theme written by a high achiever student Alhamdulillah "thanks God" bapak saya "my dad" membantu bawa sana di pasir pantai "help bring there on the beach" Interpersonal Topical Rheme Theme The student used "Alhamdulillah" which functions to share a point of view (See Emilia, 2005), expressing a religious viewpoint.
The least number of interpersonal theme indicates that the students rarely used modulation and modalization in their recount texts (see Emilia, 2005). The low number of interpersonal theme in the students' recount texts might be due to the fact that commonly interpersonal theme is used to manage spoken interaction or to manage the degree of truth in expressing opinions (see Butt et al., 2000). All interpersonal themes were used by high achiever senior high school students. There was an indication that they could manage a stance on the issue raised in the text by expressing questions and a point of view to maintain social interaction with readers. Similar to the findings of topical and textual theme in the hearing-impaired students' recount texts, the higher the students' level of education, the more dynamic the language used in the texts. Again, lack of an appropriate model of language use in certain texts has caused them to struggle in using more varieties of language, for example, more dynamic and controlled themes in recount texts.
Thematic progression in the hearing-impaired students' recount texts As shown in Table 5, of the three thematic progression patterns, namely, reiteration pattern, zigzag pattern, and multiple theme pattern as identified in other studies (e.g., Gunawan & Aziza, 2017 for academic writings), two of the three patterns occurred in the students' recount texts, which were reiteration and zig-zag pattern. The reiteration pattern was the most dominant pattern used by the students in writing a recount text (90%), followed by the zigzag pattern (10%).

More reiteration patterns
Reiteration pattern is the most dominant pattern employed by the hearing-impaired students in their recount texts. Reiteration pattern occurs when the theme of the first clause is picked up and repeated as the theme of the following clause, signaling that each clause has something to say about the theme (Paltridge, 2006). In terms of students' level, the reiteration pattern was mostly found in high achiever junior high school students' recount texts. The high frequency of reiteration pattern indicates the students' ability to achieve cohesion by maintaining the focus of the texts by allowing the thematic element in a clause to be the theme in subsequent clauses (see Eggins, 2004). As can be seen from Table 6, high achiever students at all educational levels used more reiteration patterns than low achiever students. The use of this pattern could be building a focus in their recount texts or putting forward the point discussed in their recount texts as central attention (Gunawan & Aziza, 2017), as can be seen in Extract 14.  From the example above, taken from high achiever elementary school students' recount text, the reiteration pattern was constructed by picking up the theme in the first clause to be the theme in the second and third clause, which is Ayah, ibu, kakak, dan adik "dad, mom, older brother, and younger brother." In other words, the focus of information in the example above is on Ayah, ibu, kakak dan adik. Here, a strong focus of the texts was maintained through the thematic elements. There was an effort in the recount in establishing a clear focus of the text and making it well-organized so that it is more understandable due to its cohesiveness. Theme reiteration pattern allows the readers to have a relatively easier to process information as it could lead the readers to devote greater attention to the information being conveyed in the rheme (Nwogu, 1989;Mang'oka et al., 2019).
The phenomenon where reiteration pattern dominates a recount text might not be surprising since theme reiteration pattern frequently occurs in narratives (Wang, 2007). Recount texts share some similarities with narratives, either in their generic structure or their lexico-grammatical features (Eggins, 2004). This is also in line with other studies (e.g., Nwogu & Bloor, 1991;Mang'oka et al., 2019), indicating that generally, most English texts employ theme reiteration patterns as the most dominant thematic shifts. Theme reiteration pattern is a basic theme pattern in general (Mang'oka et al., 2019). The hearing-impaired students make an effort to create their texts understandable by employing more reiteration patterns. However, the pattern was overused and lack of further development of rhemes because the writers did not expand the information introduced in the rhemes. In effect, the text was too repetitive.

Less zig-zag patterns
Zig-zag pattern was the second most employed pattern in the hearing-impaired students' recount texts. Zig-zag pattern happens when an element introduced in the rheme in clause 1 is picked up to be the theme of clause 2 (Eggins, 2004). A zig-zag pattern was far fewer employed by the students in developing their texts (see Table 7) As can be observed from the trend, a zig-zag pattern can only be found in recount texts of high achiever students of elementary school, low achiever students of junior high school, and high achiever students of senior high school. In terms of students' level, the zig-zag pattern was mostly found in high achiever elementary school students' recount texts. For example, Extract 15. Zig-zag pattern written by a high achiever elementary school student

Clause 1
Th1 Keluarga "family" Rh1 hari ini liburan naik mobil. "today holiday by going on a car" Clause 2 Th2 (=Rh1) Mobil "a car" Rh2 jalan cepat sampai di pantai. "quickly arrived on the beach" Clause 3 Th3 (=Rh2) Di pantai "on the beach" Rh3 keluarga lihat laut indah. "family saw a beautiful beach" As can be observed from the example above taken from high achiever senior high school students' recount text, the student used the element in the rheme to be the theme in the next clause. Mobil "car" in the example above, which was previously positioned as the rheme was picked up to be the theme in clause 2, making a zig-zag pattern. Di pantai "on the beach," which was also the rheme in clause 2 was reused as the theme in clause 3, also making a zig-zag pattern. Here, the student was successful in introducing new information based on the theme introduced in the previous clause. This pattern might contain more information and might seem to be less boring than the theme reiteration pattern since the rheme was developed and added more information to the text.
The low frequency of zig-zag patterns in the hearing-impaired students' recount texts might be due to the genre of the text, a recount text, demanding more reiteration pattern than the zig-zag one. The zigzag pattern is more complicated for the students in writing their recount texts since the zig-zag pattern requires larger amounts of information (Dong et al., 2016).
Finally, multiple theme was not found in the students' recount texts. A multiple theme pattern makes cohesion of a text-based on the development of different pieces of information contained in the first theme to be the theme in a number of subsequent clauses (Paltridge, 2006). However, the use of multiple theme pattern might not be cohesive in writing a recount text, as Nwogu (1989) found that instances of multiple theme progression were mostly found in the highly specialized texts.
The findings of this study have responded to other previous studies on the levels of language competence and language elements. Similar to the finding of Mang'oka et al. (2019) on lexical devices among hearing-impaired students, the students tended to use reiteration theme in their writing by mentioning viewable participants such as school, person, family, students, and teachers. The analysis is crucial to help the teachers of the hearing-impaired students develop their language that leads to their life trajectory in the future (Kilpatrick & Wolbers, 2020). Adding to these previous studies, this study has specifically offered how to construct a text in an acceptable textual manner by exploring more language elements on theme and theme progressions and less attention to segmented structural elements of language to yield on a description of meaning processes. The findings of this study point to the crucial role of reading and writing to allow them to have enough exposure to the appropriate use of language in reading and writing to mean (see Aram et al., 2015;Sugaya et al., 2019). The focus on reading and writing may open up the opportunities for hearing-impaired students to maximize their visual learning privilege (Kuntze et al., 2014), which is significant to provide more tools to support their communication skills and cognitive development.

CONCLUSION
Based on the findings of theme and thematic progression analysis, several conclusions can be drawn from this study. Regarding the types of theme, the study found that there were some purposes behind the choice of theme employed by the hearingimpaired students in organizing their ideas into recount texts. Unmarked topical theme was used by both high achiever and low achiever students to make grammatically correct sentences. Marked topical theme was used by high achiever students to highlight some elements in a clause, and used by low achiever students in less grammatically correct sentences. Interpersonal theme was only employed by high achiever students, and it was used to express a point of view and to raise questions regarding the topic in their recount texts. Textual theme was mostly employed by high achiever students, which indicates that they were aware of the need to be cohesive so that they made a connection between clauses by using either conjunctions, conjunctive adjuncts, or continuatives.
In terms of the thematic progression pattern, reiteration was the most dominant pattern that occurred in the hearing-impaired students' recount texts, followed by zig-zag pattern. Reiteration pattern was employed by the students as they attempted to make a clear focus on the topic for discussion in the recount texts. Zig-zag pattern was used by the students to develop more information based on the rheme they introduced.
These findings indicate that there was a tendency to communicate their past experiences in the recount texts to be clear to others. However, their being disabled in hearing affected the lack of access to spoken language. On the other hand, there was evidence that higher achiever students constructed sentences in better grammatical orders. It implies that the more they are exposed to written language, the better opportunities to make meanings. Therefore, hearing-impaired students need to be intervened in their language learning by providing support for academic writing, which may potentially affect their language competence in general. The findings of this study thus offer some implications for providing support for the hearing-impaired students in learning to mean like people in general. Academic writing could be more emphasized to provide them samples and exposure to making meaning in general. To support sentence construction in academic writing, the hearing-impaired student can be introduced to theme-rheme based learning samples through which they could practice using language with topical themes and more complex themes as they proceed with a higher level of education.