Original Research

Classroom plays as narrative texts that introduce preservice teachers to artificial intelligence-related vocabulary in Sesotho and isiZulu

Mosa N. Khasu

Received: 20 Oct. 2025; Accepted: 26 May 2026; Published: 16 July 2026

Copyright: © 2026. The Author. Licensee: AOSIS.
This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license (https://creativecommons.org/licenses/by/4.0/).

Abstract

Background: This study explored how the modality of classroom plays, also called playwriting, was used to introduce preservice teachers in a foundation and intermediate phase Bachelor of Education (B.Ed.) programme to artificial intelligence (AI)-related vocabulary that includes coding and robotics language in Sesotho and isiZulu. The study employed Shulman’s theory of signature pedagogy and Vygotsky’s sociocultural theory, focusing on the zone of proximal development, to explore student-teachers’ playwriting activities.

Aim: This study aimed to explore how student-teachers engaged in a playwriting process to create drama texts about AI in Sesotho and isiZulu.

Setting: This qualitative study employed a case study design and took place at a university in urban South Africa. The sample consisted of members of an intact group of 90 student-teachers in their final year of the B.Ed. programme.

Methods: Data were collected through semi-structured focus group interviews with the student-teachers and an analysis of their drama texts. An inductive thematic analysis approach was applied to the data from interviews, while a deductive approach was implemented to analyse the drama texts according to criteria for playwriting with a pedagogical purpose.

Results: Findings from the study show that playwriting is a viable method for teaching vocabulary. Through drama activities, students learned about the value of dramatic performance as a tool for making learning engaging and practical for learners.

Conclusion: Drama activities such as playwriting and meaningful feedback promote active engagement in learning. This approach supports preservice pedagogy for teaching vocabulary explicitly in a manner that is engaging for primary school children.

Contribution: The contribution of this study is the introduction of AI-related terminology into Sesotho and isiZulu. This terminology was developed by preservice teachers in a playwriting course in the B.Ed. programme.

Keywords: artificial intelligence; classroom plays; isiZulu; Sesotho; teaching; vocabulary.

Introduction

Recently, the introduction of the CAPS for Coding and Robotics Foundation phase Grade R – Grade 3, intermediate phase (Grade 4 – Grade 6) and Senior phase (Grade 7 – Grade 9) (Department of Basic Education 2024), which aims to develop science, technology, engineering, arts and mathematics (STEAM) learning. In principle, coding and robotics are now formal subjects in the curriculum aimed at developing children’s computational thinking skills. The purpose of introducing coding and robotics as a school subject is to establish a connection between the 4th (4IR) and 5th (5IR) industrial revolutions (Gorrah & Papers 2024). It remains unclear how universities are preparing preservice teachers to teach coding and robotics, and to what extent this is being done. A study by Gravett, Molaodi and Mofokeng (2025) reports on the introduction of a coding and robotics club in a South African university to formally introduce preservice teachers to coding and robotics. Firstly, preservice teachers valued the development of competencies, especially creativity; secondly, they gained valuable insights into teaching, particularly by using Scratch to enrich lessons. In his course, students are taught how to code digitally, using software such as Scratch and Octostudio. They also learn how to code manually using LEGO^® tool kits. These are technical skills related to coding and robotics that teachers must be prepared for. Bezuidenhout (2021) notes that while these technical skills are important, children must be taught the language of coding and robotics. In her study, she introduced coding and robotics to the dialogue-reading programme, which is a series of books written in English that introduce children to the language of coding and robotics. These books offer a conceptual framework for developing young children’s vocabulary in areas such as artificial intelligence (AI), coding and robotics. The story plots also present opportunities for dramatisation (Khasu 2025; Khasu & Henning 2024). The dialogue-reading books are only accessible in English, and most South African classrooms are multilingual; therefore, preservice teachers must be prepared to teach in these contexts (Mashinini 2024). The recent rollout of Mother Tongue Bilingual-based Education (MTBbE) in South Africa, which is aimed at ensuring that learners grasp complex concepts in their most familiar language while developing proficiency in English, in core subjects such as Mathematics and Natural Sciences and Technology, rather than abruptly switching to English-only instruction in Grade 4 (Mbude 2020). It is against this backdrop that I designed a course on playwriting for final-year students that formed part of one of their assessments on teaching literacy in the African languages. Preservice teachers were tasked with creating classroom plays about artificial intelligence concepts, such as coding and robotics, in Sesotho or isiZulu. The aim was to develop children’s vocabulary of AI-related vocabulary that includes coding and robotics in Sesotho and isiZulu, which can benefit children when they practically code. In South Africa, most Bachelor of Education (B.Ed.) programmes do not prepare students to teach in an African language in the intermediate phase (Grade 4 – Grade 6) of schooling (Mahlambi & Mawela 2021). Although teaching other concepts in the curriculum was not the aim of this study, the findings can suggest how preservice teachers can use classroom plays as a viable pedagogical approach to strengthen children’s understanding of difficult mathematics, science or technology concepts in an African language.

Literature review: Extending vocabulary through drama in the classroom

Fennessey (1995) describes the use of drama as an activity that stimulates the reading of literature in a variety of genres. Drama puts oral language development to use in a purposeful way that develops imagination and critical thinking, collaboration, respect for the way others think, attention and self-discipline (Gardiner & Anderson 2018). Drama enables children to journey into a particular role and to engage in learning in a way that is experienced in the imagination through the actions of another person. These are not skills that can be taught in the traditional way of teaching. Heath (1993) and Heathcote (2009) maintain that drama helps children understand others’ points of view (O’Neill 1993). Drama refers to the playwriting process as well as the rehearsal and practice aspects, leading up to the classroom performance when it morphs into classroom theatre. In this study, only the initial phase of the process of classroom plays has been included, namely, playwriting. Despite different views and types, the term ‘drama in education’ refers to a variety of modalities or ‘pedagogical genres’ of drama as a tool for learning (Anderson 2012; Heathcote 2009). Some of these may be used to stimulate creative writing or as a dramatisation of learning content across the curriculum, including science and mathematics concepts and procedures, and language teaching (Henning 1991). This usually unfolds when a teacher scaffolds learners’ roles within an imagined context but is also achieved by teaching content (Henning 1991). Advocates for drama as a pedagogical tool propose that it enables learners to combine their imagination with learning content and, importantly, take on the role of a ‘character’ (Anderson 2012). Such improvisation would be facilitated by the teacher, who builds on the actions and reactions of the learners while they are acting in a particular role. (Gray & Yang 2015; Pascoe 2014).

Drama activities are a powerful means of teaching language concepts, specifically when introducing new vocabulary (Aram & Mor 2009; Baldwin & Fleming 2003; Scharer 2022). The researchers propose that classroom play engages children in a process in which they not only acquire vocabulary but also provide an opportunity to use oral language (Baldwin & Fleming 2003; Pascoe 2014; Scharer 2022). Gray and Yang (2015) argue that classroom plays also create an opportunity for children to master and develop early academic language in a way that makes language acquisition contextually relevant because they are learning in a role as a character. Shapopi-Shingenge (2023) advises that academic language should be introduced in the early grades. In addition to oral language development, learners also read and write when they write classroom plays, on a much smaller scale, as part of the creative writing curriculum or as plays about the content of a given subject. When preparing for a classroom theatre ‘production’, learners not only read aloud for content but also with intonation, pauses and accentuation, which express meaning. I would argue that the use of classroom plays as pedagogical tools can help children not only practise the cognitive skills of reading and writing but also develop targeted vocabulary (Tunmer & Hoover 2019). The various components of language teaching are outlined in the South African curriculum policy framework. The revised curriculum also includes the inclusion of coding and robotics (Department of Basic Education 2024).

Drama pedagogy and the development of artificial intelligence vocabulary

Vocabulary is commonly conceptualised as the repertoire of words known and used within a particular linguistic system, together with an understanding of their meanings and contextual use (Butler 2019). Within educational and emerging technological domains, vocabulary extends beyond general language knowledge to include specialised and conceptually dense terminology that shapes access to disciplinary knowledge, including foundational ideas in artificial intelligence. Vocabulary knowledge underpins learning, as limited lexical understanding constrains comprehension and the construction of meaning (Hiebert & Kamil 2005; Nagy & Townsend 2012). In this regard, academic and subject-specific vocabulary, including terminology associated with artificial intelligence, is unlikely to develop through everyday interaction alone and typically requires intentional pedagogical intervention (Jalongo & Sobolak 2011; Nagy, Diakidoy & Anderson 1991). This study positions classroom plays as a pedagogical approach that can mediate access to both general and specialised vocabulary. Engagement in drama-based activities enables learners to encounter and use new terminology in context, thereby supporting expressive and receptive language development as well as creative meaning-making (Baker Sennett, Matusov & Rogoff 2007). Such an approach is particularly relevant for introducing children to emerging fields such as artificial intelligence, where abstract concepts require contextualised and developmentally appropriate representation. Although learners enter formal schooling with an existing linguistic base in their primary language, described as Language One (Cummins 2000), they must also acquire the academic register required for schooling and disciplinary engagement (Shapopi-Shingenge 2023). Drama-based pedagogy offers a context in which this transition can be supported through socially mediated interaction and purposeful language use.

Effective vocabulary instruction requires deliberate and sustained opportunities for engagement with words and their meanings (Andersen 2004). Drama texts, such as classroom plays, provide an embodied and interactive environment in which vocabulary is not encountered as decontextualised lists but as part of meaning-making practices in situated contexts (Brown 2017). In this way, learners develop deeper lexical knowledge and strategies for independent word learning. However, vocabulary acquisition remains a complex and gradual process that depends on repetition, contextual variation and opportunities for application (Leung 2008). Assessing vocabulary development, therefore, requires attention not only to the breadth of words known but also to learners’ ability to use these words productively and appropriately (Hiebert & Kamil 2005).

Henderson and James (2018) further indicate that vocabulary learning is influenced by existing knowledge, with significant variation in learners’ ability to acquire new words through exposure to texts such as stories. This highlights the need for carefully designed pedagogical strategies that account for individual differences and optimise conditions for word learning (Houston-Price, Howe & Lintern 2014). Explicit instruction has been identified as particularly effective in supporting vocabulary acquisition, as it enhances both semantic understanding and appropriate usage (Biemiller 2004; Butler 2019; Houston-Price et al. 2014). This perspective aligns with broader discussions on structured instructional approaches, including work on learners with diverse needs (Mason, Otero & Andrews 2022; Yaghmour & Obaidat 2022). Within this framework, classroom plays can be understood not merely as creative activities but also as deliberate, structured interventions that facilitate engagement with both academic and artificial intelligence-related vocabulary. Figure 1 presents the instructional implications for promoting word knowledge.

FIGURE 1: Instructional implication for promoting word knowledge.

Frequent and purposeful exposure to targeted vocabulary increases the likelihood that learners will internalise and retrieve word meanings for use in academic contexts (Alshraideh & Alahmadi 2020). In early literacy instruction, such words are often introduced as sight words, where recognition is prioritised (Nash & Snowling 2006). However, this approach may limit deeper lexical processing if learners are not supported to engage with graphemic and phonemic structures. From an academic and artificial intelligence vocabulary perspective, this limitation is significant, as understanding specialised terminology requires more than visual recognition. Learners must engage with the semantic, morphological and contextual dimensions of terms to use them meaningfully. When target vocabulary is embedded within disciplinary or thematic contexts, learners are better positioned to understand how such terms function within sentences and broader discourse practices (Shapopi-Shingenge 2023). This is particularly relevant for artificial intelligence-related vocabulary, which often consists of abstract and conceptually dense terms that require contextualisation for meaningful uptake. While the identification of key vocabulary in texts, for example, through highlighting or typographical emphasis, can support focused attention, this strategy is most effective when combined with explicit explanation and guided engagement (Biemiller & Boote 2006).

Explicit vocabulary instruction remains central to developing both receptive and productive language competencies. Providing clear definitions, examples and opportunities for application enables learners to build robust lexical networks that support reading, writing and oral communication (Razali & Razali 2013). In addition, continued access to target vocabulary in written form reinforces retention and supports independent use (Nash & Snowling 2006). Within this framework, drama-based pedagogies offer a productive space for integrating explicit instruction with contextualised practice. When preservice teachers design classroom plays that incorporate targeted academic and artificial intelligence vocabulary, they create opportunities for learners to encounter, rehearse and apply specialised terminology in meaningful and socially mediated contexts. Vocabulary knowledge is therefore not only foundational to receptive skills such as reading and listening but also to participation in disciplinary discourse (Brown 2017). The integration of targeted vocabulary within classroom plays provides a pedagogical mechanism for bridging everyday language and specialised domains such as artificial intelligence while simultaneously equipping preservice teachers with adaptable strategies for their professional practice. Conceptually, this study draws on Shulman’s notion of signature pedagogies (Shulman 2005), positioning classroom plays as practices that induct learners into disciplinary ways of thinking and acting, alongside Vygotsky’s sociocultural theory, which emphasises socially mediated learning and the co-construction of meaning through language, interaction and guided participation (Veresov 2017; Vygotsky 1987).

Theoretical framework

The theoretical and conceptual framing of this study is informed by both Shulman’s (1986, 1987) model of pedagogical content knowledge and Vygotsky’s sociocultural theory. Shulman proposes that teaching requires specialised forms of knowledge, including subject content, pedagogy and an understanding of learners. His later work on signature pedagogies (Shulman 2005) extends this perspective by emphasising the distinctive ways in which future professionals are taught to think, perform, and act within their disciplines. In parallel, Vygotsky’s sociocultural theory foregrounds the role of social interaction, language and mediated activity in the development of higher-order thinking, positioning learning as a collaborative and contextually situated process. Within teacher education, these frameworks intersect in meaningful ways. Signature pedagogies can be understood as structured, socially mediated practices through which disciplinary knowledge is constructed and enacted. As observed by Kortjass and Mbatha (2021), teacher education institutions function as spaces where professional identities are cultivated, and where preservice teachers engage with pedagogical practices that reflect the norms of their field. In this study, classroom playwriting is conceptualised as a potential signature pedagogy (Gardiner & Anderson 2018), through which preservice teachers are apprenticed into the practices of integrating drama, language and subject content.

Firstly, I drew from Shulman’s (2005) signature pedagogy framework. He identifies three dimensions of a signature pedagogy, namely surface structure, deep structure and implicit structure as noted in Figure 2. The surface structure refers to observable teaching and learning practices, while the deep structure reflects underlying assumptions about how knowledge is best taught. The implicit structure relates to professional values, dispositions and ethical commitments. In this study, these dimensions are evident in how preservice teachers engaged with playwriting as both a pedagogical strategy and a means of mediating disciplinary knowledge. Their participation in a culture module introduced them to drama, music and dance for curriculum delivery, but the playwriting experience provided a more sustained and situated engagement. From a sociocultural perspective, this process can be understood as mediated learning, where language, collaboration and cultural tools such as drama support the development of new forms of knowledge. The playwriting activity not only enhanced preservice teachers’ practical engagement with teaching through drama but also challenged their assumptions about integrating disciplinary discourse, including artificial intelligence vocabulary, into their pedagogical practices.

FIGURE 2: A signature pedagogy for playwriting in the classroom.

Secondly, I drew from Vygotsky’s sociocultural theory (Veresov 2017), focusing on the zone of proximal development to explore student-teachers’ playwriting activity. I specifically foreground the view of Vygotsky (1987), who introduced his view of children’s play in a multidimensional way. He proposed that play comprises art performance, self-expression, imagination and roleplay. This component of the framework of the study follows the ‘classic’ Vygotskian triangle (Engeström 1999). The classic Vygotskian ‘triangle’ of activity in what Engeström (1999) referred to as ‘first generation activity theory’ (AT). It is a theory that has been used to understand individual behaviour by examining how they interact with their environment. Mediation is a key theoretical idea behind activity because our everyday lived experiences are significantly mediated using tools and symbols. Therefore, activity theory helps to frame our understanding of such semiotic mediation (Veresov 2017). Semiotic mediation is concerned with the cultural mediation of mental development through acts of meaning (Hasan 2004:48). In this study, the lecturer served as the human mediator in the playwriting activity, as highlighted in Figure 3.

FIGURE 3: Vygotsky’s sociocultural theory.

Based on Figure 3, the unit of analysis of the study is the playwriting activity that preservice teachers participated in and how they made meaning of the activity (Engeström 1999¹) of their engagement with the course and the outcome of their work. The leading activity in which they were engaged was writing plays, and the outcome was envisaged to be plays with pedagogical value. In the programme module, which is the backdrop for my study, the preservice teachers were introduced to some of the terms used in the literature about play theory, educational drama, children’s theatre, and so forth. The study thus proceeded with a bifocal (Shulman’s PCK and signature pedagogy) conceptual framework to show the various fields of knowledge and practice related to the pedagogical content knowledge (PCK), specifically regarding the third component of PCK as shown in Figure 3. The system of signs (Sobkin 2017) for the framework of my study follows the ‘classic’ Vygotsky triangle (Engeström 1999).

Research methods and design

In this qualitative case study, I was informed by Yin’s (2018) definition of the various case study types; I employed a descriptive case study design. This was informed by the process of how I interpreted the phenomenon (How do student-teachers engage in the B.Ed. course about classroom playwriting on the topic of AI-related use of language?) and the real-life context in which it occurred. This phenomenon, which is also the unit of analysis, included analysing the participants’ activities and the artefacts that they had designed, using selected methods of analysis. Henning, Van Rensburg and Smit (2004) define the term ‘method’ as a systematic way of performing an activity, whereas ‘methodology’ refers to the:

[C]oherent group of methods that complement one another and that have the goodness of fit to deliver data and findings that will reflect the research question and suit the research purpose. (p. 36)

Data collection

Typical data collection methods in qualitative studies include structured individual interviews, focus group interviews, participant observations and document analysis (Collins 1992) and participant observations (Merriam 1998). The data in this study were gathered using semi-structured focus group interviews that were audio-recorded and transcribed. From the two cohorts of the final year B.Ed. students of 2020 and 2021, 90 students were placed in groups of 10. The students were purposefully sampled by their proficiency in either Sesotho or isiZulu. They worked in groups to create their play artefacts in Sesotho or isiZulu that included the AI-related vocabulary, which was the second data source as a joint project, and so I reasoned that a group interview would be suitable. The data collection was structured in this way to allow student-teachers in the first semester to learn about and begin to engage in playwriting. Measures that ensure trustworthiness in qualitative studies include credibility, transferability, confirmability and dependability. I used multiple methods of data to facilitate a deeper understanding of the findings and to ensure credibility (Oliver-Hoyo & Allen 2006). I provided detailed discussions of both the content of the research and the steps in the analysis to ensure that the study can be transferable and to enable other researchers to judge the applicability of my findings. To establish confirmability, I provided an ‘audit trail’, which highlights every step of data collection and analysis that was made to provide a rationale for the discussions in the study. To ensure the study is replicable—meaning that other researchers can repeat the study and obtain comparable findings (Shenton 2004; Townsend, Cox & Li 2010)—an audit trail was maintained. This audit trail documented the processes followed during data analysis, thereby supporting the dependability of the results.

Data analysis: Making sense of the interview data

Thematic analysis is a commonly used qualitative strategy that focuses on analysing the responses from participants (Braun & Clarke 2006, 2022; Flick 2022; Xu & Zammit 2020) by labelling units of meaning in the transcriptions or other modalities of data. It is an approach that entails searching across a data set to identify, analyse and report recurrent phenomena (Henning et al. 2004). Kiger and Varpio (2020) posit that thematic analysis is an appropriate method of analysis for data gathered from interviews so that researchers can understand the experiences, thoughts or behaviours of participants. To analyse the interview, I drew primarily on the reflexive thematic qualitative analysis method by Braun and Clarke (2022) as an analytical tool, as shown in Figure 4. This figure shows the process that I followed to analyse the interview data. Working in a ‘bottom-up’ approach allows the researcher to generate themes systematically (Lincoln & Guba 1985).

FIGURE 4: Reflexive thematic analysis process.

I began by familiarising myself (Maykut & Morehouse 2002) with the data by listening to the interview recording several times. When listening to interview recordings, it is important to check for consistency and clarify any unclear parts, including the language used in the transcript. Figure 5 is an example of the items of interest that were identified from the interview recordings.

FIGURE 5: An example of items of interest from the interview recording.

In the next step, I transcribed the interviews and coded them to assign units of meaning to the meaningful parts that were shared by the participants. During this step, the data were transcribed verbatim. Table 1 is an example from the interview transcripts.

Once the interviews were transcribed and coded, I looked for patterns between the codes and clustered them into different provisional categories (Braun & Clark 2006, 2022). For each provisional category, a rule for inclusion is provided, along with a brief description that explains why the codes were grouped (Maykut & Morehouse 1994). Sixteen categories were created. Figure 6 is an example of how the categories were analysed to look for patterns in codes. A rule for inclusion in this category: In this provisional category, I grouped codes about the participants’ views on the benefits of performing plays for children.

FIGURE 6: Analysing categories to create refined categories.

The category name is included with the examples of the recurrent codes from the transcript. An analysis of these categories led to the generation of the category labelled ‘Students’ views about their interactions with the lecturer’. A further analysis of each of the initial categories was aimed at generating the final categories. Such an approach is not conventional because, in most qualitative studies, the data analysis begins with specific clustering of codes to create categories. I clustered the categories to generate refined categories (Maykut & Morehouse 1994). A brief description of how each provisional category was analysed to generate the refined categories is included. In addition, I have also included specific examples of codes that were clustered to form each category. Eight categories were generated. Figure 7 is an example of the refined category.

FIGURE 7: An example of a category from the interview data.

This category shows the importance of performing drama plays for children. Particularly, the plays they wrote for the children. Students mentioned the benefits of watching plays that are performed on stage for children. These benefits include the development of oral language, the acquisition of vocabulary, expressive vocabulary, self-confidence (because they viewed dramatic performance as less intimidating for children), a fun way of learning, collaboration with their peers, thinking critically by reflecting on the sequence of a story, and experiencing how language can be used in different ways.

To generate the themes, I synthesised the relationship between categories. For instance, to generate the first theme, I grouped categories 1, 9 and 10 because they relate to important considerations about language when planning for and engaging in playwriting. Figure 8 shows how theme 1 was derived, followed by an explanation of this theme.

FIGURE 8: Generating Theme 1 from categories 1, 9 and 10.

The first theme was generated and identified by grouping categories that related to language when planning for and engaging in playwriting. In this theme, it was important to consider preservice teachers’ language competence for writing a play in either Sesotho or isiZulu. This was about the students’ language proficiencies, either as home language speakers or as first additional language speakers of the language. This theme highlights that the students’ language competence had a significant influence on the kind of academic language that they integrated into their plays. How they use language could also influence how they integrate the AI vocabulary into the plays. As a result, it could also be useful for evaluating how their plays could enhance children’s oral language and vocabulary knowledge of AI-related terminology in either Sesotho or isiZulu. The process of synthesising the relationships between the categories enabled me to identify salient ideas within the categories (Strauss & Corbin 1990) and culminated in five themes. The themes from the interview data (the affordances and limitations of bilingual competence and self-generated guidelines for playwriting and collaboration and generative and imaginative thinking to retain the meaning of the AI-related vocabulary) give a glimpse of the students’ experiences during the process of writing the plays.

Analysing the classroom plays

Student-teachers worked in groups to write classroom plays with an AI theme in Sesotho or isiZulu. I began the data analysis process by reflecting on the titles of the plays. I then systematically examined the content of each play according to the criteria that I had sourced from the literature about educational plays for children. After that, I identified the Sesotho and isiZulu AI vocabulary that students had included in their plays. Lastly, I gave an account of the assessment of the overall pedagogical value of each play (O’Toole, Stinson & Moore 2009). I then designed a rubric that was informed by the literature on drama and playwriting to assess the plays (Gardiner 2019; Gardiner & Anderson 2018). Table 1 is a summary of the criteria of the rubric. Furthermore, this rubric can be adapted by teachers or student-teachers when implementing playwriting as a pedagogical approach for teaching. These themes are informed by the key ideas in playwriting (Gardiner 2019). The themes are aimed at showing that playwriting can be a structured classroom activity that can be a pedagogical tool that is implemented in a specific way (Shulman 2005), aimed at teaching vocabulary or any other targeted skill or concepts.

Ethical considerations

Ethical clearance for this study was granted by the Ethics and Higher Degrees Committee in the Faculty of Education at the University of Johannesburg, where the study was conducted, and the ethics number Sem 2-2020-118 was granted. Informed consent was requested and obtained from the participants who took part in this research. The participants were also informed that participation was voluntary, and they could leave at any time (Shenton 2004; Van der Reyden 2008).

Results

Artificial intelligence-related vocabulary in the plays

An important aspect of the design of the plays was the use of AI vocabulary. Students wrote plays that incorporated some of the AI-related terminology in their classroom plays. As they created the plays, they had to ensure that the language used to describe the terminology in Sesotho and isiZulu was linguistically appropriate and that it retained the meaning of the word in English. The criteria for assessing the pedagogical value of the classroom plays in shown in Table 2. In some instances, it was easier to directly translate the words from English to Sesotho or isiZulu. I served as the linguistic expert, as I was involved in assessing the students’ work. The plays were also externally moderated by an external moderator because they were an artefact that students created as part of their highest weighted assessment. When I assessed these artefacts, I focused on the functional equivalence rather than the linguistic equivalence of the translated word (Peña 2007). Some words were rather difficult to translate directly into either of the two languages. Table 3 and Table 4 include a comprehensive list of the isiZulu and Sesotho vocabulary that was generated by the preservice teachers.

Interview data

The thematic analysis process, as outlined by Braun and Clarke (2022), led to the generation of five themes from the study. These data reflect the students’ engagement with the process of writing classroom plays with AI-related vocabulary. The narratives in the plays highlight some of the ‘dramatic collisions’ that Veresov (2017) refers to. The final themes (Figure 6) illustrate the collaborative nature of the students’ journeys during the playwriting process. There was also evidence of subjective experiences in the way that they chose topics, created scenes and described characters. In the ensuing discussion, I elaborate on the themes about the affordances and limitations of bilingual competence and self-generated guidelines for playwriting and collaboration and generative and imaginative thinking to retain the meaning of the AI-related vocabulary.

Discussion

In the discussion of the findings, the work of Shulman (2005), coupled with Vygotsky’s sociocultural theory (Veresov 2017; Vygotsky 1987), provides a theoretical lens for the interpretation of the outcome of the study. In addition, student-teachers experienced Vygotskian psychology as a component of PCK knowledge for teachers, specifically regarding the generation of the Zone of proximal development (ZPD), which is the gap between what the learner can do independently and what they can do with the help of a more knowlegdable other in learning. Even though the term ZPD is widely used, it is, according to Veresov (2004). He refers to the notion of ‘dramatic collisions’ as an example of a term from stagecraft applied to the development of understanding as the catalyst for development and a realisation that more needs to be learned. The term used by Vygotsky to show how learning increases with ‘instruction’ is the ‘zone of proximal development’. Veresov (2017:36) refers to a publication by Vygotsky – The problem of teaching and mental development: ‘We have no hesitation after all that has been said in stating the essential characteristic of teaching to be the creation of the zone of proximate development’ (in Veresov 2017:36).

The findings further indicated that classroom plays are a valuable method for introducing student-teachers to using plays as an approach for teaching vocabulary (Khasu 2024; Khasu & Henning 2024). In this study, students were introduced. In this way, students can introduce new reading material that integrates language content with AI-related vocabulary. Doing so was certainly not an easy task. During the interviews, student-teachers expressed their views about their initial experience of the playwriting activity. Most of them indicated that writing classroom plays with an AI theme in Sesotho or isiZulu was challenging. Figure 9 is a graphical representation of the responses to one of the interview questions about their experiences of the playwriting activity.

FIGURE 9: Student-teachers’ initial experiences of the playwriting activity.

Affordances and limitations of bilingual competence and self-generated guidelines for playwriting and collaboration

To make sense of the AI vocabulary, the participants referred to the importance of communicating with one another and asking for help when necessary. Guidance from their peers gave them some assurance about the terminology they used as noted in Box 1. They asked for help when needed. Reading the dialogue that their peers wrote contributed to increased understanding. Krashen and Lee (2004) found that good writers read and edit their work with the help of others to come up with new ideas, which is applicable when reading and editing other people’s writing. This process was useful for determining the functional equivalence of the terminology. The students commented on this, saying that reading their group members’ work helped them with their writing. They were also aware of their peers whom they perceived as linguistically proficient in their groups as noted in Box 1. The limitations of this process, linguistically, still require the work of a language expert or more knowledgeable other (Vygotsky 1987) to ensure that the terminology is linguistically acceptable. The lecturer in this study provided guidance in relation to terminology development. However, this process would have benefited from reviews from a language board such as the Pan South African Language Board (PANSALB).

Generative and imaginative thinking to retain the meaning of the artificial intelligence-related vocabulary

Understanding the meaning of words is important because it allows for the use of words in informal and formal language (Putranti 2018). Some words are acquired naturally through spoken language, and others are explicitly taught and are typically formal or academic language (Shapopi-Shingenge 2023). This is important because students were cognisant of how they understood the AI vocabulary, as that would determine how they included the vocabulary in their plays. Translated text must be natural and sound appropriate in spoken language (Putranti 2018). At the outset, the participants in this study relied on their conversational knowledge of the language for familiar vocabulary that was comprehensible to them. Furthermore, they described how familiar vocabulary was easily accessible and was often included in their conversational language. However, complex and unfamiliar AI vocabulary that could not be easily translated required more effort from the students. One of the important aspects was to figure out the meaning of AI terminology in English. At the outset, they mentioned that they had made use of a commonly accessible dictionary such as Google Translate, which is a typical reference for novice translators as noted in Box 2.

In the African languages, the orthography of words is transparent (Land 2015). Often, when a text is translated from English to an African language such as isiZulu or Sesotho, it does not equate to a one-to-one correspondence with the English equivalent. A significant finding from the participants highlighted how some words were described in a phrase or sentence at times as noted in Box 3, to retain the meaning of the word. Likewise, the meaning of the AI terminology was important because it broadened their understanding and gave them ideas on how to use words in the dialogue of their scripts (Brown 2017; Gardiner 2019). It was also important for students to check the meaning of the translated AI vocabulary. They did this by checking various definitions from multiple sources of information and checking the syntax in the African language to ensure that the arrangement of words and phrases was well-formed (Muin 2019). One student found that the terminology that they could not translate was simply written in English and was noted with italics in their play as noted in Box 4.

Applying different iterations of the artificial intelligence terminology in the classroom plays

When translating terminology from one language to another, there may be various iterations of the vocabulary before it is refined (Cabré 2012). Students commented that sharing ideas about the process of creating AI terminology was important. It was not easy for them to select one version of some AI terminology. They observed that they deliberated on the different versions of the AI vocabulary as noted in Box 4. They included these versions in the various scenes of their plays. In spoken language, different versions of terminology might not have a significant impact on the meaning of the words (Bowker 2015), much like the AI vocabulary in the students’ classroom plays did not affect the meaning of the vocabulary in the dialogue. The various examples of AI terminology can contribute to enhancing ‘scientific discourse’ in the African languages. One student compared the use of unofficial words on social media to the AI vocabulary they created. He mentioned that people became accustomed to new words by using them more frequently. As they used these words when communicating, they became aware of the meaning and the pragmatic use of the words, even if such words were not considered as formal language that would appear in dictionaries.

Conclusion

The findings in this study suggest that classroom plays may be a valuable pedagogical tool for teaching vocabulary (Biemiller & Boote 2006; Gray & Yang 2015). In this study, the students were challenged to think creatively to create the isiZulu and Sesotho AI terminology in their dialogue. The playwriting activity added to their repertoire of teaching methods, which could feature in their professional teaching (Shulman 2005). Furthermore, it was evident that semiotic mediation can be a complex activity when different languages are involved (Kozulin 2003). Doing such an activity in groups expanded their students’ general vocabulary of Sesotho or isiZulu and that of AI-related vocabulary, which means that the objective of the learning activity was reached. The findings of this study suggest practical uses, while also showing some limits and areas for future research. Teacher education programmes should include play-based and drama activities to support vocabulary learning, especially in multilingual settings. Group playwriting helped students expand both everyday and subject-specific vocabulary, showing the value of creative, collaborative learning. However, the study was performed in one context, so the findings may not be transferable to other contexts. Another limitation is that the short duration of the study also makes it hard to determine the long-term effects of this pedagogical approach. Working across languages is complex; learners may need more support. Future research should explore the long-term impact of classroom plays as a pedagogical tool for vocabulary learning.

Acknowledgements

This article is based on research originally conducted as part of Mosa N. Khasu’s doctoral thesis titled ‘Towards a signature pedagogy for playwriting as a language teaching tool in the primary school’, submitted to the Faculty of Education, University of Johannesburg in 2024. The thesis was supervised by Elizabeth Henning and Sarita Ramsaroop. The supervisors were not involved in the preparation of this article and were not listed as co-authors. Portions of the thesis have been revised, updated and adapted for publication as a journal article. The original thesis is publicly available at: https://ujcontent.uj.ac.za/esploro/outputs/doctoral/Towards-a-signature-pedagogy-for-playwriting/9955202007691#file-0.

This article is based on data from a larger study. A related article focusing on drama as a pedagogical tool for learning in the classroom has been published in South African Journal of Childhood Education, 15(1), https://doi.org/10.4102/sajce.v15i1.1696. The present article addresses a distinct research question, focusing on presenting terminology development through the modality of classroom plays in the primary school.

The author, Mosa N. Khasu, declares that no financial or personal relationships inappropriately influenced the writing of this article.

Mosa N. Khasu: Conceptualisation, formal analysis, investigation, writing-original draft, project administration and writing-review & editing. The author confirms that this work is entirely their own, has reviewed the article, approved the final version for submission and publication, and takes full responsibility for the integrity of its findings.

The author disclosed receipt of financial support by the National Research Foundation (grant no.: TTK240422215307).

Data sharing does not apply to this article as no new data were created or analysed in this study.

The views and opinions expressed in this article are those of the authors and are the product of professional research. It does not necessarily reflect the official policy or position of any affiliated institution, funder, agency or that of the publisher. The authors are responsible for this article’s results, findings and content.

References

Footnote

1. The classic Vygotskian ‘triangle’ of activity in what Engeström (1999), referred to as ‘first generation activity theory’ (AT). See http://iscar.org for background to the International Society of Cultural-Historical and Activity Research.