Original Research

Student responses to subtitled and dubbed educational content: Implications for university translanguaging practices

Helena C. Kruger-Roux, Muhammad Nakhooda, Ignatius K. Ticha

Received: 24 Apr. 2025; Accepted: 30 June 2025; Published: 27 Nov. 2025

Copyright: © 2025. The Authors. 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

This article examines audiovisual translanguaging as a pedagogical strategy in South African universities, where students navigate tensions between English dominance and the institutional mandate to promote indigenous African languages. Through investigating student experiences with subtitled and dubbed instructional videos in agricultural science programmes, three research questions were added: (1) students’ audiovisual language preferences, (2) how these preferences reflect linguistic tensions and (3) implications for balancing language needs. Our mixed-methods study revealed that while English subtitles are predominantly preferred, students showed greater willingness to engage with indigenous languages aurally than through text. Students strategically pair different languages across modes, often combining indigenous language audio with English subtitles to balance comprehension and comfort. However, difficulties with formal academic terminology in indigenous languages highlight the need for more accessible vocabulary development approaches. The study demonstrates how audiovisual materials create productive student spaces to leverage students’ full linguistic repertoires while acknowledging that horizontal translanguaging practices alone may not provide sufficient access to languages of power necessary for academic success.

Contribution: We make recommendations for institutional policies that recognise multilingualism as a reality, supporting both fluid multilingual practices and the development of academic registers in indigenous languages.

Keywords: audiovisual translanguaging; academic terminology; linguistic repertoires; South African higher education; student preferences; English dominance; multilingual pedagogy; epistemic access.

Introduction

South African universities grapple with the complex challenge of managing language use, seeking to balance the dominance of English with the imperative to preserve and promote indigenous African languages, as mandated by the Language Policy Framework for Public Higher Education Institutions (Department of Higher Education and Training [DHET] 2020). This tension mirrors broader societal struggles involving the need to balance a multilingual heritage with a long history of linguistic inequalities, the pressures of globalisation, complex environmental factors, current socio-economic realities and future aspirations (Spolsky 2021).

In South Africa, the legacy of apartheid in continues to cast a shadow over language policies and practices, particularly in education. Despite significant strides towards transformation, the dominance of English persists. This preference is reported to stem from, among other factors, English’s perceived instrumental value and status as a ‘global academic lingua franca’ (Bhatt, Madan & Mesthrie 2022:430).

In response to these challenges, there are growing appeals, backed by legislative prescripts, to develop and use indigenous African languages for epistemic access in higher education. For example, DHET (2020) compels universities to strengthen previously marginalised African languages through multilingualism. However, implementation remains challenging, with students expressing mixed views about using African languages for learning and teaching, often citing concerns about their practicality and economic value (Ngcobo & Barnes 2020).

Translanguaging offers a promising theoretical and pedagogical framework for addressing these tensions. As conceptualised by Heugh (2015) and Makalela (2016), translanguaging transcends rigid translation practices by encouraging the fluid use of students’ full linguistic repertoires, moving beyond narrow definitions of official languages. In the context of audiovisual materials, this approach enables the integration of subtitles, audio tracks and interactive features that incorporate multiple languages, mirroring natural speech patterns in multilingual communities. However, there remains a significant gap in our understanding of how students actually engage with and respond to such translanguaging opportunities, particularly in formal educational settings.

This study addresses the three mentioned key questions regarding student preferences and experiences with audiovisual language resources in South African higher education: (1) it identifies students’ audiovisual language experiences and preferences regarding their full linguistic repertoires, (2) it examines how these experiences and preferences reflect tensions between English dominance and indigenous language development and (3) it determines how findings can determine strategies to balance language needs. We employed a mixed-methods approach where quantitative tracking of language selection patterns is combined with qualitative exploration of students’ reasoning and experiences across two South African universities.

Theoretical framework and literature review

This study employed a mixed-methods approach to examine students’ preferences and their responses to audiovisual translanguaging in agricultural science education at two South African higher education institutions. While traditional approaches to multilingual education often impose predetermined language choices on students, our methodology was designed to observe authentic translanguaging practices by allowing students to exercise their linguistic agency when engaging with educational content. By combining the quantitative tracking of actual language selection patterns with a qualitative exploration of students’ reasoning and experiences, we sought not only to understand what language choices students make but, crucially, why they make them. This approach enabled us to move beyond documenting experiences and preferences to examining the complex interplay between practical considerations, institutional contexts and language ideologies that shape students’ educational experiences in multilingual settings.

Theoretical perspectives on translanguaging and southern multilingualism

Recent theoretical developments have shifted focus from discrete languages to speakers’ complex meaning-making practices. Lin (2019:5) defines translanguaging as the creative deployment of speakers’ entire linguistic repertoire in meaning-making, which contrasts with traditional code-switching approaches that treat languages as separate systems.

This perspective aligns with African understandings of multilingualism. Heugh (2021) advocates for reclaiming southern perspectives on multilingualism as vital to inclusive education, stressing that the recent resurgence of interest in the Global North must not eclipse Africa’s centuries-long engagement with this field. This act of reclamation holds particular significance, as fluid multilingualism has long been a cultural strength among speakers of African languages, with linguistic exchange in South Africa traceable to early Khoi and San communities (Ndhlovu & Makalela 2021:55–56).

Empirical research supports these theoretical perspectives. Aroge, Makalela and Merisi (2024) found that allowing students to use their full linguistic repertoires in the 1970 Ife Project (Nigeria) and the 2002 LOITASA Project (South Africa) led to improved conceptual understanding and better academic outcomes. These findings align with what Heugh (2021) and Windle et al. (2023) term ‘southern multilingualisms’, which recognises multilingualism’s heterogeneous nature in postcolonial contexts.

However, Heugh (2021) notes that multilingualism can also signify hierarchies of power and exclusion, particularly in educational settings where code-mixing is stigmatised in favour of ‘standard’ prestige languages. Drawing on the ubuntu philosophy, Makalela (2016, 2019) and Makalela and Da Silva (2023) conceptualise southern multilingualism as a system of interdependent linguistic relations without boundaries. As Madiba (2024:111–112) argues, this approach challenges colonial notions of languages as fixed and separate entities, instead recognising them as fluid social constructs. Tshivenda, for example, illustrates this fluidity, having emerged through translingual fusion between Shona and Sotho.

Multimodality and audiovisual translation

Research over four decades has demonstrated the potential of audiovisual translation practices such as subtitling and dubbing in language education. These approaches provide cognitive scaffolding through rich multimodal environments that support comprehension and concept acquisition (Caimi 2006, 2015; Danan 2004). Empirical studies show specific benefits: improved word recognition and vocabulary acquisition from subtitled videos (Bird & Williams 2002), adaptation to unfamiliar accents through same-language subtitles (Mitterer & McQueen 2009) and enhanced listening comprehension through dubbing (Danan 2004). These benefits can be explained through cognitive theories such as the dual coding theory of Paivio (1990), which suggests that processing information through both verbal and visual channels enhances learning and recall. Vanderplank (2016) argues that while captions may appear to provide modest effects, their true value lies in enabling learners to develop effective viewing behaviours. A translanguaging approach to multimodality encourages students to leverage their full linguistic and semiotic repertoires through multilingual and multimodal texts, collaborative meaning-making and meta-modal awareness.

Indigenous languages as additional academic languages in South African higher education

The use of indigenous African languages as the medium of instruction in South African higher education remains complex and contentious. Despite the multilingual prescripts of the Language Policy for Higher Education (Ministry of Education 2002) requiring universities to develop and use indigenous official languages as academic languages alongside English and Afrikaans, implementation has been limited and regionally disparate.

Recent research has revealed what Mbirimi-Hungwe (2024:112) terms as a ‘fear of multilingualism’ in South African higher education. Despite constitutional recognition of linguistic diversity, universities continue to rely heavily on English as a medium of instruction by requiring non-English-proficient students to undergo remedial education regardless of their competence in other languages. Nevertheless, several institutions have established substantive multilingual initiatives, as documented in the language resources audit conducted by the South African Centre for Digital Language Resources (SADiLaR) (Van Dyk et al. 2023). Experiences at the University of KwaZulu-Natal and other institutions have confirmed that the use of African languages increases epistemological access, integrates indigenous knowledge systems and advances social justice.

In response to the national mandate, all South African universities have developed language policies—in varying forms—aimed at promoting indigenous languages as a means of advancing transformation and equitable access. Some universities have taken a more proactive stance in implementing language policies, often propelled by student-led advocacy and grassroots momentum. For instance, the University of KwaZulu-Natal (UKZN) formally committed to developing isiZulu as an additional medium of instruction alongside English, as outlined in its 2006 language policy (UKZN 2006). However, Mkhize and Balfour (2017) observe that these efforts are frequently constrained by the Constitution’s conditional phrasing and the presence of escape clauses in institutional policies, which stipulate that education in selected official languages is only required ‘where reasonably practicable’.

The implementation of genuinely multilingual policies in higher education is further complicated by what De Swaan (2023) describes as the paradoxical status of English in South Africa — simultaneously regarded as a colonial language and as a vehicle for liberation, global mobility and opportunity. The complexity of South Africa’s linguistic landscape poses a significant challenge for universities striving to balance the preservation and promotion of indigenous languages with the practical advantages and perceived prestige of English-medium instruction. These policy challenges point to the need for theoretical frameworks that can conceptualise multilingualism in ways that reflect African linguistic realities.

This study adopts an expanded conceptualisation of translanguaging – one that moves beyond conventional definitions of fluid language use confined to single communicative events. Drawing on recent theoretical developments in Southern multilingualism (Heugh 2021; Makalela 2016), this study adopts an expanded conceptualisation of translanguaging. This perspective extends beyond conventional definitions of fluid language use within single communicative events to include: (1) cross-modal practices in which multiple languages are strategically deployed across auditory and visual modalities; (2) shifts between registers both within and across languages; and (3) the interplay of horizontal (fluid) and vertical (hierarchical) dimensions of language use. This expanded conceptualisation enables a more nuanced analysis of how students navigate not only between distinct named languages, but also across academic and everyday registers. It further illuminates the strategic ways in which learners integrate linguistic resources across modalities – such as auditory and textual channels – to optimise their learning experiences.

The following section outlines the implementation of the mixed-methods research design across two higher education institutions. It describes the multilingual resources developed for the study, as well as the data collection instruments employed to capture both the observable patterns of language use and the underlying motivations informing students’ translanguaging practices within the agricultural science education context.

Research methods and design

This research was conducted across two South African higher education institutions, involving second-year students from the University of Pretoria (UP) and first-year students from the Cape Peninsula University of Technology (CPUT). Data collection was conducted throughout 2023, with a focus on students’ engagement with multilingual instructional videos. The research design was intentionally structured to address three core questions: (1) What are students’ audiovisual language preferences? (2) How do these preferences reveal underlying linguistic tensions? and (3) What are the implications of these dynamics for balancing diverse language needs within higher education?

Participants and materials

The study comprised 93 participants enrolled in agricultural science programmes across both institutions. This linguistically diverse sample offered the necessary variation in language backgrounds to meaningfully address the first research question concerning students’ audiovisual language preferences. The demographic composition of the participant cohort mirrored the linguistic diversity characteristic of South African higher education, with representation from speakers of isiXhosa, Afrikaans, English, isiZulu and several other indigenous languages.

A total of 1 h, 53 min, and 2 s of discipline-specific instructional content was developed, incorporating both subtitling and dubbing options to support multilingual accessibility. The instructional videos featured original audio in academic English, accompanied by subtitles in five language options: academic English, plain English, Afrikaans, isiXhosa, and Sepedi. The multilingual design of the study facilitated close observation of how students navigated between languages and registers, offering critical insights into our second research question concerning the tensions between English language dominance and the development of indigenous languages in academic contexts.

Although the instructional materials were subsequently made available on YouTube, the research focused specifically on student engagement with two curated videos: Video 1, presented in academic English without subtitles or dubbing, and Video 2, featuring both dubbing and subtitles to support multilingual access. The instructional content centred on agricultural science topics, incorporating authentic discipline-specific terminology. This enabled a focused examination of how students engaged with academic language across multiple linguistic options – a critical dimension of our third research question concerning the balance of language needs in higher education.

Procedure

Participants engaged with both videos in a controlled setting. For the second video, they were invited to customise their viewing experience by selecting preferred language options, including the choice to activate subtitles and, if so, to determine the language in which subtitles were displayed. Students also had the additional option of accessing dubbed versions of the content in their preferred language.

This choice-based intervention was designed to examine how students exercised their linguistic agency when offered multiple language options. It specifically aimed to illuminate the tensions between the dominance of English and the development of indigenous African languages within the context of higher education (Research Question 2). By enabling students to make authentic choices instead of assigning them to predetermined language conditions, the study was able to capture their natural language preferences (Research Question 1) and generate insights into potential strategies for balancing diverse language needs in higher education (Research Question 3).

Students were allowed to switch between the subtitle options and the audio languages throughout the viewing. This approach enabled us to observe how participants navigated across languages and communicative modes – a central feature of translanguaging practices that directly informs all three research questions.

Data collection

After the students’ engagement with the audiovisual materials, they completed a comprehensive questionnaire with several sections designed to address different aspects of the research questions:

Section A included procedures for obtaining informed consent from participants (students who did not give their consent were unable to proceed with the questionnaire).

Section B collected students’ demographic information (qualification enrolled for, race, home language, age, Grade 12 performance in English, self-rated English reading and writing proficiency). These data provided the context for understanding language preferences in relation to students’ linguistic backgrounds (Research Question 1) and the potential influence of educational history on language attitudes (Research Question 2).

Section C consisted of 15 contextual multiple-choice questions on Video 1 (unsubtitled and undubbed).

Section D consisted of 15 contextual multiple-choice questions on Video 2 (subtitled and dubbed).

Students had a maximum of 25 min for each of these sections. They were also allowed to stop, rewind and forward the videos. The comprehension questions served to evaluate the impact of various language options on students’ understanding of the instructional content. This analysis was central to addressing Research Question 3, which explores strategies for achieving a balanced approach to language use in higher education.

Section E examined participants’ perceptions of the value of subtitles and dubbing, directly addressing Research Question 1 concerning students’ audiovisual language preferences. The first question was whether students used subtitles, and if they did, in which language. The latter was followed by two Likert-scale instruments: a 19-item scale assessing participants’ responses to subtitling and a 10-item scale focused on dubbing. Both sets of items were designed to evaluate perceptions of the usefulness of the audiovisual interventions.

Crucially for the analysis, the Likert-scale items were followed by an open-ended question designed to elicit more nuanced reflections from participants. The 54 responses to the open-ended question, particularly those addressing language preferences, generated qualitative data relevant to Research Questions 2 and 3. These insights shed light on how students navigate linguistic tensions and suggest potential strategies for promoting balanced language use in higher education.

In addition to self-reported data, Google Analytics was used to capture participants’ actual language selection patterns during the viewing sessions. This integration of objective tracking with subjective reflection was instrumental in developing a comprehensive understanding of translanguaging practices, thereby informing potential implementation strategies in response to Research Question 3.

Data analysis

This article presents findings from multiple data sources, analysed through complementary approaches to address the three research questions.

To address Research Question 1 (language preferences), a quantitative analysis was conducted drawing on three data sources: frequency counts from the demographic section of the questionnaire, Google Analytics data from the controlled viewing sessions, and students’ reported subtitling choices. This triangulated approach enabled us to identify patterns in language selection across various audio-subtitle combinations and to examine how these preferences aligned with participants’ linguistic backgrounds.

To address Research Question 2 (tensions between English language dominance and the development of indigenous languages), we conducted a qualitative analysis of open-ended responses from 54 participants who provided substantive commentary. Many of these responses explicitly engaged with issues of language choice and linguistic agency. It is important to note that not all participants completed the open-ended sections of the questionnaire. The analysis focused on identifying patterns in students’ reasoning for selecting or rejecting particular languages, and how these choices reflected their sense of agency and educational needs.

To address Research Question 3 (strategies for balancing language needs), the quantitative and qualitative findings were integrated to identify approaches that might effectively support both English and indigenous languages. The analysis focused specifically on students’ accounts of successful translanguaging practices, the challenges they faced and their reflections on how various language combinations shaped their learning experiences.

Thematic analysis was used to identify recurring patterns in the students’ responses. Participant comments were systematically coded and categorised into thematic areas, including language preference, challenges with discipline-specific terminology, comfort levels across different languages and reflections on their overall learning experiences. These themes were subsequently analysed through the theoretical lenses of translanguaging and transknowledging in order to understand how students navigate between languages within academic contexts and to explore the implications of this navigation for implementing balanced language approaches.

This integrated analysis approach allowed us to move beyond simply documenting language preferences. It enabled us to understand the interplay between practical considerations, institutional contexts and language ideologies that shape students’ choices.

Limitations and strengths

While our mixed-methods approach provided insights into students’ translanguaging practices, several limitations should be acknowledged. Firstly, the controlled environment of the study may not fully reflect how students would engage with audiovisual materials in natural learning settings. Secondly, while the Google Analytics data provided useful insights, it was unable to capture all instances of language switching during the viewing sessions. However, the qualitative responses helped to address this limitation by offering deeper context and participant reflections. Thirdly, as is common with self-reported data, students’ explanations for their language choices may represent post-hoc rationalisations rather than accurately reflecting their real-time decision-making during the selection process.

Despite these limitations, the integration of objective tracking data with students’ subjective reflections offers a strong foundation for understanding not only which language choices students make, but also the underlying reasons for those choices. By studying actual choice behaviour rather than hypothetical preferences, we gain a more authentic understanding of how students might engage with translanguaging opportunities in educational settings.

Ethical considerations

Ethical approval to conduct this study was obtained from the Cape Peninsula University of Technology (30016165/08/2022) and the University of Pretoria, Faculty of Humanities (HUM030/0722). Respondents provided electronic consent for participation. Data will be stored in the institutional repository for 15 years.

Results and discussion

This section presents and critically analyses the findings from our investigation into audiovisual translanguaging practices conducted at two South African universities. The results are organised to address the three research questions that guided this study. Firstly, we examine students’ audiovisual language preferences regarding their full linguistic repertoires (Research Question 1) through quantitative language selection patterns and qualitative explanations of these choices. Secondly, we analyse how these preferences reflect the tension between English as a dominant language and the development of African languages in higher education (Research Question 2). Finally, we consider the implications of these findings in order to find strategies to balance the use of English and previously marginalised African languages (Research Question 3).

Research question 1: Student audiovisual language experiences and preferences

To address Research Question 1 regarding students’ audiovisual language preferences, we examined both quantitative patterns of language selection and qualitative explanations for these choices. This section presents the quantitative findings, while section ‘ Research question 2: Tensions between English dominance and indigenous language development’ explores students’ explanations and reflections on their language choices.

Demographic context: Home language distribution

The distribution of participants’ home languages offers critical contextual insight into their audiovisual language preferences and mirrors the geographic and linguistic landscape of the study’s locations.

Figure 1 illustrates the distribution of the 93 participants’ home language across both universities, highlighting the multilingual nature of South African higher education. Indigenous African languages constitute the majority of students’ home languages.

FIGURE 1: Combined home language distribution for University of Pretoria and Cape Peninsula University of Technology.

The University of Pretoria is situated in Northern Gauteng, where the most spoken languages are isiZulu (23.1%), Sesotho (13.1%), Setswana (10.4%) and English (9.2%) (Statistics South Africa 2023). Cape Peninsula University of Technology is in the Western Cape, where the most spoken languages are Afrikaans (41.2%), isiXhosa (31.4%) and English (22%) (Statistics South Africa 2023). As both are residential universities situated in urban areas, they attract students from across the country. Among the study’s respondents, the most commonly represented home languages were isiXhosa, Afrikaans, English and isiZulu.

Quantitative patterns in audiovisual language selection

Table 1 presents students’ responses regarding the subtitles and/or audio options they selected during the viewing sessions.

While Table 1 presents students’ self-reported subtitle language preferences aggregated across all viewing sessions, Table 2 provides a more nuanced picture by showing how their subtitle choices varied depending on the selected audio language. The Google Analytics data in Table 2 reveals the percentage of viewers who chose specific subtitle languages when listening to different dubbed audio tracks (Afrikaans, English, isiXhosa or Sepedi).

It is important to note that the total number of views (109) exceeds the number of participants (93). This discrepancy likely stems from multiple factors. An electricity disruption at one of the sites forced some students to restart the questionnaire – an issue mentioned by two participants in the open-ended responses, though oral feedback suggests it affected more students. Nine students reported in the open-ended question that they switched between audio options during the session because they did not find non-English dubbing helpful. Each switch would register as a new view in Google Analytics.

Key findings on students’ language experiences and preferences

Taking this context into account, the analysis of the data in Table 2 regarding Research Question 1 reveals several key patterns in students’ language choices. English resources dominated overall, with English subtitles collectively accounting for nearly three-quarters (72.8%) of all subtitle usage, split between academic English (39.7%) and plain English (33.1%). Similarly, English audio represented the most selected single option with 41.3% unique views.

However, a more nuanced pattern emerged when examining audio and subtitle choices separately. While students showed a strong preference for reading subtitles in English, they were more open to engaging with indigenous languages through listening. In fact, indigenous languages in total accounted for a majority of audio selections, with Sepedi attracting 25.7% of views, isiXhosa 22.0% and Afrikaans 11.0%. This strategic use of multiple languages across different modalities indicates that students were making intentional decisions about how and when to draw on their linguistic resources.

The divergence between students’ home language backgrounds and their language preferences was particularly striking. Although only 15.0% of participants identified English as their home language, it remained the most preferred option for audio content. Even more strikingly, Sepedi audio drew significant engagement – 25.7% of views – despite only 4.0% of participants identifying Sepedi as their home language. This pattern indicates that language choices in educational contexts extend beyond home language preferences.

The data further indicated that subtitle preferences varied according to the audio language chosen. When listening to English audio, students showed a preference for plain English subtitles (49.8%) over academic English subtitles (38.9%). Interestingly, academic English subtitles were predominantly chosen (70.8%) when participants listened to isiXhosa audio. These varying patterns suggest that students actively adjusted their language combinations to optimise comprehension, selecting different subtitle types to complement different audio languages.

These findings reveal a nuanced interplay between audio and subtitle choices. While English-based resources remain prominent In aggregate, there is meaningful engagement with indigenous languages, particularly for audio. The observed patterns indicate that students made strategic selections among multilingual options, guided by their individual learning needs and language proficiencies.

Research question 2: Tensions between English dominance and indigenous language development

This section examines how students’ audiovisual language preferences reflect the broader tensions between English dominance and developing and promoting African languages in higher education. Through qualitative data analysis, the study explored the reasons behind students’ language choices and examined how these choices reflect the linguistic complexities of South African higher education.

The dominance of English: Evidence from student preferences and reasoning

The quantitative findings in the previous section ‘Research question 1: Student audiovisual language experiences and preferences’ demonstrated a clear preference for English subtitles across all audio types. Students’ explanations for these preferences provide an insight into the pervasive dominance of English in academic contexts.

An English-speaking participant commented:

‘I did not utilise dubbing as my home language, and my language of education has always been English. The subtitles, at times, are either a tad fast or slow, though admittedly, that would depend on the camera angle and speed of the lecturer. Overall, subtitles are easy enough to use for an English home language learner, comprehensible and to the point.’ (Respondent 19, urban origin, University 1)

While such preferences are expected among English home language speakers, the more salient tension lies in the language choices of indigenous language speakers. Several isiXhosa home language speakers also preferred English subtitles to those in their home language:

‘Using subtitles was very useful and I prefer them in English rather than my home language as it was confusing and was a bit behind trying to figure out what they were talking about. I will continue using subtitles in English as it helps me to catch up with the video.’ (Respondent 51, rural origin, University 2)

‘For me, the subtitles were useful and helpful because I tend to understand better when I am watching and reading at the same time. I did not use dubbing because it would have confused me into understanding the content in my home language. So for me, it is best that I use English to study so that I can get better understanding.’ (Respondent 63, urban origin, University 2)

These responses reflect a pragmatic acceptance of English as the dominant academic language, with students routinely selecting it over their home languages for academic engagement. While Joubert and Sibanda (2022:48) describe ‘fractures, contestations and ontological dilemmas’ generated by English as the medium of instruction, the students in this study did not explicitly express such tensions. Instead, their responses suggest an internalised acceptance of what those researchers term the ‘dislocated identities in the classroom’ (2022:56), where the use of English for academic purposes appears normalised rather than actively questioned.

The preference for English may be influenced by affective filters, as conceptualised by Krashen (1982). Students choose English over isiXhosa as it gives them a greater sense of security within an educational environment where English has always been the language of assessment and instruction. This preference may also reflect how monolingual English examinations reinforce the hegemony of English, even for students in agricultural sciences who might need vernacular competencies to engage with stakeholder communities in their future professions.

Register tensions: Academic versus plain English

In addition to the tension between English and indigenous languages, students’ responses revealed nuanced attitudes towards various registers of English. Despite Google Analytics showing higher initial selection rates for plain English subtitles in some contexts, qualitative feedback supports the value of academic English subtitles, particularly for learning specialised terminology:

‘The subtitling in plain English is distracting and unhelpful. This is because the Latin terms are important; it’s better to understand and learn with academic English subtitles.’ (Respondent 13, rural origin, University 1)

‘It was useful in understanding what the man was saying with his accent. It also assisted me in seeing the spelling of the academic words.’ (Respondent 2, urban origin, University 1)

An isiXhosa speaker even explained how the subtitles helped them understand the difference between ‘normal’ and academic English:

‘The subtitles are very useful to me because they made me understand the content which I was being taught and also they are very beneficial because they improved my understanding. Now I can understand well the differences between normal English and academic English.’ (Respondent 53, rural origin, University 2)

These comments highlight how translanguaging between different registers of the same language can develop metalinguistic awareness, an essential aspect of navigating academic discourse. This reflects the vertical dimension of translanguaging outlined by Heugh (2021), wherein students navigate the hierarchical relationship between everyday and academic registers within a given language.

Terminology challenges in indigenous languages: A barrier to full adoption

The quantitative data showed some engagement with indigenous language options, but qualitative responses revealed significant challenges with academic terminology in these languages. These challenges represent a tension in promoting indigenous languages for academic purposes.

Regarding non-English subtitles, terminology seemed to be a significant stumbling block, even in a language such as Afrikaans with an established scientific vocabulary. One Afrikaans student expressed it as follows:

‘I tried using the videos and subtitles in Afrikaans, but it confused me as I’m used to studying in English and don’t know a lot of the terms in my home language. As English is not my home language, the subtitles do help me with the spelling of certain terminology and terms. When I couldn’t hear the person clearly, the subtitles also help to understand what was said.’ (Respondent 19, urban origin, University 1)

An isiXhosa-speaking participant also remarked on the complexity of the home language subtitles:

‘When listening to an English audio, I do not focus on subtitles because I understand the language and when it came to my home language[,] it got tricky because they use the language in depth.’ (Respondent 70, urban origin, University 2)

These comments reflect what McKinney and Tyler (2024:6, 9) describe as the ‘colonial language ideologies’ that have marginalised African languages in education. However, they also point to practical challenges in implementing translanguaging approaches in higher education settings. The students’ struggles with home language terminology highlight a complex challenge: their discomfort with formal registers in indigenous languages. As Respondent 51 observed, ‘they use the language in depth’, indicating a lack of familiarity with academic registers that, while present in the home language, may remain inaccessible to students primarily educated in English. This echoes Madiba’s (2024:117) finding that students found highly formal or ‘deep-deep’ versions of their home languages problematic. This challenge reflects both limited exposure to academic discourse in indigenous languages and the historical marginalisation of these languages in formal education.

Positive engagement with indigenous languages: Glimpses of successful translanguaging

Despite the challenges, some students reported positive experiences with indigenous language options, which is promising for addressing the aforementioned tensions.

An isiXhosa speaker commented:

‘The subtitles were so helpful and made the watching and listening very fast and easy. Listening in my own language made me very comfortable in watching the video because I understood every single step in the video.’ (Respondent 67, rural origin, University 2)

This comment reflects what García and Wei (2014) describe as an ‘established translanguaging space’, where the use of home languages is legitimised and welcomed, resulting in increased comfort and comprehension.

Several students highlighted the benefits of accessing multiple languages simultaneously, illustrating active translanguaging practices.

‘It helps to understand the work better and allows me to switch languages when I do not understand.’ (Respondent 82, rural origin, University 2)

‘Dubbing can sometimes be useful because you understand and are able to explain the words in your own language.’ (Respondent 71, rural origin, University 2)

These examples of established translanguaging spaces exemplify what Makalela (2016) describes as ‘ubuntu translanguaging’, a practice in which the interdependence of languages enriches meaning-making and learning. In such contexts, multilingual individuals draw on their full linguistic repertoire, pointing to potential pathways for more balanced integration of English and indigenous languages in higher education. These students engage in ‘register-meshing’, moving between academic and everyday language across different codes to form a hybrid register (Tyler 2016:13).

In summary, the findings reveal multilayered tensions between English dominance and indigenous language development in South African higher education. While the preference for English subtitles reflects deeply embedded language ideologies and practical concerns about academic success, the strategic use of indigenous languages, particularly for audio, suggests opportunities for more integrated approaches to language in education.

Conclusion

Research Question 3 aimed to explore how the findings could inform strategies for achieving a more balanced use of English and previously marginalised African languages in higher education. The following pedagogical, terminological and institutional implications emerged from the analysis.

Pedagogical implications of students’ translanguaging practices

The findings point to pedagogical strategies that could enable a more effective linguistic balance in higher education. The deliberate pairing of indigenous language audio with English subtitles by many students reflects an organic translanguaging practice, one that educational institutions could intentionally recognise, support and expand.

The practice of pairing indigenous language audio with English subtitles aligns with Heugh’s (2021:43) third interpretation of translanguaging, which acknowledges both horizontal (convivial and fluid) and vertical (hierarchical) dimensions of language use. This interpretation ‘can offer social justice, equity and inclusion… where code-switching or translanguaging is used purposefully and systematically’ in complementary ways. Students in the present study demonstrated an intuitive approach to balance, employing indigenous languages for comprehension while simultaneously accessing English academic terminology through subtitles. However, the mixed success of this approach – with some students reporting cognitive overload when using different languages simultaneously – indicates that audiovisual translanguaging strategies need careful implementation. The findings of the study suggest the following:

Introducing translanguaging practices through a graduated framework emerges as a necessary strategy. Rather than introducing cross-language combinations from the outset, institutions could initially provide synchronised audio and subtitles in the same language. This approach would support students in developing the cognitive strategies necessary for effective translanguaging before progressing to more complex multilingual configurations. Such a scaffolded approach may also support the development of students’ confidence and competence in managing multiple languages concurrently.

The developmental potential of translanguaging was most evident when students explicitly engaged with linguistic distinctions. For example, Respondent 40’s observation about differentiating between ‘normal English and academic English’ illustrates how translanguaging practices can foster metalinguistic awareness. This underscores the importance of pedagogical approaches that deliberately address language variation and translanguaging strategies, encouraging students to actively reflect on their navigation between languages and registers, rather than allowing this process to remain implicit.

Perhaps most significantly, the wide-ranging diversity of students’ language preferences and strategies observed in this study indicates that rigid, one-size-fits-all prescriptions for language use may hinder rather than support effective learning. Students demonstrated highly individualised approaches to combining languages. Some students preferred indigenous language audio with English subtitles, while others chose entirely different configurations. This diversity in choice indicates the importance of student agency in language selection. Instead of prescribing fixed language combinations, institutions should offer a range of language options and support students in making informed choices about when and how to draw on their linguistic resources to enhance learning.

Addressing the terminology challenge

Students’ difficulties with academic terminology in their home languages emerged as a critical barrier to more balanced language use. This challenge requires targeted interventions that acknowledge both the historical marginalisation of indigenous languages and the practical needs of contemporary students.

The development of accessible academic terminology in indigenous languages demands close collaboration between disciplinary experts, linguists and custodians of indigenous knowledge. The findings highlight several key considerations for advancing this collaborative endeavour.

Firstly, terminology development must be sensitive to register accessibility. Students’ struggles with what they described as ‘deep-deep’ versions of their home languages echo Madiba’s (2024:117) findings and suggest that purely formal approaches to terminology may alienate the very students these initiatives aim to support. Terminology development should, instead, strive to balance academic rigour with accessibility, ideally incorporating familiar words and concepts where appropriate. This approach would help bridge students’ everyday language practices and academic discourse, avoiding the imposition of unfamiliar technical terminology that can widen the gap between learners and their home languages.

Secondly, the multimodal nature of the intervention also suggests moving beyond traditional print-based glossaries. Audiovisual resources that demonstrate academic terms in context could help students develop both receptive and productive competence in discipline-specific vocabulary. Experiencing academic terminology within authentic disciplinary discourse – rather than through isolated definitions – can support students in more naturally incorporating these terms into their academic repertoires.

Furthermore, the teaching of terminology itself could benefit from strategic translanguaging approaches. In line with Probyn’s (2015) concept of ‘pedagogical translanguaging’, educators could purposefully use English in conjunction with indigenous languages to introduce new terms, thereby facilitating explicit cross-linguistic conceptual links. This approach would acknowledge the reality that students often need to navigate between languages while building conceptual understanding instead of treating languages as separate compartments for learning.

These approaches align with Madiba’s (2014) work on promoting concept literacy through multilingual glossaries. However, the students broaden this perspective by exploring how audiovisual translanguaging can facilitate terminology acquisition in ways that traditional print resources may not.

Institutional policy implications

At the institutional level, the findings suggest several policy directions that could more effectively balance the use of English and indigenous languages, while remaining responsive to current sociolinguistic realities.

Fundamentally, institutional policies must explicitly affirm translanguaging as a legitimate and effective educational approach, with particular attention to its application in audiovisual resources. Current language policies often acknowledge multilingualism in principle but fail to provide concrete guidance for implementation. Effective policies must move beyond mere statements of intent to include specific implementation guidelines, resource allocation and dedicated funding for developing multilingual materials and supporting translanguaging practices.

The strong preference for English subtitles among indigenous language speakers in the study partly reflects the constraining reality of English-only assessments. When students are aware that assessments will be conducted exclusively in English, their strategic reliance on English-language resources becomes a logical and pragmatic choice. To address this, institutions could develop assessment approaches that actively value multilingual competence. This might include options for multilingual responses where students can demonstrate understanding across languages or translation tasks that explicitly assess cross-linguistic understanding as a distinct academic competency.

Supporting academic staff in the development of translanguaging pedagogies is essential to any meaningful institutional transformation. Lecturers need professional development opportunities to understand translanguaging theory, develop skills in creating appropriate audiovisual resources and cultivate strategies for facilitating multilingual learning environments. Without this support, even the best-intentioned policies will fail at implementation stage.

Finally, institutions should work within what Cenoz and Gorter (2017) term a ‘sustainable translanguaging’ framework. This approach acknowledges the importance of safeguarding and expanding opportunities for students to engage with marginalised languages, while also recognising the prevailing dominance of English in academic and professional spheres. Rather than viewing as competing goals, sustainable translanguaging seeks to develop complementary practices that strengthen all languages in students’ repertoires while preparing them for multilingual futures.

The findings suggest that effective institutional policies must acknowledge multilingualism as an inherent and valuable aspect of the educational landscape, rather than treating it as a challenge to be overcome. This entails moving beyond binary frameworks (English versus indigenous languages) towards strategies that promote both the fluid use of multiple languages and the cultivation of academic registers in indigenous languages.

Audiovisual translanguaging as a transformative strategy

The audiovisual translanguaging practices documented in this study offer particular promise for balancing English and indigenous languages in ways that transform rather than merely accommodate multilingualism. The multimodal format of subtitled and dubbed videos provides multiple points of access to content and accommodates a range of learning preferences and strategies.

The findings specifically reveal that audiovisual translanguaging offers several transformative possibilities. Firstly, it appears to develop metalinguistic awareness in ways that are not typically fostered by traditional monolingual approaches. When students reflected on linguistic differences, such as distinguishing between academic and everyday English or navigating between formal and informal registers in their home languages, they demonstrated enhanced consciousness about language as a system. This heightened awareness, fostered through exposure to multiple languages in structured audiovisual contexts, can strengthen overall language learning strategies and support more sophisticated academic literacy development.

The audiovisual format also seems to create safer spaces for indigenous language use. Several students described feeling ‘comfortable’ when engaging with content in their home languages, indicating that audiovisual resources may offer lower-stakes environments for accessing academic material in indigenous languages, particularly in contrast to the pressures of classroom discourse or written assessments. This comfort could gradually build students’ confidence in using their home languages for academic purposes, potentially reversing patterns of linguistic alienation in higher education.

Perhaps most significantly for career readiness, audiovisual translanguaging provides valuable practice in navigating between different linguistic registers and contexts. Students in disciplines like agricultural sciences, who will need to communicate with diverse stakeholders ranging from rural farmers to international researchers, can develop skills in moving fluidly between academic registers and vernacular or grassroots language use. This ability to shift between languages and registers according to context represents a crucial professional competency that is not sufficiently fostered within monolingual educational frameworks.

These transformative potentials align with what Makalela (2016, 2019) terms ‘ubuntu translanguaging’, which emphasises the interdependence of languages and their collective contribution to meaning-making and learning.

This study illuminates the complex linguistic strategies employed by multilingual students in South African higher education through their engagement with audiovisual translanguaging. Our findings reveal that while English maintains its dominance in written formats, students significantly engage with indigenous languages in aural modes – a pattern suggesting promising pathways for implementing multilingual approaches aligned with the Language Policy Framework’s requirements. Students demonstrate agency in their translanguaging practices when given appropriate tools, strategically pairing indigenous language audio with English subtitles to balance comprehension, comfort and academic requirements. This organic approach represents a practical solution to the tensions between English dominance and indigenous language development. However, challenges with academic terminology in indigenous languages – students’ difficulties with formal or ‘deep-deep’ versions of their home languages – highlight the need for more accessible academic vocabulary development. The study highlights that effective implementation depends on recognising that horizontal translanguaging practices, on their own, may be insufficient for ensuring access to dominant languages that are critical for academic achievement. Universities should develop targeted interventions that: (1) systematically support cross-modal language combinations that students find effective, (2) develop accessible academic terminology in indigenous languages, (3) establish assessment frameworks that recognise and validate multilingual competence and (4) provide professional development for staff in translanguaging pedagogies. By embracing audiovisual translanguaging as a pedagogical strategy, South African universities can move beyond deficit-based views of multilingualism and instead recognise it as a foundational reality that enriches and diversifies the educational experience.

Acknowledgements

The authors reported that they received funding from the South African Centre for Digital Language Resources (SADiLaR) and the Department of Science and Innovation. The authors have disclosed those interests fully and have implemented an approved plan for managing any potential conflicts arising from their involvement. The terms of these funding arrangements have been reviewed and approved by the affiliated university in accordance with its policy on objectivity in research.

H.K.R. led the conceptualisation and methodology development, conducted formal analysis and investigation, wrote the original draft, created visualisations and managed the administration of the project. M.N. contributed to data curation and validation, reviewed the article critically and edited the manuscript with the assistance of software implementation and resource allocation. I.K.T. supported the investigation process, contributed to data analysis, assisted with project resources, reviewed the article and provided editing feedback.

Research was funded by the South African Centre for Digital Language Resources (SADiLaR) and the Department of Science and Innovation.

The authors confirm that the data supporting the findings of this study are available within the article.

The views and opinions expressed in this article are those of the authors and are the product of professional research. They do 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