Current Issues in Language Planning ISSN: 1466-4208 (Print) 1747-7506 (Online) Journal homepage: https://www.tandfonline.com/loi/rclp20 Bilingual instruction at tertiary level in South Africa: what are the challenges? R. Wildsmith-Cromarty & N. Turner To cite this article: R. Wildsmith-Cromarty & N. Turner (2018) Bilingual instruction at tertiary level in South Africa: what are the challenges?, Current Issues in Language Planning, 19:4, 416-433, DOI: 10.1080/14664208.2018.1468959 To link to this article: https://doi.org/10.1080/14664208.2018.1468959 Published online: 02 May 2018. Submit your article to this journal Article views: 116 View Crossmark data Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=rclp20 CURRENT ISSUES IN LANGUAGE PLANNING 2018, VOL. 19, NO. 4, 416–433 https://doi.org/10.1080/14664208.2018.1468959 Bilingual instruction at tertiary level in South Africa: what are the challenges? R. Wildsmith-Cromarty and N. Turner School of African Languages and Linguistics, University of KwaZulu-Natal, Durban, South Africa ABSTRACT ARTICLE HISTORY The University of KwaZulu-Natal has a bilingual language policy (2006) where students may receive instruction in isiZulu as well as in English. An online survey was carried out for all academic and support staff to gauge their linguistic capacity in isiZulu. Staff profiles were created in relation to their respective Colleges, Schools and Disciplines, and included age, language group, language use, language of schooling and years of experience in the Higher Education sector. It also included modules currently taught by staff through the medium of isiZulu; communicative language proficiency; instructional language proficiency; perceptions of own capacity to teach in isiZulu and perceptions of the time required to develop adequate proficiency to teach in isiZulu. Respondents were also asked to translate a complex sentence into isiZulu, the responses to which were later analysed into three categories: fluent, semi-fluent and not fluent for purposes of comparison with staff perceptions of their own proficiency. Findings revealed a serious gap between current staff capacity to teach in isiZulu, and the implementation of the policy within the projected timeframes. This study also reveals the complexities of teaching when discipline experts and students do not share a common language. Received 19 July 2017 Accepted 6 March 2018 KEYWORDS Bilingual instruction; tertiary education; language policy; isiZulu as MOI Introduction Recent events in the higher education sector in South Africa regarding the #feesmustfall movement point to an underlying frustration on the part of students with the status quo (Desai, 2016). Many first-year students enter the university seriously under-prepared for their studies at this level (Desai, 2016; Magqwashu, 2014). It is well-documented (Fleisch, 2008; Spaull, 2015) that the South African schooling system is failing its children in this respect, as evidenced in the PIRLS and TIMMS results for mathematics and literacy at Grades 3, 6 and 9 levels (Howie et al., 2008). By the time learners get to Grade 12 it is too late to play ‘Catch up’ (Pretorius, 2014) and so the task is left to university lecturers to scaffold learning for their students as best they can, usually by way of academic literacy courses, mostly in English. Discipline specialists are not exempt from this task as they struggle to teach their disciplines, especially science and maths (Knoetze, 2016; Probyn, 2006, 2009; Setati, 2008) One of the ways in which both teachers and university lecturers scaffold learning in their disciplines is through code-switching and sometimes translanguaging (Garcia & Wei, 2014), CONTACT N. Turner khulisaprojects@gmail.com © 2018 Informa UK Limited, trading as Taylor & Francis Group CURRENT ISSUES IN LANGUAGE PLANNING 417 a practice that is gaining recognition in the literature and is regarded as one of the ways in which to include African languages as an essential component of instructional pedagogy, especially where the majority of learners are African, and do not necessarily speak English or Afrikaans as a mother tongue. Although many successes have been reported from studies and interventions relating to code-switching at both school and university level in South Africa (Du Toit, 2016; Wildsmith-Cromarty & Gordon, 2009), its success really depends on how it is carried out. There appears to be consensus that, for efficient concept development to occur, code-switching in class for instructional purposes needs to be systematic and purposeful; translation of terms needs to be carried out professionally (Kaschula & Wolff, 2016; Wildsmith-Cromarty, 2008), and the teacher or lecturer should be proficient in the language of the students. Various studies have reported on the successful implementation of bilingual teamteaching where the discipline specialist was not fluent in the African language but taught alongside a bilingual teacher who was familiar with the discipline in question (Du Toit, 2016; Wildsmith-Cromarty, 2013). In such cases, the discipline content was given through English, with some simultaneous translation of terms, while the African language (isiZulu) was used for conceptual clarification. Students in the team-teaching studies were free to use either language in their assessments as they were marked by both lecturers. What is crucial here is the retention of the original concept in the explanation, especially for disciplines such as Mathematics, Science, Psychology or Law, as dilution of meaning is known to occur when there are no structural equivalents for the expression of source language concepts (Dlodlo, 1999; Wildsmith-Cromarty, 2008). Kaschula and Wolff (2016) offer a word of caution in this regard: … care must be taken that quality education is indeed being transmitted through the channels of the various indigenous languages … . For a transitional period, such indigenous African languages as are selected as prospective MoI require special attention in terms of comprehensive language planning. This includes … ‘intellectualisation’, i.e. lexical expansion by terminology development and daily use in all … prestigious domains. (p. 6) We would add that the training of lecturers to use the language for instructional purposes effectively is also crucial, especially if their erstwhile training was in English. Teachers have reported difficulties in using an African language for instructional purposes if they had not been trained in the academic (and secondary) discourse (Wildsmith-Cromarty & Gordon, 2009). Discipline specialists who are isiZulu-speaking would naturally find it easier to codeswitch for instructional purposes. Such staff are also pivotal in the development of isiZulu courses for professional purposes such as nursing, health sciences, dentistry, psychology and education, which are currently offered to students at the university (Engelbrecht & Wildsmith, 2010; Kamwendo, Hlongwa, & Mkhize, 2014; Mashiya, 2010; 2014; Magqwashu, 2014; Pluddemann, Nomlomo, & Jabe, 2010). Such interventions form part of the ‘intellectualisation’ aspect of bilingual teaching, in terms of terminology development and the creation of ‘secondary’ discourses in the African language (Magqwashu, 2014). In their volume on ‘Intellectualisation: A current issue in language planning,’ Liddicoat and Bryant (2002) point out that the most difficult aspect of intellectualisation is its implementation. By this they imply that academics ‘must use the language for scholarly discourse, and it is difficult to influence such behaviour directly 418 R. WILDSMITH-CROMARTY AND N. TURNER through legislation’ (p. 2). They go on to mention the example of Sami language teacher education institutions that not only teach the language, but also train teachers in how to teach content through the Sami language. This could be what is required of UKZN for truly effective implementation of its language policy. A further complexity is the level of the students’ own proficiency in the African language. Those who have experienced it as a LoLT at school level have a far better chance of using it meaningfully as a LoLT at tertiary level. However, many children are sent to multicultural schools that use English de facto as a medium of instruction. In such cases the children are not exposed to the African language in the academic domain so for them, the use of the African language as a LoLT at tertiary level remains a challenge. As Magqwashu (2014) points out: An analysis of a series of student module evaluations based on their previous years’ experiences revealed that translating both their thoughts and reading material from English into isiZulu, as they attempted to take advantage of the ‘opportunity’, compromised their academic success as their written work lacked academic rigour and precision. (p. 92) Clearly, students will also need a solid academic foundation in the language, in this case, isiZulu, in order to take full advantage of its use as a LoLT. This further underscores the crucial importance of academic staff competence in the language used for instruction. Without it, students cannot receive the scaffolding and mediation at a ‘deep’ level of learning that is required for concept development and mastery of the secondary discourse in the African language. This has serious implications for the effective implementation of the University of KwaZulu-Natal’s language policy (UKZN Language Policy, 2006). The University responded to the South African Government’s call for tertiary institutions to develop their own language policies by designing a policy that caters for bilingual instruction. This means that, in time, tutorials and lectures will be offered to students through the medium of both isiZulu and English. To date, the language of instruction has been an ex-colonial language (i.e. English) that is not the students’ mother tongue. However, UKZN’s language policy and plan requires the University to encourage and assist students and staff to develop appropriate proficiency in both English and isiZulu. In order to determine whether the University has the operational capacity to effectively implement a bilingual language policy at the University of KwaZulu-Natal, the University Teaching and Learning Office (UTLO) commissioned an audit which aimed to find out what the language resource capacity of the university is, i.e. which subject disciplines staff on all UKZN campuses were currently able to teach through the medium of isiZulu, and how many would be able to do so in various projected time frames. The purpose of the audit was to enable the University to identify the existing language skills already in place, in order to estimate resources still required to put the University’s Language Policy into effect. In other words, the university needs to engage in ‘teacher development’ (Turner & Wildsmith-Cromarty, 2014, p. 308) in order to enable ‘subject specialists to teach their subjects through the medium of African languages.’ Furthermore, the university needs to make provision for ‘the development of glossaries, teaching manuals and other resources (which) still require the input of staff who are sufficiently proficient in the African language to guide its use as a LoLT’ (p. 308). In other words, staff need to be sufficiently proficient in the African language in order to implement the bilingual policy and teach modules in the medium of both isiZulu and English as required. CURRENT ISSUES IN LANGUAGE PLANNING 419 Bilingual/multilingual education Variations on bilingual education models can either be subtractive or additive. In Anglophone countries such as the USA, UK and Australia, they have mainly been implemented at the school level and are in essence, subtractive (Ramirez, Yuen, & Ramey, 1991). Part of the reason for this is that in the English-dominant countries, bilingual education is aimed at teaching English to immigrants or indigenous populations. Many of these programmes are transitional and usually involve a dominant, majority language such as English and a minority, immigrant or indigenous language. The minority language is used in the early years of schooling but gradually shifts to the dominant language for the later years, including tertiary education. Subtractive programmes have also been termed ‘submersion’ majority language programmes, which has been the case in Africa and in South Africa (Desai, 2016; Klapwijk & Van der Walt, 2016; Macdonald, 1990; Probyn, 2009). The irony in these contexts is that English is the minority language which the majority have to use as a medium of instruction. Bilingual programmes that are additive are implemented at both secondary and tertiary levels and are of various types such as maintenance, enrichment and heritage programmes (May, 2008). New Zealand, South America, Malaysia, India and Europe are good examples of countries who adopt an additive approach (De Mejia, 2008; May, 2008; Mohanty, 2008). Maintenance programmes aim to sustain the learning of a minority or indigenous language, alongside a ‘majority’ or dominant language, in order to strengthen learners’ cultural and linguistic identities and affirm their ethnolinguistic rights’ (May, 2008, p. 22). He further points out that this type of programme normally involves a ‘national minority group member’ such as the Welsh in England or Catalan in Spain. This could apply to the current bilingual model because the use of isiZulu alongside English is intentional and ideological. The aim is to increase the status and use of isiZulu in a prestige domain such as education, at the same time affirming the ‘ethnolinguistic rights’ of the students. However, the target group are a ‘national majority group member’ in this case, whose language has been previously marginalised. For UKZN students to truly benefit from such a model, they would need to have used it as a LoLT at school level or at least have sound FAL instruction at school in order for students to be prepared to learn bilingually. Heritage language programmes are normally associated with second or even third generation learners who are children of immigrants to a country with a different dominant language. Over time, successive generations assimilate to the dominant language and culture to the point where the languages of origin are lost to newer generations. Heritage programmes are thus attempts to introduce and sustain links to the language and culture of the ‘motherland.’ Examples are the Maori in New Zealand or the Indian communities in South Africa. The latter learn Urdu, Hindi or Tamil as heritage languages. An important point to note here regarding the above programmes is that the first language of the learners is already highly developed to an age-appropriate level in academic literacy which also facilitates transfer to literacy acquisition in the second language. This is referred to as the linguistic interdependence principle. (Cummins, 2000; Cummins & Hornberger, 2008). Enrichment programmes, such as the French immersion programmes in Canada for English speakers (Genesee, Lindholm-Leary, Saunders, & Christian, 2006) usually target 420 R. WILDSMITH-CROMARTY AND N. TURNER learners from a well-resourced language, such as English and immerse them in a second language, where they use it as a sole medium of instruction with their first language taught as a subject. Such programmes are generally regarded as elitist and are successful because both languages are well-resourced and the learners have a very strong foundation in their first language. This model is also the one with the greatest potential to educate students successfully and reduce inequalities, at the same time creating truly bilingual students. This is the model to which UKZN aspires but there is still some way to go before we get there in terms of terminology and academic discourse development and language and discourse development for staff. A more recent model proposed in the literature on bilingual/multilingual education is one which allows the use of the languages that students bring with them to the classroom. Garcia and Wei (2014) call it a ‘poly-directional bilingual education model’ with an emphasis on translanguaging in the lecture theatre, where lecturers and students move between two (or more) languages as a matter of routine. They regard this as dynamic plurilingualism. Klapwijk and Van der Walt (2016) referred to this as English-Plus multilingualism, seeing it as a new form of ‘linguistic capitalism.’ The context for this, however, would be one where both lecturers and students share the languages in question in terms of fluency. The bilingual model proposed by UKZN is a clear attempt at redressing the inequities of the past by using a previously marginalised language for instructional purposes alongside English. It is also a response to the government drive to develop the African languages in the academic domain, i.e. to intellectualise them. Finally, it is an attempt to facilitate understanding of complex subject matter for students through the medium of their own language. What is required to implement and support such a programme, however, are professionally trained lecturers who can teach through the medium of the African language, who are able to code-switch or ‘translanguage’ if necessary, and who are comfortable with the languages their students bring to class. This is a tall order in an institution where 92.3% of the academic staff are not first language speakers of the African language in question (see Figure 1). The audit was necessary, therefore, to establish the nature and level of the language capacity of the institution in terms of interpersonal communication skills and academic language proficiency. Methodology A ‘compulsory’ questionnaire was sent via email to all staff on all five campuses of the University of KwaZulu-Natal to determine: (a) Modules currently taught by staff through the medium of isiZulu (b) Capacity of staff to deliver modules in isiZulu in 2014/2015 (c) Capacity of staff to deliver modules in isiZulu by 2018–2020 Researchers with expertise in the African languages and Applied Linguistics designed the questionnaire to determine the language teaching capacity of the academic staff as well as the language abilities of the support staff. Together with a small team of post graduate students, an IT specialist and a statistician, the questionnaires were distributed via email and CURRENT ISSUES IN LANGUAGE PLANNING 421 Figure 1. Home language: academic and non-academic. the data were then collated and quantified for each of the four Colleges: Health Sciences, Law & Management, Humanities and Science, Agriculture and Engineering. The audit objectives were: . . . . To assess the current ability of the staff to teach and communicate in isiZulu. To assess the willingness of the staff to be trained in the use of isiZulu as a teaching and communication language. To obtain information on the length of time (in years) staff would be willing to be trained in isiZulu. To match self-assessed proficiency with attempts at translation of a complex instruction in isiZulu. The questionnaire asked for information on staff profiles (college, school, discipline, permanent/contract, length of employment, courses taught), age, education, home language, details about fluency in various languages, translation of phrase, school language, fluency in English and isiZulu (scale 0–10), and willingness to be trained in isiZulu. In 2013 when the audit was carried out, the staff composition at UKZN totalled 3408, of which 951 responded. This was only 28% although the sample size in itself was substantial. Of the academic staff, 594 out of 1453 (41%) responded, and of the non-academic staff, 357 out of 1955 (18%) responded. Reasons for these figures could be that isiZulu-speaking staff did not feel obliged to respond as they are fluent anyway, or that staff might have felt threatened by the ‘compulsory’ nature of the audit, given the challenging climate of transformation taking place at UKZN at the time. Any admittance to a lack of capacity to implement or support the Language Policy could have been seen as creating job insecurity. 422 R. WILDSMITH-CROMARTY AND N. TURNER Findings Figure 1 shows the percentage of staff speaking various languages as their home or first language. English is clearly the dominant language group with 72.4% for academic staff and 67.7% for non-academic staff. The next highest group for academic staff were foreign staff (14.1%). Afrikaans scored similar percentages for academic and non-academic staff. isiZulu is greater for non-academic staff (21%) than academic staff (7.7%). Other South African indigenous languages were spoken by 4.3% of non-academic staff and only 2.5% of academic staff. In sum, the majority (92.4%) of academic staff did not speak isiZulu as a first language. For non-academic staff, 79% did not speak isiZulu as a first language. Although non-academic staff need the language for interpersonal communication, they would also require some language training in the academic discourse and terminology of the disciplines they were supporting, especially when processing course notes or examination papers. Academic staff, however, would require more substantial training in both the language and the discourse of the disciplines they taught. The above percentages reveal a huge task ahead of the university in this regard as only 21% of non-academic staff are isiZulu-speaking, and only 7.7% of the academic staff. The proportion of isiZulu-speaking staff in each college is reflected in Figure 2. The College of Humanities has the highest number of isiZulu-speaking staff, with the non-academic staff double the amount of academic staff: 31.4% versus 16.3%. This college houses the language departments which could be the reason for the higher figures. There is a significant drop of isiZulu-speaking academic staff in the other colleges, however, with the lowest number recorded for Science, Agriculture and Engineering. The figures for nonacademic staff follow a similar pattern, except the decrease across colleges is more gradual. Again, the dearth of academic staff in the professional disciplines who can speak isiZulu fluently is a challenge for language training. Figure 2. Percentage of isiZulu staff per College and in total. CURRENT ISSUES IN LANGUAGE PLANNING 423 In terms of capacity to use isiZulu as a working language, staff were asked to rate their conversational ability on a scale of 1–10. A rating of between 6 and 10 defined an isiZulu speaker in this study. Figure 3 reflects the percentage of staff at UKZN who consider themselves to be fairly fluent in isiZulu. According to the chart, isiZulu capacity is greater among non-academic staff (27.6%) probably due to the linguistic composition of the staff. Other South African language speakers and foreign language speakers are the highest (73.3% and 16.7%, respectively) due to their home languages being cognate with isiZulu, especially for the Nguni languages such as isiXhosa, isiNdebele and SiSwati. Speakers from the Sotho language group would also be more multilingual in relation to other African languages. This is also true for academic staff (46.7% and 6.2%, respectively). The English speakers who know isiZulu (3.5% and 4.6%) work mainly in the African language and/or Linguistics departments. Afrikaans does not appear in this graph as the responses were zero for this particular question. In sum, as far as teaching through isiZulu is concerned, only 12.5% have full proficiency according to the responses from this survey. A follow-up question probed academic proficiency in the language, asking staff who rated their proficiency between 6 and 0 to rate their ability to write in isiZulu and to explain discipline content through the medium of isiZulu. Figures 4 and 5 show the results of this question. Not all the staff who had rated themselves between 6 and 0 for speaking isiZulu could write it well. Approximately 24% of non-academic staff and 27% of academic staff rated themselves between 0 and 5. The rest scored 76% for non-academic staff and 73% for academic staff which were very similar. Similar results were obtained for explaining it (see Figure 5). Similar to the results for writing, not all staff who rated themselves between 6 and 10 for speaking isiZulu could explain in it. For non-academic staff, 10% could not explain and for academic staff, 24% could not explain. For non-academic staff, the concept of explaining Figure 3. UKZN isiZulu proficiency 2013/2014. 424 R. WILDSMITH-CROMARTY AND N. TURNER Figure 4. Speaking versus writing in isiZulu. would refer to explanations of an administrative nature linked to the course concerned and its requirements. For academic staff, explaining would refer to explicating and elucidating complex academic concepts relating to the course content. For non-academic staff, 90% felt that they could explain through the medium of isiZulu and 75% of academic staff felt that they could do this. It is, however, the academic staff who need to be able to use the explaining function in the language in terms of actually using it as a medium of instruction. In order to verify the claims for fluency in isiZulu, the next question asked respondents to translate the following instruction into isiZulu: Could you please report that, due to problems, today’s lecture has been cancelled. Figure 6 shows the results for staff who rated their isiZulu-speaking ability between 6 and 10. Only 50 out of 73 academic staff, i.e. 68.5% attempted the translation, of which 88% were either judged to be fluent or semi-fluent. A significant number (23) of these staff did not attempt the translation (31.5%). For the non-academic staff, 67 (69.1%) out of 97 staff attempted the translation, of which 78% were either fluent or semi-fluent. Thirty-one per cent did not attempt the translation. These figures reveal that capacity for teaching through isiZulu is not equivalent to being able to converse in it fluently. Writing the academic discourse and being able to explain discipline content in isiZulu is equally, if not more, important. The following (Figure 7) shows the results regarding the actual teaching capacity in isiZulu according to College. More courses are taught through isiZulu (23) in the College of Humanities than in any other College where the average is either 4 or 3 each. This makes sense when one considers CURRENT ISSUES IN LANGUAGE PLANNING 425 Figure 5. Speaking versus explaining in isiZulu. the home language profiles of staff per College (see Figure 2) where Humanities has more isiZulu-speaking academic staff than all the other colleges combined (16% versus 12.7%) which means that it has the capacity to teach more academic courses through isiZulu. It is not the same for the non-academic staff, however, whose numbers of isiZulu-speaking staff per College are not very different except for Science, Agriculture and Engineering which has the lowest number of staff at 17.3%. The low numbers of staff speaking isiZulu in the three professionally oriented colleges is concerning especially regarding the bilingual language policy. Staff were asked whether they were willing to be trained to teach through isiZulu and the time period within which they thought they could do this. The time periods were 1, 2, 3, 4, 5, 10, 15, 20 years or never. Figure 8 shows their responses. About two-thirds (68.9%) of the academic staff and three-quarters (74.2%) of nonacademic staff indicated their willingness to undergo training. Similar scores were obtained for both academic and non-academic staff who were undecided. However, double the number of academic staff compared to non-academic staff were unwilling to be trained. None of the staff in this group had an isiZulu speaking rating between 6 and 10. In other words, they could not speak the language and they were not willing to learn it. This could be because they feel that English is sufficient for teaching their discipline, or it could be due to the longer time it would take to train academic staff to teach their discipline through isiZulu, as they would need to acquire the academic discourse. Cummins (2001) points out that it takes between 5 and 7 years to learn cognitive academic language skills in an additional or foreign language, 426 R. WILDSMITH-CROMARTY AND N. TURNER Figure 6. Translations into isiZulu. but only 2 years to develop basic interpersonal communication skills, which is what the non-academic staff would require. Finally, staff were asked to select their short-term and long-term time periods within which they thought they could learn and be trained in the language. Figures 9 and 10 show the results. Figure 7. isiZulu teaching capacity per College. CURRENT ISSUES IN LANGUAGE PLANNING 427 Figure 8. Willingness of UKZN staff to have isiZulu training. For both academic and non-academic staff one year is the most popular short-term choice. Far more non-academic staff chose this option (85) as opposed to the academic staff (59) and this could be because they believe they could master basic interpersonal communication skills within that time. It could also be that staff feel that they have to comply with the requirement, but wish to spend the shortest possible time on it. It could be that it may not be a priority for busy academics who are required to teach undergraduate students, supervise postgraduate students, engage in community outreach projects and publish at the same time. However, the academic staff did indicate a second short-term choice of 2 or 3 years (46 for both options) which could show a more serious intention to learn isiZulu and an awareness of the time it would take to learn it effectively for work purposes. Staff were also asked to indicate longer term options and Figure 10(A,B) indicates the results. Some staff chose more than one time period which poses a challenge for the analysis of the results. For the purposes of this article, we will consider time periods of 5, 10 and 20 years only. Both academic and non-academic staff preferred 5 years as a longer term choice with the next choice being 10 years. The periods from 10 to 20 years is a far more popular longterm choice for academic as opposed to non-academic staff. Perhaps this was due to the perception that training to use the academic discourse of a language for their discipline would take longer than 5 years. It is clear from the above results for the survey that the majority of staff, both academic and non-academic, are not prepared for the full implementation of the bilingual teaching 428 R. WILDSMITH-CROMARTY AND N. TURNER (a) (b) Figure 9. (A) Most popular short-term choices-academic staff and (B) most popular short-term choicesnon-academic staff. model, unless it occurs in parallel teaching with duplication of classes with different teachers in the different languages. However, this is not what the UKZN Language Policy has proposed. The policy envisages an integrated model whereby both languages are used as CURRENT ISSUES IN LANGUAGE PLANNING 429 (a) (b) Figure 10. (A) Most popular long-term choices-academic staff and (B) most popular long-term choicesnon-academic staff. complementary languages of instruction (Wildsmith-Cromarty, 2013). The survey allowed a space for staff to put their own comments concerning the implementation of the policy. These comments revealed staff concerns about the policy and have been echoed in the literature on the use of African languages in education (Desai, 2016). 430 R. WILDSMITH-CROMARTY AND N. TURNER They include concerns that the introduction of isiZulu would affect students’ development of fluency in English, which they found to be essential because English is an international language. They were also cautious about having to teach their subjects through a language which had no terminology for those disciplines. They highlighted the need for intellectualisation to precede the use of the language for instructional purposes. They were also concerned about the length of time it would take to converse in isiZulu (5 years) and teach through isiZulu which would take even longer (20 years?). They were also keenly aware of how long proficiency in writing would take to develop. Finally, a number of staff asked the question: ‘Is this audit aimed at promoting isiZulu?’. The answer to that would be: the Language Policy is, yes, but the audit is an attempt to ascertain the challenges surrounding its promotion. Conclusion The findings from the staff audit on abilities to use isiZulu as a conversational and an academic language show that most non-academic and academic staff need language training in basic interpersonal communication skills, and academic staff specifically are not yet ready to use isiZulu to teach their subjects. IsiZulu-speaking staff are in the minority, so either the University employs more isiZulu speakers in critical discipline areas, or trains the existing staff. Most require at least five years before they would feel ready to teach and the short courses that are already in place for staff ‘are not aimed at making staff proficient at teaching through the medium of African languages’ (Turner & Wildsmith-Cromarty, 2014, p. 309). Three major aspects of language planning and policy (LPP) emanate from this study which could be applicable to other contexts: 1. LPP requires financial support and resources. 2. LPP needs to be participatory and inclusive, involving all key stakeholders. 3. LPP needs to show evidence of political will at a national level. In relation to the first point, implementation of any language policy requires support mechanisms. This means that interim measures such as bilingual teaching support, including teaching assistants, team-teaching and translation of materials with training on how to use them need to be put in place. Depending on the languages in question, terminological and academic discourse development should be an integral part of this. In relation to the second point, attitudes and agency are key variables here (FentonSmith & Gurney, 2015; Liddicoat & Taylor-Leech, 2015; Wiley & Garcia, 2016). The attitudes of key stakeholders are crucial in the effective implementation of a language policy as all stakeholders are responsible for this (Wiley & Garcia, 2016). A number of staff in the current study indicated an unwillingness to be trained in the use of isiZulu as a medium of instruction or of communication. In order to ensure the success of any future training programme, staff would need to be convinced of its advantages. In addition, student attitudes towards being taught in their mother tongue also need to be monitored as they are key participants to the successful implementation of any language policy. Opinions of the students at universities and their views on indigenous language usage in the South African context has been widely researched (Dyers, 1999; Kamwangamalu, 2004; Koch & Burkett, 2005; Naidoo, Gokool, & Ndebele, 2017). Findings from such studies need to be taken into account when formulating language plans. CURRENT ISSUES IN LANGUAGE PLANNING 431 Political will is the third key variable emanating from this study. Competent bilingual staff would be more readily available had they learned the additional languages from primary school. Schools need to introduce indigenous languages as compulsory additional languages at schools in South Africa (or elsewhere) which is an important part of acquisition planning. However, they also need support from government at national level. With reference to the USA, Wiley and Garcia (2016, p. 48) make the point that explicit language planning ‘tends to be done at the state, local or institutional levels … .’ Perhaps policies need to be more explicit at national levels as well in order to create the affordances necessary for effective implementation. An example would be the South African government’s (South Africa, 2013) draft policy which proposes the incremental introduction of African languages at the foundation phase for all South African learners. While fewer English first language speakers have opted to take an indigenous language through to Grade 12 since 1994 (Turner 2012, p. 45), the new draft policy could be a longer term solution for effective language policy implementation in South Africa. Disclosure statement No potential conflict of interest was reported by the authors. Notes on contributors Prof Rosemary Wildsmith-Cromarty is an applied linguist with an interest in language policy, additional language acquisition, reading literacy in the African languages and multilingualism. She currently holds the ETDP-SETA Research chair in Early childhood Education at North-West university, South Africa. She is also Emeritus Professor at the University of KwaZulu-Natal. She has published widely both locally and internationally. Prof Noleen Turner is an Honorary Research Professor at the University of KwaZulu-Natal where she lectured in the School of African Languages and Linguistics for thirty-two years. Her speciality is second language acquisition, as well as Oral studies, Humour and Onomastics. She has published widely locally and internationally and is currently working on several research projects. She is a past secretary and chairperson of the South African Journal of Folklore Studies and was editorial secretary for the Onomastics journal Nomina Africana for 10 years. ORCID R. Wildsmith-Cromarty http://orcid.org/0000-0002-1770-6873 N. Turner http://orcid.org/0000-0003-0913-4526 References Cummins, J. (2001). Negotiating identities: Education for empowerment in a diverse society (2nd ed.). Los Angeles: California Association for Bilingual Education. Cummins, J., & Hornberger, N. H. (2008). Encyclopedia of language and education. New York, NY: Springer. Cummins, J. (2000). Language, power and pedagogy: Bilingual children in the crossfire. Clevedon: Multilingual Matters. De Mejia, A. M. (2008). Enrichment bilingual education in South America. In J. Cummins & N. H. Hornberger (Eds.), Encyclopedia of language and education (2nd ed., pp. 323–332). New York, NY: Springer. 432 R. WILDSMITH-CROMARTY AND N. TURNER Desai, Z. (2016). Learning through the medium of English in multilingual South Africa: Enabling or disabling learners from low income contexts? Comparative Education, 52(3), 343–358. Dlodlo, T. (1999). Science nomenclature in Africa: Physics in Nguni. Journal of Research in Science Teaching, 36(3), 321–331. Du Toit, M. (2016). A multilingual approach to teaching South African history. In R. H. Kaschula & H. E. Wolff (Eds.), Multilingual education for Africa: Concepts and practices (pp. 132–148). Pretoria: UNISA Press. Dyers, C. (1999). Xhosa students’ attitude towards Black South African languages at the University of the Western Cape. South African Journal of African Languages, 19(2), 73–82. Engelbrecht, C., & Wildsmith, R. (2010). Exploring multilingualism in a problem-based learning setting: Implications for classroom and clinical practice in the nursing discipline. Alternation, 17(1), 108–137. Fenton-Smith, B., & Gurney, L. (2015). Actors and agency in academic language policy and planning. Current Issues in Language Planning, 17(1), 72–87. Fleisch, B. (2008). Primary education in crisis: Why South African school children underachieve in reading and mathematics. Cape Town: Juta. Garcia, O., & Wei, L. (2014). Translanguaging: Language, bilingualism and education. Basingstoke: Palgrave Macmillan. Genesee, F., Lindholm-Leary, K., Saunders, W., & Christian, D. (2006). Educating English language learners: A synthesis of research evidence. New York, NY: Cambridge University Press. Howie, S., Venter, E., van Staden, S., Zimmerman, L., Long, C., du Toit, C., … Archer, E. (2008). PIRLS 2006 summary report: South African children’s reading literacy achievement. Pretoria: Centre for Evaluation and Assessment, University of Pretoria. Kamwangamalu, N. M. (2004). The language policy/language economics interface and mothertongue education in post-apartheid South Africa. Language Problems and Language Planning, 28(2), 131–146. Kamwendo, G., Hlongwa, N., & Mkhize, N. (2014). On medium of instruction and African scholarship: The case of isiZulu at the University of KwaZulu-natal in South Africa. Current Issues in Language Planning, 15(1), 75–89. Kaschula, R. H., & Wolff, H. E. (2016). Introduction: The multilingual context of education in Africa. In R. H. Kaschula & H. E. Wolff (Eds.), Multilingual education for Africa: Concepts & practices (pp. 2–8). Pretoria: UNISA Press. Klapwijk, N., & Van der Walt, C. (2016). English-Plus multilingualism as the new linguistic capital? Implications of university students’ attitudes towards languages of instruction in a multilingual environment. Journal of Language, Identity and Education, 15(2), 67–82. Knoetze, S. (2016). Teaching mathematics to isiXhosa-speaking students through Afrikaans. In R. H. Kaschula & H. E. Wolff (Eds.), Multilingual education for Africa: Concepts and practices (pp. 35–46). Pretoria: UNISA Press. Koch, E., & Burkett, B. (2005). Making the role of African languages in higher education a reality. South African Journal of Higher Education, 19(6), 1089–1107. Liddicoat, A. J., & Bryant, P. (2002). Intellectualisation: A current issue in language planning. Introduction to Current Issues in Language Planning, 3(1), 1–4. Liddicoat, A. J., & Taylor-Leech, K. (2015). Multilingual education: The role of language ideologies and attitudes. Current Issues in Language Planning, 16(1–2), 1–7. Macdonald, C. A. (1990). Crossing the threshold into standard 3: Main report of the threshold project. Pretoria: Human Sciences Research Council. Magqwashu, E. (2014). On developing academic literacy in the mother tongue for epistemological access: The role of isiZulu as the LoLT in a South African university. Current Issues in Language Planning, 15(1), 90–103. Mashiya, N. (2010). Mother tongue teaching at the University of KwaZulu-natal: Opportunities and threats. Alternation, 17(1), 92–107. May, S. (2008). Bilingual/immersion education: What the research tells us. In J. Cummins & N. H. Hornberger (Eds.), Encyclopedia of language and education (Vol. 5, pp. 19–34). New York, NY: Springer. CURRENT ISSUES IN LANGUAGE PLANNING 433 Mbatha, T. (2014). Experiences of foundation phase teachers qualified in a dual medium programme. Per Linguam, 30(2), 37–50. Mohanty, A. (2008). Multilingual education in India. In J. Cummins & N. H. Hornberger (Eds.), Encyclopaedia of language and education (Vol 5, pp. 165–174). New York, NY: Springer. Naidoo, S., Gokool, R., & Ndebele, H. (2017). The case of isiZulu for non-mother tongue speakers at the University of KwaZulu-Natal, South Africa – Is the compulsory language module promoting social cohesion? Journal of Multilingual and Multicultural Development, 38, 1–13. Pluddemann, P., Nomlomo, V., & Jabe, N. (2010). Using African languages for teacher education. Alternation, 17(1), 72–91. Pretorius, E. J. (2014). Supporting transition or playing catch-up in Grade 4? Implications for standards in education and training. Perspectives in Education, 32(1), 51–76. Probyn, M. J. (2006). Language and learning in science in South Africa. Language and Education, 20 (5), 391–414. Probyn, M. J. (2009). Smuggling the vernacular into the classroom: Conflicts and tensions in classroom code switching in township/rural schools in South Africa. International Journal of Bilingual Education and Bilingualism, 12(2), 123–136. Ramirez, J., Yuen, S., & Ramey, D. (1991). Final report: Longitudinal study of structured English immersion strategy, early-exit and late-exit transitional bilingual education programs for language-minority children. San Mateo, CA: Aguirre International. Setati, M. (2008). Access to mathematics versus access to the language of power: The struggle in multilingual mathematics classrooms. South African Journal of Education, 28.1, 103–116. South Africa Department of Basic Education. (2013). The incremental introduction of African languages in South African schools. Draft policy. Pretoria: Government Printer. Retrieved from https://www.education.gov.za//resources/Policies.aspx Spaull, N. (2015). Education Quality in South Africa and Sub-Saharan Africa: An Economic Approach (Unpublished PhD thesis). University of Stellenbosch, Stellenbosch. Turner, N. (2012). African languages as compulsory courses in KwaZulu-natal: Illusory initiative or inspired intervention? Per Linguam, 28(2), 26–47. Turner, N., & Wildsmith-Cromarty, R. (2014). Challenges to the implementation of bi-/multilingual language policies at tertiary institutions in South Africa (1995–2012). Language Matters, 45(3), 295–312. University of KwaZulu-Natal Language Policy. (2006). Retrieved from http://www.ukzn.ac.za/ Wildsmith-Cromarty, R. (2008). Can academic/scientific discourse really be translated across English and African languages? Southern African Linguistics and Applied Language Studies, 26 (1), 147–169. Wildsmith-Cromarty, R. (2013). A problem-oriented, form-focused course design for teaching isiZulu as an additional language. Southern African Linguistics and Applied Language Studies, 31(2), 219–233. Wildsmith-Cromarty, R., & Gordon, M. (2009). Policy versus practice: The role of the home language in learning mathematics and science in English-medium classrooms. Language Learning Journal, 37(3), 359–370. Wiley, T. G., & Garcia, O. (2016). Language policy and planning in language education: Legacies, consequences and possibilities. The Modern Language Journal, 100, 48–63.