Keywords

Against the background of the Sustainable Development Goals (SDGs) of the United Nations (UN) and education reforms since 1994, this chapter presents an overview of language in education in South Africa and highlights some of the major achievements and challenges. As one of the former anti-apartheid struggle sites, education has been transformed over the past three decades. Previously it was a highly exclusive system devised along racial and ethno-linguistic lines with variable curricula and standards across different communities; today it can be described as unified and far more inclusive, with altruistic objectives of redress, equality and social cohesion. However, access to education of comparable quality across the system remains difficult to achieve.

Infrastructural backlogs at institutions of learning and technological constraints continue to preclude many students from benefiting from quality education and the affordances of the digital era of learning. The abrupt switch of schools to online learning during the COVID-19 pandemic in 2020 excluded students from continuing their education in poorly resourced areas. Furthermore, a long-standing problem relating to preconceived ideas about English as the path to academic success at all levels has resulted in reluctance to apply policies dealing with language in education in a way that benefits the majority of learners. The result is a youth inadequately prepared to cope with the demands of a rapidly changing world. It is this matter in particular that concerns us as we believe that more focused implementation of language in education policy according to the unique context of each school could make a real difference to attaining some of the SDGs.

The 2019 voluntary national review submitted to the United Nations by the office of the Presidency (RSA, 2019) shows the South African government’s commitment to implementing the 2030 Agenda through various initiatives. We are encouraged by the many references in the review to addressing literacy issues in the early years of learning, but we remain concerned about the neglect of the benefits of additive bilingualism/multilingualism in the subsequent phases of education. Although the review expresses a commitment to increased use of the Sintu languagesFootnote 1 in primary school education (RSA, 2019, pp. 13, 49) and greater availability of literature and materials in these languages (the languages of more than 75% of South Africans), paradoxically, it reverts to ongoing “efforts to strengthen English as a subject and as a medium of instruction” (ibid. p. 49). Other than the two brief references to language issues cited above, and a third scant mention of communicating in “indigenous languages” in a section dealing with urban planning (ibid. p. 88), the voluntary review does not address how epistemological access and cognitive development can be facilitated through the choice of language of learning and teaching (LoLT), or the role that language proficiency plays in economic development. The review dryly sums up the current state of education: “… completion rates in the upper secondary grades and enrolment rates in tertiary education are low. Inadequate skill levels severely constrain growth.” (ibid. p. 12). Notably, the review is based on the findings of studies and educational planning initiatives spanning more than two decades. These studies and initiatives require a critical examination and therefore form the basis of the discussion in the rest of the chapter.

Our deliberations on inclusive and equitable education are framed by the guiding principles of the Constitution of the Republic of South Africa, 1996 (RSA, 1996a), the South African Schools Act, 1996 (RSA, 1996b) and the National Education Policy Act, 1996 (RSA, 1996c), a progressive document aimed at, among other things, advancing and protecting the rights of citizens to receive basic education, enjoy equal access to education, and receive instruction in the language of choice where practicably feasible (section 4(a)). Of particular relevance is the Language-in-Education Policy (LiEP) adopted in 1997 in terms of the National Education Policy Act (Department of Education, 1997) and further expounded in the national school curriculum known as CAPS (Curriculum and Policy Statement). Given the diverse and complex nature of the fourth SDG, our chapter will focus only on language in basic education (Grades 1–12). Separate scrutiny is required to deal with developments pertaining to early childhood education (Grade R) and higher education (post-Grade 12), although the former now also falls within the ambit of basic education.

1 Inclusivity, Equitability and Quality as SDG Goals

The fourth SDG calls for quality education that is inclusive and equitable. These notions feature strongly throughout the 2030 Agenda for Sustainable Development, but they are multi-faceted concepts that are open to variable interpretation. Although the roots of inclusive education lie largely in special education research in the 1960s, aimed at accommodating the needs of persons with disabilities (Florian, 2014), the term has a broad reach and cuts across different communities of practice. In a general sense it can be understood as a philosophical stance on what should be achieved, namely “meeting the social/academic needs of all pupils” (Göransson & Nilholm, 2014, pp. 268–269). South Africa has made considerable strides in terms of disability accommodation (see Dube, 2006; Van der Byl, 2014), but when inclusivity is considered from the perspective of linguistic access, there has been little progress. Florian (2014) considers two principles to be important when adopting an inclusive approach: (1) teaching practices that allow all children to participate in “classroom life”, and (2) the use of language in the classroom that “expresses the value of all children” (Florian, 2014, p. 290). Although Florian refers to language in the sense of ensuring conducive teacher–student discourse and respectful communication, it would be difficult to support the first principle without adequately addressing the matter of the language of learning and teaching (LoLT). Language used in the classroom affirms both the value of the language itself and the cultural identity of the learner, but it is also the key to epistemological access. Haug (2017, p. 207) reminds us that inclusion is “strongly value- and ideology-driven” and associated with concepts such as “participation, democratization, benefit, equal access, quality, equity and justice” (ibid. p. 206). It is easily included in policy and curricula documents as an intention, but practical implementation is far more complicated.

The concept of equitability is closely related to the notions of fairness and equity or the state of being treated equally. Jordan (2010) argues against attempting to define concepts such as “equity” or “equitability” with a unitary instrument since learning itself cannot be measured with the same yardstick in diverse contexts. He explains as follows:

Equity is not about providing the same education to all students regardless of race, social class, or gender. In fact, because of increasing cultural and linguistic diversity it is advantageous to define educational equity in terms of providing knowledge, skills, and worldviews which would enable social mobility … contexts shape our views of equity, and it takes on different meanings among different populations. (p. 148)

What complicates matters in South Africa is the need to use education to achieve desirable social mobility for the majority of learners, as opposed to efforts in other countries to improve the lot of minority or immigrant groups. For the purposes of our discussion on language in education, equitability will be addressed from the point of view of creating learning opportunities to provide comparable kinds of knowledge and skills that support Jordan’s ideal of “social mobility”, as cited above. This relates directly to another objective of SDG 4, namely the matter of quality of education.

Apart from the complexities of providing inclusive and equitable learning opportunities, education should adhere to a particular standard that, internationally speaking, could be considered as quality instruction. In our view, education may be considered to have quality when its effects are manifest in tangible and desirable ways in the different spheres of society, once again Jordan’s (2010) notion of social mobility. In the South African context, this would mean that on reaching adulthood (i.e. the age of maturity as a citizen and the right to vote), the majority of citizens are able to contribute to society by mobilising vocational or professional career choices in such a way that their participation and involvement in such career choices can be described as competent and rewarding. Evidence of a lack of access to quality education would then be apparent in the inability of individuals to perform with relative ease and proficiency those tasks, roles and duties traditionally expected of citizens in the private and public spheres. Obviously, a good standard of education does not always guarantee success. Personal attributes and affective variables relating to wellness and psychological mobility may obstruct social mobility. These fall beyond the scope of the chapter.

We will evaluate the extent to which education in South Africa can be considered inclusive, equitable and of quality, through a comprehensive document review of developments related to language issues in school education over the past 25 years. First, we will summarise how schooling has been made more inclusive and accessible through the formation of a unified national education department, language policy and the adoption of a new school curriculum. We will then turn our attention to how the Department of Basic Education (DBE) monitors and evaluates education progress through a series of reports, surveys and research projects. We will focus on findings and recommendations relating specifically to language and literacy issues.

2 Policy and Curriculum Initiatives

On the education front, the main objective of government shortly after assuming power in 1994 was to consolidate disparate structures by reducing the number of education departments and creating a unified system of education. The amalgamation of 18 education departments into one national department, the increased spending on education, redistribution of funds to poorly resourced schools, and the establishment of quality control organisations have all been lauded as among the most noteworthy achievements of the South African government on the education front (Jansen & Taylor, 2003, p. 2). To mention one positive outcome, the changed policy on how schools are funded (RSA, 2014) currently enables over nine million children to attend schools for free (RSA, 2019, p. 6). Many more poor children thus have access to school and the chance of an education.

The DBE is also to be commended on its sustained efforts to revise the school curriculum and to introduce new school subjects in order to prepare learners for new occupations and professions. Despite these milestones, the goal of ensuring that learning is taking place through quality schooling remains unfulfilled to a large degree, as we will see. The low status of the Sintu languages as languages of intellect and economic force has not changed much since the emergence of democracy in 1994, despite their official status. English has retained its historically hegemonic position, while the number of Afrikaans-medium schools continues to decline with the prevailing perception that, to succeed in South Africa, proficiency in English is needed above proficiency in any other language (Louw, 2004; Postma & Postma, 2011; Webb, 2013). Yet, proficiency in English appears to be problematic for both teachers and learners: students’ academic literacy levels remain low, as universities and training institutions have discovered.

The bias towards English can be seen as a continuation of the initial campaign for English as the only official language advocated by many in the ANC leadership prior to the adoption of the Constitution (see Crawhall, 1993; Heugh, 1986). It is thus not surprising to note the tendency today of more affluent parents from diverse cultural groups (in the urban areas in particular) to enrol their children at English-medium schools where their children’s first languages are not offered at all, making the envisaged national policy of additive bilingualism difficult to implement in such schools. The policy advocates that learners should continue to learn their “home” (i.e. first) languages, while learning one or more additional languages (DoE, 1997). A similar problem exists in schools where learners represent multiple language groups and the decision is taken to adopt English as the medium of instruction from Grade 1, as though this is a neutral choice and without regard to how this may have adverse effects on the children concerned. Whereas learning through an additional language is not necessarily problematic for children from middle- and upper-class families who attend good schools, it can be detrimental for children from poor families and under-resourced schools located in areas where English is barely used outside the school gates (see Heugh, 2002).

It should be acknowledged that the LiEP adopted in 1997 was a collaborative and progressive effort developed over more than 15 years. It was informed by the work of local scholars and included conceptualisations of the notions of multilingualism and “critical pedagogy” (Heugh, 2015, p. 283), as well as comments from the public. Although the term “additive bilingualism” was adopted from North America, it remains an appropriate response for the South African context when understood as follows:

A wide spectrum of opinions exists as to the locally viable approaches towards multilingual education . . .. Whichever route is followed, the underlying principle is to maintain home language(s)Footnote 2 while providing access to and the effective acquisition of additional language(s) …

The main aims of the Ministry of Education’s policy for language in education are … to pursue the language policy most supportive of general conceptual growth amongst learners, and hence to establish additive multilingualism as an approach to language in education … to counter disadvantages resulting from different kinds of mismatches between home languages and languages of learning and teaching … (DoE, 1997, pp. 1–2)

We believe the reason why we still see enormous mismatches and conceptual and literacy challenges across all spheres of public education is related to the fact that the LiEP is understood and applied differently by the DBE and School Governing Boards (SGBs). We will provide a fuller explanation later in this chapter. Suffice it to say at this point that the LiEP is not the problem, but the prescribed Curriculum and Policy Statement (CAPS) deviates from the LiEP by encouraging an early switch to English. This has negatively affected the status and development of the Sintu languages for educational purposes and entrenched subtractive bilingualism/multilingualism. It is interesting to note that the phrase “additive bilingualism” is explained in the curriculum document for the Foundation Phase (Grades 1–3), whereas in CAPS for the higher grades, “additive multilingualism” is mentioned only very briefly, in a glossary towards the end of the document. Another peculiarity we see in CAPS is the assumption that “children come to school knowing their home language. They can speak it fluently, and already know several thousand words” (DBE, 2011a, p. 8). We know this is not true: much has been published on the limited vocabulary of learners (Pretorius & Murray, 2019; Pretorius & Stoffelsma, 2017; Wilsenach, 2015). A second error is to advocate – both in schools where English is used as the medium of instruction from Grade 4 and where it is used as the LoLT from Grade 1 – that “a substantial amount of time” be “devoted to learning English in the Foundation Phase” because this develops “a strong literacy foundation in the Home Language” (DBE, 2011a, p. 9). This kind of distorted reasoning is further compounded by the following admission in the mentioned CAPS documents for the rest of the school grades:

In South Africa, many children start using their additional language, which is often English, as the Language of Learning and Teaching (LoLT) in Grade 4. This means that they must reach a high level of competence in English by the end of Grade 3 … In the Intermediate and Senior Phases … the majority of children are learning through the medium of … English …Greater emphasis is therefore placed on using the First Additional Language for the purposes of thinking and reasoning … By the time learners enter Senior Phase, they should be reasonably proficient in their First Additional Language with regard to both interpersonal and cognitive academic skills. However, the reality is that many learners still cannot communicate well in their Additional Language at this stage … (DBE, 2011b, pp. 8–9)

It is clear that the use of English as the medium of instruction is not working well since a large number of learners in the Senior Phase (Grades 7–9) remain unable to express themselves in English, despite their early exposure to the language in the Foundation Phase (Grades 1–3) and Intermediate Phase (Grades 4–6). At the same time, the Home Languages – the languages that children first acquire – have been overlooked as useful languages for teaching and learning purposes. SGBs and parents also have a say in the matter and are allowed to decide the language policy of a school. In the majority of cases, the preference for using English as the LoLT as early as possible dominates, regardless of the above confession in the curriculum document. By this stage, it is clear that there is still confusion as to the application of the LiEP and insufficient evidence that the preference for English is contributing to successful education outcomes.

3 How the Department of Basic Education Monitors Progress

It is common knowledge that public-school education in South Africa still has its limitations. The quality of schooling varies vastly between urban and rural settings and along socioeconomic lines. The matter of school-leavers’ competency and preparedness to participate in the economy – part of the social mobility referred to earlier – remains questionable (Chisholm, 2005; RSA, 2019; Solidarity Research Institute, 2015). School graduates’ abilities serve as independent external indicators that inadequate learning is taking place in public school education. To obtain an overall picture of school education and developments on the language in education front, we now turn our attention to how the DBE monitors progress in public schooling through a series of projects, surveys and reports.

The website of the DBEFootnote 3 provides a plethora of documentation. Duplication of content across the different categories necessitates a selection of items considered to be most relevant for the purposes of the current discussion. The following documents were selected owing to their comprehensive nature and currency:

2010:

The Status of the Language of Learning and Teaching (LoLT) in South African Public Schools: A Quantitative Overview (2010), in conjunction with the Organisation for Economic Co-Operation and Development (OECD)

2011:

Integrated Strategic Planning Framework for Teacher Education and Development in South Africa 2011–2025

2014/17:

National Education and Evaluation Development Unit (NEEDU) Reports

2018:

Teacher Professional Development Master Plan 2017–2020

2018:

Teachers and Principals as Lifelong Learners (TALIS): South Africa Country Report Volume I

2019:

National Senior Certificate (NSC) Grade 12 exit-level examination and diagnostic reports

2019:

A 25-year Review of Progress in the Basic Education Sector

2020:

Action Plan to 2024: Towards the Realisation of Schooling 2030

We will discuss each of the documents in the sections that follow.

3.1 The Status of the Language of Learning and Teaching (LoLT) in South African Public Schools: A Quantitative Overview (DBE, 2010)

As mentioned in the introduction, decisions about the LoLT are supposed to be aligned with the Constitution and Bill of Rights, South African Schools Act, 1996 and the LiEP. It was thus fitting for the DBE to undertake a study on the status of the LoLT, even if only a little more than a decade into the new political dispensation. Of primary importance in the report on the LoLTs is the question of what languages are used in the classroom, as well as the number of single-medium schools in existence. The first part of the document reiterates the constitutional right to be educated in the official language of choice where “reasonably practicable” (DBE, 2010, p. 6) and the obligation on the state to attempt at all costs to “promote the exercising of this right, including the establishment of single medium institutions” (ibid. p. 6). Further to this, the right of SGBs to decide the language policy of their schools is affirmed. The connection between mother-tongue education and academic success is foregrounded, and a caveat issued against a situation in which the first or “home language” is rejected as a language of learning and teaching, leading to the stagnation of the development of that language and the undermining of a student’s “personal and conceptual foundation for learning” (ibid. p. 5). These are significant statements that should continue to steer education planning. Unfortunately, the opposite has happened, hence the current struggle to attain academic language proficiency and the stagnation of the Sintu languages as intellectual tools.

The rest of the overview provides information obtained from an Annual Schools Survey and data provided by the Educational Policy Unit of the University of the Witwatersrand in 2007. From the information reported, we already see an alarming discrepancy between first/home language and LoLT in the early grades. Whereas there were only 5.6% Grade 1 learners with English as first or home language in 2007, as many as 21.8% of Grade 1 learners had English as the LoLT that same year (Table 1).

Table 1 Percentage of Grade 1 learners by Home Language (HL) and LoLT in 2007 (DBE, 2010, pp. 12, 14)
Full size table

We can see that the number of HL Afrikaans students corresponds well with the number of students who had Afrikaans as the LoLT in Grade 1. However, there is a discrepancy between the number of HL English learners and those using English as the LoLT in this important foundational year of schooling. As far as the medium of instruction in Grades 2 and 3 is concerned, English features most prominently, despite the mismatch with Home Language demographics (DBE, 2010, p. 16). Table 2 illustrates the dominant LoLT in all school grades in 2007.

Table 2 Percentage of learners by LoLT in 2007 (DBE, 2010, p. 16)
Full size table

The sudden change to English from Grade 4 as the LoLT is evident. A particularly alarming finding mentioned in the report is that although English and (to a lesser extent) Afrikaans were the dominant LoLTs from Grade 4, the majority of learners did not actually study English or Afrikaans as a school subject in Grades 1–3 before transitioning to these languages (DBE, 2010, p. 29).

Unfortunately, there are no subsequent reports on the status of the LoLT in order to compare the current situation. The Annual Schools Survey report covering the years 2010 and 2011 only mentions the total number of learners and their preferred LoLT; this is not helpful for the purposes of making comparisons and detecting trends.

There were 6000 single-medium schools in 2007 (Table 3) and 13,000 parallel-medium schools. The remainder (about 6532) offered different combinations of LoLT, especially in the Foundation Phase (DBE, 2010, p. 28).

Table 3 Number of single-medium schools by LoLT: 1998–2007; 2010–2011
Full size table

Although the document cites the Annual School Survey (ASS) as the source of the above statistics, no further ASS reports could be found after that for 2010/2011, published in 2013 (DBE, 2013, p. 17). We do find some more recent data on Grade 6 learners from a study by the Southern and Eastern Africa Consortium for Monitoring Educational Quality (SACMEQ) in 2017.

It is clear from Table 4 that most Grade 6 students do not receive adequate exposure to English at home for the purposes of using it as the LoLT, since around 75.8% use English at home only occasionally or never. Furthermore, even if learners are exposed to English at home, the level or quality of the language cannot be verified.

Table 4 Distribution of Grade 6 learners according to the frequency of speaking English at home in 2017 (DBE, 2017a, p. 18)
Full size table

From the information available on the LoLT, we can see how the preference for English as the medium of instruction was already firmly entrenched in 2007, despite the demographics of the student population. However, the competency of teachers to use the LoLT is also intimately connected to the quality and success of schooling. This is another worrying aspect. The first comprehensive document on the matter of teacher development appears to be the 2011 strategic planning framework published jointly by the DBE and the Department of Higher Education and Training (DHET). This document will be reviewed next.

3.2 Integrated Strategic Planning Framework for Teacher Education and Development in South Africa: 2011–2025 (DBE & DHET, 2011)

The objectives of the planning framework are to address the failure of the education system “to achieve dramatic improvement in the quality of teaching and learning in schools” by 2025 (DBE & DHET, 2011, p. 1). This admission by the DBE and DHET that current educational policies are not having the desired effect of ensuring quality of teaching is a positive step on the road to providing opportunities for further teacher development. The planning framework foregrounds “teachers’ poor subject matter knowledge and pedagogical content knowledge” (DBE & DHET, 2011, p. 4). The framework also refers to the “poor public image of teachers, and the status currently ascribed to the teaching profession” (DBE & DHET, 2011, p. 11). We believe this is partly related to the language proficiency of the teachers and the extent to which they can be considered articulate for teaching purposes in more than one language.

A search for language or linguistic variables in the strategic plan revealed only a few references. Referring to the NSC examination results and Annual National Assessments (ANA), the authors of the document state that priority should be given to short developmental courses for teachers of African Languages [read: Sintu languages] as Home Languages and also of English (as First Additional Language), but only for those teaching in the Foundation Phase (DBE & DHET, 2011, p. 10). This provision does not include the training of teachers in higher grades or those who teach Afrikaans as Home or Additional Language. Once again, the emphasis falls on English and those who teach it as an additional language. There is a vague reference to possibly including teachers of other subjects later. In any event, it is questionable whether the proposed short language courses will have the desired effect as language development is not a matter that can be attended to quickly or easily.

The two education departments responsible for the planning framework appear to have overlooked the role that language plays in the teaching and learning of all subject matter; by improving English L2 teaching, in their opinion, the quality of teaching across all other school subjects will improve simultaneously. We know this is inaccurate: there have been numerous reports of teachers’ inadequate English language skills (CDE, 2015; Du Plessis, 2020; Du Plessis & Els, 2019; Grosser & Nel, 2013; Nkosi, 2015). It is also strange that the DBE does not mention the teaching of Afrikaans and whether this is of a satisfactory standard.

The proposed framework for teacher development does refer to the policy on the “Minimum Requirements for Teacher Education Qualifications” – commonly referred to as MRTEQ. This document sets minimum standards for education qualifications to guide training institutions on the knowledge and practical skills that teachers need in order to be professional and effective (DBE & DHET, 2011, p. 15; also see RSA, 2015). Regarding minimum language requirements, section 8.2 of MRTEQ attempts to cover the matter of language proficiency:

All teachers who successfully complete an initial professional qualification should be proficient in the use of at least one official South African language as a language of learning and teaching (LoLT), and partially proficient (i.e. sufficient for purposes of basic conversation) in at least one other official African language [sic], or in South African Sign Language, as language of conversational competence (LoCC). If the LoLT is English or Afrikaans, then the LoCC must be an African Language [sic] or South African Sign Language. All new certificates are to be endorsed to indicate the holder’s level of competence in specific languages … (RSA, 2015, p. 13)

The matter of “proficient” is left to individual interpretation. “Partially proficient” can even be described as an oxymoron. Other countries use comprehensive frameworks such as the Common European Framework of Reference (CEFR)Footnote 4 to articulate in detail what kinds of knowledge and levels of ability would constitute “proficient”. Another problem is the required pass mark of 50% set for university language courses: can we honestly consider that mark as “proficient”? The question can also be raised whether there are any long-term benefits to be gained from a basic conversational knowledge of a language, as stipulated in MRTEQ. Usually such language courses are offered for a single semester or year. It is unlikely that students will remember much by the time they graduate if the conversational language is a new language. One year of study would definitely be insufficient to attain a working proficiency in the language.

MRTEQ does acknowledge the importance of multilingualism and the role of teachers in facilitating multilingualism. However, here too the bias towards English is clear. All Foundation and Intermediate Phase teachers have to be able to teach English as a First Additional Language (RSA, 2015, pp. 24–25).

There is also little sense in prioritising the teaching of Sintu languages for just the first 3 years of schooling – as the strategic planning framework on teacher development does – and then neglecting this important matter in the subsequent grades. This may be unintentional and related to the fact that there is a shortage of teachers of Sintu languages in the Foundation Phase, but this approach will not do much to improve the standard of language teaching in the higher grades. Our experience of university education students is that they have oral proficiency in the Sintu languages but not written or L1 proficiency, although this is the goal of the national school curriculum. This reflects poorly on the language levels of the teachers.

A subsequent document covering a master plan for teacher professional development, published by the DBE in 2019, shows that the department has no shortage of plans and good intentions. As is typical of DBE documents, this more recent publication foregrounds “literacy/English first additional language for all phases” (DBE, 2019a, p. 6) and a plan to assess teachers’ mastery of English. It mentions that practice standards need to be developed for languages in primary education in general and claims that “extensive” programmes in languages have already been implemented in all of the provinces (ibid. p. 9). Despite all of these master plans, not much appears to have changed. We return to this matter later. Apart from policy and planning documents, the DBE also relies on evaluation reports to monitor progress, especially through the work of the National Education and Evaluation Development Unit (NEEDU).

3.3 National Education and Evaluation Development Unit (NEEDU) Reports

NEEDU was established in 2009 (DoE, 2009; DBE, 2011c) with the remit to function as an independent unit to facilitate school improvement through a system of performance reviews based on empirical research studies. It focuses on all aspects of schooling and not only matters pertaining to language in education. Although it is to function independently, it reports to the Minister of Basic Education (MBE) and is monitored by the Planning and Delivery Oversight Unit (PDOU) of the DBE (Taylor et al., 2014).

Through its empirical research, NEEDU can potentially fulfil a crucial role in reducing inequality in education. The first important finding that we report relates to the promotion of learners from one grade to the next as recorded for the period 2006–2014. Throughput, which refers to the “percentage of learners in any one grade progressing to the next (higher) grade the following year” (Taylor et al., 2014, p. 20), is particularly problematic in Grades 10–12. This is disturbing since promotion requirements in South Africa are extremely low, requiring a pass mark of only “40% in three subjects, one of which is an official language at Home Language level” and 30% in the remaining three school subjects (DBE, 2009, p. 9). Despite the low promotion criteria, around 40% of learners have to repeat grades (Taylor et al., 2014, p. 89). The 2014 NEEDU report also points out that almost half of the learners who enter South African public schools in Grade 1 do not matriculate.

The drop in the number of students from Grade 1 to 12 and troubling throughput rates are indicative of a system that is not performing well. There are numerous reasons why students drop out of school, most of which fall beyond the scope of this chapter. However, the NEEDU report contains several references to the matter of language in education.

Section 3.2 of the 2014 report deals specifically with the LoLT. The prescribed school curriculum, CAPS, makes provision for either English or Afrikaans to be used as the LoLT and for the NSC school-leaving examination. However, teachers are reported “to resort to other languages where the learner, or both learner and teacher, have a better command” (Taylor et al., 2014, p. 41). At least 80% of secondary school learners study English as an additional language and do not come from backgrounds where English is used; teachers, too, are not L1 users of English (ibid., p. 41). The authors express their concern that the code-switching used in classes does not support the mastery of English and that high levels of language proficiency are needed in order to engage higher cognitive processing. In fact, “poor levels of English proficiency are undoubtedly a major – if not the largest single – cause of learners dropping out before reaching Grade 12, failing to pass the NSC, and of not completing their tertiary studies” (ibid. p. 23). Strangely, NEEDU supports the decision to introduce English in the Foundation Phase to address this problem, together with “training in English” for teachers by the British Council. A pilot project called LEAP (Learn English Audio Pilot) is being rolled out in some schools in order to improve listening and speaking skills in English. We remain dubious that this will rectify matters since a similar project using radio broadcasting was not sustainable after 2009 (see Potter & Naidoo, 2012).

A later NEEDU report, on high-performing schools located in poor areas, found that the positive culture of learning in such schools assisted them to attain good results. An interesting point mentioned in this particular study is that teachers believed the “discrepancy between language spoken at home and the language of teaching and learning at school” had a negative effect on learners’ “socio-emotional characteristics” and consequently also on their academic success (DBE, 2017b, p. 70). This is why an appropriate response to language in education demands careful consideration of multiple factors, rather than simply reaching for English as the answer. The same NEEDU study also identified the need to communicate with parents “in the language they understand” (ibid. p. 86); one of the reasons for lack of parental involvement in schools was parents’ inability to access the language of meetings and correspondence. Parental support was highlighted in the study as being another determinant of academic success (ibid. p. 170). However, if there is a clash between the parents’ language in the home and that of the school, this potential valuable support is lost.

The next document we examine involves international benchmarking of teaching and learning through participation in a survey initiated by the Organisation for Economic Cooperation and Development (OECD).

3.4 Teaching and Learning International Survey (TALIS) 2018: South Africa Country Report (DBE, 2018)

TALIS, the largest survey of its kind, investigates aspects such as sociodemographics of the teaching profession, instructional practices, teacher development, motivation and fears, which if overlooked “can lead to tensions and policy discord, which can undermine education reform” (DBE, 2018, p. 2). The DBE is to be commended for being the only African country to participate in this global initiative of the OECD which “affords teachers and principals a voice on educational policy analysis and development in key areas” (DBE, 2018, p. 11).

Amongst the key findings of the 2018 survey was that on average about 60% of South African teachers worked in schools in which more than 10% of the learners received instruction in an additional language, a much higher share than in other OECD countries participating in the study (ibid. p. 21). Closely related to linguistic and literacy challenges was the finding that 70% of South African participants in the survey reported a shortage of library materials, as compared to the OECD average of 16%, and that 71% of teachers worked in schools in which more than 30% of the learners were from poor socioeconomic backgrounds; the OECD average was 20% of teachers (ibid. p. 17). Here too we see the dire consequences of poverty.

A disturbing finding of TALIS was that around 56% of the teachers in South Africa had only completed a short tertiary programme (the OECD average was 3% for short programmes); a quarter of teachers had no tertiary qualifications as compared to the OECD average of 2% (DBE, 2018, p. 18). The practice of appointing teachers who are not appropriately qualified – and then attempting to remedy matters along the way – cannot be condoned. Consideration should be given to evaluating the standard of work of unqualified teachers and their suitability for further training to obtain the necessary qualifications within a stipulated period of time. Incentives for training in other careers could be investigated to allow gradual transition to alternative employment opportunities. South Africa cannot afford to keep unsuitable teachers in the classroom.

One surprising finding of TALIS was that only around 20% of South African teachers believed that they needed professional development to teach in multicultural/multilingual settings (ibid. p. 55). We would argue differently: too many teachers’ language skills are inadequate to assist their students and they do not serve as good language role models in the classroom, even though they may be of the opinion that they are fluent in more than one language and capable of handling multilingual teaching modes. We base our view on analyses of education students’ performance in language programmes at various training institutions (Du Plessis & Els, 2019; Grosser & Nel, 2013; Mhlongo, 2019 Van der Merwe, 2018).

An important point raised in the TALIS report is the consensus amongst researchers that “teachers and school leaders shape the quality of instruction, which strongly affects students’ learning and outcomes” (DBE, 2018, p. 25). This means that it is not enough to rely on monitoring units and programmes to ensure quality education: principals and teachers must have the required qualifications and competence before being appointed. The same of course could be said about officials in the DBE tasked with various monitoring and education responsibilities and the extent to which they have suitable linguistic and other qualifications.

The next part of the document appraisal covers learner performance trends in the annual school-leaving examination and what we can learn from the diagnostic reports.

3.5 National Senior Certificate (NSC) Examination Results and Diagnostic Reports

The DBE places a high premium on the annual results of the Grade 12 NSC examination as a means of tracking learner performance over subsequent years. Figure 1 charts education progress through examination results over the period 2008 (when the new school curriculum was introduced) to 2019.

Fig. 1
A bar graph plots the percentage versus year. It has a fluctuating trend, with 2009 having the lowest at 60.6 percent, while 2019 plots the highest at 81.3 percent.

National pass rates in the annual NSC school-leaving examination. (DBE, 2019b, p. 6)

Full size image

There appears to have been an improvement in the overall results since 2008. However, the 2008 and 2009 results were based on the previous school curriculum, not CAPS, and the respective examination papers are not necessarily of the same degree of difficulty across the different years. It is therefore difficult to make a case for education improvement based on the NSC results alone. Moreover, the low requisite pass mark of 30% obfuscates matters. There is furthermore the problem of highly unreliable school-based continuous assessment marks (37.5% of the overall mark) that contribute a substantial proportion of the NSC examination pass marks.

The 2019 NSC examination report highlights the “vacation school programmes” offered during the three periods of school holidays as the main programme to support learning in Grade 12:

The programme targets a diverse set of learners including progressed learners, learners at risk of not achieving the NSC and learners that have the potential to achieve distinctions in various subjects in an effort to focus on quality improvement. (DBE, 2019b, p. 21)

As many as 40% of the 2019 Grade 12 intake attended the vacation classes (offered through direct contact teaching or other platforms). Although this programme appears to have assisted many learners, it should not become a replacement for quality classroom teaching during the school term (teacher absenteeism averages at around 10% per day). Of particular interest to us is the fact that only English as First Additional Language was included in the vacation programme; the report contains a vague reference to extra tuition being expanded for the home languages, but no details are provided. There is still no indication of any support for the “African language” subjects (confusingly including Afrikaans, which is the language with the third most speakers in the country, most of them not Caucasian), either as L1 or L2. The bias towards English and definite lack of equivalence of standard across the different school language subjects remains, despite the disparities identified in numerous research studies (Du Plessis & Du Plessis, 2015; Weideman et al., 2017). It is not surprising to note the following “areas of concern” and recommendation in the 2019 NSC diagnostic report:

  • In most home languages, the vast majority of candidates either misinterpreted or gave limited responses to higher order questions … There is therefore a need to enhance thinking in an abstract context in languages.

  • In most languages, candidates did not understand the vocabulary used in comprehension texts … Vocabulary exercises and reading need to be promoted in schools.

  • A large percentage of candidates displayed a limited understanding of subject matter.

  • More emphasis needs to be placed on language competence since candidates often can respond correctly to questions but lack the language skills to do so. (DBE, 2019c, pp. 13–14)

The roots of these problems go back to the Foundation and Intermediate Phases and the issue of the LoLT. Despite having had at least 10 years of exposure to English, both as a school subject and as the medium of instruction, by Grade 12 many students still cannot express themselves in English. This is a recurring refrain in our discussions on education progress in this chapter. It is clear from the 2019 diagnostic report that the same concerns about learners’ English language mastery were of relevance to the remaining school language subjects. It seems that the emphasis on English has not done much to improve the students’ English and, in addition, has had a detrimental effect on the students’ competence in their first or home languages.

The next two documents we will discuss provide an overview of progress over a lengthy period and serve as a means of correlating findings and problematic aspects already identified so far.

3.6 A 25-Year Review of Progress in the Basic Education Sector (DBE, 2019d)

This report was published to mark the 25th anniversary of the democratic dispensation. Amongst the achievements hailed are the gradual increases in completion of primary and secondary schooling over the period 2002–2017 (Fig. 2).

Fig. 2
A line graph plots the percentage versus year. Grade 7 completion plots are higher than grade 12, with both graphs having increasing trends. In 2002, grade 7 and 12 completion plots were the lowest at 85 and 40 percent. By 2019 both plotted 95 and 50 percent, respectively.

National primary and secondary schooling completion rates. (DBE, 2019d, p. 4)

Full size image

It can hardly be considered an achievement when the completion rates over a 15-year period have only increased by around 10%. Moreover, the completion ages of learners are alarming: 16–18 for Grade 7 and 22–25 for Grade 12. The normal completion age for Grade 7 is 12 years and that for Grade 12 is 18–19 years. This is the same tendency we see in higher education where many students take on average 6 years to complete a basic 3-year degree programme (Du Plessis, 2020). Here too, we believe that the way the LiEP is being applied in schools to determine the LoLT is part of the problem.

The 25-year review applauds the following achievements, but acknowledges “the absolute levels of learning achieved are still substantially below desirable levels” (DBE, 2019d, p. 8):

The good news is that in recent rounds of TIMSS, PIRLS and SEACMEQFootnote 5 we have observed that the country’s levels of learning have been on an improving trend. In the TIMSS assessment (grade 9 mathematics and science), South Africa has been the fastest improving country between the surveys of 2002, 2011 and 2015. There appears to have been a significant improvement in the country’s PIRLS results between 2006 and 2011, although no significant change between 2011 and 2016. In SEACMEQ, a large improvement at the grade 6 level was noted between 2007 and 2013 in both mathematics and reading.

It is a pity that more recent data was not included. The current performance levels in the mentioned external tests are thus uncertain, but the reported improvements up to 2016 are worth noting. On the matter of learning and teaching support materials (LTSM), the review states that almost 100% of schools have been provided with textbooks and workbooks for each learner (DBE, 2019d, p. 33). This does not, however, include graded readers in home languages (ibid. p. 37). The quality of books is said to need further attention, especially in the Foundation Phase. Of concern to us is the fact that no mention is made in the 25-year review of how many schools have libraries. In 2010, only about 40% of public schools had some form of a library (ibid. p. 23). A mobile library system in 2013/14 had assisted about half a million learners, but this is not the best solution, especially in view of the promulgated minimum norms and standards for public school infrastructure which state that all schools must have libraries (ibid. p. 24). Very little progress appears to have been made here. Without well-equipped libraries and regular access to reading materials in printed and electronic format, students stand little chance of improving their language proficiency and skills, and the SDG goal of inclusive education in an increasingly digital era cannot be attained.

Notwithstanding our concerns, the DBE should be commended for participating in projects such as PIRLS, TIMMS and SEACMEQ. These serve as credible national and international benchmarks that help to identify trends in learner achievement in the areas of literacy and mathematics and are valuable tools to monitor education quality in respect of SDG 4.

3.7 Action Plan to 2024: Towards the Realisation of Schooling 2030 (DBE, 2020)

This voluminous document incorporates key elements from the National Development Plan (NDP) and consolidates the commitment of the South African government to the United Nations’ SDGs. It provides a summary of the historical origins of inequality in education and reports on planning to improve the quality of education through five priorities pertaining to early childhood development and foundational literacy, teaching professionalism, learning materials, school management, and school monitoring and support (DBE, 2020, pp. vii–viii). Much of the content of previous reports already discussed in this chapter forms part of the action plan and will not be repeated here.

We applaud the strides taken to ensure uniform access to education at all levels, improve the school curriculum, and introduce innovative assessment initiatives such as the planned comprehensive Systemic Evaluation Programme. When fully operational, the programme will enable the assessment of proficiency levels of Grade 3, 6 and 9 learners in language and mathematics every year (DBE, 2020, p. 25) in the place of the Annual National Assessments (ANA). This system could provide reliable data on trends at both provincial and national level, and also relate achievement of learning outcomes to socioeconomic realities, an important contextual element (ibid. p. 49). If the DBE manages to implement the Systemic Evaluation Programme – with the support of teachers’ unions – it would place South Africa on par with other Southern African Development Community (SADC) countries that already have international competency benchmarking tests in place.

Inasmuch as the action plan shares a vision for a “modern and decolonised schooling system” (DBE, 2020, p. v), we can see that there is still reluctance to change the ruling party’s position on the colonial language of English. What we find troublesome in the action plan is the wording used in the following statement: “Apartheid brought with it prolonged segregation by race, but also language, with a ferociousness not seen in any other country during the twentieth century” (ibid. p. 4). This is a highly subjective statement in which language, together with race, is blamed for segregating people. It is the same argument used to justify English as the dominant LoLT in schools and universities: it is perceived to be the language that can “enhance diversity in classrooms” and help to obtain “higher paid jobs, particularly in postcolonial countries where government jobs require the colonial language” (Eriksson, 2014, p. 2).

The ideological bias of government is also evident in the misapplication of research findings that effective language instruction in the first or home language in the early grades supports the learning of English and development of literacies later (Taylor & Coetzee, 2013; Wildsmith-Cromarty & Balfour, 2019). This important point, which would be applicable to the learning of any additional language, is misconstrued by the DBE as “the switch, between grades 3 and 4, from an African language [read: Sintu language] to English across most of the system remains supported by research” (DBE, 2020, p. 24). This bizarre conclusion comes directly after a series of comments in the action plan on the disadvantage of learners who have to “begin learning in an unfamiliar language in Grade 4, mostly English” (ibid. p. 24). The following extract from page 6 of the action plan is almost schizophrenic:

Around the world, much of the legacy of colonialism persists through the dominance of colonial languages. In South Africa, English, though only spoken by about 4% of public school learners as a home language, is the predominant language of the textbooks used in classrooms, as well as in the system’s policy documents. The history of marginalisation of the remaining official languages and, in particular, of the country’s nine African languages [sic] continues, despite the official position of equality between the languages as enshrined in the 1996 Constitution. The schooling system needs to pay special attention to the promotion of all official languages. Compelling research indicating that young children learn best if, during the first few years of their schooling, key concepts are taught in their home language, informs South Africa’s education policies. But beyond these pedagogical considerations, promoting all languages in the education system is a matter of national pride and of liberation [emphasis provided in original text]. (DBE, 2020, p. 6)

With regard to those schools who opted to use English as the LoLT from Grade 1 when it was not the first or home language of the students, Taylor and Von Fintel (2016, p. 77) found “a negative effect on English performance in grades 4, 5 and 6”. Based on our analysis of language and literacy issues mentioned in the documentation studied, we see little commitment to promoting languages other than English in education. Furthermore, we note a two-pronged stance in another section of the action plan that elaborates on the notion of social cohesion:

The plan envisions a South Africa where everyone feels free yet bounded [sic] to others; where everyone embraces their full potential, a country where opportunity is determined not by birth, but by ability, education and hard work. (DBE, 2020, p. 10)

The reference to birth serves as a proxy for race and language. When read on its own, the above statement may appear laudable. However, section 3.4 of the action plan deals with curriculum innovation in order to achieve “radical economic transformation” (DBE, 2020, p. 16) and advocates broad-based black economic empowerment (BBBEE). In these terms, Black persons are given preferential treatment for government funding initiatives and employment opportunities. This goes directly against the ideals of inclusivity and equality and ignores persons of other population groups and mixed descent who also suffered under apartheid, for example citizens who are classified “Coloured” in terms of the current government’s affirmative action policies. So far the BBBEE policy has done very little to uplift the majority of Black persons in the country; they continue to struggle with poverty. No wonder the “BEE” is now referred to by critics as Black Elitist Empowerment.Footnote 6

In a study on the effect of the LoLT used in primary school on labour market outcomes, Eriksson (2014) shows how language in education could potentially be used for economic advancement – in a way that in our opinion would not prejudice persons on the basis of race as in BEE. She correlates “long run effects” of language policy changes on income and educational outcomes by using “difference-in-difference estimation strategy” (ibid. p. 4) based on years of instruction in the mother tongue. Eriksson (ibid. p. 3) reports that when education in the mother tongue for Black students was increased from 4 to 6 years (in accordance with a change in policy in 1955), the effects on wages were positive, resulting in more educational achievement and “higher labour market outcomes” (ibid. p. 23). She also found evidence of higher English speaking proficiency owing to the policy, but “only in predominantly English parts of the country” (ibid. p. 3). This evidence illustrates the importance of taking the immediate community context into consideration when determining the LoLT of a school. Although Eriksson’s study used data from the Bantu Education era and the 1980 census, it highlights the importance of careful consideration of the choice of LoLT in primary school education. It also suggests that a differentiated approach to applying the LiEP in schools is needed. We discuss this possibility in our concluding comments.

4 Conclusions

Our review shows ongoing efforts on the part of the DBE to provide South African learners with quality education that indeed is inclusive and equitable. The attention devoted by the department to the monitoring of education progress through various programmes is also commendable. Unfortunately, such endeavours have not led to substantial gains. In this regard, the work of NEEDU is of crucial importance, both because of the relative independence of the unit, and also in terms of the solid empirical basis of its research.

The finding of the TALIS survey (DBE, 2018) that only about 56% of teachers had completed a short tertiary programme, and about 25% had no tertiary training at all, in effect means that it will take a long time to improve the quality of teaching. We would like to commend the DBE on its planned Systemic Evaluation Programme and for participating in projects such as PIRLS, TIMMS and SACMEQ to identify trends in learner achievement with a view to the attainment of SDG 4. However, it is clear from the voluntary review submitted to the UN that government remains perturbed about the fact that South African learners who attend public schools are not acquiring adequate skills, and that far too many youths do not complete their basic schooling. Although the DBE has reported an upward trend in national pass rates in the NSC school-leaving examination since 2008, we cannot attach much value to a system where the pass mark is 30%, and we hence cannot regard the NSC as a reliable indicator of education progress.

Given the findings of our review, we believe that by applying a differentiated approach to language in education, the social mobility and future prospects of many school students can be improved significantly. There is still insufficient recognition of our multilingual and multicultural context in education planning. Linguistic diversity should be a prominent feature of our curriculum and schooling system. The encouragement of the DBE to transition to English in Grade 4 (or even earlier) has derailed the learning of many, both in the crucial foundation phase of schooling and beyond. The repercussions are felt at tertiary level. The report from the National Benchmark Tests (NBTs) Project for the 2018 intake year shows that students who study in their first or home languages at school, such as Afrikaans first language learners, are more adequately prepared for university and tend to perform better in placement tests (CETAP, 2018, p. 39). This finding is asserted repeatedly in consecutive annual reports.

Clearly, the bias for English on the part of the DBE and governing bodies of schools cannot be deemed suitable for all school contexts, especially in areas where English is hardly used in the community. By promoting English as the LoLT in the majority of public schools irrespective of context, the DBE ignores the close connection between learners’ identity and culture, first language and conceptual growth. Inasmuch as we support the right of schools to determine their own language policies, a framework should be devised to assist SGBs to do this in a way that supports learners’ cognitive and language development. School language policies cannot be determined on the basis of popular, preconceived ideas and historical prejudices. The fact that learners are still not proficient in English by the time that they commence the senior phase of school, despite English having served as the LoLT for 5–7 years (and that has been the case since 1994), shows that a different dispensation is (desperately) needed for both learners and teachers.

It is significant that the position of Afrikaans as LoLT does not feature prominently in the reports and publications of the DBE (if at all). We believe the reason for this is the good performance of Afrikaans-medium schools owing to the close alignment of teachers and learners’ first language, the language of the community and home, and the medium of instruction – elements that facilitate the development of essential literacies and language mastery. The annual overview of matric results by the Solidarity Schools Support Centre reports that not only did Afrikaans schools contribute to a higher pass rate for the 2020 examination and perform proportionally far better in mathematics when compared to the majority of schools using English as LoLT, 12 of them are among the 20 schools with the most distinctions in the country (SOS [Solidariteit Skoleondersteuningsentrum], 2020, pp. 14–16). As in the case of the NBT reports, this finding is also of a recurring nature.

In order to ensure that the LoLT is beneficial for learners, multiple factors should thus be taken into account as part of a framework to determine a suitable language policy for schools. These include aligning the LoLT with the languages to which learners are frequently exposed in their homes and communities, adopting multilingual approaches to teaching, cognisance of the availability of literature and resources in the home, school and community, and the socio-demographics of learners, teachers and parents.

By this stage, it is clear that the LiEP is not being applied as intended in schools. The mammoth initiative of the UN SDG Agenda 30 rests on fundamental principles of international law and human rights conventions. If anything, the past months of the COVID-19 pandemic have foregrounded disparities and inequalities in societies, confirming that other measures are needed to spur governments into action to attain the SDGs. Without some form of accountability on the part of member states, it is doubtful that countries such as South Africa will start doing things differently, other than submitting reports that do not necessarily interpret correctly some of the studies they are based on. In order to pursue the language policy most supportive of general conceptual growth amongst South African learners, additive multilingualism as envisaged in the LiEP must be employed. It is unacceptable to suggest, as the country’s 2019 report to the UN states, that sectoral initiatives “should be considered” to increase “the use of African languages [read: Sintu languages] in lower school grades” (p. 13). There has been enough “consideration” over the past 25 years; we could have achieved far more had government and schools actually implemented the original policy of additive multilingualism, and had universities ensured that teachers had high proficiency levels to handle multilingual classroom contexts in the interests of inclusive, equitable and quality education.