Before Act 200 of 1993 (known as the Interim Constitution of South Africa) was passed, Afrikaans and English were the only official languages in South Africa. This Act changed this scenario by giving official status to the following nine Black South African languages or BSALs (in alphabetical order): isiNdebele, isiXhosa, isiZulu, Sesotho sa Leboa, Sesotho, Setswana, siSwati, Tshivenda and Xitsonga. Act 200 of 1993 was replaced by Act 108 of 1996 (the current Constitution of the Republic of South Africa [Republic of South Africa 1996]), and the official status of 11 languages was retained. As Rakgogo and Van Huyssteen (2018:79–80) noted, the only change between the two constitutions was that Sesotho sa Leboa was referred to as Sepedi in the latter Act.

Mutasa (2003:290) observed that although BSALs have a place in the new South Africa as official languages, their home language (HL) speakers have a high regard for English. He further indicated that many HL speakers still believe that only English can effectively serve as an official or national language. Mutasa (1999:86) demonstrated that most linguistic communication in domains of national significance remains in English, and to a lesser extent, Afrikaans. Such communication patterns have created an impression that to be part of the national economy, one needs to have communication skills in these two languages, and not in the nine BSALs.

Section 6 of the Constitution of South Africa (1996) recognises that the nine BSALs are not as developed as Afrikaans and English, therefore they need to be institutionally supported. One measure to institutionally support them was the passing of Act 12 of 2012 (the Use of Official Languages Act), whose Preamble states that the State should take practical and positive measures to elevate the status and advance the use of these nine BSALs. In taking these measures, the Act tables these four objects:

Section 3(1) of this Act states that it only applies to national departments, national public entities and national public enterprises. In other words, it does not apply to provincial departments and municipalities.

The main objective of this article is to explore how multilingualism is practised (if any) at national departments in South Africa since the passing of new language legislation called the Use of Official Languages Act. To support this main objective, the article seeks to establish the attitudes held towards language and official multilingualism by national departments’ employees responsible for matters related to language and communication. It also seeks to establish the general public’s perceptions on how public servants treat language when communicating with them. As such, the article poses the following research questions:

A mixed-methods approach was used to gather the data and that was informed by Creswell (2015) who suggested:

A questionnaire and interviews were used to gather the data. The questionnaire was used to gather quantitative data (i.e. a Likert-type scale comprising 12 belief statements related to language use) and qualitative data (i.e. five open-ended questions), and it was completed by employees (see Appendix Table 1 for their profiles) at two national departments (viz. ND1 and ND2). More qualitative data were gathered through face-to-face interviews with members of the public (see Appendix Table 2 for their profiles). Eight semi-structured questions were posed to the senior manager, and another eight semi-structured questions to members of the public.

Additional qualitative data were gathered through document analysis, that is, by reading the two national departments’ language policies. Maree (ed. 2007:82) submitted that ‘when you use documents as a data gathering technique, you will focus on all types of written communication that may shed light on the phenomenon that you are investigating’. Reading through the language policies (document analysis) was important because it helped the authors understand the national departments’ responses to the implementation of the Use of Official Languages Act, as well as how their employees viewed official multilingualism because until this Act was passed, there were no specific legal requirements or obligations to practise multilingualism. Put differently, national departments exercised their discretion to support Section 6 of the Constitution of South Africa (1996) in recognising that the nine BSALs were not as developed as Afrikaans and English, thus they needed to be institutionally supported. The reading of the two national departments’ language policies shaped the instruments used, that is, the belief statements in the Likert-type scale, as well as face-to-face interview questions.

The researchers intended to gather data from at least four national departments in Pretoria, and from between 24 and 32 employees (i.e. between six and eight per national department), they therefore targeted national departments with a mandate of providing a direct service to members of the public (e.g. health, social services, water, etc.) albeit through structures such as provincial departments. The researchers followed all the processes stipulated by the national departments they approached, including providing them with an ethical clearance letter. This proved to be a very challenging task on two fronts.

Firstly, senior managers at national departments were suspicious of the researchers’ intentions as they suspected that the researchers intended to expose them for not being where the parliament expected them to be regarding the implementation of the Use of Official Language Act, thus refused to allow the researchers access to their employees, despite being assured that the researchers would use code names (e.g. ND1, ND2, ND3, etc.) to ensure that their identities were not disclosed.

Secondly, for the national departments that granted the researchers permission to access their employees, many potential participants (i.e. employees responsible for matters related to language and communication) who initially showed enthusiasm withdrew their participation as soon as they saw the questionnaire. They were assured that code names (e.g. M1 for first male participant, F4 for fourth female participant, etc.) were to be used to ensure that their identities were not exposed. However, they declined, and the common reason was that they feared victimisation, which could see them losing their jobs.

In the end, eight employees from two national departments agreed to complete the questionnaire. Purposeful sampling was thus used because the nature of this research was that national departments’ employees were best placed to provide information needed to achieve its main objective, an approach supported by Struwig and Stead (2001:122).

Purposive random sampling was used to select participants drawn from members of the public. These participants were approached outside government facilities which serviced members of the public (e.g. health, social services, water, etc.). Palinkas et al. (2015:336) argued that purposeful random sampling increases the credibility of a researcher’s findings. Approaching potential participants after they had interacted with officials was important because had they been approached before they received services that could have influenced them in how they interacted with officials and that would have contaminated the data.

The study was granted ethical clearance by the Faculty Committee for Research Ethics (Humanities), Tshwane University of Technology, Pretoria with ethical clearance number FCRE/APL/STD/2017/19.

Data were analysed in this order. Firstly, document analysis (i.e. the correlation between language policies and observations), secondly, Likert-type scale (i.e. 12 belief statements) and thirdly, face-to-face interviews. To organise the analysis of data better or coherently, the following four themes were developed: (1) language development; (2) convenience of English; (3) language practice within government and (4) social cohesion.

Except for Theme 4 (social cohesion) where data analysed under this theme were gathered from employees of two national departments, data for the other three themes were gathered from the employees of these two national departments, and members of the public. That said, data from the two groups of participants are collated and discussed in this section.

The main objective of this article is to explore how multilingualism is practised at national departments in South Africa since the passing of new language legislation called the Use of Official Languages Act (No. 12 of 2012), and to achieve that objective, two research questions were asked and are answered in this section.

With regard to the first research question, the researchers conclude that national departments’ employees generally held positive attitudes towards official multilingualism because they supported the development and use of all 11 official languages, particularly the historically marginalised BSALs. These participants also rejected the idea of an English-only approach in government communication with South Africans, and they believed that language work (e.g. translations, text editing, etc.) should not be done by practitioners employed on a full-time basis by national departments.

Regarding the second research question, whilst members of the public pointed out that communication between them and employees at national departments was mainly in English and in Afrikaans to a lesser extent, they felt that practising official multilingualism would be inclusive of all citizens, particularly those with no communication skills in English and Afrikaans. The researchers thus conclude that these participants fully supported the promotion of official multilingualism in government communication with citizens.

Whilst the two national departments had taken steps to develop their language policies in accordance with the Use of Official Languages Act, their employees’ responses and the authors’ analysis of documents suggested that the two national departments were yet to translate their language policies into practice. As such, the researchers’ major conclusion is that, as far as these two national departments are concerned, the Use of Official Languages Act was yet to be fully implemented as per its objects set out in its preamble, specifically object (b) which says: ‘To promote parity of esteem and equitable treatment of official languages of the Republic’ and object (c) which says: ‘To facilitate equitable access to services and information of national government’.