Does Black Economic Empowerment Ownership Matter? A Decolonial Analysis of “Black Visibility” in South Africa’s Print Media Content, 1994–2014

ABSTRACT

Ownership is identified as one of the major factors that influence the production of news. South Africa’s print media transformation situation presents an important Global South case study for journalism debates about ownership effects on news. The country’s media transformation agenda is based on the premise that transformation of ownership will automatically lead to a transformation of content on these levels. This study empirically examines whether the racial changes in print media ownership facilitated by black economic empowerment in the first 20 years of South Africa’s democracy (1994–2014) led to transformation of content with a focus on racial stereotypes of blackness. It conducts a decolonial analysis of the representation of blackness in three issues that have framed post-apartheid South Africa: socioeconomic rights; labour issues and protests; black government vs. big business or “white economic elite”. The findings show that despite an increase in black ownership, “inferential racism” of blackness pervades content. The nature of representation also coincides with decolonial theory’s concept of “non-being” and its enunciation of the colonised subject being the “damned of the earth”, in the near invisibility of black people’s struggles, hypervisibility of black leadership shortcomings, and in the depiction of protestors as inherently violent, disorderly, deviants, and criminals. The study concludes that in the case of South Africa ownership does not matter, a change from white ownership to considerable black ownership since 1994 did not significantly “transform” historical racist tropes of blackness in content.

KEYWORDS:

• Ownership
• content
• print media
• blackness
• racism
• media transformation
• decoloniality
• South Africa
• Global South

Introduction

South Africa’s media transformation situation is an important Global South case study for journalism debates about the relationship between ownership and content that have not reached consensus. In post-apartheid South Africa “transformation” refers to a broad strategy that addresses the racial and gender injustices of the past, officially regulated by the Broad-Based Black Economic Empowerment (B-BBEE) Act 2003 No. 53 which is a key policy intervention aimed to address the economic disparities of the past and is applicable to companies of different sizes (Govenden and Chiumbu 2020). The Act enshrines a seven-part generic scorecard that elevates black ownership as the highest measure of transformation, the remaining parts speak to the structural aspects of enterprises (e.g. management control). As a member of the corporate sector the media is required to adhere to the B-BBEE Act, and it is the sole policy governing its transformation agenda. Hence the media in South Africa have adopted a black ownership-focused approach to transformation, clearly based on the premise that black and diverse ownership will result in transformed content on these levels. “Media transformation” in South Africa is ideally conceptualised as both ownership diversity and content diversity (Nelson Mandela 1992, as quoted in Lloyd 2013; Boloka and Krabill 2000; Steenveld 2012; Right to Know Campaign 2015 report; Govenden 2019). Scholars have debated and disagreed about this assumption that diverse ownership within the requirements of transformation will lead to transformation of content (see Berger 2000; Boloka and Krabill 2000; Friedman 2011), albeit without an empirical basis. The relationship between the racial changes in the structures of the print media and content is a pertinent yet under-researched area, as Friedman (2011: 108) posits,

While there has been racial change in the media since 1994, whites have retained significant decision making power in the press; whether it is used to perpetuate stereotypes which undermine democracy in a very direct way is an important question that deserves sustained debate.

A paucity of studies have looked at print media transformation since the advent of democracy (e.g. Tomaselli 1997; Berger 2000; Boloka and Krabill 2000; Duncan 2000; Steenveld 2012; Govenden and Chiumbu 2020); however, most of these focused on ownership in isolation.

This paper is also prompted by the print media in post-apartheid South Africa being under the spotlight for allegations of racist content despite this specialised “transformation” agenda in place, that even lead to a South African Human Rights Commission (2000) investigation. From the onset of democracy, Nelson Mandela and other ANC leaders have accused the print media of “continuing racism”, arguing that “the press seemed to delight in stories of black incompetence with an eagerness that betrayed a continuing racism” (Horwitz 2001, as quoted in Radebe 2017: 59). Numerous scholars have also accused the print media of racist practices, for exercising its watchdog role in a biased manner by scrutinising black elites and neglecting the “white economic elite” (Duncan 2009, as quoted in Radebe 2017: 61). Allegations of racism in the post-apartheid print media brought about the South African Human Rights Commission (SAHRC) full-scale enquiry into racism in the media. This investigation was prompted by a complaint to the SAHRC by two black professional bodies, Black Lawyers Association (BLA) and Association of Black Accountants of South Africa, accusing the Mail & Guardian and the Sunday Times newspapers of racially biased reporting and editorial comment (see Pityana 2000; Tomaselli 2000; Berger 2001; Durrheim et al. 2005; Steenveld 2007). The HSRC final report (2000) Faultlines concluded that the South African media was guilty of racism (as quoted in Berger 2001). However, the SAHRC investigation was tainted by accusations of methodological flaws, theoretical weaknesses, and generalisations in both the interim and final reports (Tomaselli 2000). These various accusations of racism by different societal stakeholders bring to the fore questions for the theoretical relationship between ownership and content in the post-apartheid print media that has a “transformation” agenda in place intended to facilitate transformed content through increasing black ownership.

This context prompts the focus of this study. This paper empirically explores whether the racial changes in the media facilitated by black economic empowerment in the first twenty years of democracy (1994–2014) led to transformed content that breaks away from historical racist stereotypes of blackness. It conducts a decolonial analysis of blackness representation in three issues that have framed post-apartheid South Africa that embody a racialised element: socioeconomic rights; labour issues and protests; black government vs. business or “white economic elite”. The decolonial perspective is utilised to illuminate any racist legacies of blackness stereotypes in print media content. This study contributes a unique Global South empirical understanding of the complex relationship between media ownership and content, in a country such as South Africa with a history of legalised racism. De-Westernisation and Decolonial Media Studies scholars have identified a power imbalance in global academic knowledge production, such that the Global North knowledge production is hegemonic and universalised (see Glück 2018). Empirical studies on the effects of ownership on content are Global North dominated. This concept of the Global South has recently emerged in postcolonial and transnational studies (Iqani and Chiumbu 2019), and along with the concept of the Global North are both debated and contested terms. This paper understands the Global South as countries with a shared history of exclusion and oppression whilst Global North refers to countries in Europe and North America and other countries that share sociopolitical characteristics to these Northern countries (Iqani and Chiumbu 2019; Iqani and Resende 2019).

Research background: the empirical relationship between ownership and news content

Ownership is long theorised as affecting news content (Herman and Chomsky 1988; McChesney 2004; Garnham 2011). Empirical research, however, has shown the relationship between ownership and content to be more complex and multifaceted than these linear theorisations. As Wagner and Collins (2014: 760) posit, “there is nothing approaching a universally accepted answer to the question of how ownership affects media content”. It must be noted that the body of empirical studies that explored the relationship between ownership and news content found that ownership does affect content, and notably these studies are dominantly Global North focused mostly from the USA. In some of the few exceptions, studies look at Global South countries such as Uganda and Kenya (Okech 2007; Kiwanuka-Tondo, Albada, and Payton 2012). Only a paucity of studies has documented that ownership does not play a role in effecting news content (e.g. Baer et al. 1974; Spavins et al. 2002).

Most studies have empirically shown that indeed media ownership has noteworthy effects on news content (e.g. Schaffner and Sellers 2003; Hamilton 2004; Dunaway 2013; Wagner and Collins 2014; Rohlinger and Proffitt 2017; Archer and Clinton 2018; Humprecht and Esser 2018). There has been considerable examination of how ownership types such as public vs. private, small vs. large conglomerates, ownership chains, foreign ownership vs. local ownership; effects news with regards to political party coverage, elections coverage, political issues, international conflict, agenda, tone, slant, scope, focus, diversity (e.g. Hamilton 2004; Kiwanuka-Tondo, Albada, and Payton 2012; Dunaway 2013; Humprecht and Esser 2018). However, little attention has been paid to the effect of ownership changes on content that links to the focus of this study; these have shown that it affects political coverage. For example, Wagner and Collins (2014) find a change in support of government policy, as well as attention and tone of major political parties on the editorial page after the Rupert Murdoch’s purchase of the Wall Street Journal. Similarly, Archer and Clinton (2018) demonstrate an increase in the amount of political coverage on the front page in the same Murdoch takeover. Pickard (2015: 3) emphasises the increasing importance of research into ownership effects on news in the digital age,

Even as media ownership becomes increasingly complicated by new digital media technologies, it will remain an important area for scholarly attention as long as media exist … however, much more research that connects media ownership structures with media content and effects is needed, especially scholarship that foregrounds normative considerations.

This study contributes to the generally scarce empirical research area of the effect of ownership change on content, especially in a Global South context.

An overview of the racial ownership changes, 1994–2014

The inception of the print media in South Africa took place in the country’s early European expansion period during British colonialism, it comprised of a mainstream press that was commercial as well as an alternative press (Govenden 2019). In John Mattison’s (2015) book God, Spies and Lies: Finding South Africa’s future through its past, he notes that the country’s first newspaper was conceived in sin and published in Cape Town in the 1800s by a pair of Scottish slave traders. Mattison (2015) also notes that the first newspaper group, the Argus printing and publishing company (now Independent Media) also emerged in this period of colonialism, following great economic expansion because of the discovery of diamonds and then gold in the second half of the nineteenth century. The alternative press also emerged during this era of British colonialism. Switzer (1997) observes that at the beginning of the European mission expansion, publications targeted at Africans were produced and controlled by missionaries but later written and edited by African converts. Switzer (1997: 1) also contends that from the early 1990s publications by other population groups mainly Asian and Coloured were produced and collectively were known as “South Africa’s pioneer black protest press”. Then in the apartheid period, an alternative press also appeared on the sidelines in the form of anti-apartheid and anti-capitalist newspapers (see Tomaselli and Teer-Tomaselli 2008). According to Switzer (1997) the apartheid period’s alternative press featured an early resistance press from the 1930s to 1960s that was bought out, closed down or depoliticised and merged with a new captive black commercial press controlled by white entrepreneurs, as well as a later resistance press in the 1970s to 1980s that represented the Black Consciousness movement and progressive community press. Despite the presence of an anti-apartheid alternative press, Govenden (2019: 51) contends that during apartheid the mainstream press was largely used as, “the main tool to further the agenda of the apartheid government and unfairly excluded the black majority”.

During these periods the mainstream print media ownership patterns were characterised by “white only” racial ownership that excluded the black majority. Colonial-era print media owners included European slave traders and mining big businesses such as: Anglo-American and Johannesburg Consolidated Investments (see Mattison 2015). In the apartheid period “white only” ownership continued with the same European ownership and the addition of “white” local big business ownership. The market however operated as a state-run press, the National Party government was the gatekeeper of newspaper content and stifled the free flow of ideas and ideologies only permitting the ideology of white supremacy (Govenden 2019). In 1993 just before the ushering in of democracy, there were two major Afrikaans newspaper companies, Nasionale Pers and Perskor; as well as two English ones, Times Media Limited (previously South African Associated Newspapers) and the Argus Publishing and Printing Company (Hadland 2007).

The “transformation” of South African society and its institutions began at the onset of democracy in 1994 through a politically motivated wave of unbundling of white corporations to black empowerment groups, driven by the idea of black economic empowerment. The print media paralleled these racial ownership changes and was one of the sectors most impacted by these changes (see Teer-Tomaselli and Tomaselli 2001). This marked a change in the print media market from “white only” ownership to racial diversification with the inclusion of black ownership (see Tomaselli 1997, for a detailed summary of these early deals). For example, Argus sold 52 per cent of the biggest selling newspaper of that time Sowetan (which targeted black readers) to black owned Corporate Africa Symposium consortium (Tomaselli 1997). In 1993 another significant deal took place, 34 per cent of Johnnic Communications (now Tiso Blackstar group) was sold to the National Empowerment Consortium (NEC) headed by then ANC Negotiator Cyril Ramaphosa, and then in 1996 the NEC acquired a full controlling share and Johnnic Communications effectively become South Africa’s first major black-controlled publishing house (Govenden and Chiumbu 2020).

Years later in 2013, the most significant transformation milestone for the print media since the early black economic empowerment deals occurred. As a way of context to this deal, in the early 1990 deals there was the entry of black owners as well as foreign ownership. In 1994 the local Argus group owned by Anglo American was acquired by Irish Tony O’Reilly’s Independent, and in 1999 a full control stake was acquired (Govenden 2019). In this 2013 deal the Independent Media went from being Irish owned to locally owned by the Sekunjalo Consortium whose members had apparent political ties to the ruling party ANC.

The post-apartheid print media market has been dominated by the Big four companies, which according to 2013 statistics had 90 per cent control of the market share (State of the Newsroom report SA 2013). These are: Tiso Blackstar Group, Media 24, Caxton, and the Independent media group. The Mail & Guardian newspaper owned by M&G Media (previously Weekly Mail during apartheid) holds a smaller share of the market however is a significant player. A study done by Print Media South Africa (PMSA) stated that the print media industry is a Level 5 BEE contributor with a score of 62.95 points, a Level 5 means an 80 per cent BEE recognition level (The Media Online 2011). Therefore, since the advent of democracy the print media has undergone notable racial diversification of black ownership, changing from the colonial and apartheid dispensations’ “white only” ownership to the inclusion of black economic empowerment ownership in the democratic-era print media market.

Theoretical lens: inferential racism and decolonial theory

This study draws from two theoretical frameworks to analyse print media content—Stuart Hall’s (1981) theory of racist ideologies and the media, as well as decolonial theory. Both theories are useful to the empirical focus of this study that attempts to identify legacies of racist stereotypes of blackness in print media content and link these findings to the racial diversification of ownership.

First, in Stuart Hall’s (1981: 168) theory of racist ideologies and the media, he is fundamentally of the belief that “racism and the media” directly touches the problem of “ideology” because the media’s “main sphere of operations” is the “production and transformation of ideologies”. Hall (1981: 170), in discussing ideologies and the media, distinguishes between the concepts of “overt racism” and “inferential racism”,

By overt racism, I mean those many occasions with open and favourable coverage is given to arguments, positions and spokespersons who are in the business of elaborating an openly racist argument or advancing a racist policy or view; By inferential racism I mean those apparently naturalised of events and situations relating to race, whether “factual” or “fictional”, which have racist premises and propositions inscribed in them as a set of unquestioned assumptions. These enable racist statements to be formulated without ever bringing into awareness the racist predicates on which the statements are grounded.

Hall’s (1981) concept of “inferential racism” is instructive to the South African context because the discourse of overt racism was removed from the public domain in 1976 in the interests of economic growth and investments which meant,

Western libertarian terms like “protection of minorities” and “multiculturalism” appeared, linguistically eliding the material reality of “white” political control. “Power sharing without domination”, an NP slogan, was inserted into the discourse with the impending unbanning of the ANC in the late 1980s. This discursivity encoded the idea of “minority (white racial) protection”, but in a less overt way than had occurred until then. This modified emerging multi-racialism was also reflected on television adverts and series, where blacks and whites could now inhabit the same frame without the one being subordinate to the other. The linguistically reformed apartheid ideology of the 1980s served the material interests of white bureaucrats and the middle classes (Tomaselli 1997: 16).

Given this historical context, the form of racism that would most likely be found in the post-apartheid media would not be overt racism but rather other naturalised and embedded forms. In this regard, Hall’s (1981) concept of “inferential racism” would best illuminate any subtle forms of continued racism regarding blackness stereotypes in print media content. During colonialism and apartheid the print media were a “producer” (Hall 1981), of anti-black racist ideologies (Dimitris Kitis, Milani, and Levon 2018). Hall’s (1981) theory of racist ideologies and the media is apt for the racial focus of the study and thus tracing the trajectory of South Africa’s print media. This study critically analyses whether in the post-apartheid period, black economic empowerment ownership matters and has influenced content with regards to dismantling racist ideologies of print media content where “transformation of ideologies” (Hall 1981) is seen, or are the print media “reproducers” (Hall 1981) of historical racist ideologies?

I recognise the limitations of only adopting a Western theory coined by Stuart Hall, a British Marxist Sociologist, in the study of media in a Global South and post-colonial reality such as South Africa. As Musa and Domatob (2007: 326) posit, “post-colonial societies operate with unique sets of realities and are expected to play certain roles that are alien to the norms of Western media systems”. Decolonial theory then becomes imperative in analysing the post-apartheid media. It is an emerging theory whose central premises are of particular relevance to South Africa’s 300-year period of colonialism of different types with British colonialism, and then apartheid known as “colonialism of special type” where white nationalism is a unique form of colonialism because the colonial government was inside South Africa and not in a parent country in Europe (South African Communist Party 1965, as quoted in Radebe 2017). Decolonial theory is centred on the belief that there are continued operations of colonial patterns of power after the end of colonial administration referred to as “coloniality” (Grosfoguel 2007; Chiumbu 2016). Decoloniality represents a vast and diverse body of scholarship that uses the concepts of power, knowledge, and being, as its organising principles of critique (Chiumbu 2016). This study adopts decoloniality’s “being” critique, the concept of “coloniality of being” was coined by Argentinian Walter Mignolo (1995), based on the idea that colonial relations of power left profound marks not only in areas such as authority, knowledge, and sexuality but also on the general understanding of being (as quoted in Maldonado-Torres 2007).

Fanon’s (1965) enigma of blackness and the existential traits of the colonial subject is the starting point of the “coloniality of being” (as quoted in Maldonado-Torres 2007). Fanon (1965) defines the colonial subject as the damné or condemned of the earth, as a non-being or the being who is not there or erased. Fanon’s (1965) description of the existential reality of the damné In A Dying Colonialism states,

There is, first of all, the fact that the colonized person, who in this respect is like men in underdeveloped countries or the disinherited in all parts of the world, perceives life not as a flowering or a development of an essential productiveness, but as a permanent struggle against an omnipresent death. This ever-menacing death is experienced as endemic famine, unemployment, a high death rate, an inferiority complex and the absence of any hope for the future. All this gnawing at the existence of the colonized tends to make of life something resembling an incomplete death. (Fanon 1965; as quoted in Maldonado-Torres 2007: 254–55)

Similarly, Lewis Gordon describes the hellish experience of black people in the colonised world, “Black bodies are seen as excessively violent and erotic, as well as the legitimate recipients of excessive violence, erotic and otherwise” (as quoted in Maldonado-Torres 2007: 255). Other decolonial scholars conceptualise this notion of the damné. Ralph Ellison unpacks this notion of the damné, as not a “being there” but rather a “non-being” or an invisible entity—and that what is invisible is the person of colour’s very humanity (as quoted in Maldonado-Torres 2007: 256–57). According to Maldonado-Torres (2007), invisibility and dehumanisation are the primary expressions of coloniality of being. This concept of the damné is further conceptualised by Maldonado-Torres (2007) as either invisible or excessively visible. Decoloniality’s “coloniality of being” perspective of the colonised subject offers a cogent perspective of blackness representation in the post-apartheid print media that is cognisant of the country’s decades of colonialism of different types.

Methodology

The methodology of the study consisted of both quantitative and to a lesser extent qualitative content analysis. Content analysis is the leading research method in empirical studies that analyse ownership effects on content (e.g. Okech 2007; Kiwanuka-Tondo, Albada, and Payton 2012; Wagner and Collins 2014; Rohlinger and Proffitt 2017; Humprecht and Esser 2018). The time period of analysis was approximately the first 20 years of democracy, 1994–2014. Quantitative content analysis enabled the study to draw inferences from a large sample of articles (Krippendorff 1989) and identify trends (Prasad 2008) in the representation of blackness. The study adopted qualitative content analysis as a supplementary method to provide a richer analysis. Qualitative content analysis was applied on a selected number of articles, which allowed the unpacking of meanings and associations embedded in articles (Davies and Mosdell 2006).

A total of six newspapers were selected: Business Day, Sowetan, Sunday Times, The Star, Sunday Independent, and Mail & Guardian. These newspapers belong to the historically English print media companies—Independent Media, Tiso Blackstar Group, as well as M&G Media. The English press was chosen because they have undergone far more black economic ownership changes than the historically Afrikaans press in the first 20 years of democracy, hence would be the most suitable given the research objective of the study being the relationship between black ownership and content. The sample of newspapers was selected using a four-part sampling criteria, to ensure that the diverse newspaper types and readership profiles present in the post-apartheid print media context were well represented, namely (1) newspapers that cater for each Living Standards Measure (LSM) readership profile, (2) newspapers with the highest circulation numbers, (3) newspapers that have basic and advanced language and literacy levels, and (4) different sizes of newspapers. For the first sampling criterion, the study selected newspapers that catered for each LSM readership grouping, and subsequently also considered the cover prices. The LSM categorises citizens as 1–10, where the highest living standard measure is 10 and the lowest is 1. In South Africa the LSM 7–10 is 34.9 per cent of the adult population (over 15 years): they enjoy the best standards of living, earn the highest salaries, and consume the most media due to their income and resulting available choices (South African Audience and Readership Foundation 2013). Business Day newspaper was chosen as a higher-priced newspaper that mostly the higher LSM’s can access. Sowetan was selected because it is a middle-priced newspaper that can be accessed by the higher, middle, and lower LSM groupings. This leads to the second sampling criterion: the study also selected daily and weekly newspapers with the highest circulation numbers. According to the ABC statistics January–March 2014 (which was within the time period of the sample), the Daily Sun had the highest circulation, and the Sunday Times newspaper was the top circulation weekly newspaper (as quoted in Hadland 2007), thus both were included in the sample. Regarding the third criterion, newspapers with different language and literacy levels were selected—namely, Sunday Independent, Mail & Guardian, Business Day, The Star, and Sowetan. The language level of Daily Sun and Sowetan can be accessed by a basic literacy level, while the weeklies Mail & Guardian and Sunday Independent cater for an intellectual audience. For the fourth sampling criterion, different sized newspapers were chosen, i.e. both broadsheet and tabloid sizes, namely the The Star (broadsheet) and Sowetan (tabloid). These sampling decisions do pose limitations for the study, mainly pertaining to bigger sample sizes of the English Press’ newspapers as well as the exclusion of the Afrikaans press. However, the carefully crafted sampling procedure used to select the six-newspaper sample does ensure diverse representation of the newspaper types and readership profiles present in the market. The exclusion of the historically Afrikans press newspapers (Media 24, whose parent company is Naspers; Caxton) would have yielded a richer study, especially that some of their newspapers fit the sampling criteria for the study. For example, the Afrikaans-owned newspapers have formidable circulation numbers, most notably Daily Sun being the biggest selling newspaper in South Africa, and Rapport being among the most read newspapers by the higher LSM groupings. However, the Afrikaans press played a different historical role during apartheid in comparison to the English press. The Afrikaans press was the purveyor of the conservative Afrikaans nationalist ideologies. Therefore, it is preferred that the study focus on only the English Press.

The unit of analysis were the leading front cover story as well as editorial page (editorial voice and main-ed column) because these are regarded as the most prominent and powerful framing sections of a newspaper. The front-page leading story is the issue that the newspaper regards as most important (Radebe 2017), and the editorial page can frame readers’ views (Van Dijk 1995).

Hester and Dougall (2007) point out that an important question to consider in conducting content analysis of newspapers is how many sampled editions are needed to provide adequate representation. There are many sampling techniques that can be used for content analysis of newspapers over long periods, which this study requires. There are numerous sampling techniques used for different purposes such as constructed week sampling, simple random sampling, and consecutive day sampling (Hester and Dougall 2007). For example, a previous study only used the Monday editions of European newspapers (Earl et al. 2004). Other sampling techniques involve small segments of time, such as weeks or months or randomly sampling days in a year (Earl et al. 2004). The technique of purposive sampling was used for this study, defined as “the selection of texts for analysis with a particular goal or purpose in mind” (Benoit 2010: 272). The large sample of articles (N = 684) were selected through a “purposive” sampling method that enables selection of newspaper editions with a particular purpose in mind (Benoit 2010). The purpose of this study is to get a long time-period picture of blackness representation in twenty years of democracy, to decipher if racial diversification of black ownership has affected content. The year of a major ownership change was used as the starting point for the sample period. The study analysed newspaper content that fell within the one-year period before and one-year period after major ownership changes of the English print media companies since democracy in 1994. Table 1 outlines the major ownership changes per print media company in the first twenty years of democracy.

Table 1. Major ownership changes of English print media companies in the first twenty years of democracy (1994–2014).

Print media company	Major ownership change/s and nature of change
Independent Media	1994- Change of controlling share from local (Anglo American) to foreign (Tony O’Reilly). 1995—Tony O’Reilly acquired a full control stake (58%) of newspapers and renaming of newspapers. 1999- Tony O’Reilly acquired remaining shares of 24% 2013- Irish Tony O’Reilly sold SA operations to Sekunjalo Consortium.
Tiso Blackstar Group	1993- The Johannesburg Consolidated Investments sold 34% of Johnnic Communications to the National Empowerment Consortium headed by Cyril Ramaphosa. 1996- The National Empowerment Consortium bought a controlling share in Johnnic Communications. 2008- Avusa Limited was established as a result of the unbundling of Johnnic Communications, the new company also purchased Johnnic Communications operating media and entertainment assets. Within this change the Mvelaphanda group purchased a 30% stake. 2012- There was a buy-out of Avusa by Tiso Blackstar Group that resulted in restructuring changes including a new Board, leaner structure, and significant staffing changes.
M&G Media Limited	1995- The Guardian became the majority shareholder and renamed it the Mail & Guardian. 2002- The Guardian reduced its shareholding to 10% selling 87.5% to Newtrust company Botswana LTD.

“Purposive” sampling was also applied in the selection of editions of newspapers to be analysed. Therefore, not all editions that fell within the year prior to and the year after a major ownership change were analysed. For each year identified, the first six-month period from January to June was targeted. Then the first edition of each month was included. This sampling process was purposed to ensure that a 20-year period sample of newspaper content was selected based on one set of consistent criteria of major ownership changes. In this way a biased analysis was eliminated. A total of 18 articles were analysed for each sample year—6 front covers, 6 editorial voices, and 6 op-eds. The outcome of the sampling process is reflected in Table 2. In total, the study analysed 684 articles which were made up of: 228 front cover stories, 228 editorial voice pieces, and 228 op-ed pieces.

Table 2. Sample years identified, and articles selected, per newspaper.

Newspaper	Sample years analysed	Total articles in the sample	Front cover articles	Editorial voice articles	Op-ed articles
Business Day	1992,1994,1995,1997, 2007,2009,2011,2013	144	48	48	48
Sowetan	1992,1994,1995,1997,2007,2009,2011,2013	144	48	48	48
Sunday Times	1992,1994,1995,1997, 2007,2009,2011,2013	144	48	48	48
The Star	1993,1995,1998, 2000,2012,2014	108	36	36	36
Sunday Independent	1998,2000,2012,2014	72	24	24	24
Mail and Guardian	1994,1996, 2001,2003	72	24	24	24

Note: Total articles N = 684.

The SPSS data software package was utilised to interpret the quantitative data garnered from the content analysis. A content analysis manual was utilised as the research tool.

Content analysis manual

The content analysis manual consisted of content diversity measures to capture the extent of diversity as a reflection of blackness transformation. The measures of content diversity in this study were topics, prominent voices, and tone. Topics and sources are the leading measures of content diversity in extant literature (Humprecht and Esser 2018; Masini et al. 2018). Tone has been extensively utilised in empirical studies on the effects of ownership on content (e.g. Hamilton 2004; Dunaway 2013; Wagner and Collins 2014). The tone category was purposed to decipher attitudes in print media in coverage (Masini et al. 2018), as well as to determine power dynamics in coverage (Radebe 2017). Tone can impact the way the public thinks about events and issues by assigning negative or positive attributes (Sheafer 2007). The measures of content diversity created for this study were: (1) diversity of topics in the front cover story and on the editorial page, i.e. party politics, government, socioeconomic rights, labour, business, human interest/celebrity/arts/entertainment; other; (2) diversity of prominent voices in the front cover story, i.e. government, business, political party, non-governmental organisation, civil society, trade union, expert in the field, multiple, other; (3) tones in the front cover story and editorial page, i.e. positive, negative, neutral. The tone category recorded each article’s dominant tone, and also tones exhibited towards any specific groupings, i.e. racial, socioeconomic (class), gender, sexual orientation, religion, regional, other.

The content analysis manual also allocated space for “qualitative comment” for each article, which flagged an article that was worthy of more qualitative analysis and that was not captured by the quantitative sections of the manual. For example, if an article displayed overt bias against a particular grouping such as black people, the working class, rural people, protestors, the ANC government. The details of the bias were quantitatively documented.

Post-apartheid print media’s perpetuation of blackness stereotypes: findings

This section presents the data findings of the content analysis and provides a quantitative expression of trends in the coverage. The section begins by discussing the selection rationale of the three issues used as the base to analyse the representation of blackness.

The racialised significance of the three issues analysed

The study selected three prominent issues that have framed post-apartheid South Africa that are racialised in nature. It is necessary to unpack the racialised significance of these issues to illuminate the transformation importance of the findings presented.

The first issue analysed was socioeconomic rights. South Africa is facing a severe socioeconomics struggle divided along racial lines with the triple challenges of poverty, inequality, and unemployment (Cole 2015), dominantly experienced by the black majority. Research has found that privilege is still white and poverty still black (see Cilliers and Aucoin 2016; Anwar 2017). Odhiambo (2008) notes that socioeconomic inequalities are rooted in the racialised policies of the past. The second issue analysed was that of labour issues and protests. During apartheid racial capitalism facilitated the super-exploitation of black labour by white capital (Bond n.d.). The third issue is the coverage of government vs. big business or “white economic elite”. In South Africa the “powerful” are demarcated across racial binaries of white and black. Government is majority black through the ruling party ANC. Big business is majority owned by the “white economic elite”, and black owners make up a small component of overall business shares. Black ownership of the Top 100 companies listed on the Johannesburg Stock Exchange was 21 per cent as of the year 2014 (United Nations Development Programme 2014). The lion’s share of all forms of capital is held by white people in South Africa which includes market shares (Anwar 2017). Colonialism typically viewed black leaders as incompetent, corrupt, primitive, and in need of a white saviour (Govenden 2019). The nature of coverage of black government vs. “white economic elite” thus has consequences for transformation of content.

Side-lining and stereotyping of the socioeconomic struggle of the poor

A significant amount of the coverage—30 per cent—was dedicated to socioeconomics coverage (N = 684). At surface level this considerable magnitude of socioeconomic coverage suggests that the print media attach importance to the topic of socioeconomic rights and show cognisance for human rights and citizen-related news. A deeper analysis is required in this instance because of the structural complexity of the socioeconomic situation in South Africa. Since the advent of democracy socioeconomic print media coverage has been criticised for ignoring the underlying factors of deep-seated socioeconomics issues such as poverty and inequality (Jacobs 2004; Radebe 2017).

A second-level analysis was conducted on the front-page socioeconomic stories (N = 45) to determine if the coverage was superficial or addressed the deep-seated nature of the socioeconomic challenge in South Africa. Subtopics were developed based on Jacobs’ (2004) and Radebe’s (2017) arguments and themes found in the articles. Most of the socioeconomic front-page coverage was about economics news for the elite (29 per cent); socioeconomics news as it relates to government (27 per cent); sensational news with a socioeconomic component (18 per cent); and basic coverage of socioeconomics news (13 per cent). These findings suggest that the nature of socioeconomics news is elite-driven, sensational, and simplistic, and overall omits deeper coverage (Jacobs 2004; Radebe 2017). Poverty and inequality news only garnered a meagre 2 per cent, which is inadequate given the high levels of poverty and inequality in the country. The socioeconomic struggle in South Africa is faced predominantly by black people and is a legacy of apartheid. Yet black people’s severe socioeconomic struggle mainly amounts to sensational and simplistic coverage effectively side-lined on the peripheries of the press agenda. The socioeconomic crisis has also produced an eruption of social protests. South Africa has produced a phenomenon of social protests fundamentally linked to its socioeconomic crisis. Mottiar and Bond (2011) flag the rate of social protests in South Africa as one of the highest in the world per person. Some of the reasons for social protests include service delivery, demanding justice, highlighting a cause, student protest, worker strike, political accountability, and xenophobia (Mottiar and Bond 2011). Social protests only garnered 2 per cent of the total socioeconomics coverage, which is unfairly low considering the explosion of social protests in the country.

Overall, the socioeconomic plight of poor and rural people was found to be side-lined in the mainstream press agenda. The categories of “Poverty and Inequality” and “Rural” only garnered a low 2 per cent each of front-cover stories (N = 45). Rural people were mostly made visible as violent and voiceless protestors that cause havoc during the time of social protests and were predominantly represented as causing trouble and hardly seen in a positive light. Social protest coverage also generally omitted the deep-seated coverage discussed by Jacobs (2004), such as the structural violence aspect of the socioeconomics struggle in South Africa. The socioeconomic coverage trends of this study parallel numerous other studies, which together points to a clear trend of socioeconomics issues being downplayed in print media coverage and missing its structural causes (see Chiumbu et al. 2018; Wasserman, Chuma, and Bosch 2018), as well as social protestors represented in negative ways such as being violent, barbaric, out of control, illegitimate, criminal, and irrational (see Pointer 2015; Wasserman, Chuma, and Bosch 2018).

Invisibility and demonisation of labour issues

Only 4 per cent of coverage focused on labour issues (N = 684) articles. Labour protests garnered 31 per cent of this total labour coverage, and only one article was about unemployment. No stories represented labour positively, while 34 per cent of labour stories depicted labour in a negative light, suggesting that about one-third of the print media’s labour coverage is negative. Therefore, the coverage of labour issues was near invisible in the mainstream press agenda, and when it was covered the working class grouping was depicted as mostly negative and hardly positive.

In the sample of articles about labour protests, the print media followed the “protest paradigm”, which is “a pattern of news coverage that expresses disapproval toward protests and dissent” (Lee 2014; as quoted in Leopold and Bell 2017: 721). There are five characteristics of the “protest paradigm” used to identify negative patterns in protest coverage: first, negative news frames; second, reliance on official sources and official definitions; third, invocation of uninformed public opinion; fourth, delegitimisation through a lack of information that would give context to the protest; and last, demonisation by representing protestors as social deviants (Lee 2014, as quoted in Leopold and Bell 2017). The labour protest coverage in the sample revealed a propensity to follow four out of five characteristics of the “protest paradigm”: the riot frame; privileging the official sources and position; delegitimisation through a lack of information in explaining the reasons for the protests and adopting a dismissive tone even when some context is given; and a demonisation of protestors as unruly and unmannerly. For the most part labour coverage in the sample represented labour protests negatively and demonised the labour protestors.

The first characteristic of news frame was seen in the sample of articles, specifically riot frames, which, “overemphasize any lawlessness, danger, destruction, and disorder occurring because of the protests” (Leopold and Bell 2017: 721). Many articles in the sample represented labour protestors in a single narrative as infuriatingly angry, lawless, disorderly, and the dangerous “other”.

Also evident in the coverage was the fourth characteristic delegitimisation, which is coverage that does not adequately explain the context and objectives of the protestors and as a result readers will deem the protest to be pointless (Leopold and Bell 2017: 722). The sample of articles displayed a propensity to omit key contextual information about the protest, as well as overemphasising the anger and danger posed by the protestors.

The second characteristic is reliance on official sources and the official definitions (McLeod 2007, as quoted in Leopold and Bell 2017). The fifth characteristic is demonization, which paints protestors as social deviants and reporting is episodic in nature, lacking in contextualisation (Leopold and Bell 2017). For example, common forms of demonisation are extensive coverage of destruction by protestors, listing of arrests, and highlighting altercations with the police (Leopold and Bell 2017). The elevation of the official position and demonisation narrative were seen in a significant amount of coverage. Another trend in the sample of articles was the good government and bad labour binary built in the narrative of labour protest coverage. Coverage involving labour versus government during the time of labour protests predominantly saw the print media privilege the government position. Protestors were demonised as social deviants for embarking on strike action and represented as unruly and unmannerly.

This is in agreement with previous findings that labour occupies a small and stereotypical place in the print media’s mainstream agenda. Duncan’s (2014) research about the press coverage of the Marikana massacre found that workers were subjected to Gerbner’s “symbolic annihilation” through the under-representation and non-representation of workers.

High visibility of negative coverage about black leadership

Government had high visibility in press coverage including in the most prominent section of the newspaper. The topic of “government” recorded 22 per cent (N = 684), in comparison to “business” which only made up 4 per cent of coverage (N = 684). In terms of placement, “government” garnered 30 per cent of front cover coverage (N = 284), 23 per cent of editorial voice (N = 284), and 14 per cent of main op-eds (N = 284)—which makes “government” coverage among the topics that garnered the highest coverage in these sections of the newspaper. A significant 45 per cent of stories about government represented government negatively vs. 3 per cent positive, which reveals that almost half of “government” stories depicted “government” negatively. The 4 per cent (N = 25) of business articles showed 12 per cent negative tones vs. 12 per cent positive tones, while the remainder of the stories were “neutral”. Notably, only one story was about business corruption. These tones suggest that business coverage largely amount to neutral coverage as the negative tone (12 per cent) towards “business” equates to the positive tone (12 per cent). A deeper analysis of government coverage revealed that there was a significant rise in the negative representation of the government since 1994, specifically its actions, policies and especially ANC leaders. Findings further reveal that the coverage of the ANC as a political party exhibited mostly negative tones. Of all the government stories, only one story depicted the Democratic Alliance (DA) government negatively (during the time-period of analysis the DA was the main opposition party). Therefore, the print media in South Africa over-scrutinise the black elite and overwhelmingly neglect big business which is mostly made up of the “white economic elite”.

Decolonising “blackness visibility” in print media content: discussion

The patterns of coverage inherently showed a propensity to follow “inferential racism” of blackness (Hall 1981), as well as “coloniality of being” (Mignolo 1995; Fanon 1965; Gordon 2005; Maldonado-Torres 2007).

“Inferential racism” of blackness

In the coverage of the three issues, the press paralleled the concept of “inferential racism” coined by Hall (1981). The press coverage of social protests and labour protests showed a predominantly danger and demonisation narrative, and the racist assumption that undergirded the representation of the majority black protestors were that black people are dangerous, social deviants, and are the problem. The chiefly negative coverage of the majority black government was undergirded by the racist assumption that black leadership is incompetent. It can therefore be argued that news regarding the black majority had a strong propensity to be limited to narratives of danger, incompetency, and corruption—which together painted a damaging and demeaning picture of blackness in print media content. It can also be argued that the nature of blackness racist assumptions (Hall 1981) was a manifestation of “coloniality” (Grosfoguel 2007), which is further unpacked in the subsection that follows.

Enigma of blackness and “non-being”

The negative representation of blackness in print media content found by this study also significantly resonated with the “coloniality of being” decolonial perspective (Mignolo 1995; Fanon 1965; Gordon 2005; Maldonado-Torres 2007). The findings showed that the “general understanding of being” of black people in print media coverage had “profound marks” of colonial tropes of blackness (Mignolo 1995). The representation of black people in the print media exhibited “coloniality of being” conceptualisation of the colonial subject as a “damné”, “non-being”, “hellish experience”, “excessively violent”, “invisible”, “excessively visible” (Fanon 1965; Gordon 2005; Maldonado-Torres 2007).

Black people in post-apartheid South Africa are indeed subject to Fanon’s (1965) idea of the existential reality of the colonised person’s ever-menacing death, and Lewis Gordon’s (2005) hellish experience of black people in a colonised world—they still face labour exploitation and are hardest hit by the severe socioeconomic crisis, notably unemployment and poverty. Yet this severe struggle of black people in the press discourse was equated to the “damné or condemned of the earth” in the Fanon (1965) sense, as basically “erased” and thus “non-beings”. The under-representation of labour and simplistic socioeconomic coverage showed that the plight and severe struggle of the majority black people was not newsworthy and also found to be near “invisible” in the mainstream press agenda (Maldonado-Torres 2007). Black people, on the other hand, were “excessively visible” in coverage about the incompetence and misdemeanours of black leadership (Maldonado-Torres 2007).

In addition, Gordon’s (2005) concept of black bodies seen as “excessively violent” with regards to the colonised subject was seen in print media coverage of social protests as well as labour protests, the black protestor was depicted as “excessively violent”, which points to colonial legacies in the post-apartheid print media content. This research also found that black government was represented within the colonial narrative of black leaders as incompetent, and “excessively visible” in the decolonial sense (Maldonado-Torres 2007). The incompetence of black government was regarded as far more newsworthy than the incompetence of big business or the “white economic elite”. In the coverage of business, only one story was about business corruption and business was represented in a neutral tone (12 per cent positive, 12 per cent negative), whereas the black majority government was consistently represented negatively vs. 3 per cent positive. Therefore, the post-apartheid print media coverage of news involving the black majority significantly conformed to Fanon’s (1965) “enigma of blackness” and the “existential traits of the colonial subject” as a “non-being”.

Concluding remarks: does ownership matter?

Black economic empowerment ownership has failed to dismantle colonial and apartheid era racist assumptions of blackness. There were a few stories in the large sample of articles (N = 684) that could be regarded as content that challenged hegemonic colonial and apartheid assumptions of blackness. Whilst the current “transformation” agenda is imperative for economic transformation and addressing the ills of the past that privileged ownership access to the minority white people of the country, the media is a powerful producer of racist ideas and ideologies (Hall 1981), especially as seen in South Africa’s historical context of anti-black constructs, and must be held to a deeper race standard than structural “black visibility” in ownership. The findings of the study bring to the fore important nuances for journalism debates about the relationship between ownership and content.

In the case of South Africa’s print media, it can be concluded that ownership does not significantly affect content. The content analysis revealed that ownership does not necessarily have an empirical causal link to media content. This study also importantly highlights the stark difference between Global North and Global South media contexts. Extant empirical studies comprising predominantly of Global North studies found that ownership does matter and affects content (as discussed earlier), which buttresses the theorisations of the power of media ownership on content. This study, however, exposes the limitations of the power of ownership in Global South countries where unfair colonial and racist power dynamics framed the media. The South African case study has disproved the theoretical assumption that ownership effects news content. Colonialism and apartheid did not just have a damaging effect on ownership; decades of white domination had magnitude effects on the various aspects of the print media and the system as a whole, such as news values, narratives of reporting, freedom of expression, etc. Therefore, generic racial substitution in media ownership structures in and of itself is a limited approach. Unexamined in this paper is the political economy and contextual factors that black ownership operates within. For example, the location of South Africa’s print media in the market that renders the bottom line as paramount; its libertarian media system that produces a liberal ethos idea of democratic news content; the neo-liberal context cultivating elite class continuities; journalism and training steeped in Western epistemologies; media policies and regulation. These are worthy subjects of future inquiries.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by The National Institute for the Humanities and Social Sciences (NIHSS).