The decolonial turn: reference lists in PhD theses as markers of theoretical shift/stasis in media and journalism studies at selected South African universities

The supervision and production of a PhD thesis often presents a potentially interesting tension between PhDs as conforming to disciplinary epistemologies and PhDs as breaking epistemological boundaries. No academic discipline has been left untouched by decolonial thinking in the South African university space since the eruption of radicalized student protest movements in 2015. The Rhodes Must Fall student protest movement, which quickly morphed into Fees Must Fall, precipitated a new urgency to decolonize the university curriculum in post-apartheid South Africa. A new interdisciplinary conversation in the humanities and social sciences began to emerge which challenged established orthodoxies in favour of de-Westernizing, decolonizing and re-mooring epistemological and pedagogic practices away from Eurocentrism. Whether and how that theoretical ferment filtered into postgraduate students’ theses, however, remains to be established. This article deploys a decolonial theoretical framework to explore the tension between epistemic conformity and boundary transgressing in journalism studies by analysing reference lists of PhD theses submitted at three South African Universities three years after the protest movement Rhodes Must Fall. With specific focus on media and journalism studies as a discipline, this article argues that the PhD process represents a site for potential epistemic disobedience and disciplinary border-jumping, and for challenging the canonical insularity of Western theory in journalism studies. The findings appear to disconfirm the thesis that decolonial rhetoric has had a material influence so far on the media studies curriculum, as reflected in reference lists of cited works in their


Introduction
The Rhodes Must Fall and Fees Must Fall student protest movements that began in 2015 precipitated a new urgency to decolonize the university curriculum in postapartheid South Africa (Le Grange, 2018;Mheta et al., 2018). A new interdisciplinary and transdisciplinary conversation in the humanities and social sciences received a new impetus to challenge established orthodoxies in the name of de-Westernizing and decolonizing the university curriculum, and in the quest for re-mooring epistemological and pedagogic practices away from Eurocentrism in media studies (Garman, 2015). The student protest movement set in motion a lively conversation about decolonizing the university curriculum, evidenced by special seminars, colloquiums and conferences on the topic and the establishment of committees to oversee and coordinate the process of decolonizing the curriculum at various universities in South Africa (Le Grange, 2016). Since students were at the forefront of demanding transformation of university education, it would be reasonable to expect that students' own intellectual production in the form of PhD theses undertaken after 2015 would begin to reflect the theoretical ferment in the field.
More than two decades after the end of apartheid in South Africa, the higher education system had largely remained as what Heleta (2016: 2) describes as 'a colonial outpost', existing to reproduce and disseminate an unapologetically Eurocentric curriculum. In this article, I use the term 'Eurocentric curriculum' in the sense in which Patel (2014), cited in Chambers (2020), uses it, to describe the systemic dominance of knowledge production, distribution, consumption and reproduction by institutions based in the Euro-American region. A decolonized curriculum, conversely, is supposed to aim at reducing inequalities in knowledge production.
This article examines the visibility of published decolonial frameworks relative to the traditionally accepted liberal theoretical canon in the reference lists of PhD theses submitted for examination in the media studies departments of three South African universities between 2018 and 2020, looking for evidence of disciplinary conformity or epistemic disobedience and 'transgressing the boundaries of the discipline' (Bastalich, 2015: 8). The theoretical frameworks that students use in their research studies operate like regulatory and classificatory mechanisms, by means of which conceptual boundaries of a discipline are set, contested and policed (Silverman, 2013). The process of discipline formation involves theoretical delineations of a field of study -editorial boards of leading academic journals in the field play a key gatekeeping role (Goyanes and Demeter, 2020) -and the building of a corpus of texts held sacred by the adherents of the discipline (Foucault, 1977). As Bastalich (2015: 7) points out, the theoretical framework and extant scholarly literature of a given discipline exercise: the reiterative power of discipline discourse to illuminate the phenomenon that it names and regulates, hence 'becoming academic' is not so much about discovering the new, but learning the history, language, concepts, tropes and communication styles of a field area … This fact is at the heart of academic practice, perhaps most evident in the centrality of the literature review and of research citations more broadly, which work to delineate accepted knowledge from what is judged to be a disciplinary innovation.
Disciplines in this sense can justifiably be considered to impose 'considerable barriers to free thinking' (Krishnan, 2009: 9). This study analyses the reference lists of selected PhD theses submitted in the field of media studies at three universities in South Africa to establish evidence of what Walter Mignolo (2013) calls 'border thinking' in media studies post the Rhodes Must Fall and Fees Must Fall student protest movements, which created the context in which decolonial theory supposedly gained currency in the South African academy.

Theorizing decoloniality in journalism
Decolonizing journalism implies an approach to journalism education that decentres the curriculum from the hegemonic Western model -'It refers to the everyday and ongoing efforts to challenge persistent forms of coloniality' (Zembylas, 2018: 2). Decoloniality defines a way of being in the world that is opposed to and resists colonial subjectivity in all its manifestations, not only 'in books, in the criteria for academic performance, The decolonial turn 3 in cultural patterns, in common sense' (Ndlovu-Gatsheni, 2015: 14), but also in the concrete world of practices and relationships constitutive of, and constituted by, colonial difference. Its application to curriculum reform would manifest in an expanded reading list that accommodates other ways of knowing beyond the Western episteme. To demand expansion or replacement of one reading list by another, however, is to assume that the Other enjoys the same and equal ontological space to exist democratically. In the production of new knowledge, including through PhD theses, a decolonial turn would imply a questioning of, or a shift from, privileging Euro-normativity as a prism in journalism. The reference list is a reliable barometer of the range of theoretical lenses that informs any published thesis, and it can provide a window on to the content of the curriculum on which the author draws. The ultimate concern of a decolonizing programme should, however, look beyond the reading list to interrogate the systemic infrastructure that renders possible a monologic canonization and universalization of Western perspectives on media and communication in the first place -for example, the journal article production system behind the articles that students cite in their own work.
Operationalizing decoloniality is fraught with many challenges, complexities and paradoxes. One such paradox, which has dogged the efforts to decolonize South African higher education is 'how higher education institutions continue to reproduce an epistemological hierarchy wherein Western knowledge is privileged over non-Western traditions of … knowledge-making' (Zembylas, 2018: 1), and where the university in Africa identifies as an extension of the university in the metropole and continues to benchmark itself against universities in Europe (Backhouse, 2009).

The disciplinarity of journalism studies
The content and pedagogical methods of a university subject -its ontological and epistemological status -define the curriculum of a discipline. Journalism only began to make its tentative entry into the corridors of higher education as a fully-fledged academic discipline at the beginning of the twentieth century, when Joseph Pulitzer, in an essay entitled 'Planning a school of journalism -the basic concept', made the impassioned plea for its establishment as an academic discipline with pride of place in the university, alongside the highly esteemed so-called 'learned' professions of law, literature and medicine (Adam, 2001). Pulitzer's vision of elevating journalism, from a vocation to the status of an academic profession worthy of being studied at the university, was premised on its perceived contribution in liberal democratic societies. Liberal democracies needed professional journalism as the fourth estate of the realm, and journalism needed highly trained journalists, whose single-minded study of the subject at university level would help to provide. Pulitzer saw journalism and liberal democracy as conjoined twins, where one was inconceivable without the other, and thereby they set the functionalist liberal pluralist analytic foundation from which journalism may be studied and theorized (Hanitzsch, 2019). With this conviction in mind, Pulitzer put into motion processes of establishing journalism as an academic discipline by founding one of the earliest journalism schools, at Columbia University in New York, and committing substantial resources to the project of elevating and dignifying 'the standards and to extend the influence of that calling to which he had so successfully devoted his life', which 'could best be accomplished by a university training school, placed side by side with successful training schools for the so-called learned professions and in the same academic association' (Fackenthal, 1915: 25).
The structure of the journalism studies degree curriculum as envisaged by Pulitzer was one that was open to influences from cognate academic disciplines, but without necessarily capitulating to any of them. Those other subjects, such as history, economics, politics, philosophy, law and the natural sciences, had to be articulated to journalism, and not the other way around (Adam, 2001;de Burgh, 2003). From the beginning, journalism has been one discipline where the boundaries separating it from other disciplines have been fuzzy, open and undecidable (Wahl-Jorgensen and Hanitzsch, 2009). It has been left to individual universities to decide the boundedness or openness of journalism to external theoretical influences, and this has been responsible for the variegated nature of journalisms taught around the world: References to journalism studies are often linked to media and communication studies because, although they are distinctively different in terms of curricula needs, they are also overlapping and research in each of these areas informs and enriches the understanding of the others. (Papoutsaki, 2007: 2) For purposes of avoiding clutter, I use the terms 'journalism studies', 'media studies' and 'communication studies' interchangeably to represent the broad field of journalism, media and communication studies (Pooley, 2016). The view of journalism studies as transdisciplinary is in sharp contrast with a view of it as having an identity consistent with Foucault's (1977: 141) argument that every academic discipline needs 'the protected place of disciplinary monotony'. The gatekeepers of each discipline, its professorial peerage working through a system of peer-reviewed journal publishing, define the outer limits of what may or may not be considered canonic knowledge within it. Unlike its more sure-footed neighbouring disciplines such as history and sociology, journalism is 'quintessentially cross-disciplinary' (de Burgh, 2003: 96) -a characteristic which presumably should positively predispose it to decolonization.

Contexts of media and journalism education in South Africa
Wasserman and de Beer (2009) have pointed out the yawning gap between theory and practice, and between rhetoric and action, as far as the movement away from the Anglo-American view of journalism is concerned. In a research article 'Journalism education and practice in South Africa and the discourse of the African Renaissance', Motsaathebe (2011) aptly encapsulates the tensions between globalizing and localizing tendencies on the one hand, and conformity to the dictates of industry on the other, evident in ongoing efforts at reforming journalism education curricula in the South African academy. De Beer et al. (2017) make the tacit acknowledgement that South African journalism education is in a state of flux, evolving away from its Euro-American origins towards acquiring an African orientation. They point out, however, that post-1994 South African journalism education has remained rooted in an Anglo-American professional journalism tradition.

Methodology
This study employs digital archival research to produce a corpus of research data in the form of reference lists obtained from recently completed PhD theses from three purposively selected South African universities for analysis. The digital archival research involved entering specific search terms (metadata) into the online institutional repositories of the three South African universities: the University of Cape Town (UCT), the University of the Witwatersrand (Wits) and the University of South Africa (UNISA). Archival research on the internet makes possible low-cost access to data of high documentality, as Rasmussen (2017) points out.
These three universities were selected on the basis that they were among the most racially segregated historically as Whites-only institutions (Ndlovu-Gatsheni, 2017), and arguably the most in need of decolonizing. Because most South African universities participate in the global OpenDOAR system, their institutional research output, including master's degree dissertations and PhD theses, are archived in their institutional online repositories on an open-access basis. The classificatory algorithm and architecture of each institution differs in terms of how theses are classified and archived. Thus, in order to optimize search results, search terms for each university had to include the name of the department, filtered by year of submission for theses that qualified for analysis in this research. Since the 2015 Rhodes Must Fall and Fees Must Fall student protest movements in South Africa were considered a critical marker of engagement with a discourse of decoloniality within the university space, only PhDs submitted from 2018 were considered for this study. To ensure maximum variation within the sample (Hancock et al., 2009), the criterion/quota sampling frame was used. Reference lists were then copied from the downloaded theses to a separate file, and cleaned and reformatted for easy handling and processing. A final list of 2,466 reference strings from a total of nine theses was captured and saved for coding and analysis.

Methods of data analysis
The thematic content analysis technique was used to analyse collected data in the form of downloaded reference lists. A set of 30 thematic codes was generated inductively from the data using grounded theory coding (Charmaz, 2014). The process involved rereading and reflecting to make sense of the book, book chapter or journal article title as captured in each reference string, and assigning a new code if the title did not fit in the already developed coding frame. The process was repeated iteratively with each successive data string. This initial phase was then followed by a focused sorting, synthesizing the initial 30 codes into a summary of 12 theoretical constructs. The inductively drawn set of codes was particularly suited to this study in that it checked against author bias. The data were captured on an Excel spreadsheet, and descriptive tables and charts were generated to describe the general patterns in order to respond to the research questions. The second level of analysis focused on the geographical diversity of institutional affiliation of editors-in-chief and editorial boards of referenced journals. This analysis sought to map the geopolitical division of labour in knowledge production in media studies.

Ethical considerations
Although all the theses are published on university websites, and are therefore in the public domain, the contributors of the work and their supervisors have been anonymized, thesis titles have been removed, and any identifying details have been kept to the minimum to avoid unnecessary and uninvited attention to the authors of the work being analysed without their consent (Crossen-White, 2015; Sugiura et al., 2017). As the quality of the theses was not at issue in the analysis and evaluation, the present discussion is confined to the reference lists as indexing the literature and theoretical frameworks engaged with in the theses.

Presentation and discussion of findings
The nomenclature of departments that offer degrees in journalism differ from university to university, but they are generally located in the humanities or social sciences faculties. At the University of Cape Town, journalism as a discipline is hosted within the Centre for Film and Media Studies. At Wits, the Department of Journalism is one of the three departments that make up the School of Literature, Language and Media Studies (SLLM) in the Humanities Faculty. At UNISA, journalism is offered within the College of Communication Science in the Faculty of Social Sciences. The name and location of the discipline are important markers of theoretical orientations that inform research in and around the discipline. The reference lists of the study sample tend to reflect the different theoretical inflections of research focuses at each university. The research focus tends to correspond to, or closely reflect, the research expertise of members of faculty. In light of the master-apprentice model most prevalent in PhD supervision relationships in South Africa (Backhouse, 2009), it is reasonable to argue that the reference list of a thesis indirectly indexes the supervisor's own theoretical preferences and areas of research interest and expertise. The corollary may also hold true -that those theories and literature sources of little interest to supervisors would tend to feature less prominently in students' work. The summary of data parametres, with candidates and supervisors anonymized, is shown in Table 1.
The year of first registration of the submitted thesis was inferred by counting at least three years back from date of submission, given that it is standard practice for universities in South Africa to allow a minimum of three years for completion of PhD studies. Since the 2015 'Fallist' movements were an important reference point for this study, theses submitted before 2018 were excluded, as the starting year for those studies would have been earlier than 2015. Only theses submitted in 2018 and after could have been first registered in 2015 and were therefore conceived under the intensified environmental pressures to decolonize the curriculum in South Africa. Given that literature points to 2015 as the year that marked the beginning of robust theoretical engagements with decoloniality within the South African intellectual space and beyond (Fataar, 2018;Mheta et al., 2018;Ammon, 2019;Jansen and Achebe, 2019), it would be logical to expect a time lag (Said, 2014;Mpofu, 2017) between the arrival, emplacement and incorporation of such a theory into the circuits of knowledge production and dissemination, including through dissertations and theses in a particular discipline. Reference lists are fairly reliable indices of the theories that frame students' studies and the key source texts that they cite in the body of the thesis. Beyond looking at thesis subtitles, the theories used in a thesis can be inferred from the reference list.

Reference lists as a reflection of disciplinary conformity in the media studies discipline
Departmental nomenclature, as can be deduced from thesis titles, provide good pointers for the general theoretical focuses of theses at each of the universities. Theses from UCT tend to focus more on visual media and cultural studies topics, those from Wits focus on journalism and media studies, while those from UNISA tend to be inclined towards communication theory, public relations and marketing communication.
Application guidelines for the PhD programmes on the respective web portals of the universities carry generic statements cautioning would-be applicants that successful application to the degree programme depends on availability of supervision expertise in the host department. Prospective students are therefore encouraged to submit 'an acceptable research proposal for which supervision is available' (UCT, 2021). Academics generally perceive the task of PhD supervision as one that entails striking a balance between a heavy hands-on and a complete hands-off approach (enforcing standards and monitoring progress), advising (guiding, coaching) and promoting (mentoring and initiating) candidates into full membership of the discipline. Some scholars have imagined the PhD process as a 'rite of passage' through which the doctoral candidate is apprenticed to an expert who deftly guides them until they attain full membership of a community of practice. It is a two-way back-and-forth dialogical and negotiated disciplining process by means of which the initiate attains the status of a peer in their chosen subject (Cullen et al., 1994;Delany, 2009). In selecting a corpus of research literature for review, evidence suggests that PhD candidates negotiate between their own and their supervisors' world views, as well as how they locate themselves in the discipline. Practical considerations of scarcity or availability of literature on decolonizing media studies cannot be ruled out as a probable factor in whether or not students deploy a decolonial theoretical framework in their work. Scholarly articles and book-length monographs on decolonizing media studies only began to appear in reasonable numbers after the cohort of PhD theses under review had been submitted.

Analytical frameworks used by journalism/media students
The theoretical kernel of the journalism studies discipline, as reflected by the frequency of use within and between different theses, is shown in Figure 1.
The theoretical compass of media, communication and journalism, as reflected in the data, plots its cardinal points, as well as its marginal points, to illustrate media theories that have attained canonical status and dominance in the field. Figure 2 illustrates the pattern of occurrence of the 30 analytical constructs inferred from the data.
The least often occurring analytical frameworks included decolonial theory, postcolonial theory, Afro-feminism and critical theory. The variations in the data could be attributed to the broader theoretical orientations of the media and communication departments at the three universities, as well as to the different research topics investigated. For example, there was a strong cultural studies element in the theses from UCT, while there was a strong journalism element in the theses from Wits, and communication theory, including strategic and marketing communications, was prominent in those from UNISA. Mass communication theories tended to be more cross-cutting. It could be argued from the data that although an element of transdisciplinarity was quite evident, journalism as a discipline also maintained a system of epistemic boundary policing. The metaphor of the PhD journey as a 'rite of passage' would cease to apply if disciplines lost their identity-conferring function by allowing themselves to be buffeted by every theoretical wind that comes. A possible explanation for the poor showing of a decolonial theoretical perspective in the data set could be the scarcity of a rich corpus of published literature engaging with the theory at this early stage of the decolonial theoretical ferment in South Africa. Hasselberg (2013: 34) argues that 'a well-researched area should lead to more references … than an area which is relatively new'. The PhD students were still clearly operating within the orbit of a Eurocentric media universalism, from Siebert et al.'s (1956) Four Theories of the Press, to Hallin and Mancini's (2004) Comparing Media Systems. The PhD theses in the sample thus tend to be located at the conjuncture of critique and conformity regarding extant methodological and theoretical research agendas, as narrowly defined and framed by a hierarchized institution of academic journal publishing in the field (Chakravartty et al., 2018). In the following section, I zoom in on the geopolitical representation of academics affiliated with institutions in the Global South, particularly Africa, vis-à-vis those from the West on the editorial boards of journals cited in the PhD theses in the sample. Literature on the subject points to the fact that editors-inchief, and their editorial boards as an institution, influence knowledge the most by 'determining what is published' (Metz et al., 2016;Goyanes, 2020). They also act as 'a good predictor for the national diversity of their publication output' (Demeter, 2018(Demeter, : 2914 and, in consequence, author diversity. Thus, while the main focus of this study was The decolonial turn 9 on whether or not decolonial thinking manifested in the literature that PhD students in media studies cited, it was also important to shine a light on the nature of the journal editorial structure, which may be responsible for the production of the content that students cite. Literature has pointed to a significant and positive correlation between the structure of editorial boards and the diversity of the published content of a journal (Goyanes and Demeter, 2020;Metz et al., 2016).

The geopolitics and body politics of journal editorial decision-making in media studies
The editorial board of a journal arguably plays a key gatekeeping role in the production of new knowledge, and consequently in the production of the theoretical corpus of any academic discipline. Any attempts at modifying the curriculum would necessarily need to engage with this structure by means of which a curriculum reproduces, maintains or transforms itself. The debate about decolonizing the media curriculum needs to be situated within the broader context of existing epistemological hierarchies set in place by an international intellectual division of labour, itself deeply rooted and implicated in global capitalist modernity (Bhambra, 2007). It is not by accident that knowledge production in the field of media and communication studies enunciates itself epistemically largely from its North American and European bases, because, as Grosfoguel (2007: 213) has argued 'the locus of enunciation, that is, the geo-political and body-political location of the subject that speaks' does matter. Wasserman   Beer (2009: 430) opine that 'journalism in Africa often displays an uneasy relationship between its colonial heritage and post-colonial appropriation, between globalized, Western influence and local resistance'. The sample was further analysed in terms of whether it falsifies or confirms the thesis posited by Chakravartty et al. (2018) that knowledge production in the discipline has been, and continues to be, dominated by academics domiciled in Western universities in 'splendid oblivion' of African perspectives (Schoon et al., 2020: 2). Figure 3 summarizes data on the geographical diversity of institutional affiliations of editors-in-chief of journals cited in the PhDs. It illustrates that publishing in the discipline continues to be directed from the Euro-American locus, in what Weiler (2009: 4) has described as: one of the most salient features of the international knowledge system … its peculiar division of labor, in which key intellectual tasks, such as setting theoretical agendas and methodological standards, are the prerogative of a relatively small number of societies and institutions that play a disproportionately important role in this system -societies and institutions which are, almost without exception, located in the economically privileged regions of the world.  (Demeter, 2018). This demonstrates that the knowledge that drives much of higher degree scholarship in media studies at the South African universities in this study remained firmly Eurocentric in that regard. In addition, an analysis of the institutional affiliation of members of editorial boards shows that Africa- based academics were only listed in 32 per cent (UNISA), 47 per cent (UCT) and 46 per cent (Wits) of journals listed. The struggle to decolonize media and communication, and to decentre it away from its Western moorings, as Wasserman (2020) and Garman (2015) have argued, would require addressing systemic barriers that foreclose the Global South as a site of theory generation: The high concentration of academic and textbook publishers in the journalism/media studies field in Europe and the US has contributed to the dominance of the Anglo-American model. This is partly due to the fact that it was Western, particularly American, higher education systems that first incorporated journalism and communication into their curricula and had thus more time to develop. (Papoutsaki 2007: 4) A programme of decolonizing the media curriculum ought of necessity to pay attention to how the editorial mechanism of scholarly journals operates to reproduce Western perspectives as the norm. It should be about moving the centre or shifting the locus of epistemic enunciation (Wa Thiong'o, 1993;Grosfoguel, 2011) by pushing for greater geographical diversity of the editorial boards of leading journals in the discipline. Extant research has established a positive correlation between geographical diversity of members of an editorial board and diversity in the geographical origin of published content (Lauf, 2005;Demeter, 2018;Goyanes and Demeter, 2020), an important step towards pluriversality in theoretical perspectives in the field. Arguments in support of the status quo often cite the double-blinded peerreview evaluation system as an inbuilt mechanism for ensuring that 'good' theories from any part of the world can enjoy an equal opportunity of being ventilated in scientific journals, irrespective of their locus of articulation or the composition of the editorial board. Therefore, the identity or geographic origin of an article should ordinarily not influence the review outcome. The peer-review system is often presented as a fail-safe mechanism for a just pre-publication arbitration on the scholarly merits of an article, a structure that guarantees epistemic neutrality. Yet, as Kerr (2014) points out with reference to Canadian educational contexts, knowledge production is far from neutral. Studies have shown that there are manifold biasing factors that skew publication in favour of research work originating in the geographic North rather than that originating from the developing world (Skopec et al., 2020;Metz et al., 2016). What the defence, however, fails to acknowledge is the amount of influence that the editor-in-chief, together with the editorial board as gatekeepers of last resort, are capable of exerting in shaping 'what is published and thus what informs theory development, research and practice' (Goyanes and Demeter, 2020: 1). Demeter's findings in a study in 2017 replicated Lauf's (2005) study on the influence of editorial boards on published content, and concluded that 'not just the virtues of the anonymous article matter, but the location of the publisher and the editor (or even the affiliations of the review board)' (Demeter, 2018(Demeter, : 2897. The task of shifting the biography and geography of knowledge to embrace subjugated knowledges, as Mignolo (2007) argues, therefore remains legitimate.
When one peers behind the curtain of African visibility on editorial boards, one notes a clear hierarchy in which academics affiliated with South African universities tend to dominate, with the result that South Africa metaphorically becomes the standin for a whole continent of 53 countries (see Table 2).
What remains arguable, however, is the significance of their influence on the conceptual and theoretical agendas of the journals (Mudimbe, 1988). While the argument for a shift in the geopolitics and body politics of knowledge production in journalism may have its merits, the presumption that inclusion of scholars based in the Global South will necessarily reorient journalism away from the Euro-American model has no basis in the literature. Affiliation with a university in the Global South does not necessarily assign criticality, as Grosfoguel (2011: 5) argues: 'The fact that one is socially located in the oppressed side of power relations does not automatically mean that he/she is epistemically thinking from a subaltern epistemic location.' So far as can be inferred from students' work, the process of decolonizing the journalism curriculum cannot be done asynchronously from every other pillar that supports that curriculum. The present study brings to light that decolonizing the journalism curriculum would require that equal attention be paid to the supply side of the knowledge production and dissemination value chain. The journal publishing ecosystem is clearly one of the chief means by which 'the enunciative privilege of the Renaissance and Enlightenment European institutions, men and categories of thought' is maintained (Mignolo, 2013: 133).

Limitations of the study
While the post Rhodes Must Fall period provided a unique opportunity and a conducive environment for a scholarly engagement with a conversation about decolonizing university curricula in South Africa, the time frame of the present study imposed its own limitations in terms of how much epistemic shift could reasonably be measured using reference lists in students' work. The absence of citations referencing a decolonial theoretical framework could be the result of scarcity of literature on decolonizing media studies, because it was still a relatively little researched and published area by media scholars at the time that the students were working on their PhDs. In this sense, students therefore probably had no option but to engage in what Hasselberg (2013: 41) metaphorically refers to as 'a ritualistic drowning in texts that do not give rise to a will to become submerged'. Another important limitation of this study arose from the inbuilt and inevitable anonymity around the peer-review process behind the published and subsequently referenced journal articles. The picture of geographical and racial monopoly over journal article publishing thus remains incomplete and inconclusive. An area of possible further research might include an analysis by gender, race and national origin of the African-affiliated members of the editorial boards in those limited instances when they occur. This may be particularly useful in the South African context, where middle-class white male professors continue to dominate top-class educational spaces (Albertus, 2019;Conway, 2017).

Conclusion
The evidence considered, and the discussion above, have revealed that the inbuilt flexibility and self-reflexivity within the Euro-Western paradigm of journalism forecloses the radical epistemological shifts suggested under decoloniality. The data considered evidence a transformational trajectory involving piecemeal 'inclusion of Africans into a long-existing "European game" without changing the rules of the game' ( Ndlovu-Gatsheni, 2017: 61). Through this mechanism, internal theoretical contradictions are resolved in a continuous process of self-renewal of the Eurocentric dominance of the media studies curriculum. The question that future research needs to confront head on is: What would journalism look like if the current reading lists were totally substituted by knowledges from the margins? Study of the reference lists of PhD theses produced by students under supervision leads to the conclusion that either media and journalism academics in South Africa lack the expertise and confidence to lead their supervisees along new decolonial theoretical alleys, or that literature on decolonizing journalism is unavailable. On the basis of the evidence considered, this discussion has endeavoured to show how the predicted failure to decolonize university curricula will be predicated more on failure to confront a deeply conservative self-entrenching and hegemonic knowledge production system, as Vandeyar (2019) argues. Decolonizing higher education curricula remains just as urgent in South Africa today as ever before (Albertus, 2019: 4).