Using Critical Policy Historiography in Education Policy Analysis: A South African Case Study

This article foregrounds the value of using “critical policy historiography” as an analytical/methodological tool in undertaking policy formulation research, highlighting the importance of taking a long-term historical perspective. Using school-funding policies in South Africa as a case study, it argues that while there was wide consultation with stakeholders, power dynamics within the policymaking process underscore the influence of policy elites, such as experts and powerful lobby groups, in shaping policy outcomes. In particular, it critiques the foundational moment of school-funding policymaking in the democratic era, that is, the South African Schools Act of 1996, focusing on key actors that shaped the Act, and whose interests were ultimately served. The article then reviews changes made to school-funding policies to date, and, building on the work of Salim Vally and others, argues for a policymaking framework that strengthens the influence of poor and marginalised communities on policy outcomes if the education social justice project in South Africa is to be advanced. in Silver


Introduction
While education policy has been the subject of considerable research in South Africa, limited attention has been paid to education policy methodology, especially in the recent past. By drawing on critical policy historiography (CPH) as an analytical tool, this article attempts to reclaim the importance of history as methodology. Research since the early 1990s, while concerned with policy dynamics in general, tended to focus more on the implementation of policy during South Africa's transition to democracy, with a focus on the relevance of the wider economic, ideological and political contexts (e.g., Kallaway et al. 1997;Unterhalter, Wolpe, and Botha 1991), the gap between policy and practice and problems and challenges related to policy capacity (e.g., Sayed and Jansen 2001), and, importantly, a concern with education quality, and theoretical and methodological approaches underpinning education change and reform (Sayed, Kanjee, and Nkomo 2013). Hoadley (2018) and Chisholm (2019) emphasise the value of taking a "long view" in education research, the former in relation to curriculum policy and pedagogy, and the latter in relation to teacher education and preparation. De Clercq (2020), on the other hand, offers a theoretical lens premised on a post-structuralist framework, highlighting the critical role of agency, to better understand lingering inequalities and weak efficiencies in the education system. Others, such as Spaull (2019), have tackled squarely issues of equity in South Africa in relation to funding allocations as an issue of historical justice; likewise, Motala and Carel (2019) stress the importance of recognising education as a public good, with equity and justice as key drivers of education reform. Overall, these publications veer towards a focus on theorising education policy reform and implementation. Although they inherently recognise the importance of history and the importance of historical span in education policy research, there is little focus on history as method, which this article sets out to do. In so doing, it refocuses attention on the power dynamics in the policymaking process by recalling the work of Webster (1998) on elites impacting policy settlements, and more specifically, the work of Fataar (1999), who has argued that participation in education policy processes has given rise to a technicist approach that excluded many interest groups. The article then fast-forwards to the present, with its persistently exclusive and narrow policymaking processes, by drawing attention to the need to empower the poor and explore alternative discourses (Vally 2020). It builds on Fataar's (1999) critique of exclusive policymaking and Vally's recent emphasis on privileging the poor and the working class through "forging a pedagogy of possibility, social class analysis and effective community participation in education policy deliberations" (Vally 2020, 1). Moreover, with its emphasis on historical methods, and critical policy historiography, the article seeks to reclaim the value of a historical analytical and methodological lens in education policy research.
Using school-funding policies as a case study, the article foregrounds the significance of CPH in addressing issues of democracy and social justice, especially questions on whose voices are most influential in policy formulation, who benefits and who is marginalised. In so doing, it foregrounds the unequal distribution of power among policy actors in policymaking as a critical factor in shaping policy outcomes. It thus seeks to refocus policy research on the origins of policy and the power dynamics that characterise policymaking at a particular historical conjuncture. In so doing, it draws attention to the influence of policy elites on policymaking and why such influences have persisted over time.
Crucially, little attention has been paid to how marginalised voices can be empowered to have a greater influence in shaping policy outcomes. This article argues that refocusing attention on the power dynamics in policymaking is as important now as it was in the 1990s; further, following Vally (2020), it proposes that empowering marginalised voices in policymaking remains a critical project in pursuit of social justice and democracy in South Africa. This is done through an analysis of school-funding policies in South Africa post-1994 that have been at the centre of persistent inequality in South African schooling. At the heart of the conundrum is the effectiveness of schoolfunding policies to contribute to redress and equity for a large majority of learners faced with debilitating learning and living environments. By drawing on CPH, the article argues for the adoption of a longer-term historical lens to help illuminate persistent inequality and equity challenges. It examines the role of history in illuminating issues of democracy and social justice in policy formulation research, and subsequent policy choices.
The article begins by briefly locating the importance of history in education policy research. Thereafter, the notion of critical policy historiography is examined to reflect on the political and ideological dimensions of policymaking/formulation, as well as the dominance of the discourse of policy as the "domain of experts". It then reviews key funding policies, specifically using as a case study, the South African Schools Act (SASA) (RSA 1996), as the foundation stone of ongoing education inequality and equity challenges. The article concludes by reviewing recent research on subsequent schoolfunding policy choices, highlighting their impact on the quality of education in schools and on South Africa's social justice project, with specific reference to the poor and marginalised. While many of the school-funding and historical research debates are not new, this article reclaims the importance of history as methodology in education policy analysis research, with specific reference to South Africa's social justice project.

History and Education Policy Research
Several scholars have highlighted the importance of history in education policy research (Griffiths 1998;Ozga 2000;Silver 1983;Taylor et al. 1997). Silver (1983, 240) directs attention to an approach to policy and theory that has a strong commitment to detailed historical description and analysis: By historically, we … refer to two main features of the best historical work: first a concern with close and detailed description and analysis, firmly set in time and place, and second, a preoccupation with continuity and especially with change, with crisis and with transformations. There is, in our view, a close association between historical understanding in this sense anda hopeful, progressive, politics. (Barron et al. 1981, 20, cited in Silver 1983;emphasis added) While historical research is about investigating a phenomenon over a period of time (Layder 1993), the critical issues are about change and continuity, and the hope of a better future. More specifically, attention to the temporal dimension is particularly relevant in policy research because "policies are often a continuation of existing practices"  and seldom emerge in a vacuum. But, as Mills (1959, 154) stresses, it is not enough to "explain something as a 'persistence from the past', we ought to ask 'why has it persisted?'" The historical approach, therefore, can be particularly useful in undertaking studies of educational change and development as it compels a review and analysis of how phenomena changed (or did not change) from one epoch to another. This is especially relevant to research that focuses on issues of social justice and democracy, which beg analysis of the wider context and forces at work.

Critical Policy Analysis and Historiography
The importance of reflecting on issues relating to theory and methodology in education policy research has often come under the spotlight (e.g., Ball 1990;Taylor 1997;Troyna 1994). Much reflection has been framed by the discourse of "critical policy analysis" or "policy sociology" (Gale 2001). Others, such as Jie (2016), are concerned with the different phases of the policy process, while acknowledging the influence of powerful lobbies and the broader social context in policy.
A strength of critical policy research is its emphasis on advancing the course of emancipation, democracy and social justice. For example, researchers can facilitate "bringing in" many marginalised voices to the policy development process, the central focus of this article. As Griffiths (1998, 32) extols, "social justice is concerned to right both individual disempowerment and structural injustices of gender, social class, race, sexuality and disability". This article highlights, in particular, how South Africa's school-funding policies have been mediated along class and racial lines, affecting mainly the poor in society.
The notion of critical policy research is also to be found in the work of Gale (2001), specifically his proposal of three alternative and overlapping historical lenses with which to explore policy research: policy historiography, policy archaeology and policy genealogy. Policy historiography is associated with the substantive issues of policy at particular hegemonic moments; policy archaeology is concerned with conditions that regulate policy formations; and policy genealogy is connected with social actors' engagement with policy (Gale 2001, 384-5). In this article, I draw on Gale's analysis of "historically informed" policy research and I reflect essentially on his notion of "policy historiography".
The term "historiography" may be used to refer to "the specialist study of the writings of [particular] historians" (Marwick 2001, 29), or more generally to a "range of historical discourses" (Gale 2001, 384). Policy historiography, according to Gale, asks three broad questions: 1) What were the "public issues" and "private troubles" (Mills 1959, 8 cited in Gale 2001) within a particular policy domain during some previous period and how were they addressed?
2) What are they now?
3) What is the nature of the change from the first to the second?
Critical policy historiography (CPH) adds two further questions to these: 4) What are the complexities in these accounts of policy? and 5) What do these reveal about who is advantaged and who is disadvantaged by these arrangements? (Gale 2001, 385).
These questions, couched in the notion of historical span, lend themselves to addressing issues of democracy and social justice. Policy historiography, as depicted in questions one to three, draws attention to the importance of historical analysis in highlighting critical social issues affecting policy actors in various milieu. In tracing the nature of change of key social questions from one epoch to another, the historian is able to identify crucial moments of crisis, transformation and stagnation. The "critical" dimension in CPH, as manifested in questions four and five, provides the depth of inquiry needed to illuminate those forces and structures impacting the lives of individuals and social groups. It is in seeking answers to these questions that notions of inclusion/exclusion, domination/emancipation and the dynamics of power relations can be exposed, thereby making the important link between education policy research and democracy and social justice issues.
However, complexities in the accounts of policy are not confined to the political and ideological dimensions of policy formulation. There is an often-hidden methodological complexity that raises questions about whether the analytical framework itself is inclusive or exclusive, whether the analytical conventions and tools might privilege particular individuals and groups, and yet estrange others. In this regard, the article highlights the tendency in policymaking towards a reliance on "experts". This tendency recognises the enormous influence that policy analysts and policy professionals wield in public policy development, especially their utilisation by governments to satisfy the scientific legitimation of policy decisions (Dye 2000;Magasela 1998). The "expertdriven" approach to policymaking has come to be characterised by an emphasis on methodological expertise, through the use of highly sophisticated and technical methods, such as statistical techniques, economic modelling and cost-benefit analysis, to name a few (Weiss 1992). A significant consequence of this approach is that policy experts tend to have a disproportionate influence in policymaking, very often at the expense of ordinary citizens and civil society organisations (Magasela 1998).
Without discounting the value of statistical modelling and policy expertise, this article argues for the empowerment of poor and marginalised stakeholders in shaping policy outcomes as a key component of the social justice project in South Africa. As will be discussed, while education policymaking in South Africa sought to reconcile experts with the masses through consultation, a reconstituted civil society and the unequal distribution of power in policymaking privileged some voices while marginalising others (Fataar 1999).

The South African Schools Act: More Questions Than Answers
The data reported on in this article draws from a study on teachers' participation in policymaking conducted between 2002-2005. 1 The study examined the forms of teachers' participation, predominantly through their unions, that have been most influential in shaping the content of the South African Schools Act (SASA) of 1996, the first policy to focus on school funding, organisation and governance post-1994. It further sought to unearth the factors that limited or enhanced teachers' participation in policy formulation in the context of South Africa's transition to democracy in the 1990s, of which the emergence of a very conservative, neoliberal framework of policymaking was significant (Fataar 1999;Vally 2020). The study was therefore concerned with notions of "democratic participation", "effective participation" or forms of participation that have an "empowering capacity". It was also concerned with questions of power and influence in policymaking, whether teachers and other policy actors were able to shape the outcome of policy processes. For the purposes of this article, the notions of power and influence are considered in the context of democracy and social justice, and are extended to incorporate both historically marginalised voices and "the new marginalised" of the post-apartheid era. The historical span covers the apartheid epoch at the height of the struggle for democracy and social justice during the 1980s and early 1990s. The early 1990s also marked the beginning of a new, democratic epoch in South Africa. The study thus straddles these two epochs: apartheid in its twilight and democracy in its infancy. This historical contextualisation forms the backdrop for a review of more recent research on school-funding policies and interventions, and the extent to which the interests of marginalised voices have found expression in policy outcomes.
In responding to the "what to look for" question in Gale's conception of a "historicallyinformed" research design (Gale 2001, 383), the study focused on the "participation" of those teachers who were actively involved in the struggle for democracy and social justice in South Africa, as well as the involvement of teachers belonging to organisations that historically had been branded as "conservative" and interested only in "professional" matters. Thus, the South African Democratic Teachers' Union 1 As part of the author's PhD.
(SADTU), "the teacher activists", and the National Professional Teachers' Organisation of South Africa (NAPTOSA) and the Suid-Afrikaanse Onderwysersunie (SAOU), "the conservative professionals", were included in the study. This was important, too, as teachers and teachers' organisations of various ideological persuasions and beliefs had seemingly been "participating" in policy processes post-1994 on level terms. As will be discussed, while all teacher unions were represented in the SASA process, policy distortions occurred due to the unequal distribution of power among them.

School Funding and the Model-C Poser
Arguably the inequality and equity debates around the apartheid-spawned white Model-C school system and the question of funding dominated deliberations with regard to SASA (Govender 2008). This was because the white community was concerned over possible reduced state funding of their well-resourced schools. The fulcrum of the debate was about achieving funding equity to ensure that black schools would have the necessary funding to enable them to provide the kind of education quality that would at least be on par with their white counterparts (Govender 2008, 186). It soon became apparent to the new government that finding the necessary financial resources posed an immense challenge.
A second issue that was to be hotly contested, with implications for redress and equity, was around language policy. As part of the negotiations for a new political dispensation in the early 1990s, the contending parties had agreed on the drafting of a new constitution. The National Party (NP), the then apartheid-era ruling party, had demanded an "education clause" that would allow parents and students to choose the language of instruction in state schools. After much posturing and contestation by opposing sides, the African National Congress (ANC), representing the black majority, agreed to a compromise clause guaranteeing such a right where it could be "reasonably provided". Similarly, Section 247 of the Interim Constitution provided for bona fide negotiations with existing school governing bodies (primarily white schools) if any changes to their powers were to be contemplated by a new government (Govender 2008, 186-87). These provisions would eventually have a significant impact on the participation and influence of interest groups in the development of SASA.
A quick survey of the more than 1,500 submissions made by members of the public with regard to SASA reveals a predominance of inputs by the white ex-Model-C constituency (Department of Education [DoE] 1995), which provided the main opposition to the new schools' policy. Interview data and other document sources attest to this predominance (Govender 2015). The "personal troubles" of this constituency, "temporarily settled" in the form of racially defined education policies under apartheid, were quickly and adroitly turned into a "public issue" in the making of SASA, which constituted a critical point in South Africa's transition to democracy in the educational realm. In the process, the voices of the historically privileged in South Africa were more influential in shaping policy outcomes. For example, the strategic thrust of the educational leadership of the privileged white community was to highlight the good of the existing Model-C school system for all of South Africa's schools. They argued that historically disadvantaged schools must be brought up to the standard of their schools, which would be in the best interest of education in South Africa (Interview, Koos Steyn, 11 June 2002). Their arguments had an educational soundness to it, made more acceptable because it was couched in the prevailing discourses of democracy and social justice of the time. Particularly, the arguments resonated with the main economic and political theses that had started to gain currency within state administrative and political circles. It was not surprising, therefore, that the Model-C lobby found common ground with the neoliberal economic position advanced by the government's foreign consultants, Luis Crouch and Christopher Colclough, which was to crystallise in the now famous "Fourth Option".
Of course, the "settlement" around the funding of schools would not have washed politically if it was not apparent that significant sections of the then expanding black middle class would also benefit. In the view of Luis Crouch, one of the education department's consultants (or experts), it was critical to retain the "personal support" of key elements of the middle class, such as editors and Members of Parliament, for the public school system, otherwise these influential decision makers would rather send their children to private schools, a tendency borne out by international experience, especially in countries "with big income differentials, such as Mexico, Brazil and South Africa" (Interview, Luis Crouch, 10 July 2002). This is clearly a recognition of the influence of "policy elites", especially those with economic and political clout, and underlines the unequal distribution of power in policy formulation. This was in spite of intense lobbying on the part of the ANC Education Alliance within the Parliamentary Portfolio Committee for Education, chaired by Blade Nzimande, for education to be free and compulsory; eventually the ANC and its Alliance partners, including SADTU, had to settle for a compromise wherein the setting of compulsory school fees could only be made subject to national norms and standards established by the Minister of Education (Govender 2008).
Moreover, the democratic movement in the education sector, such as the South African Democratic Teachers' Union (SADTU), was unable to shape the funding policy outcomes of the SASA process as it would have liked. As an alliance partner of the ANC-led government, SADTU had an ambiguous relationship with the state: on the one hand, its policy strength and influence were derived from its close relationship with, and participation in the ANC Education Alliance; on the other hand, developing policy positions as part of a collective undermined its independence as a union, and the leverage to impact policy outcomes in its own right (Govender and Cross, forthcoming). As such, the union struggled to "introduce new ways of thinking and operating within the state policy cycle (from formation to implementation to review)" (NALEDI 2006, 18 Significantly, scholars such as Sayed and Carrim (1997) have criticised democratic notions of participation as encapsulated in SASA for further marginalising black working-class and rural families, and they have argued that community representatives should have been included in the composition of the SGBs as a way of challenging power elites in school governance decision-making. Arguably, from a social justice perspective, the "new, democratic" cohort of policymakers lost the historical plot because key aspects of SASA, notably issues of school financing, governance and language as a medium of instruction, contrived in their settlement to benefit those already privileged and not the historically disadvantaged black majority in spite of widespread consultation and participation. As Jansen (2001, 279) argues: "Participating groups have unequal power and expertise in different policy forums, leading to different kinds of emphases in policy outcomes." Consequently, not much had changed from the previous epoch to the new, democratic epoch as the beneficiaries were still middle-class parents, predominantly white, and not working-class parents or the rural poor (see Parker [1993] for similar findings in relation to policy development in the early 1990s).
And, certainly, there were other complexities in such accounts of policy (recalling Gale [2001]), such as the nature of political negotiations and reconciliation in the transition phase, the continuing influence of apartheid-spawned civil servants in the education bureaucracy, the legal and constitutional framework, especially Section 247 of the South African Constitution, which made consultations on the powers and functions of governing bodies obligatory, and the macro-economic policy framework that influenced particularly decisions around funding (see Kallaway et al. 1997;Motala and Pampallis 2001). While these complexities provide an explanation for the making of policy history at a particular moment in time, it nevertheless begs the question: What about policymaking for social justice? Part of the answer lies in the dominant policymaking discourse and framework that has characterised South Africa's policy processes.
The Discourse of Policy as the "Domain of Experts" A key factor in the limited influence of many constituencies in policy formulation in South Africa has been a growing reliance on an expert-driven approach, including the use of foreign consultants. This has been evident from the development of the National Education Policy Investigation (NEPI) process, which was "dominated by a heterogeneous grouping of intellectuals" (Parker 1993, 221), to the development of the National Qualifications Framework (NQF), which had been formulated within "a marketised, commodification-based vision of competency and training", resulting in expert-driven innovations more suited to disempowering people from the policy and planning arenas than liberating them (Donn 1998, 84) and to the development of Curriculum 2005, which saw an implicit technicist model of policymaking take hold, resulting in a shift in education policymaking from an apparently stakeholder-driven process to one driven by bureaucrats and experts (Kruss 1998, 104). The formulation process of SASA also mirrored this trend, as evidenced by the role played by foreign consultants and academics. Certainly, teachers' organisations were able to relate to the technical and legal discourse in the formulation of SASA, especially NAPTOSA and SAOU, both of which found it within their broad ambit of experience and "know-how". The young SADTU, established in 1990 and in the midst of organisational development, though, was severely constrained. Despite having policy analysis support from the Centre for Education Policy Development (CEPD) and the Wits University Education Policy Unit at the time, the union had recognised that the more technical and legal aspects of policy analysis work were not an area of traditional strength for them (Interview, Duncan Hindle, ex-President of SADTU, 5 June 2002). This was compounded by the flight of many leaders from organisations such as SADTU and civic associations to government, with their new leadership unable to develop adequate strategies to organise their members to continue demanding for pro-poor education policies (De Clercq 2020). The one area where SADTU did make an important contribution to policy development during its early years was in the area of teacher appraisal and development, specifically the development appraisal system (DAS). Nevertheless, DAS did not escape criticism, including its downplaying of performance monitoring by "rendering teachers as competent professionals" (Sayed and Kanjee, 2013, 21), thus underlining the self-serving role of teacher unions in the interest of its members, and not necessarily for the public good.
Large sectors of the public were estranged from the "technical discourse" in the formulation of SASA, especially working-class parents, students and the rural poor (Govender 2008). As Vally (2020, 11) contends, "what has been missing from most analyses of transitional policymaking in South Africa is a careful examination of social class, and particularly how and why social movements and social actors on the ground, who were initially central to policy formulation and critique, became largely marginalised once policies were institutionalized". In reality, an elite group of individuals and organisations, privileged by an "expert-driven" approach to policy formulation, was able to exert considerable power and influence in the shaping of SASA.
Arguably, then, the collective agency power of the white Model-C lobby, including teachers, parents and academics, characterised by thorough preparation, sound organisation and political savvy, had as much to do with the mediation of the new government's democratic and social justice agenda as did factors such as the nature of political negotiations and the prevailing neoliberal macro-economic climate. So too did the exclusionary nature of the dominant policy discourse that had come to be associated with policy development in South Africa in the 1990s, namely policy as the "domain of experts". In combination, all the above colluded to ensure that social justice was kept on the agenda, but largely to benefit those least in need. This was in spite of the efforts of organisations representing the historically marginalised in the education policy domain, notably the National Education Coordinating Committee (NECC), and later the ANC Education Alliance (see Govender and Cross, forthcoming). As such, the achievement of political power had little to do with the achievement of social justice for a large majority of South Africans. Ironically, a new struggle for "social justice" was legitimated in the policy formulation of SASA for an already privileged white minority. As observed by Motala (2001, 243), "the decline in the levels of organization of particular social forms, and the rise of other strong and articulate-even if minoritysocial, business and other interest groups in 'civil society'" is a matter to be reckoned with. Portraying themselves as the "new marginalised", with the interests of the majority at heart, the white, middle-class constituency was able to protect important economic, cultural and ideological gains made under apartheid. For the Afrikaans-speaking white community, this meant the preservation of a particular culture and ethos at their schools, not least of which was the protection of the Afrikaans language.
Funding Policies Decades Later: What Has Changed?
Following the legislation of SASA in 1996, several pro-poor education policies were promulgated. These included the National Norms and Standards for School Funding (NNSSF) (DoE 1998) and the Amended National Norms and Standards for School Funding (DoE 2006(DoE , 2008. In broad terms, these policies represent aspirations towards social justice and equity, given their provisions on no-fee schools, school fees exemptions, and other pragmatic interventions (Mestry and Ndhlovu 2014). Specifically, the NNSSF policy was intended as an equity instrument to benefit poorer schools through a ranking system where quintiles 1, 2 and 3 are declared no-fee schools and are provided with substantial funding, and quintiles 4 and 5 schools are affluent schools with significantly reduced funding (DoE 2006(DoE , 2008. Overall, while the above policies have contributed towards equalisation of resources in South African public schools and improved resource distribution in pursuing its propoor strategy, inequality and inequity persist (Mestry and Ndhlovu 2014;Motala and Carel 2019;Spaull 2019). Essentially, though, the net of beneficiaries has narrowly widened to include mainly black, middle-class children fortunate enough to attend highfee schools. As Spaull (2019, 6; emphasis added) has argued, today black and "coloured" learners make up 60% of those attending former white-only fee-charging schools as part of "a small, separate and functional school system, created to privilege one section of the population and exclude others".
Spaull (2019) offers compelling reasons why the majority of black learners continue to be excluded from enjoying quality learning outcomes based on their inability to afford high school fees. He points to, for example, informal mechanisms of exclusion (social networks, biased admissions, interviews and policies, waiting lists), but also "feeder zones" where schools have significant latitude to define the geographical areas from which they will accept learners, resulting in former "white" schools in "white" neighbourhoods selecting learners from their immediate surrounds. Motala and Carel (2019) stress historical inequalities between provinces, the national finance redistribution model, the distribution of teacher qualifications in terms of the postprovisioning model, and private expenditure in public schooling utilised to fund significant quality and efficiency differentials as some of the causes. They claim that changes in funding alone will be insufficient to equalise education outcomes, but more pro-poor funding will be necessary over succeeding decades if there is to be any chance of turning around the vast, failing and largely poverty-stricken schooling system (Motala and Carel 2019).
A key argument advanced by Motala and Carel (2019) is that both social equity and education equity need to be addressed simultaneously to ensure quality education does not continue to exclude the poor, suggesting that equity in school funding has progressed at a faster pace than equity in South African society. This is not surprising as broader social and economic inequalities, linked to poverty, unemployment and a weakening of state organs through corruption and governance crises, are more difficult to fix than reprioritising education budgets. Moreover, many scholars have argued for a greater focus on improving the overall quality of education, as this is likely to have more sustained social benefits over the long term. This includes recognising that the most poorly resourced schools are the ones with the least competent teachers (Shalem and De Clercq 2019), and the need to improve teachers' disciplinary knowledge and pedagogic competencies (Taylor 2019). A notable development relating to those able to afford minimal fees, and which could negatively impact education quality, is the market response through the creation of a range of low-cost private school options catering for families who cannot get their children into good public schools, but who can afford to pay fees. Motala and Carel (2019) argue that there are risks attached to this approach, particularly regarding how private forms of schooling affect the social functions of education, or the idea of the public good, with the implication that the state is relinquishing its responsibilities around the social rights of citizens. As De Clercq (2020, 17) asserts, the state has not been "pressurised by the education sector to ensure that its education reforms and programmes of redistribution and support benefitted the disadvantaged majority". De Clercq (2020) attributes this to the unequal power relations in society and the absence of collective agencies in challenging unequal structural educational inequalities.
It is clear from the above that school-funding reform proposals in the democratic era have been mediated by racial and class factors. The proposals in the main constitute policy tinkering, with the status quo favouring largely an expanded middle class that includes black citizens. However, the poor and predominantly black working class in South Africa still struggle to access quality education, irrespective of whether it is provided by the public or private sectors. It is also clear from the above that academics still dominate funding policy reform proposals, almost 25 years since SASA was legislated. There has been little public engagement with the poor and working-class communities in particular. School governing bodies (SGBs), provided for by SASA and established as organs of democratic participation in education, have struggled to fulfil their promise. As Vally (2020, 12 citing Lewis and Naidoo 2004) highlights, SGBs have served "to reinforce existing patterns of power and privilege in schools … and the technocratic character of school governance in South Africa makes it inaccessible to the majority of its communities and disempowers the poor". SGBs, moreover, have become sites of contestation for power, where various school stakeholders conflict and struggle for their interests as opposed to the interest of the school, resulting in many disadvantaged schools failing to develop democratic governance to support school development (De Clercq 2020). Thus SGBs, ostensibly granting decision-making power to key stakeholders, such as parents, have proved largely ineffective as a vehicle for meaningful participation of poor and working-class communities. It is not surprising that organisations representing marginalised voices, such as Equal Education, have struggled to shift the power dynamics in policymaking, often having to mount legal challenges against the state to be heard.

"Bringing in" Marginalised Voices: Towards a Socially Just Policymaking Framework
Drawing on Gale's (2001) critical policy historiography, and specifically the question of who is advantaged and who is disadvantaged, there is a greater imperative today for addressing the educational access needs of the poor. This is especially relevant given that issues of redress and equity in education had become somewhat peripheral in recent years, but have now resurfaced in the context of the COVID-19 pandemic (see, for example, Fataar and Badroodien 2020). Equally, the contestation for state resources has intensified in the face of mounting economic and social challenges. It is opportune for marginalised communities to lobby not just to be consulted, but to be recognised as voices with power to effect change.
A related, but more complex task is to challenge the hegemony of the expert-driven approach to policy formulation with a view to facilitating a broader understanding and involvement in the policy development process. Critical policy historiography can, at the very least, expose the constraining effects of "dominant" policy discourses and perhaps begin to explore alternative discourses and tools of analysis with more widespread appeal. It can do this by locating itself more firmly within the tradition of critical theory. As Ranson (1995, 443-4) argues, there is a need to develop a framework of values that stress the notion of "public" in education policy: The appropriate values for public policy and its analysis are those of democracy and citizenship for the learning society. … It is only when the values and processes of learning are placed at the centre of the polity that the conditions can be established for all individuals to develop their capacities, and that institutions can respond openly and imaginatively to a period of change.
One learning possibility is to promote the development of policy analysis capacity among civil society organisations, such as governing bodies and students. Although this might not guarantee the influencing of final policy outcomes, it will at least ensure that marginalised sectors of society will have the necessary knowledge and skills to engage in policy formulation. As Govender (2008, 546) argues, building policy capacity and skills of the intended policy beneficiaries is critical for promoting a learning societythe construct of "policy learning", for example, could help bridge the gap, as a way of empowering citizens in policymaking. This is also important in the context of reclaiming the educational legacy of participatory democracy and an alternative, more empowering and inclusive policymaking framework. Vally (2020), in a similar vein, suggests revisiting the praxis of South Africa's social movements of the 1980s and 1990s, notably the "people's education movement", "workers' education", the "popular adult and/or community education movement", and "education with production" as well as collaboration between researchers and communities to help communities acquire "tools to inform, direct, own and use research to claim their space and voice" (Vally 2020, 9).
It is essential to affirm and embed the notion that education is a public good, and foreground the view that equality and social justice must drive educational reform. This has implications for current and future processes of policymaking. As suggested earlier, the "final" shaping of policies represents temporary settlements at particular hegemonic moments-consequences of the historical confluence of political, economic and ideological forces. The contestation around issues of funding in the South African Schools' Act can be regarded as one such moment. Critical policy historiography suggests that the education policy community, broadly defined to include the poor and working class, will continue to be confronted by significant "hegemonic moments" in the future.
Indeed, the national review of education policy in South Africa, including the Schools' Act, in November 2002, constituted another moment and, since then, has generated a new wave of struggles to capture the "sociological imagination" (Mills 1959), leading to yet more temporary policy settlements. Today, the "language question" has reemerged as a critical education policy issue in the context of learner achievement and education quality, particularly ongoing debates on the continued privileging of English and Afrikaans as mediums of instruction while the development and use of African indigenous languages are only beginning (Taylor et al. 2017). And herein lies the value of holding on to the notion of historical span-it ensures that critical policy researchers do not lose sight of the wave-like rhythm of policy change and development. In the process, "unresolved" issues in the quest for social justice can remain within the possibility of hope and resolution.
However, policy changes are not made easily, especially those with huge financial implications. This suggests that a pro-poor funding approach must be matched by a wellorganised civil society that is capable of successfully lobbying for education reforms, given COVID-19-related fiscal priorities of the health sector, and a deepening socioeconomic crisis characterised by high unemployment. While non-governmental organisations (NGOs) such as Equal Education and Section 27 have had some lobbying impact in pushing a pro-poor education agenda in South Africa, they will need to form larger, more powerful civil society alliances. At the very least, they will need to garner the solidarity of teacher unions, who have stronger bargaining power, in spite of competing interests, notably around teacher salaries. Indeed, teacher unions such as SADTU, ostensibly representing the interests of the working class, appear to have become too inward-looking; moreover, SADTU's autonomy in the policy arena has been compromised given its alliance with the ruling ANC (Govender and Cross, forthcoming). Other NGOs, such as Solidarity and Afriforum, representing mainly privileged white Afrikaner interests, have also staked a claim in the education policy arena, particularly in the context of COVID-19-as such, civil society interests have become quite diverse, further deepening policy contestation.
Alternatively, smaller but influential NGOs need to secure a seat at policymaking forums, both at the national and provincial levels, given their policy and technical acumen, which is often on par with teacher unions, and certainly more robust and critical of government (see, for example, Cross, Govender, and Essop, forthcoming). As Veriava (2010, 45) has suggested, this could be done through a public participation process whereby the Department of Education works together with education activists and lobby groups to define the elements that would constitute an adequate education in respect of non-personnel expenditure and then funding this accordingly. Moreover, establishing working relations with the health sector will be equally important to secure policy trade-offs to benefit education. Joint, collaborative lobbying of policymakers is likely to be more effective.

Conclusion
Adopting a long-term historical lens, and drawing on critical policy historiography as an analytical tool, this article has highlighted the reality that a vast majority of citizens remain estranged from matters of policy in the democratic era; and, in spite of a vocal civil society, school-funding policies and reforms have remained inadequate in addressing the needs of poor and working-class communities. Significantly, the unequal distribution of power in the genesis of South Africa's school-funding policymaking process in the 1990s and the continued dominance of academics and experts in policymaking initiatives since then remain important factors in compromising the influence of the country's poor and working-class constituencies on education policy outcomes. In order to address the persistent inequality challenges, an important area of focus should be the crafting of a policymaking framework that places the empowering of the poor and marginalised, such as SGB parent and community constituencies, at the centre. Civil society organisations, such as Equal Education, in partnership with other collective agencies, including teacher unions, can be supported to coordinate and lead policy literacy programmes. Teacher unions, moreover, will require considerable soulsearching on their part, so that they are able to balance serving the interests of their members as well as that of the public good.
While school-funding policies are seen as an important instrument in the quest for social justice, there is no simple panacea. This is largely because funding policies are inextricably linked to the broader social, economic and political contexts of policymaking. These include government fiscal constraints often leading to policy trade-offs and competing stakeholder interests. At present, though, government and policymakers find themselves in a policy quagmire with regard to school-funding policies given the multiplicity of economic and social policy demands in the wake of the COVID-19 pandemic. This will no doubt further hamper the advancement of social justice goals in education in the short-term at least.