A Foucauldian Discourse Analysis of Regional Balance in the Financing of Higher Education in Uganda: The Uganda Students’ Higher Education Financing Policy Perspective

In Uganda, there have been historically glaring disparities in access to higher education, where the majority of the students that join universities come from good secondary schools, which are mostly afforded by rich individuals. This has created a situation of ‘inherited merit’, where students from particular backgrounds and regions dominate access to HE. Because children of the poor who cannot afford to go to good schools hardly access higher education, whilst many of the students who join higher education institutions (HEIs) fail to complete their study programmes due to the failure to meet the costs involved. The Ministry of Education and Sports 2012 introduced the Uganda Students’ Higher Education Financing Policy to address the problem of inequitable access to higher education. Thus, the third objective of the policy aims to ensure ‘regional balance’ in the provision of HE through awarding of student loans. Because there are strong intra-connections between policy and language, the language used in discourse plays a critical role in the way the term Regional balance was constructed in the policy. This paper performs a Foucauldian discourse analysis perspective of elite existing regional social the how power is within the guise of regional balance. As the to a discursive understanding of regional balance the an appreciation of the role of power relations in policy rhetoric as a form of exemplary theatrical government, and to the politics of regional balance. The findings revealed that the term


BACKGROUND
Traditionally, discourse analysis has been associated with linguistics (Buchanan, 2008). However, several authors call for its application to educational policy research discipline (Ferreira-Neto, 2018;Warriner & Anderson, 2016;Lester et al., 2017;Rogers et al., 2016;Dunn, 2006). The rationale for the application of discourse analysis in educational policy research is that policy is made of language (Majone, 1989). Thus, there are strong interlocking connections between policy and language (Hewitt, 2009) hence making "language important for public policy analysis" (Pohle, 2013, p. 1). Gee (2015) has elucidated the basic types of discourse analysis by talking about "small d" discourse analysis (language-in-use) and "big D" Discourse analysis (the enactment of socially and historically significant identities and social structures). Another lens from which to discern discourse analysis is to conceptualize discourse as linguistic and textual vis-à-vis conceptualising discourse as socio-historical knowledge construction (Fairclough 2003;Foucault 2000a). Thus my goal for using discourse analysis to interrogate the regional balance discourse in the Uganda Students' Higher Education Financing Policy is to introduce not only an ontological approach to discourse that construes social reality in higher education policy studies but also to test different textual methods for the analysis of the term 'regional balance' as used in the policy and, thus epistemologically help to understand the ways in which we can understand the role of language in policy-making. But first, I will briefly recapitulate the core of Discourse Analysis (DA). Discourse analysis is generally an umbrella term for the diverse traditions by which discourse may be examined. It aims to critique cognitivism that developed in the 1970s, albeit it is deeply rooted in the 'turn to language' during the 1950s (Morgan, 2010).
As noted by Morgan (2010), cognitivism speaks of "objective, observable, knowable reality," whilst discourse analysis speaks of "multiple versions of reality, multiple 'truths', which are constructed through texts" (p. 1); thus, there are correspondingly multiple versions of analysis. In this regard, language is viewed as a social performance-it is productive and constitutive (language both creates social phenomena and is representative of social phenomena). This analytical lens aims to explore power relations from a critical point of view in an attempt to make sense of the social world by providing new critical insights. Scholars that have used discourse analysis have identified about five major competing traditions of discourse analysis, namely: Critical Discourse Analysis (CDA); Conversation Analysis; Interactional Sociolinguistics; Bakhtinian Discourse Analysis (BDA); and Foucauldian Discourse Analysis (Lester et al., 2017;Morgan, 2010). It is thus imperative that a distinction among these traditions of discourse analysis is drawn from the onset with a view to showing how they differ. Conversation analysis is an objectivist, realist position in which inductive, data-driven activity is achieved and whose main aim is to find patterns within language (the text) and describe what is there (Morgan, 2010). Thus within this inductive process, "patterns in language are discovered, not interpreted or constructed" (ibid, p. 2).
On the other hand, Interactional sociolinguistics is an analysis of power within linguistic practices, implying an exploration of patterns within language as a system (Morgan, 2010). Although similar to conversation analysis, interactional sociolinguistics posits that "members' interpretations of language form methods of dominance, not the words themselves" (Morgan, 2010, p. 2). Thus the focus on power is more pronounced than in conversation analysis. In Bakhtinian discourse analysis, language is considered to be alive and fluid, a struggle between centripetal (authoritative, fixed, inflexible) discourse and centrifugal (genres, professions, historical specificity, cohorts) forces (Brait, 2017;Breeze, 2011;Morgan, 2010;Gatt, 2004).
Accordingly, "everyday speech is patterned in speech genres -themes, constructions, and styles" (Morgan, 2010, p. 2). This model incorporates very specific concepts, and social conflict and ideology can be evidenced in evaluative accents or judgements, which are conveyed by words. Bakhtin makes explicit three linguistic yardsticks upon which to analyze and judge text or speech. First, the concept of heteroglossia identifies a dynamic multiplicity of voices, genres and social languages that project interactions within a social world (Morgan, 2010). Bakhtin asserts that reported speech can either be linear (reported verbatim) or pictorial (the reported voice is infiltrated with the voice of the reporter). Second, the concept of dialogicality, which explains that there is always one other voice implicit in any utterance; and third, chronotopicity, which infers the spatiotemporal nature of language (Morgan, 2010;Erskine, 2008;Smith, 2008).
Fairclough's Critical Discourse Analysis (CDA) is an interdisciplinary field that includes linguistics, semiotics and discourse analysis (the field of discourse studies) and is concerned with theorizing and researching social processes and social change (Curtis, 2014;Morgan, 2010;Breeze, 2011;Hart, 2007). Fairclough opines that language-in-use, which is socially and historically placed, and other facets of the social domain are in "a dialectical relationship that is 'socially shaped' but also 'socially shaping' or 'socially constitutive'" (dialectic-discussing the truth of opinions) (Snowdon & Karlsson, 2021, p. 4). Critical discourse analysis begins with the formulation of a perceived or real social problem, such as regional disparities in access to higher education. Thus the role of the analyst is to explore the tension between these two faces of language use rather than one side or another. Its primary focus is to "analyze how discourses are formulated, maintained and challenged" (Snowdon & Karlsson, 2021, p. 4). On the other hand, Michel Foucault's Discourse Analysis (FDA) is a system of language use involving the production of power/knowledge through language (Curtis, 2014). FDA posits that language plays a crucial role "not only in our understanding but also in the formation of our social world" (ibid, p. 7). Foucault's preoccupation understood the relationship between knowledge and power; and interrogated what and how people communicate through discourse (Snowdon & Karlsson, 2021).
However, across these divergent discourse analytic perspectives, there are several shared assumptions. First, language is assumed to be performative, as it is understood as the medium by which social life is accomplished. In other words, language is actionoriented. Second, many scholars using discourse analytic perspectives take up a social constructionist position and thereby position reality and knowledge as constructed in and through language. Finally, discourse analysts "take up a critical understanding of the world as they critique taken-for-granted knowledge and practices" (Lester et al., 2017, p. 7). Thus given that regional underrepresentation in access to higher education still persists in Uganda (Ministry of Education and Sports, 2012; 2019; Equal Opportunities Commission, 2019) despite the various affirmative action measures that have been undertaken by the government over the years, the paper suggests that new theoretical perspectives are needed to better understand the problem. To that end, this paper suggests that Foucauldian notions of power, which are currently under-engaged in the higher education literature, especially in the Ugandan context, could advance research on the regional imbalance in access to HE in Uganda from a policy perspective. I used Michel Foucault's discourse analytic perspective because it befits my epistemological assumptions that I hold about the use of language: that those who have the knowledge and power to formulate policies conceal their selfish interests and aspirations in language; that such power elites always ensure they maintain a grip of control and dominance on certain sections of society through policy rhetoric; that political architects use their power and authority to influence the policymaking process and implementation. Thus certain regions of the country end up benefiting more from government resources at the expense of other regions of the country.
The above epistemological assumptions enable me to hold the view that in order to emancipate those individuals marginalized by power elites, there is a need to critically analyze the language and structure of policy so as to reveal power relations of policy making and to highlight gaps between rhetoric and practice of policy (Sharp & Richardson, 2001;Flyvbjerg, 2001;Hajer, 2006) and expose the power relations hidden in the discourse of regional balance. Moreover, Foucault's discourse analysis enabled me to have insight into the functioning of bodies of power and knowledge as they are situated into the term regional balance as used in the policy. This enabled me to generate interpretive claims with regard to the power effects of discourse within regional balance discourse without claims of generalizability to other policies on (higher) education. In other words, as Foucault (1970) argues, this particular discourse analytic perspective enabled me to uncover the diverse influences that define a policy problem.

Statement of the Problem
Higher education accrues enormous benefits both to a country in general and to the individual graduate in particular (Wanti et al., 2022). This makes fair regional representation in higher education imperative. However, in many countries, there are glaring inequities that characterize HE sectors rooted in domains between "rural and urban areas, between rich and poor people, between majority and minority groups, and between male and female" (Guiake, 2020, p. 2). In the case of Uganda, due to the Government's introduction of Universal Primary Education (UPE) in 1996 and Universal Secondary Education (USE) in 2007, there was an upsurge in the number of students qualifying to join higher education institutions (Ministry of Education and Sports, 2012). Earlier in 2005, the Government had introduced the District Quota system, whose intrinsic objective was to ensure regional balance in access to HE by supporting academically outstanding students from each district, allocating a certain number of slots in public universities (Arinaitwe et al., 2015). Furthermore, the Government 2012 introduced the Uganda Students' Higher Education Financing Policy, whose underlying objective was to "increase equitable access to higher education in Uganda" (Ministry of Education and Sports, 2012, p. 2). The government observes that "due to the costs involved, many students fail to join universities or tertiary institutions and even those who are admitted to the universities fail to sit for the examinations due to failure to pay the required fees" (Ministry of Education and Sports, 2012, p. 3). Thus, "the majority of the students who join universities are those who are able to access the best secondary schools in the country and these are not regionally balanced" (ibid, p. 13). Basing on this finding, objective three of the Uganda Students' Higher Education Financing Policy aims to "ensure regional balance in the provision of higher education services in Uganda" (ibid, p. 14) through advancing loans to needy students from all regions of the country. Furthermore, the Government has established and encouraged the establishment of new universities or liaison university colleges and other institutions of higher learning across all regions of the country (Eton et al., 2020) with the aim of promoting (regional) access to higher education. Consequently, about 50 universities and 150 other tertiary institutions have been established in Uganda both private and government (National Council for Higher Education [NCHE], 2020) although the biggest percentage (72%) of tertiary enrolment are enrolled in Universities (Uganda Bureau of Statistics [UBOS], 2020). Plausible as these government interventions may be, they have not significantly translated into equitable access to university education in Uganda as regional disparities are conspicuously apparent in Uganda's higher education sector (Kimoga, 2021;Ministry of Education and Sports, 2020;UBOS, 2020;Equal Opportunities Commission [EOC], 2019).
In light of this, this paper aims to critically interrogate and unmask the power relations and the discursive practice of the regional balance discourse as produced in the Uganda Students' Higher Education Financing Policy. Regional imbalance in access to higher education that is already prevalent in Uganda may worsen due to a lack of evidence in research and poor decisions in the policymaking and implementation process. These discourses may influence each other, allowing a misuse of power that produces and perpetuates regional disparities in access to students' loans to access higher education. Thus this paper aims to address a significant gap in the existing academic literature about the discourse of regional balance in access to students' loans to facilitate access to higher education by using Michel Foucault's discourse analytical framework. Through examination of the language used to construct the discourse of regional balance, the paper aims to interrogate and uncover the political effects of the discourse of regional balance in access to higher education and seeks to determine whether this truth presented in objective three of the policy creates unequal power relations between the different regions of Uganda when it comes to accessing students' loans so as to facilitate access to higher education.

RELATED LITERATURE
Literature on access to higher education indicates that in many countries [including Uganda] there are usually geographical disparities that leave some regions of the country flush with traditional agedstudents and some areas seeing a reduction when it comes to accessing HE (Alphin et al., 2017; Asian Development Bank [ADB] Report on HE, 2012). Geography, combined with other factors, contributes to the disadvantage of certain populations (Altbach et al., 2009), where disparities between regions continue to jeopardize equity and access issues of many students (Koshy & Seymour, 2014), who eventually end up being disadvantaged and excluded from attaining HE. Unfortunately, however, geography is easily ignored as a factor that contributes to inequitable access to HE (Altbach et al., 2009). Thus, raising educational aspirations and educational attainment in under-represented communities should be part of the widening participation agenda by governments and HEIs (Purcell, 2008). This perhaps justifies the third policy objective of aiming to ensure 'regional balance' in the provision of student loans to facilitate access to HE by the historically underrepresented or marginalized regions of Uganda.
However, a myriad of literature on policy studies shows that the term 'regional balance' is largely a political term used by governments (Balaam & Dillman, 2011) and "it can be attractive to many different political and economic actors for different reasons, and to serve many purposes at once" (ibid, p. 301). Thus policies determine politics (Lowi 1964;Smith 2002) in which one social group benefits from public funding at the expense of others (Lipsky, 1980). Hence every category of policy has its own political dynamics (Lowi, 1964;McCool, 1995). Thus to better understand how the discourse of 'regional balance' constitutes power, I engaged Michel Foucault's discourse analysis as an analytical framework. Moreover, Foucault argues that it is not possible to study the discourses without an analysis of the political rationality underpinning them" (Foucault 1997b, p. 67;Lemke, 2001, p. 2).
Moreover, there is an emergent and increasing use of discourse analysis within educational research (Warriner & Anderson, 2017). The fundamental reason why I employed a Foucauldian approach was a desire to understand regional balance as something more than a neutral concept or scientific term employed on behalf of the public good -higher education, but also to avoid reducing it to an ideological mystification or piece of cynical hypocrisy that merely entrenches existing power relationships. My choice to apply Foucault's analytical lens was also a desire to examine what regional balance actually does and how it fits into contemporary relations of power/knowledge, discourse and regimes of truth.

Power/Knowledge and Regional Balance
The concept of power is as old as the history of philosophy (Moghadam & Rafieian, 2019). The scope of power theory can be traced back starting from Plato and Aristotle and brought to the contemporary period through Machiavelli's power theory to Gaventa's (1982) theory of power based on a model of power and powerlessness; Mann's (1986) theory of power whose postulation is that power relations go to those who have a better organization; Clegg's (1989) triple circuit of power comprising of the obvious power circuit, social power circuit and power systemic and economic circuit; and Giddens' (1984) power theory which augments social theory that he termed as 'structuration' (Moghadam & Rafieian, 2019). However, Michel Foucault's theory of power represents a crucial analytical framework from which to understand the concept of power. Foucault conceptualised power as relational and was largely "interested in power relations between individuals, social structures, and institutions" (Mills, 2003, p. 23). Therefore, Foucault was interested in the relationships between knowledge and power and was particularly interested in what and how people communicate through discourse [such as regional balance] (Snowdon & Karlsson, 2021). Foucault's thought radically reconceptualised how we think of power. Power, in this thinking, is no longer seen merely as a repressive force that commands obedience to a set of rules but as a complex system dynamically linked to knowledge . Foucault posits that knowledge and power are closely intertwined and thus denotes this interweaving as power/knowledge (Geerlings & Lundberg, 2018;Assche et al., 2017). Thus, power is a topic which suffuses throughout all of Foucault's works. In its archaeological phase, knowing was considered a composer of power forms, which were instituted into the social body. In its genealogical phase, power had a relational perspective with knowing and with the institutions, leading to the appearance of human mankind (Brito et al., 2020). In the ethical domain or methodologically denominated arch-genealogical, the "relationships of power were presented before an ontology of the present to set aside the individual, their subjective constitution and the practices of freedom" (ibid, p. 3). For Foucault, "it is not possible for power to be exercised without knowledge, it is impossible for knowledge not to engender power" (Foucault, 1977, p. 52). In light of this, this paper submits that regional balance as a discourse refers to power and implies power effects.

Discourse and Regional Balance
The ways we think and talk about a subject influence and reflect the ways we act in relation to that subject. This is the basic premise of discourse theory (Foucault, 1972. Foucault famously described discourses as "practices that systematically form the objects of which they speak" (Foucault, 1984, p. 54). In this regard, discourses shape not only what is said but also what is thought and what can be known and accomplished. In that way, "discourses are intrinsically bound up in the operation of power" (Beddoes et al., 2013, p. 3). Thus discourse, knowledge and power are inextricably linked and mutually reinforcing. As Foucault explained, "discourse is not simply that which translates struggles or systems of domination, but it is the thing for which and by which there is a struggle [e.g. regional balance in accessing students' loans to access higher education]; discourse is the power which is to be seized" (Foucault, 1970, pp. 52 -53). In light of this, formations of power/knowledge such as regional balance may be referred to as discourses, and they constitute certain ways of thinking about, representing and acting upon the world. It should be noted that the notion of discourse plays a key role in Foucault's theory of power/knowledge. Discourses can be defined as "the domain of subconscious knowledge that guides the games of truth which define which things become articulable as knowledge" (Foucault, 1984, p. 40).
For example, in his three volumes of The History of Sexuality (1978,1985,1988), Foucault elucidates how in the eighteenth and nineteenth centuries, "a discourse of sexuality came to define how people looked at pleasure, kinship relations and interpersonal relations from a particular point of view related to sex, and how this discourse excluded other views on these topics" (Geerlings & Lundberg, 2018, p. 7).
Thus, discourses are lenses through which people view the world and which account for what is considered knowledge and truth. Discourses thereby produce statements of knowledge and certain realities. Based on these realities, specific social imaginaries are constructed, which affect our epistemology and ontologyour ways of knowing and being in the world. In other words, discourses create specific types of subjectivities: "we do not speak discourse, discourse speaks us" (Ball, 2012, p. 40). Essentially, discourse analysis may uncover the diverse influences that define a policy problem (Foucault, 1970). Moreover, the aim of Foucauldian discourse analysis is to construct a critical narrative of text or speech (Hajer, 2006). The idea of power/knowledge is something that can reveal the discourse of regional balance in the awarding of student loans through the Uganda Students' Higher Education Financing Policy.
Foucault's analysis is fundamentally based on the hypothesis that knowledge has an inseparable connection with power networks (Ure & Testa, 2018). Every power relationship creates its own sphere of correlated knowledge and every piece of knowledge harbours a unit which requires and binds power relations (Moghadam & Rafieian, 2019). Thus in light of the regional balance discourse, this paper argues that it is not with the distribution of power, but indeed among the various players or key stakeholders in the implementation of the Uganda students' loan scheme, the "induction of knowledge and the production of the desire to know" (Foucault, 1980, p. 54). Foucault opines that it is not possible to divide or enforce power. Rather, power exists in space and the will to know can extend the range of power of each player (Foucault & Ewald, 2003).

Foucault's Regimes of Truth and Regional Balance
Foucault interrogated "truth" through epochs of epistemologies referred to as 'epistemes' (Geerlings & Lundberg, 2018). In other words, he sought to "discover what was considered truth at different times, without seeing the present as a causal, linear progression from the past" (ibid, p. 7). An important epistemological consequence is that "truth can never be more than a thing of this world, a social construction" (Foucault, 1994, p. 133). Foucault argued that "truth is linked in a circular relation with systems of power which produce and sustain it, and the effects of power which it induces and which extend it"; thus "truth is already power" (Foucault, 1980, pp132 -133). Foucault's (1978) notions of regimes of truth, along with the associated constructs of will to knowledge and will to truth, are taken up to deconstruct the binary stances and contradictions within these domains with the aim of a reconstruction that captures the affordances of both. According to Dunkerly-Bean & Bean (2016), such structured systems "promote the varying will to truth dimensions" (p. 2) and accordingly contribute to the regional balance discourse. With the concept of what constitutes knowledge as a whole, this paper examined the notion of the will to knowledge and the will to truth in the discourse of regional balance.
Foucault conceptualised the will to knowledge as a general desire to know and the will to truth as the desire to determine the difference between truth and falsehood in a particular discipline (Ball, 2012). However, as Gore (1993) postulates, "the 'will to truth' which characterises much of the intellectual work is such that the need, desire, or willingness to question one's own work is often lost in the desire to believe that one has found the 'truth,' that one is right" (p. 11). Although the term 'one' can refer to individualistic or idiosyncratic belief,  perceived it as the "many struggling within a politic of truth" (p. 20), which is how it was utilised in this paper. It is in the seeking, debating, and codifying of truth about regional balance in the provision of students' loans to facilitate access to HE that in turn, creates the regimes of truth that characterize sites of power relations. Thus  original conceptualization of regimes of truth is that: Each society has its regime of truth, its general politics of truth: that is, the types of discourse that it accepts and makes function as true; the mechanism and instances which enable one to distinguish true and false statements, the means by which each is sanctioned; the techniques and procedures accorded value in the acquisition of truth, the status of those who are charged with saying what is true (p. 131).
In examining the interplay between truth, power, and knowledge in any discourse such as regional balance, an examination of the ways in which power, knowledge, and truth are named and enacted is necessary. To , "power and knowledge do not operate in isolation of, or in opposition to the other" (p. 3). Rather, the two are interconnected. As  explains, "there is an administration of knowledge, a politics of knowledge, relations of power which pass via knowledge, and which, if one tries to transcribe them, lead one to consider forms of domination designated by such notions as a field, region or territory" (p. 9). Given that the discourse of regional balance embodies power relations between policymakers, policy practitioners, human rights activists, social justice researchers, the political elite, academics and students' loan scheme beneficiaries in HEIs and what is counted as knowledge, it can be inferred that regional balance constitutes a regime of truth.  viewed societal discourse as a structured system for the production, regulation, distribution, and circulation of statements that are sustained by power relations. In this sense, "truth is generated by this system and sustains itself via a circular system of power" (Foucault, 1980, p. 133). Yet, it is important to note that for Foucault, there are no inherently liberating practices (or discourse). Rather, it is in discourse that power and knowledge are joined together: We must not imagine a world of discourse divided between the accepted discourse and excluded discourse, or between the dominant discourse and the dominated one; but as a multiplicity of discursive elements that can come into play in various strategies.
Discourses are not once and for all subservient to power or raised up against it, any more than silences are. We must make allowances for the complex and unstable process whereby discourse can be both an instrument and an effect of power but also a hindrance, a stumbling block, a point of resistance and a starting point for an opposition strategy (Foucault, 1978, p. 100).
In Uganda, several studies have been conducted on the current student loan policy. For instance, a study conducted by Onen et al. (2015) found that students loan schemes in Africa have a cross-cutting challenge of inadequate funds to fully operationalise them; as well as the challenge of using an effective means-testing to determine true needy children that deserve the loans. Wanyama et al. (2016) study found that Uganda's student loan scheme is a major predictor of access to HE although it is narrow in the scope of coverage as it provides loans to students pursuing science-related programmes, science, technology, engineering, and mathematics (STEM). Kyaligonza's (2017) study explored gaps in the implementation of Uganda's students' loan scheme and found that Uganda's Higher Education Students' Financing Board (HESFB) is facing difficulty in recovering the loan repayments from the beneficiaries of the loan scheme and thus recommended a robust awareness campaign on the obligation to repay the loans by the beneficiaries. From these studies, it is evident that discourse analysis, let alone Michel Foucault's discourse of power/knowledge, remains under-engaged in the current students' loan policy. This paper employs a Foucauldian discourse analysis as an analytical lens to examine the power relations imbued in the discourse of regional balance.

METHODOLOGY
The study used an interview guide with semistructured open-ended questions to interview one Higher Education Students' Financing Board executive official (hereinafter HESFB Official) who was sampled purposively, and four students' loan scheme beneficiaries (Simon, Mary, Maureen, and Mathew) who were sampled through snowball sampling in one Public university. Also, the study employed a document analysis of the regional balance discourse as used in the Uganda Students' Higher Education Financing Policy. Most importantly, the paper used an analytical lens inspired by Michel Foucault to interrogate the power relations embedded in the discourse of regional balance. The justification for engaging in Foucault's discourse analysis was the desire to examine what regional balance actually does and how it fits into contemporary relations of power/knowledge, discourse and regimes of truth.
Within the construction of the policy, the study explored the distinction between the intended purpose of the policy and the way it is being implemented or how it functions. This was done by critically analyzing how and why the discourse of regional balance was used in the policy. Document analysis aided the distinction between "manifest" (or "obvious") content and "latent" or underlying and inferred meaning (Wood et al., 2020, p. 5) of the term regional balance, and this helped me to deconstruct contradictions and power relations within the policy rhetoric. Ward & Wach (2015) contend that when doing document analysis, "other sources of information that could be used to triangulate and strengthen confidence in the findings of the study should be employed" (p. 6).
Thus, the study also employed a document review method in order to supplement the other methods of data collection. Whereas document analysis involves the rigorous and systematic analysis of the contents of written text (Ward & Wach, 2015) aimed at "obtaining deeper and underlying meaning" (Wood et al., 2020, p. 5), as well as implications (Ward & Wach, 2015) of the text, document review, on the other hand, is a way of "collecting data by reviewing existing documents" (ibid, p. 6). This study employed a document review, specifically, the Minutes of the Parliamentary Committee on Education at the time when the Uganda Students' Higher Education Financing Policy was formulated. This document is technically referred to as the Hansard (a document that includes all the official verbatim records of debates in Parliament). I reviewed the Hansard because after the Ministry of Education and Sports drafted the policy, it was discussed and debated by the Parliamentary Committee on Education at the time, upon which the full Parliament debated and enacted the Uganda Students' Higher Education Financing Act (2014) that gave legal backing to and operationalised the policy, which effectively commenced in 2014/2015 academic year with the first cohort of beneficiaries.

Regional Balance and Foucault's Analytical Perspective
Foucault's writings prior to Discipline and Punish (Foucault, 1977) are often referred to as 'archaeological' texts. In these textsnotably Madness and Civilisation, The Archaeology of Knowledge, and The Birth of the Clinic (Foucault, 1965;1973) Foucault mapped out a methodological guide that explored historical conditions that had warranted particular ways of thinking and acting. Foucault's re-examination of the role of statements and discourses in the construction of knowledge was realized through a number of methodological imperatives, principles and rules that Foucault elucidated quite sporadically through his later writings (Foucault, 1981a;1981b). In this framework, "statements form the basic unit of analysis in the archaeological inquiry because they make objects, subject positions, concepts and strategies visible, and consequently they become amenable to analysis" (Nicholls, 2014, p. 3).
At the same time, in exploring these statements, we learn something about the ways in which they are made visible in the first place: the matrices of power relations that make certain forms of knowledge authoritative (and thereby acceptably expressed as statements) (Nicholls, 2014). Here the paper sought to use Foucauldian steps of deconstructing political concepts to deconstruct regional balance as a discourse. In order to interrogate the power relations imbued in and understand the discursive construction of the regional balance discourse, the study employed Foucault's archaeological strategies that he laid down in his The Question of Methods (1981). These include; the Rule of Discursive Formation, the Rule of Discursive Correlation, and the Rule of Discursive Transformation.

Rule of Discursive Formation
The first archaeological rule relates to the Rule of discursive formation and it considers the way in which some discourses are formed by particular statements and not others (Nicholls, 2014). Foucault encouraged analysts to interrogate the ways that "certain objects, subjects, concepts and strategies make particular thoughts, actions and behaviours possible, and the way these relate to the construction of knowledge and the formation of texts" (ibid, p. 4). Foucault explored the surface of the emergence of concepts, the authorities that gave weight to these discursive constructions (what Foucault called the 'authorities of delimitation'), and the ways in which the concepts are classified, organised, divided and regrouped (or 'grids of specification') (Foucault, 1972). In the current study, here the fundamental question that was explored is; what was the socioeconomic situation at a time that could have warranted the construction of the term regional balance in the policy? The findings revealed that the Minister of Education and Sports at a time or Government generally had observed that: The majority of the four-thousand (4000)  Exploring the surface emergence of regional balance, it is evident that regional balance is an equitable and inclusive term aimed at empowering the less privileged regions to enable them access higher education so that they acquire knowledge, which is a reflection of power, hence the adage, 'knowledge is power'. The Minister's submissions allude to this empowerment: "The Board [HESFB] is allowed to take into consideration regional balance" (Parliament of the Republic of Uganda, 2013, p. 10164); and "the Ministry [of Education and Sports] will ensure that the beneficiaries of the [loan] scheme are generated from across the country, and all the children of Uganda will benefit from this scheme, including those from Bugisu, especially Mbale" (ibid, 2013, p. 10234). To that, various Members of Parliament (MPs) added emphatically thus: "We want something, which is transparent, which will treat all Ugandans as Ugandans; put them on the same platform and those who will meet the criteria should be given" (Parliament of the Republic of Uganda, 2013, p. 10156); and "the privileged of this country should be disqualified from this loan scheme. There are very many brilliant young people in this country in the remote areas who have no uncles in Kampala who deserve university education" (Parliament of the Republic of Uganda, 2013, p. 10106).
From the above excerpts, it can be seen that both the Minister and Members of Parliament wished for a more inclusive loan policy in order to enable children from apparently marginalized regions also access higher education. There is an acknowledgement that the policy is a step toward promoting equity and social justice to those needy students who may not be able to afford higher education. For instance, phrases such as; "the privileged of this country should be disqualified from this loan scheme"; "we want something, which is transparent, which will treat all Ugandans as Ugandans." These statements about the envisaged policy suggest a strong connection with a notion of regional balance and social justice towards those perceived as powerless or marginalized but at the same time, reveal that there should be a system which requires specific actions of HESFB officials to be "transparent", "non-discriminatory" (Parliament of the Republic of Uganda, 2013, p. 10164); and "unbiased treatment of all Ugandans as Ugandans" (ibid, 2013, p. 10156) for the policy to be successful in ensuring regional balance. These are examples of the operation of meaning in the discourse, where the beliefs and practices of the institution are a discourse that conveys certain ways of acting and thinking, resulting in a power-relation that normalizes these actions. This normalizing process is an example of the productive nature of power-relations. Nevertheless, regional balance becomes a political term as far as it was uttered by a political person in her capacity as Minister of Education and Sports, at the same time a Member of Parliament (politician), and supported by fellow politicians (MPs) in a political context (Cabinet of Government Ministers and then debated in Parliament).
Foucault also spoke of the formation of subject positions. Here, his preoccupation was to explore the ways in which discourses benefit certain sections of the population whilst marginalizing others. For instance, Members of Parliament from apparently marginalized regions may have had the knowledge and experience of such marginalization for they had represented such regions for a long time and thus, their voices exhibited the imbalance that exists between different regions. For instance, on the issue of the policy supporting Science, Technology, Engineering and Mathematics (STEM) programmes, many MPs, especially from Northern Uganda and West Nile objected to it. Supported by a couple of MPs, one MP argued: In the history of our education system, we are going to marginalize  Similarly, another MP argued that:

When I look at the performance in my District last year, you wouldn't see any child qualifying [for loan scheme] and that is not only in Agago
District, it is in most of the rural districts in this country. When you visit all our Secondary schools in the rural districts, you will never find science laboratories; you will never find science textbooks; you will never find enough science teachers; you will find a lot missing (Parliament of the Republic of Uganda, 2013, p. 10220).
The above voices suggest that there may be institutional processes that promote and sustain regional imbalance thereby exacerbating social injustice that some of the regions feel have suffered.

Rule of Discursive Correlation
The second Foucault's archaeological rule is the 'Rule of discursive correlation' and is concerned with the fluid inter-relationship between discursive formations. This rule concerns the way "discourses intersect, abut, compete, overlap, dominate, marginalize or negate one another" (Hook, 2001, p. 41). Foucault argued that this interconnectedness between discourses needed to be explored at a microscopic level (between subjects, objects, strategies and concepts) and at a macroscopic level (between discursive formations, competing knowledge and power effects), and so this rule encourages us to pay kin attention to the relational qualities of discursive formations (Powers, 2001).
Here the question that this paper explored is; in which other way was the term regional balance used? In studying the concept of regional balance, the correlations between discursive formations played an important role because they provided a means for interpreting the changing context in which Members of Parliament conceived of the term regional balance. In other words, by exploring the correlations between the discourse of regional balance, it was possible to analyze the changing context in which MPs conceived of the term regional balance.
For instance, MPs first conceived of regional balance from the equity and emancipatory point of view aimed at enabling children from underprivileged regions (who cannot afford elite schools to qualify among the 4000 Governmentsponsored students in Public universities) also access to HE. The voice below alludes to this emancipatory worldview: These loans should go to the most-needy Ugandans; the privileged of this country should be disqualified from this loan scheme. There are very many brilliant young people in this country in the remote areas who have no uncles in Kampala who deserve university education. If we do not put in that restriction, we are going to end up with our own children getting these loans, yet we have other means of financing our children (Parliament of the Republic of Uganda, 2013, p. 10106).
Another context within which the MPs may have conceived of regional balance regarded the political discontent among the opposition MPs who had knowledge of how the State House scholarships had for a long time, largely benefited children from Western Uganda and the privileged ruling political party -National Resistance Movement (NRM) cadres, which discontent may have informed their conception of regional balance: Thus, the MPs' conception of regional balance is that the policy ought to treat all Ugandan children as 'equals' "put on the same platform." Therefore, the MPs wished the policy was implemented with social justice and fairness to benefit needy children from all regions of the country.
Another worldview that may have informed the MPs' conception of regional balance pertained to the study programmes that the loan policy was meant to finance, that is programmes and courses that were regarded as "critical to national development" and were to fall mainly in the categories of "Science, Technology, Engineering and Mathematics" (Ministry of Education and Sports, 2012, p. 23) which some MPs supported while others strongly objected as it would create inequitable access to the loans and thereby perpetuating regional imbalance in access to higher education as some regions do not pass science subjects in Secondary schools so as to qualify for the loans, yet they may be vulnerable and wish to study Arts and Humanities. For instance, the voices below are evident of this conception: In the history of our education system, we are going to marginalize  From the above voices, it appears the MPs' conception of regional balance was for the policy to support both Sciences and Arts and Humanities programmes so that beneficiaries can be drawn from all regions of the country -those that pass science subjects as well as those that pass Arts subjects in High school. Thus, it is evident that the MPs' conception of the regional balance discourse was informed by three changing contextual orientations, namely regional balance informed by social justice and emancipatory imperatives; regional balance informed by the political discontent against State House scholarships that were traditionally perceived as privileging Western Uganda and NRM cadres; and regional balance informed by Programmes to be funded by the loan policy.
Foucault spoke of the formation of subject positions. Here, his concern was to explore the ways in which discourses privilege certain subject positions whilst marginalizing others. The study participants intimated that West Nile and Greater North regions are put in a feeble position when it comes to awarding the study loans, thereby marginalising those regions. Simon alluded to this marginalisation: Thus the above participant forms beliefs and actions within a framework of what he perceives as a regional imbalance. To Foucault, this is a normalizing process where the HESFB has continued to award more loans to students from Western Uganda, yet Government has continued to convince the stakeholders that the policy is achieving regional balance. For instance, it appears students from "West Nile and that part of Greater North" (Simon) or generally "Northerners" (Mary) see themselves as underprivileged and marginalised and unless "you have a God-father" (Simon, Maureen) or "you have money to influence the Board" (Simon) otherwise "the poor are left [out] and the money is given to the rich people who are able to influence the board" (Simon) whilst, the HESFB also perceives these marginalised regions as being responsible for their own marginalisation. The HESFB Official opined that:

But in this country, information dissemination has become a major challenge. For instance, how many people have read this Newspaper today, Ugandans who have accessed this New vision [newspaper] today? How many people have accessed the radio? The radios are very many but what is on that radio that people listen-to to know what is going on in the country? It is also subjective. TV [Television]
, how many people have access to TV? A lot is going on and very few people know it. So the few who know benefit, the many who do not know cannot benefit. Moreover, regional capital may be contributing to such regional disparities in favour of Western Uganda. The HESFB Official may have implied this when he posited that: The West has got the biggest number of people working in government and government institutions. So, they know government programmes and they know which government programmes can benefit which people. So they promote the programme in their area. Members of Parliament (MPs) and people working in government Ministries, Agencies and other government Bodies. So they know and they promote it.

Rule of Discursive Transformation
Foucault's third archaeological rule is the 'Rule of discursive transformation' and it implores analysts using this tradition of discourse analysis to explore how discourses shift and change over time (Nicholls, 2014). Where are shifts occurring? What changes are happening 'internal' to the discourse? What effects are these changes having on the relationships with other discursive formations? Thus the analyst's task is to map the transformations in these discourses over time and explore the changing contexts in which they operate. From the findings of this study, the discourse of regional balance appears to have mutated seismically especially since the promulgation of the 1995 Uganda Constitution which forms a bedrock on which government policies, laws, standing orders and modus operandi are anchored. At the height of the Constitutional reforms that took place and the eventual promulgation of the 1995 Uganda Constitution, it was vital that the Government took "affirmative action" [Art. 32, p. 52] to ensure "balanced development of the different areas of Uganda and between rural and urban areas" [Obj. As the Millennium Development Goals (MDGs) and Sustainable Development Goals (SDGs) have gradually taken center-stage in government policymaking processes and national development plans, especially the Second National Development Plan [NDP II], ensuring regional balance in the provision of HE meant government "increasing the number of scholarships [not loans] for disadvantaged areas to enroll in higher education" (Republic of Uganda, 2016, p. 12) so as to ensure 'inclusive and equitable quality education and promote life-long learning opportunities for all' as per SDG 4. Specifically, Target 4.3 of the SDG 4 calls on Governments to ensure "equal access [not affirmative action] for all women and men to affordable and quality technical, vocational and tertiary education, including university" (United Nations, 2015) so as to address the disparities in access to HE.
Furthermore, the government brings on board a new paradigm shift of the construct of regional balance in the provision of HE in Uganda, which focuses on the government establishing and/or encouraging the establishment of new universities and other higher education institutions in all regions of the country so as to ensure balanced regional access to HE in the country (Eton et al.,, 2020). However, Government adds insult to injury by turning against these same regions and sub-regions and lambasting them in the NDP III for frustrating government efforts to eradicate poverty in the country. The government aims at these regions by postulating that, "the overall national efforts in poverty eradication are being derailed due to some regions dragging the poverty-reduction gains" (Uganda National Planning Authority [NPA], 2020, p. 206). Such regions include "Bukedi, Busoga, Karamoja, Bugisu, West Nile, Acholi, and some parts of Bunyoro" (ibid, p. 206).
However, emerging perspectives on the concept that constitute the notion of regional balance seem to have been taken over by neo-liberal economic imperatives, which require the government to ensure "regional balance in development" by "targeting poverty-eradication measures in these [less privileged or less developed] areas" (NPA, 2020, p. 206) so as to achieve "balanced and equitable development" (ibid, p. 29). Foucault did not consider that discourses could be 'defined'since this might reinforce the view that they were monolithic entities warranting descriptioninstead, he argued that their relationships, tactics, operations, oppositions, etc., should be 'mapped' across a broad terrain of events . In this paper, it became necessary to map an array of discursive formations within which the meaning of regional balance changed over time, making it almost a transitory term. Keeping faith with Foucault's methodological intentions, this paper made visible the various statements that cohere around the discourse of regional balance and in so doing, it explored the relations of power that made these statements visible.

DISCUSSION
Objective three of the Uganda Students' Higher Education Financing Policy aims to "ensure regional balance in the provision of higher education services" (Ministry of Education and Sports, 2012, p. 14). However, study findings have revealed that the implementation of the policy, or indeed, awarding of student loans is marred with glaring regional disparities. The study findings have revealed that since its inception, the Western part of the country has had a significant share of the students' loan policy, followed by Eastern Uganda and Central region while regions of West Nile, Northern Uganda and the Greater North have not benefited much from the loan policy. This finding resonates with the finding of the Uganda Equal Opportunities Commission Report which found that between 2014/15 and 2019/20 academic years, "the Western region had the highest number of student loan scheme beneficiaries (37%), followed by Eastern Uganda (28%), followed by Central region (24%), followed by the Northern region (10%)" (Equal Opportunities Commission, 2019, p. 287). The study findings have revealed that this regional imbalance in awarding of student loans is in most part explained by the dominance of people from Western Uganda in most of the Government Ministries, Departments and Agencies (GoMDAs) who often use their political capital to influence the HESFB. The HESFB official alluded to this when he said that; The West has got the biggest number of people working in government and government institutions. So, they know government programmes and they know which government programmes can benefit which people. So they promote the programme in their area. Members of Parliament (MPs) and people working in government Ministries, Agencies and other government Bodies. So they know and they promote it. Of course, not withstanding that political influence also comes. [But] we have tried our level best to be as objective as we can. It is not national players but individuals within the political environment. Yes, they come but we safeguard ourselves by being objective in what we do. Cochran and Malone (2014) argue that policy making process is part of politics and political action in order to keep the issue of contestation [such as regional balance] always live and like an on-going concern. Thus unequal access to political structures represents the efforts of the government to create a controlled society, and this is how government tries to control the behaviour of free subjects and thus a certain social group [or region] continues to dominate society (Lemke, 2015). The findings have demonstrated that people from Western Uganda who are more represented in GoMDAs, and thus virtually control the political landscape, have benefited more from the students' loan policy compared to other regions which may be less represented in political structures. This finding supports Balaam and Dillman's (2011) postulation that in policy discourse the term regional balance is largely a political term used by governments as a camouflage to cover up their real political agenda. Thus, the term regional balance "can be attractive to many different political and economic actors for different reasons, and to serve many purposes at once" (ibid, p. 301), where policies determine politics (Cairney, 2012) in which one social group [or region] benefits from public funding at the expense of others (Lipsky, 1980). Hence every category of policy has its own political dynamics (Lowi, 1988;McCool, 1995, cited in Cairney, 2012. Nevertheless, Cochran and Malone (2014) opine that once power elites hijack the policymaking process and implementation, they "permit the assimilation of some non-elites into the elite category, but only after they [non-elites] accept elite values, in the process encouraging system stability and reducing the threat of revolution" (p. 9) or fear of "conflict expansion," which occurs when "an increasing number of disgruntled people mobilize around a policy issue" (Masse, 2018, p. 2). In light of this argument, it can be understood that although the HESFB has awarded enormous loans to students from Western Uganda, and many from Eastern and Central regions, a few students from West Nile, Northern region and the Greater North are selected (or assimilated) into the bracket of beneficiaries to mitigate possibilities of violent sentiments (conflict expansion) that would otherwise emanate from severe disgruntlement. Such nostalgic sentiments are apparent in Mary's submission: For them [HESFB] they just see your name starting with 'N', 'A', so it is not fair, I know majorly they take those complicated names, not Northerners, I don't know why. Even if you get the newspaper after putting the list out you will see this percentage, you will get most of the people who have gone through, they are not from that side of the North. Dunn (2006) posits that because policy making is inherently a political process, "imaginative policy ideals" [such as regional balance] (p. 7), or "suppressions" [marginalising certain regions or social groups] (p. 10) are never innocent. Rather, they are usually used in "policy and political discourse to assist in legitimisation of a social practice" (ibid p.10) hence making the social actor [underprivileged regions] "powerless" or virtually "non-existent" (ibid, p. 10). This position is also echoed by Nikolai et al., (2022) who contend that State politics is a dominant cause of regional disparities in access to higher education in many countries.
The findings have demonstrated that the marginalized regions are subjugated on three main fronts. First, the fact that such regions and/or subregions are less represented in Government Ministries, Departments and Agencies and therefore these regions have limited political capital. Second, the policy's focus on mainly STEM programmes in which case many students, especially from upcountry or rural regions may not qualify for funding due to "relatively poor academic performance in most of the rural-based schools" (Parliament of the Republic of Uganda, 2013, p. 10102) especially in science subjects in High school, particularly in the national examinations-Uganda Advanced Certificate of Education (UACE) examinations, resulting from lack of adequate science equipment as well as well trained and qualified teachers to teach science subjects. This sentiment was shared by a section of Members of Parliament. This finding supports previous studies which stress that in many developing countries [such as Uganda], there are usually geographical disparities that leave some regions of the country having many students attending higher education institutions and some areas seeing a reduction or non-attendance (Alphin, 2017;Asian Development Bank Report on HE, 2012;Altbach et al.,, 2009). Thus disparities between regional and remote students continue to jeopardize equitable access to HE of many students (Koshy & Seymour, 2014), who eventually end up being disadvantaged and excluded from attaining higher education. Meusburger (2017) argues that "deepening access to higher education is about overcoming social differences and often also about overcoming regional differences" (p. 10). In this regard, the fears expressed by some Members of Parliament about the policy funding of STEM programmes are justified because "in many developing countries [such as Uganda], regional educational disparities are partly due to regionally diverging supplies of educational and training programs, the dismantling of educational infrastructure [such as lack of wellequipped science laboratories] in rural areas, as well as the expansion of private schooling in urban areas" (Nikolai et al.,, 2022, p. 1).
The third front from which the underprivileged regions are marginalized concerns the elite information channels and communication pathways through which legitimate information regarding the students' loan policy is disseminated. Information about the policy (adverts, eligibility, selection criteria, courses to fund, etc.) is by law published in the most widely circulated newspapers -The Monitor and New Vision; Government Gazette (Parliament of the Republic of Uganda, 2013, p. 10212); and the internet (HESFB Website). However, given the limited access to internet in Uganda, that is 25% of Ugandans accessing internet (NPA, 2020, p. 152) out of the 41.6 million Ugandans by Mid-year 2020(UBOS, 2020; limited circulation of newspapers, radios and televisions especially in upcountry districts where potentially most-needy students are concentrated, the elite means used by the HESFB to disseminate information regarding the policy have potentially hindered many needy Ugandan children from accessing HE via the student loan scheme. Such elitism imbues power effects which are masked up by an array of rules, rituals, systems and bureaucratic procedures that then project truth as taken-for-granted or commonplace. As argued by Middleton & O'Keefe (2001) in their chapter Polite Meaningless Words, the findings in this study have revealed how linguistic strategies and political capital have reinforced existing power relations, arguing that elites in positions of authority "tinker with our consciousness in order to conceal their real agendas" and that words [such as regional balance] are "often used to conceal a disagreeable reality" (p. 32). Thus in the context of this study, it can be inferred that the term regional balance is a polite meaningless word, indeed a mere political rhetoric that is intended to mask the real intentions of policy architects as regards awarding of students loans to access HE in Uganda.
Anchoring this into Foucault's discourse analytical lens, the construct of regional balance as used in the policy includes the presence of knowledge that demonstrates power relations. It thus resonates with Balaam & Dillman's (2011) conception of regional balance (as a politically rhetorical concept) which is synonymous with Foucault's theory of power with its top-down nature. This view also considers sources of power as lying between the "traditional power theories where there can also be found the power of religion, law and force" (Moghadam & Rafieian, 2019, p. 5), which accompany Foucault's view of power that considers knowledge as the main producer of power. Foucault argued that: "If power is nothing but a repressive force, if power has nothing to do but just to say no, in this situation do you really think that it was compulsory for us to obey it?" (Foucault, 1985, p. 17). The idea of power/knowledge is something that can reveal the concept of regional balance in the awarding of student loans through the current loan policy. Foucault's analysis is clearly based on this assumption coming from Nietzsche that knowledge and power are inseparable (Ure & Testa, 2018). "Power produces knowledge (And not just by promising and encouraging the production of knowledge, because knowledge serves power, or with the use of knowledge because of its usefulness), power and knowledge both imply the other" (Ure & Testa, 2018, p. 127).
In light of this argument, it can be said that because people from Western Uganda are dominating Government Ministries, Departments and Agencies, they have first-hand information (knowledge) regarding government programmes, which knowledge gives them an upper hand (power) to influence the direction of policy implementation. Moghadam & Rafieian (2019) argue that there is no power relationship which does not create a sphere of its own correlated knowledge and there is no knowledge in which the unit requires and does not bind power relations.  perceived this as the "many struggling within a politic of truth" (p. 20), which is how it was utilised in this paper. It is in the seeking, debating, and codifying of truth about regional balance in the provision of the students' loans to facilitate access to HE that in turn creates the regimes of truth that characterise sites of power relations. Thus Foucault's original conceptualization of regimes of truth is that: Each society has its regime of truth, its general politics of truth: that is, the types of discourse which it accepts and makes function as true; the mechanism and instances which enable one to distinguish true and false statements, the means by which each is sanctioned; the techniques and procedures accorded value in the acquisition of truth, the status of those who are charged with saying what is true (Foucault, 1980, p. 131).
However, this paper suggests that it is not enough to unearth and expose power relations hidden in discourse. Rather, the analysis should go deeper and reveal the hidden intention of why the power imbalances exist in the first place. For instance, often times power relations are imbued in policy and political discourse because the architects of the policy want to change the course of discussion of the issue of contestation (Iqbal et al.,, 2020); to exert socio-economic and political control (Li, 2019), and for domination (Maleya et al.,, 2018). In light this argument, since knowledge is power and one who has power may have easy access to knowledge, the accumulation of which strengthens control and domination, it can be inferred that as more people from Western Uganda continue to reap from the students' loan policy to access higher education (accumulation of human capital), they will inevitably accrue higher education benefits (knowledge, skills, productivity, employment, income, savings etc.) which may further bolster their control, influence and domination within the political landscape (regional capital) as well as national resources.

CONCLUSION AND RECOMMENDATION
Section 4 (v) of the Uganda Students' Higher Education Financing Policy stipulates that "majority of the students who join universities and HEIs are those who are able to access the best schools in the country and these are not regionally balanced" (Ministry of Education and Sports, 2012, p. 13). Thus under objective three of the policy, government is committed to "ensure regional balance in higher education services in Uganda" (ibid, 2012, p. 14), which may suggest that the Uganda Students' Higher Education Financing Policy is inclusive. However, according to the findings of the study, it is remarkable that the social justice ideals of regional balance and the real policy practice are scarcely related. The hijacking of policy implementation by political elites (political influence-peddling), the study programmes funded by the policy, and the communication channels used by the HESFB to disseminate information regarding the policy potential are all aspects that have perpetuated regional disparities in access to higher education. As a result, these have promoted a kind of inherited merit in Uganda's higher education sector. As argued by Foucault (1970), discourse, knowledge and power are all inextricably linked and mutually re-producing. Thus "discourse is not simply that which translates struggles or systems of domination, but it is the thing for which and by which there is a struggle, discourse is the power which is to be seized" (pp. 52 -53).
In light of the above Foucault's argument, formations of power/knowledge such as regional balance may be referred to as discourses, and they constitute certain ways of thinking about, representing and acting upon the world. From the findings of the study, it can be concluded that the term regional balance as used in the policy is a mere policy rhetoric and a discourse that reproduces power imbalances between regions of the country. Thus the hegemonic established order maintains dominance through unequal power relations between regions of the country (regional political capital), keeping the dominant discourse unchallenged. This dominant discourse of GoMDAs, expressed through systems of networks, as described by Foucault, "infuse policy with an expert ideology" (Foucault, 1994, p. 30) to suppress equity and social justice for the underprivileged regions in Uganda and thus maintain the hegemony of the established order. Thus the optimistic, synergistic tone exhibited in the promise of ensuring regional balance in the awarding of student loans to facilitate access to higher education appears not carried forth to the policy implementation as the study findings have revealed that regional disparities are glaringly apparent when it comes to accessing the student loans.
In order to promote and achieve regional balance in the provision of higher education via this loan policy, this paper recommends that the HESFB should use a targeted approach when advancing loans to the successful applicants. In this regard, the HESFB should provide regional quotas where relatively more slots are awarded to the less privileged regions and sub-regions. This will not only help to make up for the marginalization and exclusion from higher education that these regions have historically suffered, but it will also increase their access, participation and successful completion in HEIs. In this regard, Crosby (1994), a social and educational policy analyst, uses the allegory of athletics to justify this kind of approach: In a fair system, every runner has an equal chance of winning the race. To equalize chances, all runners must have the same number of meters to run. In a straight track, the finish line is pulled straight across the lines, at a 90° angle to them. In an oval track, however, the finish line must slant to compensate for the fact that the inside track has fewer meters than the outside track. Treating every runner on the oval track fairly requires that 'an adjustment' be made to the finish line, so that it is placed at a 60° or 70° angle and not at a 90° angle. A very simple-minded person would see the slanted finish line of the oval track as giving advantage to the runner on the outside track, but a more sophisticated thinker would understand that the slant in the line simply compensates for unequal conditions of running (p. 29).
Besides, the less privileged, vulnerable and generally marginalized and/or excluded regions and sub-regions in Uganda with "poor or no access to basic social services [including higher education]" (NPA, 2020, p. 206) are fairly documented in Government policy statements. Some of these regions and sub-regions include, "Bukedi, some parts of Busoga, Karamoja, Bugisu, West Nile, Acholi, and some parts of Bunyoro" (NPA, 2020, p. 206; Ministry of Gender, Labour and Social Development [MGLSD], 2011; Equal Opportunities Commission [EOC], 2019). Although the Government turns against these same regions and lambastes them for "dragging the povertyreduction gains in the country" (NPA, 2020, p. 206), such rhetorical ethos are in the words of Snowdon & Karlsson (2021) "synonymous with the neoliberal ideology of blame the victim", (p. 9), where power elites "ignore" the social, cultural, economic, and political realities that make certain social groups (and regions and sub-regions) victims and continue being marginalised owing to the decisions that such power elites explicitly or implicitly make, which end up sustaining the power imbalances. Nevertheless, because education policy implementation is a "complex, evolving process that involves many stakeholders, it can result in failure if not well targeted" (Viennet & Pont, 2017, p. 8).

Limitations and Direction for Future Research
A threat to this study's validity is that, since discourse analysis is mainly qualitative in nature, interpretations may vary. Thus, the scope of generalization of the findings may be limited. Generalisation precisely means, "the extent to which you can come to conclusions about one thing based on information about another" (Vogt, 2005, p. 131). However, Dahler-Larson (2018) warns against generalisation of qualitative research findings, arguing that "issues of causality and generalisation are important, but their meaning is not legislated by the philosophy of science. Instead, their meanings flow out of debate, argument, institutionalised rules, and power" (p. 5). Thus, generalisation in research is "no longer defensible or functional" (Donmoyer, 2012, p. 46;Flyvbjerg, 2001;Cronbach, 1975) because human beings generate meaning "by social interaction rather than external causes" (Donmoyer, 2012, p. 46). More so, in educational research, since too many contextual variables can shape the findings (Labaree, 2004), generalizability should not be a concern in qualitative research (Darlington & Scott, 2003).
Besides, in this paper my emphasis was on providing insights into an under-engaged area of research into the discourse of regional balance and how it was constructed in the Uganda Students' Higher Education Financing Policy. The study was intended not to offer a comprehensive account of the complex factors that shape policy formulation, practice and implementation in Uganda but to expose the power imbalances, victim blaming, and policy rhetoric that can form and maintain the discourse in which the underprivileged regions that have traditionally been marginalized and excluded from HE is constructed within the broader (social, political, and economic) context in Uganda.
Another limitation is that Foucauldian discourse analysis interpretations are subjective, relying on understanding the interdiscursivity of discourse types formed in discursive practice. Thus, the analyst's grasp of the orders of discourse and the discourse types running through the text are reliant on the authors' ability to derive meaning from the communicative events. This, in turn, depends on the knowledge of the social, political, economic and historical contexts in which these events take place, so it is essential to have a wide perspective of the processes that underpin the formation of the discourses (Snowdon & Karlsson, 2021). Therefore, future studies may use other research methodologies and research instruments to measure the efficacy and effectiveness of the policy for instance, using the Likert scale.