On the absent ground of transnational partnerships in education: A post-foundational intervention

ABSTRACT Transnational multi-stakeholder partnerships (MSPs) are proliferating in the domain of education and development. This paper intervenes in scholarly discussions on such partnerships by pondering what post-foundational political thought might add. It is argued that both liberal and critical literature, albeit from very different viewpoints, tend to foreground order and stability. Drawing on post-foundational political theory, and a case study of the Global Partnership for Education, the present paper explores transnational MSPs in education as incomplete hegemonic projects. By bringing attention to fissures and disruptions, the paper exposes how the absent ground of ‘the political’ ultimately haunts the stability of partnerships.


Introduction
Multi-stakeholder partnerships (MSPs), cutting across the traditional public/private binary, are becoming commonplace in the domain of education and beyond. These governance arrangements, assembling very different types of actors for a supposedly common good, can also be found on a variety of scales ranging from the local to the global. This paper is primarily concerned with the latter, that is transnational multi-stakeholder partnerships, more specifically as they pertain to education and development. There is every reason to take interest in such partnerships as they occupy a central position in the 2030 Agenda, even constituting one of the 17 sustainable development goals (SDGs) in their own right (UN 2015), and as they have been mushrooming in the education sector over the past decade (e.g. Robertson et al. 2012).
Indeed, public-private partnerships (PPPs), of which MSPs constitute a certain kind, have also attracted attention in the academic community. At the risk of some simplification, and as will be shown in greater detail below, two contrasting images prevail in the literature. One body of scholarly work greets (albeit with varying degree of enthusiasm) the new partnerships, depicting them as venues with potential for deliberation and deepening democracy, mutual learning, greater transparency, enhanced efficiency, and, increased mobilization of resources. In the other camp, more critical studies abound. Here the typical picture portrayed is one of arrangements that, under the facade of 'partnerships' between fellow equals, rather serve to conceal, or even consolidate, power relations.
The present paper intervenes in these scholarly discussions by considering what post-foundational political thought might add to them.
Drawing on post-foundational political theory, and an in-depth empirical case study of the Global Partnership for Education (GPE), this paper aims to explore transnational MSPs as incomplete hegemonic projects. By bringing attention to fissures and disruptions in the workings of the GPE, the paper argues that the absent ground on which such partnerships rests, haunts both the conditions of possibility for establishing 'rational' deliberative order and the solidity of power configurations. In other words, borrowing from Macgilchrist (2016), while much of the literature on partnerships, liberal and critical alike, tend to foreground, albeit from radically different viewpoints, stability and order, the present paper is concerned with instability and disorder. Lest we should be misunderstood, we are not suggesting that it is irrelevant to explore order in the context of partnerships. As will be evident below we have ourselves engaged in similar undertakings. Nor are we suggesting that the present intervention will somehow miraculously make the picture complete. Our more modest message is that post-foundational political thought has something important to contribute to current academic discussions on MSPs in education. Hence, whilst debatable, our contention is that the findings of this paper are of relevance to broader scholarship in the domain, and that our arguments extend beyond the singular case of the GPE.
Empirically, the paper draws on an in-depth case study of the GPE. Fieldwork was carried out in Brussels, Kigali, London, Stockholm and Washington D.C. in 2015 and 2016. The data includes 35 interviews with partners to the GPE, including bilateral and multilateral donors, developing country governments, civil society organizations (CSOs), teacher unions, private foundations and private enterprises, alongside staff at the GPE secretariat. Interviews normally lasted between 45 and 90 min, and informants were purposively selected to represent all categories of major partners to the GPE. This way, we managed to capture different actors' perspectives on the workings of the GPE. The data further includes policy documents, reports, news media articles, social media posts, web pages, and other printed material. All the data was subject to a thematic analysis where the data was coded and recoded. We initially approached the material with a broad interest in how the GPE operates and how the plethora of 'partners' understood this global governance arrangement as well as their own and others' role in it. Yet, our attentiveness was soon drawn to fissures and disruptions that clearly destabilized the consensual order of the partnership. This prompted a turn to post-foundational political thought. The post-foundational methodology that underpins the present study is further elaborated in a recent methodological article (Knutsson and Lindberg 2017).
The paper is organized as follows. The next section first outlines some general features of MSPs and accounts for their role in the post-2015 agenda, before turning to a brief description of the particular case of the GPE. The second section summarizes much of what has been written on educational PPPs and MSPs over the past years, including liberal and more critical work. The third section accounts for the paper's theoretical perspective. Drawing on this perspective, the fourth section offers a theoretically informed reading of fissures and disruptions that emerged from our empirical data on the GPE. The final section concludes.
Multi-stakeholder partnerships in education: general features and the case of the GPE The terms PPPs and MSPs can be conceptualized in different ways (see e.g. Draxler 2016; Verger and Moschetti 2017). In this paper, we will draw on Robertson et al. (2012, 1), who broadly defines PPPs as 'cooperative institutional arrangements between public and private sector actors.' The term MSPs, on the other hand, denotes a specific kind of PPPs that includes a broader collection of actors, particularly CSOs, alongside the public sector and private sectors entities that also form part of PPPs (Draxler 2012). MSPs can be found on different levels of scale. In this paper, we are mainly concerned with transnational MSPs, which tend to operate across scales and often involve both global and local actors. Such arrangements have been described as hybrid ways of mobilizing, pooling and managing resources, competencies and commitments across sectors, in attempts to achieve shared objectives ). As will be seen below, the term MSPs is well suited when looking at the case of the GPE with its multitude of actors. In addition to the GPE, other examples of transnational MSPs in education and development include the Education Cannot Wait Fund, the United Nations Girls Education Initiative, and the International Education Funders Group.
In the 2030 Agenda, partnerships between different stakeholders are identified as a necessary complement to government action for achieving sustainable development (UN 2015). In the SDG17for examplethere is a call to enhance MSPs that mobilize and share financial resources, knowledge, and technology, in order to achieve the other SDGs, particularly in developing countries. In line with this, a recent parliamentary committee report aiming to steer the Department for International Development (DfID) to a more effective implementation of SDG4 (which specifically concerns education), argues that DfID should channel even more of its education aid funds through MSPs, particularly the GPE (IDC 2017).
The GPE is the largest MSP in the field of aid to education, and it has as its overall aim to achieve Education for All (EFA). To reach this aim the GPE supports the development of education sector plans in 65 developing countries, while also working to increase and harmonize aid flows to implement these plans. The GPE comprises of a variety of partners ranging from the donor and developing country governments, multilateral donor agencies, private companies and foundations, to CSOs (including INGOs, NGOs and teacher unions). The GPE is, however, rather unique among MSPs in education and development in that almost all of the funding still emanates from government sources (Menashy 2016). The GPE is administered at two levels: the country level and the global level. At the country level it operates through so-called Local Education Groups (LEGs), that is, national multi-stakeholder forums that develop, implement, monitor and evaluate education sector plans. At the global level, in turn, there is a board of directors, five board committees, and a secretariat. The board of directors comprises representatives from developing countries, donor countries, multilateral agencies, the different types of CSOs, and private companies/foundations. Each board member represents a highly diverse self-governing constituency and at the time of our main field research in 2015, board representatives included actors as diverse as the World Bank, Pearson, Action Aid, and Education International, in addition to donor and developing country representatives and NGOs.
All partners to the GPE have to sign up to a Global Compact (GPE 2013), whereby developing country governments are obliged to implement 'sound', sustainable and broad-based education sector plans, increase domestic funding for these plans, and exhibit results on key performance indicators. The other partners are in turn requested ̶ through the compact ̶ to increase funding to these plans, align with the developing country's policy priorities, and harmonise activities among themselves as far as possible. Despite the diversity of actors and interests, the GPE Global Governance Manual further instructs that the Board should attempt to make all decisions by consensus (GPE 2015b). A key governance strategy for the GPE is hence to convene and build consensus among its diverse partners (GPE 2015a). In this way, the GPE constitutes a very ambitious MSP and, in our interpretation, it reflects the liberal idea that it is possible to establish a consensual order, among diverse actors, founded on reason and rational deliberation (cf. Knutsson and Lindberg 2018). In the following section, we will discuss what the previous literature have had to say about PPPs and MSPs.

Liberal and critical literature on partnerships
Over the past years, PPPs and MSPs have attracted increasing interest in the research community. The literature is vast and as signified above the varying applications of the terms pose challenges. Hence, without claiming to be comprehensive or immaculate, but indicative of the general picture, this section provides a brief overview 1 of much of what has been written about such partnerships with reference to education and development, including the specific case of the GPE. In our interpretation, the literature can be divided into two main camps. It should be emphasized that this division relates to studies that have been conducted and authors can thus be found on both sides of the analytical divide. Most notably Francine Menashy (2015Menashy ( , 2016Menashy ( , 2017Menashy ( , 2018, whose kaleidoscopical investigations of the GPE, approaching it from different theoretical perspectives, provide rich accounts of this particular entity. The first camp consists of studies that are liberal in orientation and broadly optimistic about PPPs and MSPs in terms of their ability to serve a common good. For example, as regards resource mobilization, efficiency, accountability, transparency, mutual learning and democratic governance. Many of these studies are economically oriented and rather unconcerned with issues of power. The most cited of these contributions, stressing benefits in terms of increased access to schooling for disadvantaged groups, as well as better results and more efficient use of resources, is Patrinos, Barrera-Osorio, and Guaqueta (2009). The purported evidence of the book is gathered from different country-contexts around the globe and the overall conclusion is that, given the right design, partnerships can yield success. Chakrabati and Peterson's (2008) edited volume, collecting studies informed by human capital theory, likewise contends that although circumstances might differ, generated outcomes have overall proved to be very promising. Similar conclusions are offered by LaRocque (2008) and Fielden and LaRocque (2008). Several subsequent studies report further support for the partnership model, thus reiterating the optimistic sentiments (e.g. LaRocque 2008; Wolf et al. 2013;Chattopadhay and Nogueira 2014).
In the optimistic camp, we also situate studies of partnerships that are conducted from the particular perspective of deliberative theory. Deliberative studies are obviously very different in theoretical orientation and unlike the abovementioned literature, they are not indifferent to issues of power. Yet, they ultimately share a liberal worldview and the associated belief in the possibility of governance arrangements that serve the best interest of everybody. Deliberative studies typically have a positive outlook on MSPs as these entities, at least on a superficial level, reflect many of the criteria for participatory democracy proposed by the likes of Jürgen Habermas, John Rawls, James Fishkin, John Dryzek and Thomas Risse. Deliberative theory assumes the possibility of establishing, through dialogue and reason, a 'rational' order beyond hegemony. Empirical studies in the deliberative tradition are certainly not blind to 'real-world' problems of unequal power relations but they still cling on to the assumption that rational consensus is possible in the long run. For example, one of Menashy's (2015) studies of the GPE identified a serious lack of deliberation on the contentious issue of private schooling and further proposed that avoiding to deliberate on this matter might risk the legitimacy of the partnership (see also The Freshed podcast 2016). 2 Deliberative studies of MSPs in the domain of environmental governance (Bäckstrand 2006) and HIV/AIDS governance (Brown 2010) provide similar arguments. While these studies all recognize a 'deliberative deficit' they still make a case for deliberation by arguing that it can be improved through more robust dialogue and participation. In our view, however, these studies tend to attribute the problem to an imperfect reality rather than to their own theoretical assumptions. This, we argue, is an impasse and as will be evident in the next section post-foundational political theory offers a different take on the matter at stake.
Ultimately what unites studies in the first camp, albeit their different focus and theoretical orientation, is a liberal belief that a good and rational 'order' which serves all concerned parties can be established through partnerships. While some studies argue that such consensual orders have already materialized in local educational settings around the globe, others suggest that they can, given 'proper' deliberation, be accomplished in the longer run.
The second camp harbours a myriad of critical studies which provide a gloomier picture. Broadly speaking, these studies are concerned with how partnership discourse conceal unequal power relations, how power operate in the context of partnerships, and how ideological forces have propelled the partnership model globally. As indicated above it is impossible to account for the entire literature in a short section like this, Hence, what follows is a selection of some influential critical works.
An important anthology, collecting a range of studies by different scholars concerned with the twopart question of who wins and who loses, and who has the power to decide, in educational PPPs, is Robertson et al.'s (2012) edited volume. Most of these contributions display a concern with how increasing private sector involvement affect (in)equality, 'citing the associated erosion of citizens' voice as a contributor to change' (Robertson et al. 2012, 14). It is argued that the partnership model is strongly associated with neoliberal economic globalisation and that the new partnership arrangements run serious risk of amplifying power hierarchies and of undermining education as public good and human right. Another edited book, gathering studies specifically focused on educational PPPs in low-income countries, similarly raises concerns about aggravated inequality (Rose 2011).
A number of contributions further discuss how the semiotic use of the term partnership conceals power relations. According to some, it serves to mask a neoliberal ideological agenda, pushed worldwide by the World Bank and its allies (Klees 2010(Klees , 2012Nordtveit 2012; see also Klees 2008). These contributions stress that partnerships are a misnomer since what follows in their wake is essentially privatization and public cut-backs with devastating consequences for the most vulnerable. In a similar vein, Miraftab's widely cited study argues that PPPs is a Trojan horse of neoliberal development, seemingly arriving like a gift, only to further dispossess the poor (Miraftab 2004). Inequality and the neoliberal tenets of partnership discourse are also discussed in Vavrus and Seghers (2010). Samoff and Carrol's (2004) study chiefly situates the concept in a (post-)colonial context, arguing that it conceals the continuities of colonial dominance in international development. Another type of critique is offered in Srivastava's (2010) study which suggests that the Indian government has failed to specify the term and that this conceptual confusion, in turn, has allowed for private actors to take advantage of the situation.
Other studies have focused on how education policy networks have contributed to the global dissemination of the partnership model, arguing that certain actors in these networks have been particularly influential in setting the agenda and framing the idea (Ball 2012;Verger 2012) and that networks can, in contrast to conventional wisdom, in fact, reinforce power hierarchies (Faul 2016).
Finally, and of particular interest in the context of the present paper, we should mention Menashy's analysis of power asymmetries in the GPE (Menashy 2018). Drawing on the concept of structural power, the study exposes a hierarchical structure within the partnership which substantially determines the influence and capacity of different actors. It is moreover argued that the current organisational arrangements of the GPE, in fact, serve to reinforce structural power and the associated dominance of certain actors, most notably the World Bank and some bilateral donors (Menashy 2018).
A common denominator of the critical studies in the second camp, despite their different focus and theoretical tenets, is obviously their significant unveiling of power relations in the context of partnerships. However, inspired by Macgilchrist (2016), what strikes us is that these hierarchies, and the ideological order of which they form part, often come across as something fairly stable. Arguably, less attention has been paid to how the power of partnerships is disrupted and challenged. Given our previous work on post-political (Knutsson 2014; Knutsson and Lindberg 2019) and biopolitical (Knutsson 2016) ordering in the context of international development, we certainly recognize the importance of exploring governmental order. Indeed, it should also be acknowledged that there are studies that have explored resistance (e.g. Verger and Novelli 2010;Stromquist and Sanyal 2013), as well as limitations (e.g. Menashy 2017), in the context of global governance of education, including the particular case of the GPE (Menashy 2017). In comparison, however, this literature is small. Hence, we believe that more energy should be devoted to exposing how the order of educational partnerships is disrupted and destabilized (cf. Lindberg 2017, 2018) as this can teach us something about the possibility and presence of political action. Post-foundational political theory invites us to do so.

Post-foundational political thought
This section accounts for the paper's theoretical approach which stems from post-foundational political thought. Post-foundational political theory has gained prominence over the past decades and the tradition assembles a range of critical thinkers such as Alain Badiou, Ernesto Laclau, Chantal Mouffe, Jean-Luc Nancy, Jacques Rancière and Slavoj Žižek, only to mention a few. In this paper, we draw mainly, but not exclusively, on . Although there are important differences between the abovementioned scholars in terms of emphases, nuances and political conclusions (for good overviews see e.g. Marchart 2007;Wilson and Swyngedouw 2014), they unanimously embrace a post-foundational ontology. In other words, they refute the possibility of any ultimate ground to social orders, be it truth, reason, God, the economic base, the state of nature, or anything else. Hence, all social orders are taken to be 'profoundly contingent, and structured to conceal their own absent ground' (Wilson and Swyngedouw 2014, 10). Of note, though, is that this position is not tantamount to crude anti-foundationalism or extreme post-modernism (Marchart 2007, 10-14). What post-foundational political scholars assume is the absence of an ultimate ground, i.e. they stress the radical contingency of all foundations (Marchart 2007).
This understanding of social orders and their absent ground is furthermore reflected in the tradition's peculiar conceptual differentiationcanonized as 'the political difference' (Marchart 2007(Marchart , 2011)between on the one hand politics, and on the other, the political. This is not the place for philosophical genealogy, but it is worth noting that post-foundational political thought is sometimes referred to as 'left Heideggarianism' (Marchart 2007) and that 'the political difference' closely corresponds with Heidegger's analytical distinction between the ontic and the ontological levels of being ; see also Heidegger [1927Heidegger [ ] 1962. While the ontic is the level of concrete and specific realities, i.e. factual matters that are empirically observable, the ontological level concerns the deeper structure that underpins the existence of these matters, i.e. their very conditions of possibility. Accordingly, for post-foundational scholars, politics is situated at the ontic level. It refers to the arrangements, institutions and practices through which social order is established (e.g. Mouffe 2005, 8-10). Such an order is always hegemonic in the sense that while it might appear as 'natural' or 'rational' it, in fact, reflects a particular configuration of power relations. 3 Hence, social order is never a manifestation of something that is 'exterior to the practices that brought them into being' (Mouffe 2013, 2). This particular conception of politics as a hegemonic apparatus of order is obviously quite radical and, as discussed by for example Beveridge (2017), not sole or dominant in the social sciences. However, although hegemonic order can be powerful and pervasive it is never complete nor stable. Things could always be otherwise and, as pointed out above, there is no essential ground on which to consolidate social order. This absent ground is the political. The concept of the political, thus, refers to the ontological level. It is, in the words of Mouffe (2005, 9), the 'dimension of antagonism' that is 'constitutive of human societies'. Whether we like it or not, Mouffe argues, the impossibility of reconciling all interests and of instituting a social order that will benefit everyone forms part of our ontological condition. By way of summary, then, politics is the ceaseless hegemonic efforts to establish order on the absent ground of the political.
Empirically exploring transnational MSPs as incomplete hegemonic projects are of course not a straightforward enterprise, and to us, two questions come to the fore. Firstly, how can the political be exposed in the context of partnerships, i.e. where to search for its manifestations? Secondly, and interrelatedly, what qualifies as 'truly' political action, i.e. how is such action different from the everyday 'game' that resides within the realm of politics? Starting with the latter, it should be recognized that there is no consensus amongst post-foundational scholars of what qualifies as 'truly' political action. Of note though, is that many thinkers in the domain have rather high demands, as they are exclusively concerned with revolutionary moments of insurrection where the entire order is challenged (e.g. Badiou 2009;Swyngedouw 2009;Swyngedouw and Wilson 2014;Žižek 2002). The implication of this is that many scholars in the tradition, while recognizing that the political is constitutive, in fact, argue that we live in a world where political subjects and actions are utterly rare (Badiou 2009;Swyngedouw 2009;Swyngedouw and Wilson 2014;Žižek 2002) Elsewhere we have elaborated on the risks that are associated with overstating the case for a 'post-political condition' Lindberg 2017, 2018) and similar critiques have been offered by others (e.g. McCarthy 2013; Darling 2014; Larner 2014). Therefore, without being naïve about the prospects for significant change, we would propose a less grandiose conception of political action (see also Marchart 2011). On this account, we find the work of Mouffe helpful. Apart from stressing the conflictual dimension of the political, which is always articulated in the form of an adversarial relationship, she also has a deeper, psychoanalytically inflected point about political identities always being collective and affectively invested . Accordingly, inspired by Mouffe, we conceptualize political action as entailing four basic components: (1) a conflict of interest that lacks a 'rational' solution, i.e. demands a choice between conflicting alternatives; (2) which is articulated in the form of a we-they relation, i.e. involves the construction of collective identities based on the affirmation of difference; (3) that is invested with affect and expressed in adversarial terms; and (4) which entails some kind of concrete strategic action intended to change the state of affairs or, at least, to expose a perceived wrong.
Obviously, the whole idea of 'partnership' appears to hinge upon the assumption of the possibility of constructing a shared 'we' of partners and of instituting 'rational' order, grounded on reason, deliberation and notions of a joint struggle towards a common good. This brings us back to the first question. If we want to expose the political in the context of transnational MSPs we must, as argued by (Macgilchrist 2016, 264), explore 'the fissures in what had seemed to be a certain and fixed ground'. This involves bringing attention to moments of breakdown and dissonance (Macgilchrist 2016), and in our case, that would imply instances where the partnership ceases to run smoothly and where 'rational' deliberation runs into an impasse. In other words, we must search for political action that destabilizes the partnership as a hegemonic apparatus of order. So, where to look? Given that the self-descriptions of transnational MSPs typically, as for example in the case of the GPE, entail the anything but timid epithet 'global', we might, arguably, look pretty much anywhere. However, in this study, accounts of political action, as conceptualized above, emerged through, on the one hand, in-depth interviews with formal partners to the GPE, and on the other, through publicly published material, including reports, debate pieces in news media and social media posts. In interviews like these, as we have argued at length elsewhere (Knutsson and Lindberg 2017), it is important to be attentive to mundane, everyday political struggles and articulations of adversarial we-they relationships, in order for the political to surface. And in addition to this, different kinds of written material is undoubtedly important for our purpose here since, as argued by Mouffe (2013, 143), media is one of the important terrains where political subjectivity is articulated today (see also Macgilchrist 2016).
Guided by these ontological, theoretical and conceptual tenets derived from post-foundational political thought, we will now turn to an exploration of the empirical case of the GPE.
On the absent ground of the global partnership for education As indicated above the GPE is keen to present itself as a consensus-oriented and coherent transnational partnership, bringing together a diverse set of actors for a common good. In the theoretical vocabulary of this paper, the GPE can be understood as a hegemonic project, located at the ontic level of politics, seeking to establish a consensual order among diverse partners, stabilize their conduct and, thus, facilitate global governance for implementing EFA worldwide. However, as this paper is mainly concerned with the impossibility of grounding such a consensual order, this section will analyse fissuresor 'moments of the political' (Marchart 2007, 160) where the apparent ground of the GPE dissolves. Hence, by focusing on political action pertaining to two particular issues, the instability of the partnership and its incompleteness as a hegemonic project will be exposed. The first of these issues concerns whether the GPE should support so-called low-fee private schools (LFPS), while the second is about whether the GPE should embrace the notion of global tax justice. As will be argued, these issuesboth of which concern political economy 4illustrate the impossibility of reconciling the interests of capitalist actors and of their political adversaries, in spite of the GPE's efforts to subjectivate them as 'partners'.
LFPS are privately run fee-charging schools that specialize in attracting children from less affluent families by keeping fees relatively low (Verger, Fontdevila, and Zancajo 2018). A key achievement of the EFA movement during the past decades has been the abolishment of school fees in many poor countries. In this light, LFPS are controversial since they imply that the poor once again is supposed to pay for their own education, but also because they tend to employ untrained teachers whom they equip with very detailed scripts, and thus, arguably, contribute to de-professionalization. Within the GPE some partnersnot least private companies like Pearson, but also a few bi-and multilateral donorsare in favor of using GPE funds to support these schools. An informant from the private sector/foundations constituency, for example, told us that many partners in this constituency are interested in the 'explosion' of LFPS in developing countries, as these schools are seen as 'a potentially interesting way for very, very poor people to get their children an education when the public system does not offer a great alternative' (interview, private sector/foundations, 20 May, 2016). This is also linked to an economic interest, especially on the part of the private sector partners. According to a blogpost posted on the website of Pearson a few years back, the company's investment in Bridge International, which is one of the big chains of LFPS, forms part of Pearson's attempt to expand its education business in Africa. 5 Other partners, most notably teacher unions and other CSOs, are forcefully against this new type of private schools: … part of the way they [the LFPS] make money is that they hire unqualified teachers, young teachers, and they give them scripts, and they do not pay them very well. And part of their argument is, 'we can make money because they are not public employees'. So … there is an ethical dimension to it, an inequality dimension, because, if all the research is saying that you need to have the most qualified teachers for the most needy, most marginalized children, that can make instructional decisions based on different learning styles, and be able to make formative assessments in real time, and then the best you can give them are cheap teachers without any labor rights, then somebody is benefitting off somebody else. (Interview, CSO, 28 May, 2015) The controversy has led to a situation where there is no clear policy with regard to LFPS, or to the privatization of education at large, in the GPE (Interview, CSO, 23 April, 2015; Interview, private sector/foundation, 20 May 2016). Contrary to the official self-image of the GPE as a consensusoriented and dialogue driven entity, our data show that there is no deliberation in the GPE on this issue: it is being avoided. Similar findings have been reported by Menashy (2015;. However, drawing on post-foundational political theory our contention is that such deliberation would have been fruitless as the issue is 'truly' political (see also Knutsson and Lindberg 2018). Hence the contentious issue of LFPS exposes a fissure in the supposedly coherent partnership. From a post-foundational perspective, this fissure exposes the ontological instance of antagonismthe politicaland thus illustrates the impossibility of instituting a social order that will benefit everyone. In other words, it points to the GPE as a hegemonic project that is neither complete nor stable.
Political action, as conceptualized in this article, with regard to LFPS has largely taken place in the public domain. One informant explained that they were working intensively to mobilize and coordinate action between teacher trade unions and other CSO in different countries against pro-LFPS forces (Interview, CSO, 20 November, 2015). Even if, as argued by Menashy (2018), these CSO partners operate from a position of structural subordination, this does not imply that they silently accept such a position. The structural order is not all-powerful or stable as it is continuously challenged and destabilized through political action. One such example is when a number of GPE's CSO partners published a public response to a speech by the World Bank president, in which he spoke in very positive terms of Bridge International (Action Aid 2017). The straightforward message in this response was that the World Bank president should make a 180 degree turn and recommit to the idea that basic education should be free of charge. In addition, on the 1st of August 2017, several CSO partners to the GPE joined hands with a large number of other organizations in a public call on investors to cease support to BIA (GIESCR 2017). Many CSO partners have also published reports and articles heavily criticizing LFPS (e.g. EI andKNUT 2016, Results 2017), with the aim to change the parameters of the entire discussion. Finally, in late 2017 and early 2018, representatives of GPE partners like EI, Action Aid and Oxfam have been active in heated Twitter debates with representatives of Bridge International and other pro-LFPS actors, with regard to the activities and results of LFPS in Uganda, Liberia and Kenya. All of these examples illustrate Mouffe's point about media as one important terrain for political articulation.
Another issue which destabilizes the GPE regards proposals for global tax justice. These proposals are brought forward by several of our civil society informants (Interview, CSO, 23 April 2015;Interview, CSO, 28 May, 2015;Interview, CSO, 20 November, 2015). The key point of these informants is that even if developing countries would spend the recommended 20% of their national budgets on education, the goal of reaching quality EFA will still demand much more resources. The size of the pie, the argument goes, must be made bigger. This should be done by introducing more progressive tax systems and, in particular, by making it more difficult for multinational corporations to avoid taxation in poor countries. Today, it is argued, 'developing countries lose billions of dollars because corporations are not transparent about where they report their profits' (Interview, CSO, 23 April, 2015). In pushing for tax justice these CSO partners thus bring in the deeply political issue of how large corporations use the global capitalist system to their own benefit. This, however, is challenging the whole idea of what the partnership is, and the CSOs and other partners arguing for it have been met with the response that tax justice does not lie within the realm of the GPE (Interview, CSO, 20 November, 2015). Since attempts to put tax justice on the agenda of the GPE board largely have failed, much of the political action on this issue has turned to the public domain, through the publishing and promotion of reports as well as of both scientific and news articles (e.g. GCE 2013, Action Aid 2017; Balsera, Klees, and Archer 2018).
The suggestion to bring tax justice into the GPE has not only generated conflicts about what the partnership should be. Importantly, it also challenges the idea that the vision of the GPEto ensure inclusive and equitable quality education and promote lifelong learning opportunities for all (GPE 2016)is achievable within the frames of the current strategy. In adversarial terms, it has been argued that the present 'tax consensus' in developing countries is a result of policy advice by the World Bank and the IMF, who allegedly have paid little attention to whether tax systems are progressive or not (e.g. Archer 2016). In interviews with CSOs, the example of Pearson is furthermore mentioned as a multinational corporation which is actively avoiding taxes itself: 'I think they [Pearson] are aggressively avoiding tax which is undermining … public education, and then they aggressively support private provision' (Interview, CSO, 20 November, 2015).
The issues of LFPS and global tax justice expose deep conflicts of political economy which clearly destabilizes the consensual order of the GPE. The struggles surrounding these issues qualify, in accordance with our definition above, as 'truly' political action in that, to start with, they entail conflicts of interest that lack 'rational' solutions which can satisfy all parties. They furthermore involve articulations of adversarial we-they relationships as many of the CSOs position certain actors within the GPE, most notably the World Bank and Pearson, as political 'others' (Interview, CSO, 23 April, 2015;Interview, CSO, 28 May, 2015;Interview, CSO, 20 November, 2015). Thirdly, these issues are affectively invested. Several informants expressed serious antipathy towards LFPS, portraying them as threats to the right to education. Deep concerns were also raised of how poor people are being wronged by unjust tax rules in the global political economy. These informants articulate moral indignation and a 'staging of equality' which is reminiscent of the work of Rancière's (1998). Fourthly, the issues involve processes of collective strategic action, in a contingent situation, intended to expose wrongs and alter the state of affairs.
In summary, while much of the existing literature on MSPs in education, liberal and critical alike, tend to foreground stability and order, our study shows, in line with post-foundational political thought, that attentiveness to disruptions in the ordinary workings of such partnerships can help expose their absent ground.

Conclusions
Transnational partnerships are proliferating in the domain of education and development. They have also been subject to extensive, both liberal and critical, research. This article has intervened in contemporary scholarly discussions about such partnerships by showing how post-foundational political theory can be put to use to help foreground their underlying instability. Drawing on an indepth case study of the GPE, the article has brought attention to political action, pertaining to issues of private schooling and global tax justice, that destabilizes the consensual order of the partnership and which exposes its incompleteness as a hegemonic project. These fissures and disruptions in the GPE machinery can be seen as 'failed attempts to domesticate the field of differences' (Laclau and Mouffe 1985, 96) and thus serve to remind us of the impossibility of any final closure or 'suture' to social orders . In other words, they illustrate how the absent ground of 'the political' haunts the stability of the partnership.
Our contention, whilst debatable, is that these findings are of broader relevance to scholarship on PPPs and MSPs. While this paper has focused on the GPE, and on conflicts articulated around two specific issues related to political economy, we believe that there is every reason to explore other partnership arrangements, and their inability to fully suppress conflictuality, in a similar vein. Whether such articulations relate to, as in our case, political economy, or to, for example, political ecology, (post)colonial relations, gender inequlity, or any other form of conflictuality is, in our view, of secondary importance. Hence, what we suggest, is that the research community should pay more attention to different kinds of political moments where the order of ubiquitous partnership arrangements is being disrupted. As we have argued here, post-foundational political thought lends itself well to such analyses.
Why, then, are such studies important? In our view, they cannot only help exposing what is politically at stake underneath the elastic veil of partnership discourse, but also generate significant insights about the possibility of political action in the context of partnerships. In times that have frequently been described, in rather defaitist terms, as post-political and where some post-foundational thinkers appear to see no other option that to resort to revolutionary romanticism (e.g. Badiou 2012;Žižek 2002 we believe that such research is worth while. If we as (post-foundational) scholars recognize the radical contingency of all social orders, we should also be perceptive to the more mundane and less grandiose efforts of altering them.
Notes distinction made is not between politics and the economy, but between politics as efforts to establish hegemonic order and the political as the ontological void underpinning that order. The hegemonic order, in turn, 'includes the entirety of the 'political economy' with which orthodox Marxists concern themselves' (ibid, p. 10). 5. https://www.pearson.com/corporate/news/media/news-announcements/2011/03/bridge-international-academies-in-kenya-quality-schooling-for-les.html (last accessed 18 Januari, 2019)