Getting down to business? Critical discourse analysis of perspectives on the private sector in sustainable development

Critical discourse analysis is employed to examine discourses of the private sector within key texts of the Millennium Development Goals and the Sustainable Development Goals. Taking a genealogical approach, four discourses are identified in the liter-ature: the dominant, pro-private sector discourse, showing unconditional support for the private sector; the sceptical discourse; the middle-ground discourse with new approaches, specifically designed to leverage development relevance; and the antiprivate sector discourse, which considers that current approaches to the private sector will not be conducive to sustainable development of the global South. The pro-private sector discourse was found to be predominant within the global goals, reflecting the post-Washington Consensus as well as the role of the developed countries and the corporate sector in their formulation. All discourses on the private sector, however, place an emphasis on economic and social development at the expense of the key environmental component of sustainable development.


| INTRODUCTION
Can, for example, profit-motivated businesses really make a meaningful contribution to achievement of the [Sustainable Development Goals] or are we likely to see "business as usual," which results in greater profits for some, and lost opportunities for many? (Scheyvens, Banks, & Hughes, 2016, p. 372).
In September 2015, the member states of the United Nations  (UN, 2015). For the first time, these global development efforts in the economic, social, and environmental spheres for both developed and developing countries are being integrated (Cummings, 2015). To achieve this ambitious agenda, global efforts will need to make the most of the role of the private sector.
The way in which development policy incorporates the private sector is a particularly important issue because sustainable development will need to use all possible means at the disposal of the international community, going beyond business as usual (Spangenberg, 2013). As many commentators have already noted, the private sector has received considerable emphasis in the SDGs (see, e.g., Scheyvens et al., 2016;Spangenberg, 2017), probably because of "a widely shared assumption that the 2030 Agenda for Sustainable Development provides momentum for a renewed engagement of the private sector in the service of the United Nations' goals and objectives" (Dumitriu, 2017, p. 2). Indeed, Scheyvens et al. (2016, p. 372) consider that the private sector has been "foregrounded" in the SDGs in which "businesses, governments and civil society actors are equally called upon to pursue a more sustainable path forward." In his investigation of agency within the SDGs from the perspective of multiple actors, Spangenberg argues that: Business is treated as a per se benevolent actor for the public good, instead of a market based, profit seeking undertaking; the objectives and targets include no criteria to distinguish between a positive and a negative role of business for sustainable development (the fact that many sustainability problems have been caused by business activities is not mentioned at all). (Spangenberg, 2017, p. 316).
Since the 2012 United Nations Conference on Sustainable Development (Rio + 20), there has been a growing trend towards multistakeholder partnerships, defined as partnerships between donors, governments, non-governmental organizations, and the private sector (Pattberg & Mert, 2013). However, this trend often lacks a deeper understanding of different types of private sector actors and what they could potentially contribute (Pattberg & Widerberg, 2016). When the private sector ranges from individual entrepreneurs to multinational corporations, encompassing a wide range of practices and implications for development, a critical appraisal of the contributions of the private sector is required. Indeed, as development researchers, the authors have noted that in academic and policy discussions of multistakeholder partnerships, there is increasingly a knee-jerk reaction from colleagues who ask "have you involved the private sector?" often without reference to a deeper understanding of what types of private sector actors should be included and what they could potentially contribute.
In this paper, we aim to take a post-structuralist approach, drawing on Foucauldian understandings of power, governmentality, and discourse to reveal more of the concealed dimensions underlying development policy. In particular, we see the creation of international development policy as an example of governance and thus "conduire des conduites" (Foucault, 1994, p. 237), translated "conduct of conduct." This perspective aims to counteract the tendency of development organizations to "flatten political problems into technical issues" (Telleria, 2017(Telleria, , p. 2144, which also takes place through "fetishization of concepts" (Corbridge, 2007, p. 193). This approach guides the paper through the questions of how the private sector is approached, whose perspective on the private sector is heard, how it is used to govern, and how it relates to wider global structures. Poststructuralist Foucauldian analyses have been employed in the analyses of both policy and governance in the past, particularly in order to examine the prominence of neoliberal governmentality (see, e.g., Lemke, 2016 andLövbrand &Stripple, 2015). In contradiction of many interpretations of neoliberalism, which see it as a manner in which the state shrinks its direct control, the Foucauldian analysis of governmentality considers that neoliberalism is simply another governmentality discourse in which governance's conduct of conduct remains but where individuals are instructed to care for themselves (see, e.g., Lemke, 2016). We consider that this can be referred to as the "conduct of conduct of conduct," although the only earlier reference to "conduct of conduct of conduct" to our knowledge concerns the relationship between the state, the human being, and the nonhuman animal (Howell, 2018, unpaginated). This phrase is reminiscent of Spivak's "it is the machine, as a machine of a machine" (Deleuze & Guattari, 1977quoted in Spivak, 1988. It is here also worth paraphrasing Spivak's assertion that it is crucial to go beyond Cox's "[knowledge] is always for someone and for some purpose" (Cox, 1981, p. 129) and to specify that Western academic thought has Western economic interests attached (Spivak, 1988). This contention is particularly relevant to development policy, which generally still entrenches historical global South-global North relations.
Using the methodology of critical discourse analysis (CDA), this paper investigates the manner in which discourses of development policy currently focus on the private sector, how this has changed since the year 2000, and how the different interpretations are embodied in the policy documents governing international development. In order to do this, this paper compares the texts of the MDGs and the SDGs by focusing on their key documents, namely, the "United Nations Millennium Declaration" (UN, 2000) and "Transforming our world: The 2030 agenda for sustainable development" (UN, 2015), respectively. This focus on policy texts is justified because "politics appears as a 'mentality of rule' [...] largely evidenced in the texts of government" (O'Malley et al., 1997, p. 510). We aim to answer the following three-part research question: Which discourses on the private sector are evident in these key policy documents; whose discourses are they; and how do they relate to sustainable development? The paper first considers current approaches to development policy and perspectives on the private sector. Next, the methodology of CDA is described in detail. As part of CDA, we undertake a genealogical analysis of past discourses of the private sector. This is followed by an analysis of the main texts of the SDGs and the MDGs, together referred to as the global goals. Finally, we discuss how the discourses evident in the global goals relate to wider turns in global governmentality and whether the discourses' perspective on development will be sufficient to address the existing challenges facing sustainable development.

| DEVELOPMENT POLICY AND THE PRIVATE SECTOR
Since the 1990s, development policy has been dominated by the two specific, neoliberal approaches, namely, the Washington Consensus and the post-Washington Consensus, which replaced it. The Washington Consensus, dominant from the 1990s onwards, was a term first used by the economist John Williamson in 1989 to describe "ten policies that more or less everyone in Washington would agree were needed more or less everywhere in Latin America" (Williamson, 2008, p.14), based on approaches proposed by Washington-based institutions, such as the International Monetary Fund and the World Bank. Stiglitz describes this approach to development as "strategies focusing around privatization, liberalization, and macro-stability (meaning mostly price stability); a set of policies predicated upon a strong faith-stronger than warranted-in unfettered markets and aimed at reducing, or even minimizing, the role of government" (Stiglitz, n. d., p. 1). As discussed above, this focus on markets and reduction of government is not considered a move away from governance, but is considered a neoliberal governmentality encouraging states to encourage individual self-governance, namely the conduct of conduct of conduct.
The Washington Consensus was characterized by multiple humanitarian failures, such as the inability to provide increases in the standard of living or to promote democratic development (Stiglitz, n. d.) as well as slow growth in income, employment and productivity, worsening inequality, and increased absolute poverty (McCleery & De Paolis, 2008). In this way, the Washington Consensus represented a governance failure, which was responsible for mass human suffering.
As a response to these failures, the post-Washington Consensus replaced the Washington Consensus in the years prior to the 2008-2009 global economic crisis (Güven, 2018). As with the Washington Consensus, the post-Washington Consensus also included a bias against direct management of the economy by the state, intolerance for statecontrolled development enterprises, and emphasis on private sector development (van Waeyenberge, 2006). As van Waeyenberge argues: The [private sector development] strategy has been a vehicle for the promotion of a role for the private (and, often, foreign) sector from infrastructure to basic service provisioning. Whatever the changes in [World] Bank rhetoric, the underlying imperatives, if shifting, remain pervasive (van Waeyenberge, 2006, pp. 37-38).
Fine and Saad-Filho make a similar point arguing that the difference between the Washington Consensus and the post-Washington Consensus is "overblown and, in particular, that they share much the same conception of development and attachment to neoliberalism, and the same limited commitment to democracy" (Fine & Saad-Filho, 2014, p. 154). Although they also argue that the post-Washington Consensus is more able to withstand criticism, its flexibility makes it more difficult to dislodge. Other commentators also argue that there has been no real change: … there was not the slightest hint from the IMF or World Bank that the policies they advocated during the heyday of the so-called Washington Consensusausterity, privatisation and financial liberalisationhave contributed to weak and unequal growth, with all the political discontent that this has caused (Elliot, 2016, unpaginated).
Having sketched the background of development decision-making on which the MDGs and SDGs were created, we now turn to specifically examining these documents through CDA. In order to do this, we will first detail the particular policy-oriented CDA and the variety of discourses on the private sector that we see-one of which is consistent with the conduct of conduct of conduct described above. Given the dominance of this post-Washington Consensus discourse, we expect to see some evidence of this within discourses related to the private sector and within the main texts of the MDGs and the SDGs.

| CRITICAL DISCOURSE ANALYSIS
Discourse analysis is a collective name for a number of scientific methodologies for analysing semiosis, namely, how meaning is created and communicated though written, vocal, or sign language. Discourse analysis is used in many disciplines in the social sciences, each with its own methodologies and assumptions. 1 CDA is one type of discourse analysis that aims to "understand, expose, and ultimately resist social inequality" (van Dijk, 2005, p. 352), focusing on the dialectical relationships between discourse and other elements of social practices.
According to Fairclough (2012), networks of social practices constitute a social order and "one aspect of this ordering is dominance: some ways of making meaning are dominant or mainstream in a particular order of discourse, others are marginal, or oppositional, or alternative" (Fairclough, 2012, p. 2). Here, Fairclough appears heavily reliant on previous work on understanding marginalized knowledges, as well as on work emphasizing the exclusive nature of knowledge creation. Here again, we emphasize the importance of Foucauldian post-structuralism, although we would argue it is crucial to also bring in both feminist and postcolonial theory's understanding of domination, marginalization, and exclusion. In particular, we aim to include the understandings of Spivak (1988), Said (1985), and Mohanty (1988) because they offer a critique of Foucauldian post-structuralism as itself being complicit in these exclusions through its disregard and domination of non-Western knowledges and knowledge producers. This is something which we, the authors, wish to be particularly aware of, considering our own places within key Western institutions focused on knowledge and policy creation.
This study employs an adapted version of transdisciplinary CDA (Fairclough, 2012), comprising a four-phase research process (see Textbox 1). The first phase involves the selection of a research topic that relates to a social question that can be productively approached by a focus on semiosis. The second phase involves the identification of obstacles to addressing the social question based on the analysis of dialectical relations between semiosis and other social elements.
This involves the identification of a suitable text as well as information on how the text was created. Once an appropriate text (or texts) have been identified, the text is analysed. We accept here a shortcoming in CDA as requiring a "suitable" text given the exclusions it creates, but given our focus on policy documentation, this seems to be necessary for the purposes of this paper. The third phase considers whether the social order "needs" the social question, namely, whether it is inherent to the social order, whether it can be addressed within it, or whether it can only be addressed by changing the social order. The fourth phase is based on the identification of possible ways past the obstacles with a semiotic point of entry through the use of discourses, narratives, and arguments. In short, this methodology uses words in texts and speech to identify social questions and then considers how words in texts and speech could be employed to contest these questions.
In this paper, the methodology has been adapted to make it more suited to the analysis of key policy documents. We also place much greater, explicit emphasis on the genealogy of past discourses, which now receives a full step in its own right, following a previously employed methodology (Cummings, Regeer, de Haan, Zweekhorst, & Bunders, 2018). Although Fairclough refers to the genealogy of past discourses as an important issue, it is not explicitly included in the original methodology. We consider that making this stage explicit is an important amendment to the methodology, making it particularly effective for identifying the subdiscourses, which post-structuralist, feminist, and postcolonial theory enables us to lay bare.
The use of the methodology is justified because "critical attention to the discursive elements of [private sector development] frameworks is a necessary concern for those wishing to understand the full relevance of private sector development" (Langan, 2009, p. 88). In Step 1: Select a research topic that relates to a social question that can be productively approached with a particular focus on the dialectic relations between semiotic and other elements Step 2: Undertake a genealogical study of past discourses in which the current discourse can be located Phase 2: Selection and analysis of texts Step 1: Select texts appropriate to the object of research Step 2: Describing how the text was created Step 3: Analysis of texts in terms of vision, strategy, means of implementation, and goals and targets at the level of: • individual words and phrases, • how the words and phrases relate to each other in the text, and • the priority given to different themes.
Step 4: Identify discourses in the text, based on the prior discourses identified in Phase 1 Phase 3: Consider whether social order "needs" the social question, namely, whether it is inherent to the social order, whether it can be addressed within it, or whether it can only be addressed by changing the social order

| DISCOURSES OF THE PRIVATE SECTOR: A GENEALOGICAL APPROACH
Through an analysis of the literature, we have identified four main discourses of the private sector with associated subdiscourses. Our literature search was not exhaustive but was designed to identify main trends and categories.
The first discourse we have identified sees the private sector as a crucial player in development; we designate this as the unconditionally pro-private sector discourse. The second, sceptical discourse is characterized by reservations about the pro-private sector discourse.
Next, the middle-ground discourse stresses the importance of a more pragmatic, nuanced approach to the private sector, focusing on methods that aim to leverage the development relevance of the private sector and are able to support the private sector's role in development. The fourth discourse, the antiprivate sector discourse, sees the private sector as fundamentally contradictory to sustainable development (see Figure 1). In the following discussion of discourses, we aim to provide a general overview of the characteristics of the discourses to be identified in the key documents we will be analysing later.
4.1 | The dominant discourse: Unconditionally proprivate sector Langan (2011) charts the emergence of approaches that incorporate the private sector into development as part of "recent donor strategies aimed at making free markets work for 'the poor'" (p. 84). This development strategy reflects the post-Washington Consensus described above and is, in essence, the development strategy that accords with this form of governance. It is centred on ensuring that individuals are able to become the type of subjects that thrive within the existing structures, with characteristics such as entrepreneurial, F I G U R E 1 Discourses of the private sector independent, and self-governing. Development, within this governmentality, in essence is making "poor" people into neoliberal subjects who will then be able to do well within the system (see, e.g., Lemke, 2016). This appears to be the embodiment of what we have above defined as the conduct of conduct of conduct.
Proponents of this discourse perceive the private sector as able to contribute to "pro-poor" development. They tend to be the govern- Given such powerful advocates, these discourses are increasing reflected in international agreements related to development (Scheyvens et al., 2016). They are, for the purposes of this analysis, considered the dominant discourses, namely, the primary way in which current systems of governmentality would see development as functioning. In fact, Adams and Martens (2015), writing for the German Global Policy Forum, identify this dominance as also being reflected in the business model of the United Nations-although they use the terms "stakeholder discourse" and the "multi-stakeholderism" discourse to describe it.
Proponents of this discourse see the private sector as having the potential to contribute to international development for a number of reasons. First, the private sector is seen as having the potential to

| Sceptical discourse
A number of academics are sceptical of the win-win scenario espoused within the pro-private sector discourse. Langan (2009Langan ( , 2011, for example, considers that this emphasis on the private sector by development agencies has a number of strategic implications, which are advocated by the global North because it is in its interest. First, "developmentally questionable market opening" (Langan, 2011, p. 84) (Langan, 2009, p. 426). This argument is also supported by Siles-Brügge (2014) who argues that EU's developmental trade agenda is being increasingly determined by the EU's own commercial imperatives. Third, it offers a credible alternative to the interventions, which failed to facilitate "meaningful" business development in former African, Caribbean, and Pacific colonies (Langan, 2011). Fourth are the de facto subsidies to European firms, which disadvantage businesses in the Global South as well as their workers and host communities (Langan, 2011).

| The middle-ground discourse: New approaches for private sector development
The middle-ground discourse appears to be based on the understanding that the development sector is in need of "better evidence of what works and what does not in marrying commercial and sustainable development success" (Horn-Phathanothai, 2013). The discourse is pragmatic and appears to include three main elements: (a) approaches that are specifically focused on the poor or development relevance; (b) methodologies that focus on context and provide for better understanding of relationships between the public and private sector; and (c) calls for better regulation, including legislation and certification.
These elements of the middle-ground discourse appear to be based on a subtler understanding of the relationships with the private sector, social goods, and a greater caution, in an effort to avoid previous mistakes. Jeffrey Sachs, for example, observes that "many large companies are also lobbyists for policies antagonistic to sustainable development, so engagement with business has to be done cautiously" (2012, p. 2211). In many ways, however, this is also a development approach that reflects the previously discussed understandings of contemporary governmentality in disciplining people to become entrepreneurial and self-governing. This development approach is, however, slightly different, as the examples below of social entrepreneurship, corporate social responsibility, and regulation show. This approach considers itself more cautious and more considerate of the human dimension but, in practice, its conduct of conduct is focused on similar disciplines and, perhaps, given its attention to the individual dimension, represents a more effective approach of the same governmentality narrative.
Social entrepreneurship within this discourse is seen as combining social and economic objectives, putting social value above profit and aiming to generate revenue for sustainable development (Zahra, Gedajlovic, Neubaum, & Shulman, 2009). In this way, social entrepreneurs are seen as providing essential goods while also aiming to integrate vulnerable groups into economic activities through access to microcredit and other inputs, such as fertilizers and technologies (Draperi, 2010). Social entrepreneurship is considered to be "development beyond aid" (Fowler, 2000, p. 638) by "defying the obstacles that have prevented businesses from providing services to the poor" (Seelos & Mair, 2005, p. 242).
Methodologies that facilitate work with the private sector in international development include the use of ethical perspectives, which yield the following criteria: exclusion of industries or businesses from partnership; inclusion of businesses who are socially responsible; developing a due diligence review of the partner's track record; risk assessment and management process for the partnership; and determining parameters of accountability in contractual agreements (Carney, 2014). Another example is Corporate Social Responsibility, which, within this discourse, is considered to be "corporate activities that reflect and address both the social imperatives for business success and the social consequences of business activity" (Moon, 2007, p. 299). An example of this approach is a recent publication by Oxfam, which has developed a framework to assess the development relevance of projects and programmes supported by the main bilateral donors (Jeune, 2019).
The final example comprises regulatory frameworks that include a variety of certifications that aim to harness market forces for social and environmental purposes, making partnerships more transparent (Taylor, 2005). Other stakeholders, particularly civil society, are calling for increased regulation because "corporations have a fundamentally different primary interest from that of governments, UN agencies, [civil society organizations], and social movements: their prime interest-enshrined in their fiduciary duty-is to satisfy the interests of their owners and shareholders" (Adams & Martens, 2015, p. 126). Within Fairclough's understanding of discourses, the antiprivate sector discourse would be considered a marginalized or excluded discourse, as well as one that requires a change to the social system.

| The antiprivate sector discourse: The conflicting interests of the private sector
In this manner, this discourse is more of a critique on the way bodies have been governed and disciplined, namely, the conduct of conduct, than an aspect of this governance. That said, Spivak's (1988) critique of Western-produced knowledges and their complicity in the structures of domination that exist between the global South and the global North-as well as others that exist between people due to differences in class, gender, sexuality, disability, and so forth-demonstrates that even this antiprivate sector discourse is produced in the same structures and by the same hierarchies it wishes to dislodge. Due to this, this antiprivate sector discourse is also complicit in current neoliberal governmentalities, particularly due to the locality of its production as being largely global Northern institutions, and its own critique is limited by its complicity and involvement in the conduct of conduct it is critiquing. This discourse is thus produced by and productive of the governmentalities it desires to dislodge.
It is also important to note that, from the perspective of this antiprivate sector discourse, it is not only the pro-private sector approach that is criticized. In fact, middle-ground approaches are generally considered to be similarly exploitative within this discourse, with the added problem that they sanitize the exploits of the private sector. Draperi (2010), for example, considers that social entrepreneurship is unable to contribute to welfare aims because it takes place within capitalist frameworks. Similarly, others see corporate social responsibility as incompatible with development because "there is a danger that organizations or activity are described in these terms simply for image-making purposes" (Moon, 2007, p. 297). For example, Gilberthorpe and Banks consider that "the emergence of [corporate social responsibility] in the extractive industries represents a bid to legitimize the sector after decades of environmental disasters and the trampling of indigenous rights" (Gilberthorpe & Banks, 2012, p. 185).
Researchers have also been describing how certifications have been co-opted by the corporate sector "to further capital accumulation and to defuse threats to their profit margins and to status quo production, pricing, labour, trading and retailing practices" (Jaffee & Howard, 2010, p. 27).

| DISCOURSES OF THE PRIVATE SECTOR
To examine discourses of the private sector within the SDGs, this arti- have been "formed, disseminated and legitimized within complex chains and networks of events (committee meetings, reports, parliamentary debates, press statements and press conferences, etc.)" (Fairclough, 2013, pp. 244-245). In this section, the text is analysed at the level of individual words and phrases related to the private sector, how the words and phrases relate to each other in the text, and priority given to different themes.

| The private sector in the MDGs
The Millennium Declaration (UN, 2000) is a relatively short document of only nine pages. It focuses on "development and poverty alleviation" (UN, 2000, p. 4), specifically emphasizing the role of the UN and members states, as well as globalization. It makes very little reference to actors outside of the UN and national governments. In total, there are three references to the private sector, one reference to industry, and one reference to information and communication technologies (ICTs), which has been identified elsewhere as having a strong link to discourses of the private sector (Cummings et al., 2018) because of "a strongly Western-centric and universal idea of the contribution of technology to economic growth and development" (Mansell, 2012, p. 3).
The first two references to the private sector are "to develop strong partnerships with the private sector and with civil society organizations in pursuit of development and poverty eradication" (p. 5) and "to give great opportunities to the private sector, nongovernmental organizations and civil society, in general, to contribute to the [UN's] goals and programmes" (p. 9). In these references, there is the emphasis on the important role of the private sector in partnerships. The final reference to the private sector, however, refers to the problematic role of the private sector because the Declaration aims to "encourage the pharmaceuticals industry to make essential drugs more widely available and affordable" (UN, 2000, p. 15).

| The private sector in the SDGs
The document "Transforming our world" (UN, 2015) is a longer document than the Millennium Declaration, totalling 40 pages. It has five main sections: the preamble, the declaration, the SDGs, the means of implementation and, finally, follow-up and review. Within the whole document, there are a limited number of references to the private sector (15) and to business (5). The first mentions are in the "Means of implementation" within the Declaration: In the first quote, we can see how the private sector has been included as a key actor, second after government, in the actors responsible for the SDGs, receiving mention even before the UN itself. In the second, the document recognizes the diversity of the private sector. These statements appear most closely aligned with the pro-private sector discourse. In the next reference to the private sector, the emphasis is on private finance (p. 15): Means of Implementation 67. Private business activity, investment and innovation are major drivers of productivity, inclusive economic growth and job creation … We will foster a dynamic and well-functioning business sector, while protecting labour rights and environmental and health standards in accordance with relevant international standards and agreements and other ongoing initiatives in this regard … (p. 34) This reference to regulatory frameworks does bear some similarity to the middle-ground discourse.
The SDGs include two references to entrepreneurship in Targets 4.4 and 8.3, but these refer to the private sector more generally and are not evidence of the middle-ground discourse. There are also passing references to microfinance (Target 1.4), microenterprises, and small and medium-sized enterprises (Target 8.3) and two reference to microenterprises in the "Means of Implementation." This is combined with the many references to the importance of international trade and, in particular, the continued need for removal of trade regulation.
In this way, the pro-private sector discourse is linked to the post-Washington Consensus.
Although there is relatively scant reference to the private sector, the SDGs' text does refer frequently to ICTs. Given the strong links between ICTs and the private sector, this also links strongly with the pro-private sector discourse in which ICTs can alleviate poverty, despite the fact that very few studies appear to show a causal link between ICTs and poverty alleviation (May & Diga, 2015).

| Whose discourses?
In order to understand whose discourses are represented in the MDGs and the SDGs, it is also important to engage with the way in which these documents were developed as policy statements.
According to Hobbes (2015), the MDGs were developed without the involvement of developing countries, although developing countries Despite the complexity of process appearing to indicate that the creation of the SDGs was significantly more democratic and collaborative, many commentators have had wide-ranging complaints both about the process and the outcome. This is partly as, even in the final stages, priorities and agendas differed between countries as a journalist reported in the New York Times: "Priorities differ, agendas differ.
The willingness to take on commitments differs" (Porter, 2014, unpaginated). In addition, it appears that the negotiations around the SDGs were full of suspicion with and conflicts in perspective between developed and developing countries: The tense climate in the discussions revealed scepti-

| DISCUSSION
In our analysis, we found an emphasis on what we have called the pro-private sector discourse within the global goals, presenting an unproblematic understanding of the private sector in development and most probably reflecting the post-Washington Consensus.
Although we have not found any dramatic changes between the MDGs and the SDGs, it does appear that the private sector is more in the foreground in the SDGs than in the MDGs, something also noted by Scheyvens et al. (2016). Indeed, the MDGs do feature one reference to poor practices by the private sector, namely, the pharmaceutical industry, indicating a greater understanding of the potentially negative impacts of the private sector. This pro-private sector discourse has been espoused by the national governments of developed countries and the private sector itself, probably a reflection of who was most influential in developing these policy texts. The MDGs, for example, were known to have been elaborated by the national governments of developed countries. In the negotiations around what stayed in the text of the SDGs and what was removed, the national governments of the developed countries and the private sector were apparently dominant (Briant Carant, 2016;Cummings et al., 2018). In this process, the global goals present an apparently immutable, aspirational vision or imaginary reflecting dominant discourses as advocated by the dominant actors, reflecting contemporary governmentalities focused on creating self-sufficient, self-governing neoliberal subjects.
The aspirational, imaginary nature of the SDGs also reflects what has been said by some key UN actors: "High-level political commitments contained in the 2030 Agenda still represent an aspiration; they do not describe the existing reality" (Dumitriu, 2017, p This seems similar to the way in which discourses around knowledge society have been constructed by the governments of developed countries "for either legitimising ongoing programmes or pushing future economy and technology-oriented activities" (Hornidge, 2011, p. 32). In both processes, the development of aspirational imaginaries appears to play an important role. Analysis of discourses around the private sector not only demonstrates Fairclough's understanding of discourses, but also the ways in which the different discourses related to the private sector were included or excluded when the MDG and SDG policy documents were created. This is mirrored by statements that stress the importance of being critical of the SDGs because they "remain abstracted from an understanding of why the world now faces the entwined crises of ecological collapse and gross economic inequities" (People's Health Movement et al., 2017, p. 17). These commentators are emphasizing the limitations and the need to take a critical perspective.
In this paper, we have taken a post-structuralist approach that provides considerable emphasis on the fact that the private sector has played an important role in the inequalities which have created the need for development and development policy. If we were to take this to its logical conclusion, we might need to argue for greater prominence of new and existing antiprivate sector discourses. However, the complete overthrow of the pro-private sector discourse, which represents a social problem, proposed by Fairclough, appears to be unrealistic because of the dominance of the pro-private sector discourse, entrenched by the post-Washington Consensus. In the current policy environment, the antiprivate sector discourse remains marginalized and there appears to be little prospect for its increased dominance. In addition, throwing the private sector baby out with the bathwater might also not be desirable when development requires resources and the private sector has these. The most pragmatic way forward might be the creation of discourses, which place a greater emphasis on the diversity of the private sector and on a more pragmatic, nuanced approach to the private sector found within the sceptical and middleground discourses. For example, given the diversity of the private sector actors, it would be better to emphasize those private sector actors that have direct development relevance, such as endogenous small and medium-sized enterprises and the informal sector. For example, recent research from South Africa emphasizes that: … the informal sector is a significant component of the economy that provides employment, livelihoods and income for millions of workers and business owners.
This has, however, largely been overlooked by economic analysts and policy-makers in South Africa (and elsewhere). The informal sector has remained forgotten or, at best, in the margins of economic analysis and policy consciousness. (Fourie, 2018: p. 2) We also recognize that discourses around efforts to regulate the private sector, such as certifications and benchmarking, also represent a pragmatic way forward. In addition, new discourses should also interrogate the pro-private sector discourse in order to emphasize the sceptical and middle-ground discourses which are based on a more realistic, analytical, and nuanced perspective of the development relevance of the private sector.
Given that we are advocating for a more nuanced approach, we still feel that the post-structuralist perspective has been valuable for three main reasons. First, we consider that its fundamental focus on power and politics has helped us to resist the tendency to approach development as a technical problem. Second, it supports us as authors in the awareness that we are creating knowledge in a global system which entrenches historical global South-global North relations, and that we are also enmeshed in this reality with a concomitant emphasis on our personal positionality. Third, we consider that the emphasis on neoliberal approaches to the private sector, which aims to create selfsufficient neoliberal subjects has also enriched our understanding on the subject under discussion. Within this context, we consider that current approaches to the private sector could indeed be classed as fetishization in the sense described by Corbridge (2007), citing Chatterjee (2004), which obscures meaning from ordinary people, reinforcing the bureaucratic and technical aspects of development.
Finally, based on our analysis, do we feel that current approaches to the private sector are sufficient to bring about sustainable development? Within all the discourses discussed above, there appears to be an emphasis on the potential economic contribution of the private sector, with the social advantages of the private sector only directly included in the middle-ground discourses. Given the integrated nature of sustainable development, the private sector will also need to make a positive contribution to environmental concerns. It is thus striking that this paper is being written at a time when The Guardian newspaper reports that "New data from world-renowned researchers reveals how [a] cohort of state-owned and multinational firms are driving the climate emergency that threatens the future of humanity, and details how they have continued to expand their operations despite being aware of the industry's devastating impact on the planet" (Taylor & Watts, 2019), again indicating the role of the private sector in climate change. It therefore appears that new discourses are needed, which question the potential of the private sector to contribute to the environmental component of sustainable development. Bieler and McKenzie's (2017) accommodative, reformative, and progressive categories of response to sustainable development within the strategic plans developed by higher education institutions in Canada, might, for example, prove useful when applied to the private sector's contribution to sustainable development.

| CONCLUSIONS
The social issue that this analysis aimed to address was the extent to which approaches to the private sector would be able to optimize the contribution of the private sector to sustainable development in a transformational manner. We consider that the unproblematic emphasis of the private sector, which is consistent with the proprivate sector discourse espoused by governments of developed countries and the corporate sector, is evidenced within the MDGs and to an even greater extent within the SDGs. The method of CDA therefore calls for new discourses, but we also need to remember that "it is not enough to talk about the private sector. We need to engage with the private sector, listen to business leaders and understand what drives them; and we need to learn from their vast experience" (Moreira da Silva, 2017, unpaginated). New narratives need to be created not only by better in-depth research but also by challenging unproblematic approaches to the private sector in both the written and the spoken word. When development colleagues ask "Have you involved the private sector?" the answer should be "Which private sector did you have in mind? And how do you think they will contribute to the economic, social and environmental aspects of sustainable development?"

ACKNOWLEDGEMENTS
We acknowledge that Sarah Cummings' contribution to this research has been undertaken as part of the project "An unusual suspect: The role of the private sector in knowledge brokering in international development," funded by the Science for Using Research (SURe) programme, NWO-WOTRO Science for Global Development, The Netherlands. Final revisions were undertaken after Sarah had been awarded the NWO-WOTRO NL-CGIAR project on 'Improving the effectiveness of public-private partnerships within the CGIAR: knowledge sharing for learning and impact' (W 08.240.301). Anastasia-Alithia Seferiadis' contribution has also been funded by the Fondation Croix Rouge, France. Leah de Haan's contribution was not undertaken as part of Chatham House's research programmes or projects. The authors would also like to acknowledge with gratitude the role played by the anonymous reviewer. We are particularly grateful that he/she challenged us to be more explicit on our political economy perspective.