Socially-Tolerated Practices in Environmental and Social Impact Assessment Reporting: Discourses, Displacement, and Impoverishment

: Normative guidelines for addressing project-induced displacement and resettlement have been successful in coercing companies and practitioners to comply with international standards and local requirements. However, good practice has not always been effectively implemented, leading to reduced social wellbeing of people in local communities. We assess how the reciprocal relationships between institutional norms and practitioners’ situated perspectives about company-community interactions can improve social management practice. Drawing on Hajer and Versteeg’s method of environmental discourse analysis, discussions and storylines about a mining project in Mpumalanga in South Africa were assessed against contextualised discursive conventions in the mining industry. It was found that practitioners learn to manipulate legislative requirements, which ultimately perpetuates the impoverishment of project affected communities. The question is not whether or not practitioners understand the requirements of environmental and social management, but the extent to which such understandings are manipulated for corporate gain as opposed to social good. We consider practitioner rationalities about the purpose and function of environmental and social management, and how it is implemented. We suggest that practitioners and companies should construct positive aspirational identity perspectives about social management that would transcend from their current limited view (that achieving minimum compliance is sufficient) to aspiring to achieve better social development outcomes for all, especially the most disadvantaged. This requires a genuine commitment to obtaining and maintaining a social licence to operate, perspective transformation, a commitment to inclusiveness, and increased capacity for critical reflection.


Introduction
The technical-rational nature of environmental impact assessment (EIA) procedures [1,2] has meant that adequate attention is not accorded to the forms and ways by which existing social structures and relations are transformed [3,4] by changes resulting from planned interventions [5,6]. Increases in the number [1,7], procedures [8,9], and technicality [2] of EIA requirements, and its lack of innovation [3], are much reported in the literature. However, little effort has been devoted to understanding the effects of practitioners' and developers' discursive approaches to international standards [10][11][12] on actual practice and social justice. Nevertheless, social impacts resulting from planned interventions are symptoms of deeply-rooted [2] and multi-linear complex issues [5]. In this paper, we seek to unravel emergent social realities by interrogating patterns of situated discursive practices of environmental and social impact assessment (ESIA) requirements and the challenges they pose to impoverishment in mining-induced displacement and resettlement. We do this through a critical discourse analysis of the discussions and storylines relating to a specific mining project located in Mpumalanga, South Africa.
Given that discourses are institutional and language-based modes of meaning-making and ordering people, objects, and practices [13], the meanings people make of ESIA can be fundamental to its incremental improvement. Worldviews and mechanisms of identity formation provide the cognitive background in which practices evolve [14]. The entrenchment of identity perspectives of corporate social responsibility (CSR) into ESIA thinking can, therefore, encourage critical reflexive practices in impact assessment. CSR [1], ethical professional practice [15,16], and a human rights based approach [17] are at the forefront of strategic deliberations on project-induced displacement and resettlement. Inherent in the routine application of environmental regulations [10], organisational relations (e.g., project proponents and ESIA practitioners), and social relations (affected communities, civil society and government) are contradictions [11] that perpetuate projectinduced displacement and impoverishment. Of particular note is the ability of new cognitive frames to enable changes in worldviews and identity formation to a social justice consciousness.
In line with the 'environmentalism of the poor' movement [18], impoverished populations have increasingly been contending with the very state and private companies that threaten their health, livelihoods, culture, and autonomy, given the prevalence of social and environmental injustices [4]. Social impacts arising from planned interventions, such as mining projects, are not usually adequately addressed in ESIA reports [19]. Mining activities, usually paraded as catalysts to economic development [7], often turn out to be sources of considerable negative environmental and social impacts [20], as well as the cause of the physical and economic displacement of communities [21]. In fact, it is estimated that around 15 million people are resettled every year [22]. Project-induced displacement and resettlement is expected to increase, especially on lands that have low acquisition cost, weak land tenure arrangements, or support vulnerable people who are politically weak and powerless [23].
In South Africa, the communities most affected by mining activities are smallholder farmers, many of whom have had their farmland taken from them without adequate compensation [24]. The agricultural sector contributes to 5.2% of employment in South Africa, with an increasing population of subsistence and small-scale farming, of which 30% are reported to have no income or basic schooling [25]. Smallholder farmers struggle given their inability to participate in the modern agricultural value chains [26], which remains one of the key challenges to food security [27].
There is usually inadequate provision for sustainable livelihood restoration programs in resettlement action plans [28]. Furthermore, CSR initiatives do not adequately address community development needs [29]. In South Africa, CSR primarily emerged as an approach to appease political pressure arising from the social exclusion established by the apartheid regime [30], which resulted in South Africa being under-developed [31]. This has implications for the way ESIA is implemented [28], given its overlap with CSR [32]. Benefit sharing in resettlement and livelihood restoration practice in South Africa requires particular attention [24], with land use and land reform being at the forefront of economic and political instability [33]. Land is a symbolic representation of cultural heritage [34,35] and often the only means of livelihood for communities affected by mining projects, especially historically-disadvantaged populations [23]. Displacement caused by mining projects can exacerbate the already-contested terrain of poverty and impoverishment [2].
Following Boyce's [36] 'power weighted decision rule', Scharber [37] reiterates that the benefits and costs of mining, sugar plantations or other major projects are not the only decision rules in such situations, the power of potential winners in relation to losers is a key factor in project negotiations. Increasing environmental injustice around the world can normalise power manipulation and perpetuate project-induced displacement [4]. Since affected communities are less powerful than project proponents, such manipulation will likely persist. Project proponents operating in countries where social activism and civil societies have legitimised voices are more accountable and receptive to livelihood restoration programs than those in countries where such activism is delegitimised by social and discursive practices [38]. The prevalence of the triple dilemma of inequality, unemployment, and poverty in South Africa [39] can normalise impoverishment in resettlement and livelihood restoration interventions.
Antecedents of political struggles permeate into the present through the reproduction of discourses, in that current political and social realities are a function of the revolutionary but often empty actions of the past, which are subject to manipulation for political or economic gain [40]. In an effort to be recognised as being socially responsible, companies strive to demarcate boundaries and a sense of identity [41]. This process of identification inevitably leads to the creation of multiple identities, which organisations struggle to manage in order to build and portray a coherent identity, especially in relation to discharging their broader corporate citizenship duties [42]. Corporate citizenship in the mining industry in South Africa and elsewhere entails compliance with international standards (such as the International Finance Corporation (IFC) Performance Standards and the World Bank Environmental and Social Framework), national environmental regulations (such as the South Africa Mining Charter 2018), and to the tenets of CSR and the notion of social license to operate [43]. The IFC requires mitigation of risks to acceptable limits [44]. However, acceptable impacts and risks are socially-constructed and are based on contextualised patterns of prevailing socio-economic challenges [5], which are often negotiated and sanctioned in discursive practices through ESIA reporting [11]. The espoused identities of project proponents are diverse, often in an attempt to conceal conflicting identities that threaten their social license to operate [45].
The technical documentation of a resettlement process tends to be given more priority than its actual implementation [7]. There have been attempts to address project-induced displacement and resettlement from a human rights perspective, especially its involuntary nature [17,46,47]. However, storylines in ESIAs often position mining activities to be in the public interest, as the only likely catalyst of rural development, and/or as representing the optimal use of the natural resources [11].
From a discourse analysis perspective, the first step towards social justice is consciousness [48]. The manner in which environmental arguments are deliberated in ESIAs can reveal prevailing thresholds for social issues and the corresponding approaches to bring about change [10]. The application of interpretation derived from the sense-making of an experience into decision making turns meaning-making into learning [49]. An important insight that can be drawn from this analysis is that a social justice consciousness applied to the classification schemes used to portray affected communities that are implicit in ESIAs can foster sense making of the threats to social wellbeing and highlight the confining identities that perpetuate project-induced displacement and impoverishment.
We consider the Umsimbithi eMakhazeni mining project in Mpumalanga, South Africa, and assess the interplay between the institutional and contextualised classification schemes in the discussions and storylines present in the ESIA from this project and elsewhere. The analysis revealed that normative attempts to enforce the alignment of legislative frameworks to international good practice are not effective in addressing the impoverishment resulting from mining activities. The excessive focus of proponents on the techno-scientific nature of ESIAs encourages manipulation of legislative frameworks for corporate gain. There is an implicit 'culture of atonement' for loss of life or livelihoods caused by mining projects that is rhetorically built into ESIA reports but which is not fully implemented in social management practice. A holistic community development approach is pivotal for addressing livelihood restoration [7], demanding that the resettlement process be conducted with a genuine social justice consciousness. Critical reflection on CSR-based identity perspectives and ESIA practice can facilitate transcendence from cosmetic allegiance to the process of resettlement to an enduring value system that takes responsibility for better social development outcomes. We offer an approach grounded in self-reflexive practice and that integrates resettlement practice into the wider CSR portfolio.

Developing a Framework to understand CSR-based Identity Perspectives about ESIA
The use of ESIA in South Africa has been limited, is contested [28,29], and best practice ESIA remains aspirational [29]. One complication is that in South Africa there is a conflation of CSR with corporate social investment (CSI) [31]. CSI initiatives tend not to adequately address community development needs [29], even though South Africa continues to battle with economic challenges such as high unemployment, high poverty levels, and inequality [39].
History, culture, and strategy are the core determinants of corporate identity [41], and give an organisation centrality, distinctiveness, and endurance [50]. An assessment of the CSR-based corporate identity construction of three Johannesburg Stock Exchange listed manufacturing companies revealed a lack of commitment to, or coherence in, espoused CSR values [51]. An assessment of the sustainability reports of 246 large South African companies revealed that, rather than a genuine commitment to sustainability, integrated reporting was merely a symbolic attempt to acquire organisational legitimacy [52].
Historically, the approach to gaining a social license to operate has been project focused and based on the notion that the acceptance of the project by the local community is essential for the success of the project [1]. Compliance with the IFC Performance Standards is intended to help project proponents find ways to enhance local benefits, encourage good corporate citizenship practices, and gain genuine community acceptance of projects [53]. The ability to acquire a social license to operate is based on the extent to which affected communities perceive the practices of project developers to be legitimate, credible, and trustworthy [54]. Actionable requirements for benefit sharing need to be made in resettlement legislation, as the loss of assets and insufficient compensation are not sufficient to restore livelihoods [55].
Disregard of the prior social and political context of host communities, legacy issues, or of any ethnic or political tension can trigger negative reactions towards a project [2]. Consequently, for projects likely to result in physical or economic displacement, project proponents need to exceed IFC's minimum public disclosure and consultation requirements [21]. The IFC Performance Standards address the environmental, social, health, human rights, and labour issues that are likely to be encountered throughout the life of a project [44]. They require effective implementation of an environmental and social management system (ESMS), comprising: (i) policy; (ii) identification of risks and impacts; (iii) management programs; (iv) organisational capacity and competency; (v) emergency preparedness and response; (vi) stakeholder engagement; and (vii) monitoring and review [44]. The Performance Standards underscore the importance of managerial responsibility for effective implementation of the ESMS. The organisational capacity and competency has implications for the ability of the project staff to integrate the ESMS actions into wider community development initiatives, such as CSR, given the centrality of CSR to organisational identity [50].
A robust CSR portfolio demonstrates allegiance to society and local communities through fulfilment of good corporate citizenship activities that surpass corporate gain [56]. Carroll [57] provides a framework for approaching CSR as a pyramid that comprises (in order) economic, legal, ethical, and philanthropic responsibilities. In the African context, however, it is argued that the order of elements in Carroll's pyramid would be different: economic, philanthropic, legal, and ethical responsibilities [58]. The positioning of ethical responsibilities at the top of the pyramid limits the likely uptake and implementation of a social justice conscious CSR portfolio in Africa [56]. The capabilities and competencies required to institutionalise a social justice conscious identity perspective applied to CSR can be inherent and emanate from espoused corporate values [59].
Organisational goals and objectives are manifested in corporate identity through corporate philosophy, structure, ownership, history, culture, visual images, and symbols [60]. Identity perspectives [51] applied to CSR can be instrumental in initiating a shift from the technical implementation of resettlement plans to a relational approach, which extends beyond mere legal compliance [61]. An earlier framework for understanding identity perspectives [59] demonstrated the overarching role of a social justice consciousness in the construction of CSR-based corporate identity and governance. That framework also explicated how past configurations (antecedents) of CSR learning can be used as a tool to assess beliefs that feed into present and future practices, namely core and aspirational components of corporate identity attributes, as outlined below:  Antecedents of CSR Corporate Identity: A compliance-based approach to CSR engagements can be reinforced by lifelong learning values;  Core Components of CSR Corporate Identity: Efforts are constantly geared toward ensuring coherent organisational identities;  Aspirational Components of CSR Corporate Identity: To gain loyalty, the process of acquiring a license to operate should be altruistic. From a reflexive learning perspective, a process in which problems are defined and solved becomes a context for learning [62]. Deeply-held assumptions retard learning, active reflection of practice, and effective decision making, which consequently leads to non-reflexive, habitual action [49]. Therefore, we offer a conceptual framework that assists in understanding how the antecedents of situated learning about CSR foster project-induced displacement by encouraging a distorted focus that embeds strong allegiance to only minimum legal requirements. Our framework considers how identity perspectives in CSR can foster a focus on regulatory practice or on social justice conscious practice, depending on the domain from which learning is drawn. Learning based on reflection of presuppositions facilitates retrospective questioning of deeply-held assumptions, which can enable reflexive interpretation of practice, reflexive actions, and, ultimately, social justice conscious decisionmaking (see Figure 1). We encourage all ESIA practitioners to undertake a social justice consciousness appraisal of their beliefs, and a thorough examination of the assumptions they hold. Our framework promotes a relational approach to ESIA, which is grounded in a social justice consciousness.

Analysis of Discursive Practices in ESIA
Language permeates human life experience, and at times, language is at the root of severe social problems [11,63]. Realising how language works can be a first step in addressing such problems [64]. For example, discourse legitimises and secures the dominance of capitalist society [65]. Institutionalised ways of talking often regulate and reinforce actions and exert power [66], which can reveal and redress socially-regulated patterns of impoverishment in resettlement and livelihood restoration practice.
Discourse is an ensemble of ideas, concepts, and categories through which meaning is given to phenomena, which is produced and reproduced through an identifiable set of practices [10]. The use of critical discourse analysis in social research can reveal patterns of social wrongs, resistance, and possibilities of overcoming them [67]. Since ESIAs contribute to granting project proponents the permit to mine [11], analysis of ESIA discussions and storylines that relate to communities affected by mining projects is instrumental for revealing implicit discursive practices that can inform the criteria for regulating and/or redressing impoverishment caused by such projects.
As Hajer and Versteeg [10] note, the focus of interpretative environmental policy research is on the way society makes sense of environmental phenomena. Foucault [68] (p. 54) declared that discourses are 'practices which systematically form the objects of which they speak'. Foucault [69] (p. 394) further argued that power consists of a 'series of particular mechanisms, definable and defined, that seem capable of inducing behaviours or discourse'. Foucault's ideas have informed contemporary social sciences and linguistic research in relation to discourses and power relations, the understanding of realities in political conflict, public debate, and scientific knowledge-making in the governance of risk populations [13]. The environmental discourse analysis of Hajer and Versteeg [10] has the capacity to reveal the role of language in politics, the embeddedness of language in practices, and answer the 'how' questions to illuminate the mechanisms that can be operationalised through analysis of discussions and storylines. Discussions are the oral and literal utterances that form the object of analysis, while storylines are narratives on social reality through which elements from different domains are combined providing actors with a set of symbolic references that suggest common understanding [10].
Fairclough [48,70,71] argues that text, social interaction, and social context are central elements of a discourse. Written or spoken text, processes of re (production) and consumption of text (social interaction), socio-cultural, economic, political, and environmental conditions of production and interpretation (social context) constitute the description, interpretation, and explanation phases of critical discourse analysis. Fairclough also argues that the description phase involves the generation of text and its categories, interpretation entails the search for meanings of described text, whilst the explanation phase explicates the implications of the meaning of social practice.

Background to the Case Study and Methods
This paper is based on a discourse analysis of material related to a particular mining project, the Umsimbithi eMakhazeni project, which is an extension of the Wonderfontein colliery in Mpumalanga, South Africa. Although mining activities at the Wonderfontein colliery commenced in 2013 without a social license to operate, there was proof of public participation [72]. This project was selected for this research because it was an expansion of an existing mine that has been controversial, especially with regard to the multidimensional stress experienced by affected communities due to the physical and economic displacement. The EIA and public participation reports for the Umsimbithi eMakhazeni project were used in the discourse analysis. Furthermore, given that ESIA is a social and cultural process that cannot be approached in a vacuum [3], we also drew upon prominent mining-related discourses, such as the Marikana address [73], the Mining Charter [74], the debate on the State of the Nation Address on the Marikana killings [75,76], as well as the significant influence of the 'stakeholder discourse'.
Umsimbithi Mining Pty is the project proponent. It is jointly owned by the Shanduka Group and Glencore, with Glencore being the managing shareholder with a 49.99 per cent share [77]. Glencore is a mining giant headquartered in Switzerland [78]. It has a long history of allegations of tax evasion, poor conduct, and corruption from various social and environmental activist groups [79]. In 2018, Usimbithi Mining Pty proposed to expand operations of the Wonderfontein colliery, calling it the Umsimbithi eMakhazeni mining project. They held discussions with the Department of Environmental Affairs, NGOs, social and environmental activists, and community members and farmers affected by this expansion. Kogiwe Environmental Pty was the private consultancy firm in charge of public participation and the EIA. Data (i.e., text for discourse analysis) were gathered from secondary sources available in the public domain. Data context (how the text relates to what is going on), the intended target audience for the text, and alignment of text sources with the broader mining debates were considered. The EIA and public participation reports of the Umsimbithi eMakhazeni mining project formed the core dataset, which was supported by corresponding storylines and discussions: the Marikana address in 2012 [73], the 2018 debate on the State of the Nation Address on the Marikana killings [75,76], and the Mining Charter 2018 [74].
Emergent coding [80] was used to inductively bring to the surface prevailing themes (hereafter, discourse strands) from the data. Given the vital role of contexts of discourse construction [81] and medium of communication [82] in shaping meaning, a preliminary analysis was conducted on the Mining Charter using N-Vivo 12 to identify patterns of socially acceptable language and prevailing narratives. Subsequent analyses were informed by emerging discourse strands. First, collocation analysis [83] was used to examine the embeddedness and sense-making of discourse strands in discursive conventions. Collocation principles for the lexical co-occurrence of words reveal the unit of meaning as construction since pre-existing political dynamics feed into future patterns [83]. Second, an analysis of the linguistic strategies used in mining-related keynote address was conducted to examine socially-regulated notions of sense making of the 'threshold of acceptable impacts' in storylines. Third, intertextuality was used to further assess socially-regulated ways of producing and interpreting the prevailing discourse strands in relation to its trajectories in practice, based on an inductive approach. Intertextuality entails how discourses are produced and interpreted [84] and reveals the implications of prior texts for the restructuring and possible transformation of prevailing conventions [85]. A lexical word search was conducted on the EIA and public participation reports, using MAXQDA.

The Role of Language in ESIA
Preliminary analysis of discussions in the Mining Charter 2018 revealed the following central discourse strands: 'compliance', 'impact avoidance and mitigation' and 'acceptable impacts', as well as 'social licence to operate' and 'stakeholder engagement'. The IFC Performance Standards advise that stakeholder engagement is an important component in building and maintaining relationships with affected communities and other relevant parties [44]. Consistent with the IFC requirements and given the centrality of the stakeholder discourse strand in the Mining Charter 2018, a word frequency analysis for 'stakeholder(s)' was done on the 649-page EIA and the 363-page public participation report, in which the word occurred 79 and 95 times, respectively. A word tree analysis was conducted, which showed patterns of how the 'stakeholder(s)' concept is constructed in line with other recurring discussions. The word tree analysis systematically showed the structure of the word and its meaning composition in the reports through the recurrence of nodes and phrases. The word, 'stakeholder', was correlated with 36 and 66 nodes alongside recurring phrases in the EIA and public participation reports, respectively (see Boxes A1 and A2 in Appendix A). The analysis also showed correlation of the word with many other sub-nodes. Nodes such as comments, database, engagement, meetings, networking, dialogue were predominant in the EIA. The analysis revealed that names of provincial environmental officers, competent authorities, project managers, meetings, stakeholder engagement, and engagement were the most recurring nodes in the public participation report, with the node 'engagement' being most recurrent. These preliminary findings revealed that deliberations that involve stakeholders and/or affected local communities have more weight in the public participation discussion than in EIA negotiations, although the analysis showed no pattern of actual participation with local communities. The analysis also revealed that issues that are not stakeholderrelated issues are seldom addressed beyond the public participation section of the EIA. Nevertheless, prevalence of the word 'stakeholder' in ESIA reports does not necessarily translate into identification with or allegiance to the concept, nor to the project actually conducting effective grievance redress mechanisms or meaningful consultation with affected communities. Impact assessment not involving people can undermine the social realities faced by affected communities. The greater prevalence of the use of the word 'stakeholder' in the public participation report than in the ESIA has implications for consent-seeking behaviour. It implies that there is a tendency to bypass the social problems caused by mining projects.

Text Description Phase
Insights dawn from the preliminary analysis necessitated further investigation in the discourse analysis. For example, it was particularly important to interrogate the potential ideological impacts of the meaning-making of the stakeholder concept and how such impacts can deepen understanding of the challenges posed to the representation of local communities affected by mining activities. As such, it was deemed fit to analyse the co-occurrence of the word 'engagement' with the stakeholder concept. Analysis of modality was thereafter conducted on the Executive Summary of the EIA to bring to the surface prevalent ideologies perpetuated in the entire report.

Collocation Analysis
Excessive use of nominalisations, i.e., words that end with 'ment', 'tion', 'ting', can have ideological meanings in discourse analysis, as they often conceal actions and people, and convert actions into objects [86]. The collocation analysis shows that the nominalisation, 'engagement', was mostly collocated with the noun, 'stakeholder', especially in the public participation report. This was followed by words such as meetings, comments, inputs, dialogue (see Boxes A3 and A4 in Appendix A). The rhetorical occurrence of the nominalisation, 'engagement', in the business language of 'stakeholder engagement' conceals agency, although it signals inclusivity, but suggests a buried ideology of reducing the 'otherness' effect. In other words, repeated use of the term can exonerate the speaker from excluding relevant parties from project consultations. This questions the genuineness of participatory decision-making and responsibility for decisions in the resettlement process. Stakeholders are broadly defined, and it is not clear which particular stakeholder group is in question, their stake, level of vulnerability in the project, the agent responsible for the engagement, or the nature and proposed outcome of the engagement.

Analysis of Modality
The analysis of modality in discourse analysis is based on the ideologies implicit in the use of modals (e.g., can, could, will, might, may, should, must, would) in written or spoken text as they reveal the degree of obligation and commitment to truth [86]. Modality can also be associated with hedging terms such as 'I think', 'kind of/sort of', 'seems' or 'often'. Modality can be epistemic, deontic, and dynamic. Analysis of modality was conducted on the Conclusion section of the Executive Summary of the EIA report, as shown in Box A5 in the Appendix.
The statement, "Based on the information contained in this report, it is the opinion of the EAP [environmental assessment practitioner] that the negative environmental impacts resulting from the eMakhazeni Project can be mitigated to within acceptable limits and that the project should be authorised", contains the modal verb, 'should'. The tone of the modal verb in the statement relates to deontic modality and showcases the speaker's power and authority to endorse the commissioning of the project. It is noteworthy that the phrase, "the project should be authorised", as well as the next sentence, appeared in bold font in the text (see Box A5 in Appendix A).
The second modal verb used in the text was 'would': "The findings of the impact assessment have shown that the eMakhazeni Project would conclusively result in certain negative impacts to the environment". This reveals another angle to the discourse. The modal verb 'would' here relates to epistemic modality, which expresses the likelihood of the occurrence of negative impacts. The speaker constructs a discourse, which appears to reduce the 'otherness' effect, in that the 'certain' negative impacts in question were not enumerated or highlighted in the text. The text, however, concludes with a list of potential positive impacts arising from the proposed project. In addition, the use of an assertive modal, 'will', which conveys a considerable degree of certainty of future occurrence, was used twice in the last sentence of the paragraph: "The eMakhazeni project will thus facilitate the planned mining activities and will have rollover benefits in terms of local employment, local economic development and, increased government revenue and taxes." The co-occurrence of 'modality' and 'modality without modals' was also observed in the text, suggesting a tendency for consent-seeking behaviour. For example, inherent in the speaker's conflation of the hedging terms, 'seems unlikely' and 'likely', with the modal, 'will', in the statements (1) "it seems unlikely that the mine will be able to operate on local groundwater resources" and (2) "It is likely that water will need to be imported to operate the mine, e.g., from the ELM" is the tendency to downplay the complexities of water management in coal mining, potential risks, and associated negative impacts in order to facilitate approval of the project.

The Interpretation Phase: Embeddedness of Language in Practice
The ideologies implicit in the stakeholder concept and prevailing storylines in the EIA report suggest that the conventional language used in policy documents to regulate practice can be used to manipulate the system. It is, therefore, appropriate to map how such ideologies permeate sociopolitical discourses and normalise the appalling treatment of vulnerable mining communities. Even though the 2012 Marikana killings were perpetrated by the police, the ideologies that permeated during and after the killings of Marikana miners [87] reveal how the killings were interpreted in South Africa. Our data for analysis of storylines comprised: the address of a prominent political leader (PPL1) shortly after the Marikana killings; the Mining Charter 2018 (MC); and the address of another prominent political leader (PPL2) in 2018 in response to the Marikana killings (see Box A6 in Appendix A).
As shown in Table 1, analysis of storylines in the mining industry reveals the centrality of the ideological underpinnings of the tenets of 'Ubuntu' (social capital) [88,89] on the notion of economic prosperity for the atonement of blood and affirmative action. Certain linguistic strategies, specifically deictic centre of time, collectivisation, nominalisation, epistemic, and deontic modality, were embedded in the data revealing how the Marikana incident was downplayed and manipulated to justify economic development and prosperity. Taken together, the analysis of all three sets of textual data shown in Box A6 reinforce the values of resistance to hardship (PPL1), resilience (PPL2), and tolerance (MC) (see Table 1).

The Explanation Phase: Illuminating Processes and Mechanisms in Wider Stakeholder Discourses
Intertextuality in non-literal works was conducted to assess how the stakeholder discourses transcend into discursive practices to illuminate the socially-regulated uptake of ESIA requirements in line with international standards [84,85]. The ideological positioning of the stakeholder discourse in wider discourses was mirrored in the Usimbithi eMakhazeni project. Insights were also drawn from the discourse constructed by advocates of the stakeholder concept: in theory, Ansoff [90], Freeman [91], and Hutton [92]; and in practice, Mandelson [93] and Blair [94]. Using an inductive approach, a lexical word search of stakeholder related concepts was conducted on the EIA and public participation reports using MAXQDA.
When the spirit of the Mining Charter is not fully embraced and compliance is seen only as means to protect social license to operate, discourses will tend to be constructed to protect a project's social license to operate. Our intertextual analysis sought to gain an understanding of the patterns of intertextual weaving in the various documents.
The prevalence of socially-regulated language in reports hides agency and hence the responsibility for actions. Practitioners often have to respond to queries from NGOs and activist groups. The Umsimbithi eMakhazeni project was no exception. Practitioners and developers are usually required to provide detailed explanations and clarity on the practicality and relevance of project interventions, and adequate predictions of likely project impacts. For example, groundWork, one of the activist groups that responded to the EIA of the Umsimbithi eMakhazeni project, queried the adequacy of the proposed interventions to compensate affected communities, in the absence of commensurate resettlement and livelihood restoration programs [95].
Ensuing queries arise from the prevalence of jargon-laden language and socially-regulated ways of producing the ESIA discourse, which typically fail to adequately account for the realities of affected communities. As demonstrated in our findings, an example of such language is the 'stakeholder' concept, which tends to reinforce and signal allegiance to the core values of inclusivity in discursive and non-discursive practices.
The term 'stakeholder' was popularised by Ansoff [90] in corporate strategy and governance, and by Freeman [91] in academic parlance. While both Ansoff [90] and Freeman [91] conceptualised the word in business terms as anyone who can affect or be affected by business activities, such as employees, suppliers, shareholders, consumers, communities, interest groups, and government [91]. Tony Blair [94] adapted the concept to labour relations and politics. Intertextual analysis reveals the entanglement of discourse strands, the transfer and reproduction of ideologies, and the fragments that develop as a result. Blair constructed a discourse which shifted the ideology of stakeholders from its conventional business genre to a political discourse. The understanding of the term moved from individual to collective categorisations with collective bargaining, inclusivity, and the dignity of labour. The stakeholder ideology permeated by Blair may have influenced global labour relations and negotiations. Blair's notion of the stakeholder economy resonates with the ideology of Ubuntu in South Africa, which is grounded in a collectivistic notion of 'a person being a person through other people' [89,96]. This notion could have implications for expectations and the application of a legislative framework for consultation with stakeholders.
Another strand of the stakeholder discourse, 'stakeholder engagement', emerged later as a managerial and professional discipline [97]. The term 'stakeholder engagement' has become influential in business and political discourse and is used by social actors to negotiate legitimacy in corporate communication through rhetorical language manipulation to acquire a social license to operate. From the lens of discourse analysis, repetitive reference to people as 'stakeholders' and to planned activities interacting with people in 'stakeholder engagement', especially in the context of mining activities with major legacy issues, can be a 'distancing strategy' that suggests a lack of commitment to tangible actionable interventions.
Using the word, stakeholder(s), as a primary source, the lexical word search on MAXQDA, was based on a 10 item wordlist: people, property, properties, affected, community, local, lives, farm(s), blasting, and resettlement. This analysis revealed situated patterns of the ways in which ESIA arguments are constructed to appease regulatory requirements. It was found that the socio-economic development discourse was frequently used to respond to and moderate the sense-making of negative project outcomes. For example, affected communities were frequently positioned to be 'employed', 'moved', 'identified', and 'contacted', while concerns raised by civil societies and activist groups regarding blasting and compensation for lost properties were generally 'noted' and promised to be acted on in the future. Another pattern in the reports relates to the strategic use of premeditated and endorsed phrases such as universally acceptable principles, worst-case scenario principle, precautionary principle, good mine planning, and water management principles, all of which reveal situated discursive practices of the application of ESIA.

Conclusion
Practices that perpetuate impoverishment in mining-induced displacement shed light on prevailing worldviews and mechanisms of identity formation. Situated discursive practices in ESIA can be grounded on constructing compelling arguments that seek to overcome institutional barriers to projects, while acknowledging socially-regulated levels of harm that can be inflicted on affected communities. Skewed efforts geared toward transcending institutional obstacles can normalise the exclusion of affected communities and hence perpetuate impoverishment in project-induced displacements. Capabilities required to promote inclusiveness in the resettlement process can be enabled through reflexively interrogating practices that may have been overlooked or misunderstood. The willpower to accept and take responsibility for negative impacts caused by mining projects is key to redressing the ills of project-induced displacement.
Being 'less powerful' than companies and government, vulnerable mining communities are often made worse-off by the pre-determined compensation arrangements for loss of livelihoods. Desperation for buy-in perpetuates the notion of 'otherness' and undermines meaningful consultations through the use of proper grievance redress mechanisms. Stringent normative attempts at promoting social wellbeing and inclusiveness in ESIA can coerce practitioners to comply with legislation but can be incapable of fostering tangible outcomes and sustainable development. Situated assumptions often influence interpretation and application of legislation, discursive practices, and thus, attitudes toward project-induced displacement and impoverishment.
Mastery of environmental policy influences the manipulation of legislation, which undermines the wellbeing of project affected communities, especially those who have less power to challenge authorities. Reflections on presuppositions can foster the unlearning of compliance with the regulatory minimum. We propose that the capacity for legislation to nurture inclusiveness and critical reflections of assumptions in ESIA communication and practices is linked to aspirational identity perspectives to CSR, based on social justice consciousness in the assessments and reporting of policies, procedures, and plans.
If the relative power of winners and losers determines the conditions of and responsibility taken for harm caused to communities affected by mining projects, and ESIAs grant the permit to mine, what implicit social networks regulate the threshold of acceptable negative environmental impacts and qualify such impacts to fall 'within acceptable limits'? Such implicit networks enable cosmetic regulatory approaches and can be equally linked to deeply-seated factors, which this study addresses by offering a framework that promotes reflexive ESIA practice.
Efforts targeted at intentionally altering belief systems can be cosmetic, producing momentary results, which may look good in reports but are too short-lived to make significant impact bringing about the desired change. Ad hoc efforts toward changing belief systems do not necessarily translate to a change in a problem. The will to change, therefore, rests in the hands of project developers and practitioners, as long as they surrender their self-interest for the social good. Thoughtful actions and reflections on presuppositions will not only promote wellbeing and sustain livelihoods for affected communities but will also lay an enduring foundation for social justice. This demands that agents be intuitive and critical of their own way of thinking and of the ways through which thinking evolves into practice. Of utmost importance is the need to consistently resist thoughts that perpetuate despair or jeopardise the wellbeing and safety of affected communities.