Fossil-fueled stories: an ecolinguistic critical discourse analysis of the South African government’s naturalisation of fossil fuels in the context of the climate crisis

ABSTRACT

In recent years, aging coal power plants, lack of maintenance, and issues of poor governance have resulted in a high frequency of rolling scheduled blackouts, throughout South Africa. This has led to greater urgency being placed on switching to renewable energy sources, which South Africa has great potential for Oyewo et al. Despite this, and the current reality of the global climate crisis, South Africa continues to rely heavily on coal, not only as an energy source at home, but also as a key part of the country’s export economy. This paper utilises a corpus-assisted, ecolinguistically-informed critical discourse analysis framework to investigate the ways in which members of the South African national government construct the use of coal and other fossil fuels in light of a global climate crisis. The data for this paper consists of a variety of public-facing texts from the South African national government on the topic of climate change. Our analysis revealed that despite a stated commitment to prioritising renewable energy by the government, its implicit messaging normalises the continued use of fossil fuels and minimises their serious negative impacts.

KEYWORDS:

• Ecolinguistics
• climate change
• fossil fuels
• critical discourse analysis
• South Africa

Introduction

We are currently in the midst of an escalating planetary crisis of rapid anthropogenic climate change and global heating. There is a high degree of consensus among climate scientists regarding the direct cause of climate change, namely greenhouse gas emissions (IPCC, 2022). There is also strong consensus that the climate crisis will continue to be disastrous for human and nonhuman life (Gregory, 2023; Hopkins, 2019; Pierrehumbert, 2019). Despite the commitment of world leaders to limit warming to 1.5°C (Paris Agreement, 2015), the 2022 report by the Intergovernmental Panel on Climate Change (IPCC) projects warming of between 3.3°C and 5.7°C by the end of the twenty-first century (IPCC, 2022). Noting the effects already occurring at a rise of below 1°C, an increase in the above range will be catastrophic.

While there is strong scientific agreement on the direct causes of global heating, there is less agreement regarding the social factors that underpin it. The ‘common sense’ discourse around climate change tends to place the blame on humanity in general, exemplified by the notion of the Anthropocene Era, a proposed ‘new geological epoch’ (Cronin, 2017, p. 9) reflecting humanity’s impact on Earth’s geology (Laurance, 2019). However, this broad focus flattens the causal factors of the crisis, failing to account for the role of capitalism and the hegemonic ideology of perpetual economic growth (Satgar, 2018). It also erases the role of colonialism in the development of a society based on extractivism, as argued by scholars like Liboiron (2021), who encourage us to interrogate the inherent power relations between coloniality and land. In a similar vein, Mirzoeff conceptualises the Anthropocene as ‘“white” (Euro-American) domination of colonised and enslaved … populations of the world’ (Mirzoeff, 2018, p. 124). There is a direct line between the expansionist project of empire-building and colonialism, built around extracting wealth from colonised and enslaved communities, to our contemporary growth-obsessed society still oriented around extracting wealth from the earth and from the Global South. As our article will show, many discourses of climate change in the realm of contemporary South African politics are driven by hegemonic stories of consequence-free economic growth.

Context and research aims

South Africa is the highest greenhouse gas emitter in Africa, and the 12th highest worldwide (Global Carbon Atlas, 2020). This is largely owing to the centrality of coal to the country’s export economy and as an energy source at home (Williams, 2018). South Africa is the biggest producer and exporter of coal in the Sub-Saharan region, and the 5th largest consumer of it globally (Adebayo et al., 2021). Nevertheless, in addition to abundant coal, South Africa also has an abundance of potential renewable energy sources (Williams, 2018). This makes it a prime location for a transition to an economy based on renewable energy, though this must be done with the utmost care to provide employment in the renewable energy sector to those currently employed in mining and related industries (Ashley, 2018). However, although members of the government frequently make references to the goal of a ‘just transition’, namely the transition to renewable energy sources without causing the widespread loss of employment and income, action on this front has been minimal.

The centrality of coal as an energy source in South Africa has never been more evident as it has in the past 15 years. The country has been experiencing periodic rolling blackouts of electricity supply known as ‘load shedding’ since 2007. This has impacted the country’s central and essential processes in complex ways, from disruptions to mining, railway and harbour operations, to the failure of water and sanitation systems due to outages. In addition, and perhaps more pertinent to this study, are the economic ramifications. In a study reporting on the economic cost of load shedding between 2007 and 2019, Walsh et al. (2021) estimated that load shedding cost South Africa R35 billion. It is in this context that the discourses around renewable energy sources are presented. While there is a global need to use clean energy, South Africa is already grappling with a shortage of energy supply. As argued by Van Niekerk (2020, p. 132), load shedding is ‘the visible tip of a deep crisis in the South African electricity system’. Indeed, South Africa’s reliance on fossil fuels that are both environmentally destructive and toxic to human health is at the heart of its energy shortage crisis as well as its high greenhouse gas emissions (Hanto et al., 2022; Van Niekerk, 2020).

As such, this study aims to investigate the fossil fuel discourse of the South African government, to uncover how it construes fossil fuels and the ‘stories’ it tells about them (Stibbe, 2015). This will shed light on the discursive and ideological scaffolding that supports the country’s destructive love affair with fossil fuels.

It is in this vein that our paper endeavours to answer one major question:

1. What stories are being told by the South African government about the continued use of fossil fuels and what are their potential ecological impacts?

The section that follows offers a definition of ecolinguistics and its relation to discourse studies. In this section, we also outline the ecosophy of this paper. We then present an ecolinguistic lens to critical discourse analysis, focusing on Arran Stibbe’s (2015) concept of ‘stories-we-live-by’ as a technical term for the myths and ideologies that influence people’s perception of the natural world. What follows is a detailed methodology of how the research was conducted. Here we discuss the selection and collection of data as well as how the data were processed and analysed. Following this, we present a discussion of the data from a corpus-assisted critical discourse perspective, and finally we draw conclusions to the paper.

Ecolinguistics

This study is theoretically and methodologically informed by the growing field of ecolinguistics (Penz & Fill, 2022; Poole, 2022; Stibbe, 2015; Zhou, 2022). Ecolinguistics draws on a variety of linguistic theories and analytical methods, though it is generally considered a branch of critical discourse analysis. However, where CDA’s focus is on demystifying unequal social relationships and power through critical analysis of texts, ecolinguistics broadens its scope to systems of power that incorporate the natural world. This conception is strongly informed by posthumanist approaches to linguistics and the social sciences (Fox & Alldred, 2020; Kinefuchi, 2018; Pennycook, 2018). Most significantly it is concerned with the interrelationship of human society with the natural world. Human ideologies and discursive practices have ramifications for nonhuman species and ecosystems, and the natural world also influences human culture (Norton & Hulme, 2019). Moreover, the ways in which our society is unequally structured is closely tied to the causal factors and impacts of the climate crisis (Davis et al., 2019; Mirzoeff, 2018; Ritchie & Poser, 2020). Thus, ecolinguistic analysis does not discount the social, but rather sees humans and their discursive practices as situated in an ecological as well as social context.

Fossil fuel discourse

Despite the central role that fossil fuels play in perpetuating the climate crisis, there is a dearth of research on fossil fuel discourse that uses an explicitly ecolinguistic approach. However, there has been some scholarship in this arena utilising a more traditional critical discourse analytic approach. For example, Megura and Gunderson (2022) analysed the corporate responsibility reports of major fossil fuel companies, uncovering naturalisation strategies that constructed them as essential to human life and economic growth and presented technological innovation as the cure to the climate crisis. Wright et al. (2022) found similar legitimising narratives among Australian industry stakeholders, as well as discourses that posed decarbonisation as a threat to economic prosperity and that projected a ‘fossil fuels forever imaginary’ (Wright et al., 2022). Si et al. (2023) examined the discourse of fossil fuel companies via their Twitter posts which revealed a pattern of mentioning natural gas in tandem with renewables in order to position it as equivalently ‘green’. The authors also noted that the use of constructions such as ‘cleaner’ rather than ‘clean’ and ‘low carbon economy’ perpetuates nebulous understandings about the risks and benefits of fossil fuels (Si et al., 2023). Interestingly, in comparison with corporate and media discourse, there has been comparatively little attention paid to the fossil fuel discourse of politicians.

Although political discourse has long been a key area of investigation for CDA research, within ecolinguistics it has not received as much attention, though this is beginning to change (see Al Shboul, 2020; Cunningham et al., 2022; Laurie, 2022). In addition, some CDA scholars have examined political discourse on climate-related issues without an explicitly ecolinguistic approach (Govender, 2015; Kukkonen & Yla-Antilla, 2020). However, there is clearly a need for more ecolinguistically-oriented research on political discourse, particularly from the global South (Demeter, 2020). If there has been little ecolinguistic analysis of political discourse, there has been even less attention paid to the fossil fuel discourse of politicians in the Global South. Furthermore, much of the existing research on fossil fuel discourse comes from a public policy or sociology perspective, which suggests that a robustly (eco)linguistic approach would be worthwhile.

The ecosophy of this paper

The notion of an ‘ecosophy’ or ecological philosophy is key to ecolinguistic analysis. Each analyst must develop their own ecosophy that reflects their positions regarding the interrelationships of people with the nonhuman world and informs their critical analyses. Broadly, the ecosophy of this paper places value on protecting the ‘systems that humans and other forms of life depend on for their well being and survival’ (Stibbe, 2015, p. 9). More specifically, it values long-term resilience of people and other organisms and active care for the ecosystems that support life over short-term gains (Laurie, 2022). Thus, stories that position the land and the systems of life that exist upon it as resources for human extraction will be evaluated negatively. In contrast, discourses that encourage significantly scaling back practices that harm ecosystems will be evaluated positively. In addition, social justice and equity is an important element of this paper’s ecosophy. In this vein, our ecosophy values wellbeing over wealth accumulation and protection of life-supporting ecosystems over exploitation.

The stories we live by: a corpus-assisted Critical Discourse Analysis

Although the ‘eco’ prefix in ‘ecolinguistics’ has been interpreted variously to refer to natural, symbolic, sociocultural or cognitive ecology (Li et al., 2020), it is now most commonly understood in reference to natural ecology (Fernandez-Vasquez, 2021; Norton & Hulme, 2019; Poole & Spangler, 2020; Stibbe, 2015; Zahoor & Janjua, 2020). Contemporary ecolinguistic research thus examines the potential impacts of discursive processes on the natural world. It does not limit itself to discourse on topics related to climate and ecology, but rather on the implicit ‘stories’ communicated through text that may have ecological effects. ‘Story’ is used here as a technical term, referring to the myths, ideologies, frames, and metaphors that exist in the minds of individuals and influence how they perceive the world (Stibbe, 2015, p. 6). When stories become widely communicated in public discourse, they have the potential to become ‘stories-we-live-by’ (Stibbe, 2015, p. 6), or entrenched mental models shared between members of a society.

Stories-we-live-by are thus the key dialectical link between discourse and its effects on social and ecological structures. In other words, stories as we find them in texts can have effects on the natural world, by encouraging either ecologically destructive or beneficial ways of thinking and behaving. As such, when analysing a text, ecolinguists are concerned with determining whether the stories it communicates are ecologically destructive, beneficial, or ambivalent (Stibbe, 2015). The researcher does so by evaluating the stories against their own ecosophy, as we have outlined above.

Within ecolinguistics and CDA as a whole, it has become increasingly common to use a corpus-assisted approach to data collection and analysis (see Bernard, 2015; Dayrell, 2019; Grundman & Krishnamurthy, 2010; Poole, 2022; Willis, 2017). Using corpus-assisted analysis makes it possible to investigate much larger quantities of data (Subtirelu & Baker, 2018). Moreover, it can provide information that the analyst may not notice through their own impressions of the data alone, such as the comparative frequency of lexical items and their collocational environments. Nevertheless, in a study that is corpus-assisted rather than corpus-based, the corpus analysis plays more of a supporting role in guiding the qualitative CDA (see Bernard, 2015 and Willis, 2017).

Our analysis embarks from the foundation of critical discourse analysis as inherently politically oriented, and with the knowledge that discourse, particularly that of politicians, can be used to manipulate and obscure (Van Dijk, 2006). In this mind, we seek to uncover the hidden ideologies and obscured truths within the national government’s communications about the use of fossil fuels. In other words, what are the stories being told around the use of fossil fuels within the context of the climate crisis? In what ways do these stories naturalise, obscure, obfuscate, or perpetuate the destructive and continued use of fossil fuels?

Materials and methods

Data collection

The data for this paper consists of public-facing texts from the South African national government for the period April 2018–2021. This period begins when Cyril Ramaphosa took over from Jacob Zuma as president, and ends soon after the Virtual Leaders’ Summit on Climate Change that president Ramaphosa took part in. April 2021 was also close to the time that we began data collection. This data was collected via the South African government website and the website for the Department of Environment, Forestry and Fisheries. As the data for this study comes from a larger study on the climate change discourse of South African politicians, we used the search term ‘climate change’ to find texts. The texts were then included or excluded based on whether they fell into the relevant time period and whether primary recipients were the general public. To determine this, we utilised Reisigl and Wodak’s (2016, p. 29) ‘fields of political action’. These fields are defined by different functions of discursive practices in the political sphere. For this study, only texts that belonged to ‘public-oriented’ political fields, including formation of public attitudes, opinion and will; organisation of international/interstate relations, and political advertising were included. After this selection process, the final corpus consisted of 199 texts in a wide range of text genres and 228,386 tokens. See Table 1 for a full inventory of text genres.

Table 1. Overview of all texts and text categories in the corpus.

Text categories	Number of texts
Blogs	5
International speeches	26
Local speeches	52
Media advisories	21
Media statements	89
Statements by politicians	3
Informational pamphlets	3

Data analysis

The corpus analysis software AntConc (Anthony, 2020) was used to perform a collocation analysis on researcher-selected fossil fuel-related words. As our focus was the construction of fossil fuel extraction and use, the words energy, coal, and gas were chosen for analysis. Coal still accounts for close to 85% of the electricity used in South Africa (Ember, 2023), while explorations for gas, both onshore and offshore, have been frequent in recent years. We initially included oil in our analysis, but ultimately excluded it due to limitations of space and because there were few instances of it in the corpus. The collocation analysis allowed us to uncover each instance of these word types and the environments in which they occurred. We utilised a collocational window of 3L to 3R (three words to the left and right of the lexical item under analysis) and recorded the most frequent collocations. 3L-3R is the default collocational window on the version of AncConc used, namely AntConc359, and it was deemed sufficient for our purposes. It allowed us to see which words occurred immediately before and after the lexical items, as well as their extended lexical environments.

To select texts for in-depth analysis, we used another AntConc tool, namely File View (Anthony, 2020), which shows the number of tokens of each word type in each text. The texts with the most tokens of each word type were preselected for qualitative analysis. We then briefly examined the preselected texts to remove duplicates and to determine whether its uses of the word in question referred to fossil fuels. For example, the word gas was frequently used to refer to greenhouse gas emissions rather than gas as a fuel source. This final selection resulted in 15 texts to be subjected to in-depth analysis. Table 2 illustrates the word types analysed as well as how many tokens of each were in the corpus and how many texts they appeared in, and Table 3 presents an overview of the texts chosen for qualitative analysis.

Table 2. The word types under analysis and their number of tokens and texts appeared in.

Fossil fuel word types	No. of tokens	No. of texts
Energy	1001	83
Gas	265	60
Coal	118	26

Table 3. Texts for qualitative analysis and the number of tokens for each word type they include.

Texts chosen for qualitative analysis	Energy	Gas	Coal
Minister Gwede Mantashe Approval of Integrated Resource Plan	28	13	9
Minister Mmamoloko Kubayi-Ngubane Mineral Resources and Energy Dept Budget Vote 2020–2021	35	15	6
Minister Jeff Redebe Energy efficiency walk about in-store activation	44	0	0
Minister Jeff Radebe Women in Energy dialogue	58	2	0
Minister Jeff Radebe South African National Energy Development Institute Energy Breakfast	65	0	0
Minister Jeff Radebe South Africa Renewable Energy and Energy Storage Systems Conference	58	2	2
Minister Jeff Radebe SADC Ministerial Workshop on Regional Gas Infrastructure and Market Development	21	75	3
Minister Jeff Radebe Renewable Energy Independent Power Producer Programme Bid Windows 3.5 and 4 contractual close signing ceremony	41	1	1
Minister Jeff Radebe Renewable Energy Independent Power Producer Procurement programme	78	1	24
Minister Jeff Radebe Links between energy and mining sectors	46	3	10
Minister Jeff Radebe Launch of Black Energy Professionals Association	35	17	1
Minister Jeff Radebe Energy Dept Budget Vote 2018–2019	39	32	2
Minister Gwede Mantashe Approval of Integrated Resource Plan 2019	28	13	9
From the desk of the President 28 September 2020	23	2	4
Environment, Forestry and Fisheries publishes Geographical areas for development	12	7	1

Results and discussion

Collocation analysis

Energy

The word type ‘energy’ had 1001 tokens across the corpus, making it the most frequent of the words under analysis. It was also the 6th most frequent content word in the corpus and occurred in 83 out of the 199 texts that made up the corpus. Of these occurrences, 206 occurred in the phrase ‘renewable energy’, 74 in the phrase ‘energy sector/s’, 41 in the phrase ‘energy efficiency’, 25 in the phrase ‘energy storage’, 24 in the phrase ‘energy mix’ and 21 in the phrase ‘energy technology/ies’. Table 4 illustrates the most frequent collocates of the word ‘energy’.

Table 4. Most frequent collocates for ‘energy'.

Collocational phrases for ‘energy'	No. of occurrences
Renewable energy	206
Energy sector/s	74
Energy efficiency	41
Energy storage	25
Energy mix	24
Energy technology/ies	21
Energy supply	19
National energy	18
Department of Energy	18
Energy system/s	18

This shows that the topic of renewable energy is a pertinent one in the national government’s discussions of climate change, pointing to a recognition of the need to expand renewables as a source of energy and their crucial role in mitigating the climate crisis. Across the texts, ‘Renewable Energy’ is often capitalised in the phrases ‘Renewable Energy Development Zones’ and the ‘Renewable Energy Independent Power Producer Programme’. There is thus an emphasis on renewable energy and projects to expand it, albeit ones which are strongly situated within a market-based approach (Muller & Claar, 2021).

However, despite the emphasis on renewable energy, the phrase ‘energy mix’, which occurs 24 times in the corpus, indicates a reluctance to move away from coal entirely. It is clear from reading the extended environments in which this phrase occurs that the energy mix is seen to comprise a range of energy sources, including renewable energy and coal, though the proportions of each are not specified.

Gas

The word type ‘gas’ occurred 265 times in the corpus across 60 texts. Its most frequent collocate was ‘greenhouse gas’, which occurred 72 times but was not considered relevant for analysis as this paper is concerned with gas as a fossil fuel rather than greenhouse gas emissions. ‘Natural gas’ was also a frequent collocate, with 30 occurrences, followed by ‘shale gas’ with 13 occurrences, ‘gas reserves’ with 9 occurrences and ‘gas pipeline/s’ with 8 occurrences. These collocates are summarised in Table 5.

Table 5. Most frequent collocates for ‘gas'.

Collocational phrases for ‘gas'	No. of occurrences
Greenhouse gas	72
Natural gas	30
Shale gas	13
Gas reserves	9
Gas pipeline/s	8
Gas market/s	7
Regional gas	7
Gas to power	6
Oil and gas	6
Use of gas	5

The two most frequent collocates for gas as a fossil fuel refer to different types of gas, namely natural gas and shale gas. The next most frequent collocate, ‘gas reserves’ refers to South Africa’s unmined stores of gas. Referring to the gas potentially located under the earth as ‘reserves’ suggests that they exist to be used and are available for extraction. ‘Gas pipeline/s’ then refers to ways in which such gas, once accessed, could be extracted and transported, and ‘gas markets’ to the buying and selling and economic value of this gas. This suggests a positive attitude towards gas and gas extraction as a source of economic enrichment, but a more in-depth investigation is required to take into account the wider textual context of these phrases. This will be explored in the qualitative analysis section.

Coal

The word type ‘coal’ had 118 tokens in the corpus, and was used across 26 texts. Its most frequent collocate was the compound ‘coal-fired’, which occurred 14 times. Other relatively frequent collocates included ‘from coal’, which occurred 5 times, as well as ‘coal power’ and ‘coal technology/ies’, which both occurred 4 times. There were many more phrases that occurred three times in the corpus, which are illustrated in Table 6. Overall, the collocates for ‘coal’ were more distributed than those for ‘energy’ and ‘gas’, with many that only occurred once or a few times.

Table 6. Most frequent collocates for ‘coal'.

Collocational phrases for ‘coal'	No. of occurrences
Coal-fired	14
From coal	5
Coal power	4
Coal technology/ies	4
Coal bed methane	3
Coal for energy	3
Clean/er coal	3
Coal resources	3

Considering the fact that coal currently provides the lion’s share of South Africa’s electricity and is a major economic export, it is intriguing that ‘coal’ occurs the least frequently of the three lexical items under investigation. Indeed, the phrase ‘renewable energy’ occurs much more frequently across the corpus than the word ‘coal’. However, this should not be taken at face value to mean that the national government values renewable energy more than coal. Rather, the scarcity of explicit mentions draws the audience’s attention away from coal and consequently from the fact that it is still the largest supplier of energy in the country. In contrast, renewable energy is much more frequently signposted, despite making up a minimal proportion of the country’s energy supply at the time of data collection. As argued by Stibbe (2015, p. 188), what is absent from a text is equally relevant for analysis as what is present. Coal is almost universally recognised as an extremely environmentally harmful source of energy, and it is thus possibly harder to present it in a positive light than gas, which is widely considered a ‘cleaner’ fuel source. Once again, this is a matter for further investigation and in-depth analysis, which follows in the next section.

Stories revealed through qualitative analysis

For the purposes of this paper, the word frequency and collocation analysis served as a basis for selecting texts for qualitative CDA. Critical analysis revealed four stories about fossil fuels in the government corpus, namely (1) Fossil fuels are a part of life, (2) Fossil fuels are a wonderful gift, (3) Fossil fuels can be environmentally friendly, and (4) Technology can save fossil fuels (and the climate). These stories are constructed through the discursive strategies of naturalisation and mystification, as well as framing, euphemism and obfuscation.

Story 1: fossil fuels are a part of life

One of the key stories operating in almost every discussion of fossil fuels examined was the story that fossil fuels are a part of life. This story serves to naturalise the continued use of fossil fuels by constructing their extraction and use as inevitable and eternal (Megura & Gunderson, 2022), and is often exhibited via strategies of mystification and obfuscation. This is evident in the phrase ‘energy mix’ which was revealed in the collocation analysis and refers to an energy system consisting of a mixture of fossil fuels and renewable energy sources. On the one hand this seems to promote the use of renewable energy, but by subsuming all energy sources into one ‘mix’, it implicitly naturalises the continued use of fossil fuels.

1. As a department we have been categorically clear from the onset that ours is not to be a lobby group for a particular energy technology, but rather to execute our mandate of ensuring security of energy supply, using all available resources. Now that the energy mix has been outlined, we must work with the necessary speed and resolve to ensure its implementation. (Minister Gwede Mantashe – Approval of Integrated Resource Plan 2019)

2. South Africa recognises the role of all forms of energy in ensuring security of energy supply and meeting the challenge of climate change. We promote an energy mix of coal, gas, renewables and nuclear. (Minister Jeff Radebe – Launch of Black Energy Professionals Association)

While the notion of a ‘mix’ evokes ideas of balance and fairness, there is seldom any specificity regarding how much of the ‘energy mix’ will come from coal, gas, solar, wind or other energy sources. Although text 2 lists the components of the proposed energy mix, there is no mention of what proportion of the mix each component will supply. Instead, the components are simply listed, creating a false sense of equivalence between each one. More frequently, the components are not mentioned and instead all subsumed into a single mass, thereby obfuscating the extent to which the country’s energy system is still based on fossil fuels. Moreover, mentions of ‘the energy mix’ usually take the form of a presupposition, as the use of the definite article or similar determiner presents it as a given. Presuppositions are a key naturalisation strategy when used to present an idea the author wants the audience to accept as if it is already a common assumption, thus helping it become so (Fairclough, 1989). There is thus a dual discursive strategy of naturalisation and mystification, as the use of coal is normalised but the extent of its use is obscured.

In text 1, the evocation of fairness displayed by the term ‘energy mix’ is emphasised by the assertion that the Department of Energy should not be ‘a lobby group for a particular energy technology’. While the impression given is one of fairness and a commitment to ensuring energy access to South African citizens, the urgent need to move away from fossil fuels is obscured. Instead of mentioning any specific energy technologies, each one is presented as equivalent as all are subsumed into the same category. Mu et al. (2023) puts forward that mystification can be achieved through ambiguity, nominalisation, and generalisation in discourse. In this extract, Gwede Mantashe utilises this discursive strategy by mentioning a category but leaving the audience mystified as to what the items within the category are. The text further continues to argue that securing energy supply will involve using ‘all available resources’, but does not specify these resources. This creates a mystifying ambiguity regarding which resources will be used, while covertly endorsing the continued use of fossil fuels like coal and gas, as these are undeniably available resources. Thus, through projecting an image of apolitical benevolence, the continued use of fossil fuels is construed as neutral (Kraushaar-Friesen & Busch, 2020).

The shift between active and passive voice and the strategic implementation of agency in text 1 is also worth noting. The first sentence, which refers to a commitment to ensure a secure energy supply, is entirely in active voice with ‘we’ in the agent position. It is also made clear that ‘we’ refers to the ‘department’ [of energy]. The second sentence switches to passive voice to refer to the outlining of the energy mix in the Integrated Resource Plan, and omits the agent entirely. The text then switches back to active voice when speaking about the plan’s implementation, using the same ‘we’ in agent position. There is thus no agency assigned to any human actors for crafting the country’s energy policies, which is a form of mystification as it obscures the role and identity of the people responsible for making the policy (Hart, 2020). Cunningham et al. (2022) discovered a similar strategy of devolving responsibility onto inanimate actors such as acts and commissions in their study on the climate change discourse of activists and politicians in the UK. Nevertheless, the strategic use of active voice creates a sense of proactiveness, which is reinforced by the phrase ‘execute our mandate’ and the modal ‘must’. In this way, the text creates an impression of committed action when it comes to supplying energy and implementing plans, while mystifying the role of human actors in creating the policies within the plan itself. The above extracts illustrate how strategies of naturalisation and mystification can work together to appease the dissonance between two narratives: namely that fossil fuels are causing the climate crisis and that they are a natural part of life.

Story 2: fossil fuels are a wonderful gift

In several instances, fossil fuels are not merely positioned as a natural part of the country’s energy supply, but as a bountiful gift. This was particularly evident when investigating the extended textual contexts surrounding the words gas and coal that were revealed by the collocation analysis.

• (3) South Africa has an abundance of coal resources which plays a major role in the South African energy mix. Over 75 percent of energy used in South Africa is generated from coal. The use of coal for energy generation is underpinned by availability, accessibility, reliability and affordability to South Africans. (Minister Jeff Radebe – Links between energy and mining sectors)

• (4) Abundant natural gas resources and rising production which includes supplies of tight gas, shale gas, and coal bed methane contributes to the strong competitive position of natural gas. Recent discoveries of new supplies of gas, the remarkable speed and scale of shale gas development and decreasing gas prices below expectations has heightened the awareness of natural gas as a key component of energy supply mix in a carbon-constrained world economy. (Minister Jeff Radebe – SADC Ministerial Workshop on Regional Gas Infrastructure and Market Development)

The word ‘abundant’ or ‘abundance’ is sometimes used to describe both coal and gas resources, as seen in texts 3 and 4. This positive framing constructs an image of bounty and a sense that the country is lucky to have these resources. In addition, there is a clear, though implicit, argument here that if we have fossil fuels, we must use them (Kraushaar-Friesen & Busch, 2020). This reinforces the construction of fossil fuels as an inevitable part of life, and displays a highly positive appraisal pattern (Stibbe, 2015). For example, in text 3, the mention of coal is accompanied by many positively coded lexical items, namely ‘availability, accessibility, reliability and affordability’, which work to create a positive impression surrounding coal and its usefulness to South Africans. The fact that ‘over 75 percent’ of South Africa’s energy comes from coal could be framed negatively, however, it is instead presented as a matter of fact, further naturalising coal by presenting its continued use as commonplace and unchangeable while also promoting a positive image of it as a useful and important fuel source (Wright et al., 2022).

Text 4 exhibits a similarly positive appraisal pattern toward gas, referring to it with phrases like ‘strong competitive position’, ‘new supplies’, ‘remarkable speed and scale’ and ‘decreasing prices below expectations’, in addition to ‘abundant’. These phrases are evidence of a neoliberal-oriented economic framing in which gas is construed as highly beneficial to the South African economy (Kraushaar-Friesen & Busch, 2020). Once again, the fact that shale gas development is proceeding at ‘remarkable speed and scale’ could be framed in a negative light due to its environmental consequences and the likelihood that it will hinder the rollout of renewable energy infrastructure (Costa et al., 2017; McGlade et al., 2018). However, it is framed positively here, thus obfuscating the environmental harms of gas extraction and use. In addition, we should also note the obfuscation of the climate crisis itself and of the need to transition away from fossil fuels in text 4 as this is euphemistically referred to as a ‘carbon constrained world economy’. Yaseen (2017, p. 41) concurs that the use of ‘euphemism is often resorted to in order to mislead listeners and to effect deception’. In this context, it serves to minimise the seriousness of the climate crisis while also reifying the central position of the economy in the way policy decisions are made.

While the collocation analysis suggested a more equivocal construction of coal when compared to the explicitly positive framings of gas, coal is still portrayed uncritically in several instances as a resource to be exploited and profited from, as in texts 3 and 4. Moreover, in the texts selected for in-depth analysis, coal is presented positively or at most neutrally in almost every instance.

Story 3: fossil fuels can be environmentally friendly

In addition to portraying fossil fuels in a generally positive light, there are many cases of fossil fuels being reconstructed as environmentally friendly energy sources.

• (5) With regard to Gas to Power: South Africa has recognised this global shift and has set itself the vision to enter the global gas market and promote the development of a gas market, not only locally here in SA but also in the Southern Africa Region. For emerging economies, switching to gas as a competitive, cleaner and more flexible source for power production is a game changer. (Minister Jeff Radebe – Launch of Black Energy Professionals Association)

• (6) The low carbon emission profile of gas-to-power and its modular configuration provides an opportunity for use of gas for power generation in the region. With an average electrification rate of 35%, the SADC region has an opportunity to develop a cleaner, reliable power generation sector from the use of gas. The gas to power can also support the renewable energy projects in various SADC member countries. (Minister Jeff Radebe – SADC Ministerial Workshop on Regional Gas Infrastructure and Market Development)

• (7) ‘Coal will continue to play a significant role in electricity generation as the country has the resource in abundance. New investments will be directed towards more efficient coal technologies (High Efficiency, Low Emissions), underground coal gasification and the development of Carbon Capture and Storage to enable us to continue using our coal resources in an environmentally responsible way’ (Minister Gwede Mantashe – Approval of Integrated Resource Plan 2019).

In text 5, gas is promoted explicitly and described as a ‘game changer’. The text uses a market framing when discussing gas, reinforcing the primacy of the goal of economic growth and shifting responsibility away from human actors. As in text 4, this falls in line with a neoliberal discourse in which policymakers are compelled to follow the market in every decision (Mercer et al., 2014). However, in addition to being ‘competitive’, gas is also described as a ‘cleaner’ and ‘more flexible’ source of power. The equivocal ‘cleaner’, also seen in text 6, allows the author to create the impression that gas is environmentally friendly, without committing to it being fully ‘clean’ (Si et al., 2023). Text 6 also describes it as having a ‘low carbon emission profile’, thus construing it as environmentally beneficial.

However, these constructions display a strategy of mystification. When gas is described as ‘cleaner’, the implication is that it is better than coal, but there is no clarity on what the parameters are or how much greenhouse gas it emits. Similarly ‘low carbon emissions profile’ suggests that the extraction and use of gas does not involve harmful carbon emissions, but again it is not clear how much ‘lower’ the carbon emissions involved in gas are in comparison to coal or to renewable energies. Indeed, gas is technically ‘cleaner’ than coal in terms of particulate pollution, but it is still an ecologically harmful fossil fuel which does not reduce carbon emissions to the extent required to avert climate catastrophe (Costa et al., 2017). It is certainly not the ‘game changer’ text 5 presents it as. The descriptor ‘flexible’ also contributes to mystification due to its semantic ambiguity. It is not clear what it means for a source of power to be ‘flexible’, beyond a general sense of advantageousness. However, the overall impression is that gas is both economically and environmentally beneficial, which obfuscates its environmentally destructive nature.

Like gas, though admittedly not to the same extent, coal is also sometimes positioned as an environmentally friendly resource. This echoes the findings of Biddau, Rizzoli and Sarrica (2024) in their study on media discourse on fossil fuels in a coal mining region of Italy and the implication of the collocation ‘clean coal’ revealed by the corpus analysis. Given that approximately 40% of global carbon emissions come from coal (Jakob et al., 2020), the fact that its continued use is positioned as not incompatible with efforts to tackle climate change is noteworthy. In text 7, coal is again presented as ‘abundant’, and this is presented as the reason for its continued role in electricity generation. Like in texts 3 and 4, the implicit assumption is that any resource the country has must be fully utilised regardless of its consequences.

As well as being abundant, coal is also constructed in text 7 as potentially environmentally friendly, as ‘High Efficiency, Low Emissions’ coal exists and it may be used ‘in an environmentally responsible way’. Thus, coal is reconstructed as a ‘green’ fuel, as long as it is used ‘responsibly’. This is highly misleading, but avoids being explicitly deceptive as these ‘more efficient coal technologies’ are not claimed to be in use but are positioned as having ‘new investments directed’ towards them. However, the text creates the impression that these technologies are reliable and able to sufficiently mitigate the negative impacts of carbon emissions from coal, which most climate scientists agree is not the case (Asayama & Ishii, 2017). Moreover, the use of the future construction downplays any urgent need to curb emissions from coal, thereby obscuring the seriousness of the climate crisis.

Story 4: technology will save fossil fuels (and the climate)

There is a strong techno-optimist thread running through the data, which was hinted at by the frequency of the phrase ‘coal technology/ies’ revealed by the collocation analysis.

• (8) ‘Given the need to increase energy supply in a globally carbon-constrained environment South Africa is investing in the development of clean coal technologies such as: – carbon capture and storage, Coal Fluidized Bed Circulation combustion, integrated gasification combined-cycle plant, underground coal gasification and ultra-super critical technologies’ (Minister Jeff Radebe – Links between energy and mining sectors).

• (9) Battery energy storage is undergoing particularly rapid techno-economic advancements with costs declining significantly. The large-scale uptake of battery energy storage would accelerate the stable implementation of cost-effective renewable energy technologies, permit the decentralisation, modernisation and digitalisation of the electric grid, and enable improved electrical power system flexibility, security and affordability. (Minister Jeff Radebe – South Africa Renewable Energy and Energy Storage Systems Conference)

• (10) The South African National Energy Development Institute (SANEDI) deemed it fit that together with the Department of Energy, they should present possible collaborative partnerships in the energy efficiency and renewable energy space, that can establish win-win cleaner technology solutions into mining value chain after thorough consideration of the mining production processes; operations and maintenance; research and development; technology growth, development and advancement in contributing towards a low carbon economy in the short, medium and long terms. (Minister Jeff Radebe – South African National Energy Development Institute Energy Breakfast)

Text 8 is an example of the techno-optimist thread which presents technological innovation as a solution to the climate crisis and as a means to continue burning fossil fuels, particularly coal, indefinitely (Megura & Gunderson, 2022). Phrases like ‘High Efficiency, Low Emissions’ and ‘underground coal gasification’ are used to explicitly argue that technological innovation will mean that coal can continue to be used in the future. Their jargon-like nature is also part of a discursive strategy of mystification, as they create an impression of expertise that most members of the public are not able to verify (Brombal et al., 2023). They also evoke a sense of almost futuristic technological innovation which feeds into techno-utopian imaginaries. This mystifying effect is heightened even further in text 8, with incomprehensible yet excitingly high-tech sounding phrases like ‘Coal Fluidized Bed Circulation combustion’, ‘integrated gasification combined-cycle plant’, and ‘ultra-super critical technologies’.

Texts 9 and 10, while less overtly techno-utopian, also link technology with climate solutions in a way that evokes an exciting sense of technological progress and innovation. This is evidenced by phrases like ‘rapid techno-economic advancements’, ‘decentralisation, modernisation and digitalisation’, ‘win-win cleaner technology solutions’ and ‘technology growth, development and advancement’. These excerpts also illustrate a discursive link, not only between technology and sustainability, but also between technology, sustainability and economic growth. This is conspicuous in the phrase ‘techno-economic advancements’ and also evidenced by mentions of the ‘mining value chain’. The notion of ‘win-win cleaner technology solutions’ also implies that technological developments can lead to both sustainability and wealth. These discursive linkages serve to evoke idealised imaginaries of techno- and eco-futures in which we can have it all (Jochum, 2020, p. 8). This is highly misleading, as the technology we currently have is not nearly sufficient to avert climate catastrophe without drastic cuts in carbon emissions. Moreover, the conflation of technological innovation with economic growth reveals a conception of technological development in service of accumulation and consumption (Acosta & Abarca, 2018). This approach is not consistent with the protection of earth’s finite resources and the planetary systems that support life.

Conclusion

In many of the texts discussed above, we have seen the use of a neoliberal economic framing to legitimise the continued use of fossil fuels. This is part of a prevalent and dangerous story that economic growth is the single most important objective to which every other goal, including ensuring a livable climate, is necessarily subordinate (Stibbe, 2015). The implicit naturalisation of the primacy of economic growth is evident across the data, and an underlying discursive thread that fossil fuels are necessary to achieve it, both nationally and globally, is also apparent. The economy is thus metaphorically constructed as a fossil fuel-powered machine that must continue running at all costs, which serves to naturalise its position as the central force around which decision-making revolves.

Naturalisation is not only evident in the way fossil fuels are linked to the economy, but also in how they are constructed as a part of life. The notion of fossil fuel use as inevitable and fossil fuels as abundant resources that will continue to be drawn on, is a prudent example of how discourses can obscure ‘the fact that “the way things are” is not inevitable or unchangeable’ (Cameron, 2001, p. 123). In addition to this, fossil fuels are perpetually constructed positively, as beneficial endowments that we are privileged to have access to. These potentially fatal stories not only show how discourse is politically driven, but how it can also perpetuate neoliberal capitalist, extractivist ideologies that prioritise economic growth over a sustained ecology. While the texts in our analysis obscure this by constructing an idealised future in which both are possible, they simultaneously naturalise the goal of economic growth as more important than the goal of a livable climate.

In conclusion, through an ecolinguistically-informed CDA, this paper has shown that the South African government utilises discursive tactics to construct the use of fossil fuels as inevitable and beneficial, to erase their negative impacts, and to create false dichotomies in which they can be ‘green’ energy sources – in other words, to discursively maintain a fossil fuel hegemony (Kraushaar-Friesen & Busch, 2020). In addition, the use of jargon, euphemism and mystification as naturalisation strategies echo Willis (2017) argument that politicians ‘do not respond to scientific evidence on climate change in a simple or linear way … instead, whether consciously or not, shape an issue to fit with their ideology’ (p. 214).

We identified four key stories the South African government tells about fossils fuels, namely fossil fuels are a part of life, fossil fuels are a wonderful gift, fossil fuels can be environmentally friendly and technology will save fossil fuels (and the climate). These stories are, as per our ecosophy, ecologically destructive. The reality of climate change is that it is no longer an imminent cautionary phenomenon; we are unequivocally in the midst of a climate emergency, so any stories that perpetuate their use are harmful to human and nonhuman life and must be challenged.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by National Research Foundation.

Notes on contributors

Julia Laurie

Julia Laurie is a PhD student and ad hoc lecturer in the Department of African Studies and Linguistics at the University of Cape Town. Her PhD research focuses on discourses of load shedding in the context of the climate emergency and energy inequality in South Africa. Her research interests include ecolinguistics, critical discourse studies and digitally mediated communication.

Miché Thompson

Miché Thompson is a senior lecturer in linguistics at the Department of African Studies and Linguistics at the University of Cape Town, South Africa. Her current area of research focuses on critical race theory and the politics of identity in South Africa, in which she critically engages with categories of race and belonging in post-Apartheid South Africa.

Other key areas of research are broadly within the sociolinguistics of migration. This research focuses on the Chinese diaspora in South Africa in the economic domain, while also drawing on recent debates around China's political engagement in South Africa.