2. Definitions

To begin with, it is important to establish the main categories and definitions that we use in this article (Table 1). We define intent as the judgement of whether or not an actor is purposefully spreading climate misinformation, based on their discourse in context. This follows the main objective of our article, which is to develop criteria for informed judgements of intent, while also pointing to the main information that should be analysed: actors and their discourse. Importantly, we agree with Ramaswami et al. (Reference Ramaswami, Brunnermeier, Jenkins, Lane, Oppenheimer, Pacala and Ewalt2022) that intent should be judged from observable actions, affiliations, or behaviour and is thereby different from motive (the reasons for taking action), which is an internal state that cannot be observed. We acknowledge that definitive conclusions are difficult to reach on intent. It is therefore useful to consider it rather in terms of confidence (e.g. highly likely, more likely, and unlikely) as derived from one’s analysis.

Table 1.

Categories and definitions used in this article

Table 1 lists further categories and definitions that are related to intent. We consider misinformation in a broad sense, capturing not just false information but also the use of misleading information and rhetorical techniques such as cherry picking evidence or oversimplification (see e.g. Cook et al., Reference Cook, Ellerton and Kinkead2018), insofar as these can lead an audience to false conclusions. These definitions are specific to our review, but are informed by the wider fields of climate politics and climate policy discourse studies, which are admittedly not always consistent in terms of underlying epistemology and terminology.

3. How have discourse studies on climate mitigation evaluated intent?

In this section, we review how prior work has evaluated intent in climate policy-focused discourse analysis. To obtain a relevant sample, we screened the complete set of articles citing Lamb et al. (Reference Lamb, Mattioli, Levi, Roberts, Minx, Müller-hansen and Steinberger2020) and Coan et al. (Reference Coan, Boussalis, Cook and Nanko2021) up to March 2024 in Google Scholar. We then reduced this set to articles that applied a stated form of discourse analysis on subjects related to climate change policy or mitigation. It should be noted that not all of these studies directly apply the frameworks of Lamb et al. (Reference Lamb, Mattioli, Levi, Roberts, Minx, Müller-hansen and Steinberger2020) or Coan et al. (Reference Coan, Boussalis, Cook and Nanko2021), but instead cite these works to situate themselves in the field of climate policy discourse studies. We then conducted an interpretative review, focusing on how authors either implicitly or explicitly evaluate intent. That is, we started by looking at the methods they apply and the actors they chose to focus on, both of which may carry implicit notions of intent, before summarising the relevant categories and approaches in studies that more directly address this issue.

Most articles in our sample utilise techniques such as discourse analysis (Chen, Reference Chen2023; Christiansen et al., Reference Christiansen, Hajdu, Planting Mollaoglu, Andrews, Carton and Fischer2023; Hanson, Reference Hanson2024; Kukkonen & Malkamäki, Reference Kukkonen and Malkamäki2024; Lin, Reference Lin2024; Lowes et al., Reference Lowes, Woodman and Speirs2020; Markard et al., Reference Markard, Rinscheid and Widdel2021; Neubauer et al., Reference Neubauer, Graham and Krobath2023; Ohlendorf et al., Reference Ohlendorf, Löhr and Markard2023; Trimmel et al., Reference Trimmel, Kriechbaum, Lazou and Brudermann2024; Van Ryneveld & Islar, Reference Van Ryneveld and Islar2023; Vowles & Hultman, Reference Vowles and Hultman2021), rhetorical analysis (Sassan et al., Reference Sassan, Mahat, Aronczyk and Brulle2023), or frame analysis (Moreno et al., Reference Moreno, Kinn and Naberhaus2022; Supran & Oreskes, Reference Supran and Oreskes2021) to do the methodological work of untangling climate policy discourses. Hence, the common theoretical basis on which they build their interpretations is that it is possible to discern meaning from narrative, that meaning is constructed by both the author and the audience of a given piece of communication, and that communication is often an attempt by the communicator to convey information in a way that will have an influence on the worldview or opinions of the intended receiver. In other words, these works attempt to move beyond describing what people say, to evaluate what they mean (Hajer, Reference Hajer, van den Brink and Metze2006; Hajer & Versteeg, Reference Hajer and Versteeg2005).

Our first observation is, therefore, that most studies contain at least an implicit evaluation of intent, as captured in their underlying epistemology. Studies have largely adopted qualitative methods, often by categorising content and frames into codes – either drawn from the literature or developed deductively – to identify consistent patterns across text or media sources. Others have taken quantitative approaches or combined them with qualitative analysis, exploring the formation of narratives through the occurrence of specific keywords, or using methods such as topic modelling (Chen, Reference Chen2023; Si et al., Reference Si, Desai, Bozhilova, Puffer and Stephens2023; Supran & Oreskes, Reference Supran and Oreskes2021), cluster analysis (Meyer et al., Reference Meyer, Peach, Guenther, Kedar and Brüggemann2023), or manual counting (Neubauer et al., Reference Neubauer, Graham and Krobath2023; Painter et al., Reference Painter, Ettinger, Holmes, Loy, Pinto, Richardson and Wetts2023).

We find that in terms of actor focus, many studies deal with discourses propagated by incumbent fossil fuel interests, including industry advocacy groups (McKie, Reference McKie and McKie2023, Reference McKie2021; Neubauer et al., Reference Neubauer, Graham and Krobath2023; Sassan et al., Reference Sassan, Mahat, Aronczyk and Brulle2023), oil and gas firms (Kuhl et al., Reference Kuhl, Stephens, Arriaga Serrano, Perez-Lugo, Ortiz-Garcia and Ellis2024; López, Reference López, Fowler-Watt and McDougall2023; Lowes et al., Reference Lowes, Woodman and Speirs2020; Ohlendorf et al., Reference Ohlendorf, Löhr and Markard2023; Si et al., Reference Si, Desai, Bozhilova, Puffer and Stephens2023; Tilsted et al., Reference Tilsted, Mah, Nielsen, Finkill and Bauer2022), and electric utilities (Williams et al., Reference Williams, Bartone, Swanson and Stokes2022). Others focus on ‘aligned’ interests, such as right-wing think tanks (Moreno et al., Reference Moreno, Kinn and Naberhaus2022; Plehwe, Reference Plehwe, Abelson and Rastrick2021; Samoilenko & Cook, Reference Samoilenko and Cook2024) and right-wing populist and far-right political actors (Hanson, Reference Hanson2024; Selk & Kemmerzell, Reference Selk and Kemmerzell2022; Vowles & Hultman, Reference Vowles and Hultman2021). Since these types of actors may have a direct interest in delaying climate policy, intent could be argued to be implicit in the focus of such studies. However, interestingly, a significant set of studies also examines broader media discourses in newspapers, on television, and in social media, where such motives are not as easily traced (Andersen, Reference Andersen, McAllister and West2023; Lin, Reference Lin2024; Meier & Fugl Eskjær, Reference Meier and Fugl Eskjær2023; Meyer et al., Reference Meyer, Peach, Guenther, Kedar and Brüggemann2023; Nielsen, Reference Nielsen2023; Painter et al., Reference Painter, Ettinger, Holmes, Loy, Pinto, Richardson and Wetts2023; Sylla et al., Reference Sylla, Glawe, Braun, Padev, Schäfer, Ahmetaj, Calero Valdez, Spezzano, Amaral, Ceolin, Fazio and Serra2022; Trimmel et al., Reference Trimmel, Kriechbaum, Lazou and Brudermann2024).

Importantly, few studies explicitly discuss intent. This topic is rather dealt with on a spectrum, from studies that tend not to deal with it even indirectly (e.g. descriptive social media analysis reporting on the range of different climate policy discussions that take place), to analyses where the evaluation of intent is implicit but captured in other analytical concepts (e.g. discourse coalition studies of newspaper material, where the political leanings and histories of different actors are traced), to those that make a more direct claim of intent (but may nonetheless refrain from stating it outright). In the latter categories, we observe that most studies focus their attention on the actors themselves, typically evaluating their interests, behaviour, and track records.

4. Characteristics of intent: interest, behaviour, and track record

Interest emerges as a key characteristic of intent in our review, on the basis that someone who actually benefits from misinformation is arguably more likely to spread it (Fallis, Reference Fallis2015). As such, it is unsurprising to find that many studies explore the connections between those propagating climate delay and groups that are materially threatened by measures to address climate change. These connections may consist of employment, funding, shared personnel, or other support received from various parties linked to the fossil fuel industry, including companies (e.g. oil and gas firms), special interest organisations (e.g. industry trade associations), or non-profit organisations (e.g. foundations and think tanks) (Andersen, Reference Andersen, McAllister and West2023; Christiansen et al., Reference Christiansen, Hajdu, Planting Mollaoglu, Andrews, Carton and Fischer2023; Holder et al., Reference Holder, Mirza, Namson-Ngo-Lee, Carbone and McKie2023; Moreno et al., Reference Moreno, Kinn and Naberhaus2022; Plehwe, Reference Plehwe, Abelson and Rastrick2021; Tilsted et al., Reference Tilsted, Mah, Nielsen, Finkill and Bauer2022; Williams et al., Reference Williams, Bartone, Swanson and Stokes2022). On the other hand, we often find that the focus on particular actors or organisations and their interests is not fully elaborated in the discourse literature, for instance, only briefly pointing to connections with the fossil industry, without a deeper explanation of how discourses may be shaped by material alignments.

In other areas of the climate obstruction literature, it has been shown that material interests can intersect with ideological stances, values, and beliefs. Much work, for instance, has gone into examining think tanks such as the Heartland Institute, the Cato Institute, and the Institute for Economic Affairs, which espouse a broader program of neoliberal ideology and limited state intervention that is highly opposed to climate policy and action (Plehwe, Reference Plehwe, Abelson and Rastrick2021; Plehwe et al., Reference Plehwe, Goldenbaum, Ramanujam, McKie, Moreno, Ekberg and Hrubes2021). These and other institutes have been linked to a decades-long effort of breathtaking scope by Charles and David Koch to promulgate libertarian ideas and support conservative political causes in the United States, backed by personal wealth accrued from fossil fuel industries (Bennett & Livingston, Reference Bennett, Livingston, Bennett and Livingston2020; Doreian & Mrvar, Reference Doreian, Mrvar, Tindall, Stoddart and Dunlap2022; Gibson & Brulle, Reference Gibson and Brulle2024; MacLean, Reference MacLean, Bennett and Livingston2020).

A second analytical area concerns behaviour. This consists of concrete acts that may be consistent or inconsistent with the rhetoric or claims of a specific actor, such as financial investments (Li et al., Reference Li, Trencher and Asuka2022), voting patterns in parliaments, or acting as a signatory of a known misinformation document (Spampatti et al., Reference Spampatti, Brosch, Mumenthaler and Hahnel2023). The logic here is that rhetorical opposition to climate policy is a manifestation of actual opposition to climate policy, which can be observed through someone’s behaviour.

One example of how to leverage behaviour in the analysis of intent is to demonstrate an action-rhetoric gap. For example, one could assess whether industry claims with regard to their climate actions are consistent with what we can observe and verify, as in Li et al. (Reference Li, Trencher and Asuka2022)’s comparison of oil major claims versus actions and investments with respect to renewable energy. In a corporate context, this has commonly become known as the analysis of ‘greenwashing’ (Nemes et al., Reference Nemes, Scanlan, Smith, Smith, Aronczyk, Hill and Stabinsky2022). Similarly, one could compare discrepancies between private and public communications, such as when ExxonMobil internally discussed the drivers and risks of climate change, but externally downplayed them (Franta, Reference Franta2018; Supran & Oreskes, Reference Supran and Oreskes2017). The stark difference between these communications suggests an active choice to present information to the public in a certain way. Indeed, perhaps the strongest example here would be an actual statement of intent to mislead, which explains the continuing interest among climate obstruction scholars in archival research of internal communications (Franta, Reference Franta2018).

The focus on behaviour can draw closer attention to the processes of climate discourse production and propagation. Here, it should be acknowledged that fossil interest groups mobilise vast resources on public relations efforts (Brulle et al., Reference Brulle, Aronczyk and Carmichael2020; Brulle & Werthman, Reference Brulle and Werthman2021), including instances where several million dollars have been spent by the American Petroleum Institute on large-audience ‘Super Bowl’ advertisements to greenwash oil and gas products (Sheehan, Reference Sheehan and Rimmer2018). Indeed, interest groups are known to engage in sophisticated chains of activities that produce and propagate discourses, such as funding oppositional science and issuing press releases to the media; establishing ‘grassroots organisations’ that manufacture community opposition to policy proposals; or coordinating industry-wide organisations to produce educational material and deliver training to key stakeholders (Ulucanlar et al., Reference Ulucanlar, Lauber, Fabbri, Hawkins, Mialon, Hancock and Gilmore2023). The scale, reach, and messaging of these campaigns – which are unambiguously supportive of the fossil industry – ought to be a clear measure of intent.

A final characteristic that cuts across the analysis of interest and behaviour is track record. This captures whether or not someone has had connections to fossil interest groups in the past, or has had a history of behaving in ways inconsistent with a desire to mitigate climate change. Such a perspective is strongly embedded in the epistemology of discourse analysis, which aims to situate discourse actors within the wider (historical) context of their actions and networks. It is also reflected in online resources such as DeSmog’s Climate Disinformation Database, where detailed biographies of individuals and organisations that have repeatedly engaged in climate contrarianism over an extensive period of time are gathered.

An interesting aspect of this longer-term perspective is whether actors and organisations that propagated climate denial or obstructive behaviour in the past are now shifting towards climate delay discourses. In other words, has someone been active in propagating climate denial in the past? This shift has been observed at an aggregated level (Coan et al., Reference Coan, Boussalis, Cook and Nanko2021), but only one study in our review traced it for individual actors or organisations (Ekberg & Pressfeldt, Reference Ekberg and Pressfeldt2022). (Conversely, a shift in the other direction – from delay to denial – is of course also plausible, and was identified for one of Ekberg & Pressfeldt’s cases). In our view, explicitly considering such historical contexts and whether such shifts occur is of deep interest to the analysis of intent, even if some public figures have purported to ‘change their minds’ about climate change (Walker, Reference Walker2021).

5. Additional avenues to evaluate intent: density, ambiguity, professional context

So far, our review has primarily focused on actor-based categories that are used to judge intent. But in addition to understanding who is propagating climate discourses, we propose that it may also be useful to observe how arguments are structured and where they may be delivered. We therefore suggest three further categories: density, ambiguity, and professional context.

First, in our experience, certain actors have a tendency of deploying many climate delay claims in fast succession – which we call density. This pattern of argumentation stands in stark contrast to when someone carefully engages with a given normative aspect of climate action, such as whether policies do indeed impact on social equality and ought to be redesigned in a given way. The use of dense argumentation is a known debating technique referred to as Gish Galloping or Firehosing, where many individually weak arguments are used to overwhelm efforts at rebuttal. In our review, we did not identify any studies that observed density or reflected on it when contrasting alternative discursive positions. But given that climate contrarian content is known to combine many different categories together – including denial, impact scepticism, ad hominem, and climate delay claims (Flack et al., Reference Flack, Cook, Ellerton, Kinkead, Coan, Boussalis and Dargaville2024) – we think that there should be closer attention to this phenomenon.

Second, we would argue that it is a fundamental task of those who engage in public discussions on climate change not to leave any ambiguity as to whether we should take action. If one’s intent is not to undermine climate policy action, it is arguably counterproductive to continuously field arguments that attack such policies while failing to offer constructive alternatives. In addition, one always has the opportunity to explicitly clarify intent and therefore prevent others from being misled (Fallis, Reference Fallis2015). Usually, it is not in the interest of political actors to reduce ambiguities, as leaving room for interpretation can help them to evade responsibility or smooth over potential conflicts (Geden, Reference Geden2018). This underlines the important moderating role of journalists and other third parties in challenging political actors on their concrete positions and intentions. Ambiguity can also hide inconsistencies in policy preferences, which two of our articles suggest may be related to underlying obstruction strategies (Hanson, Reference Hanson2024; Vowles & Hultman, Reference Vowles and Hultman2021). For instance, it has been shown that far right party manifestos enthusiastically support local environmental policies (presumably to tap into nativist sentiments), but generally oppose climate change policies, even though the two issues are clearly interlinked in practice (Forchtner, Reference Forchtner2019; Schwoerer & Fernandez-Garcia, Reference Schwoerer and Fernandez-Garcia2024). It is therefore important to observe whether or not actors reduce ambiguity and clarify their positions, and whether these are consistent with the claims being made.

Finally, we would argue that the professional context of climate discussions matters. There is a difference between the everyday conversations of private actors, versus public statements from professional communicators such as politicians, public relations experts, or media commentators. Actors in the latter category have a deep awareness of and agency over their messaging strategies, and sufficient resources that they cannot claim error or ignorance with respect to their likely effects (Ramaswami et al., Reference Ramaswami, Brunnermeier, Jenkins, Lane, Oppenheimer, Pacala and Ewalt2022). Their messaging is, by nature, intended to influence large swathes of the public. In turn, these public messages can be reproduced in private discussions and among social media users, while interacting with and shaping broader cultural phenomena, such as the low salience of climate issues in some countries (Brulle & Norgaard, Reference Brulle and Norgaard2019). These may indeed be the objectives of professional communication efforts. Thus, while climate delay can be undoubtedly present in private conversations, judgements of intent ought to take a more critical stance for those operating in professional contexts where strategic and well-honed messaging is the norm.

6. Comparing approaches to identify and debunk climate delay discourses

The original claim in this article is that an analysis of intent is instrumental to identifying climate delay discourses. Here, we set this discussion in the context of two other approaches that are used in the fields of cognitive psychology and communication studies to classify and counter climate misinformation – namely, fact and logic-based approaches (Table 2).

Table 2.

Fact, logic, and intent-based approaches to identifying misinformation or disinformation

In the fact-based approach, misinformation is interpreted as the spread of claims that are contrary to ‘common understandings of verifiable facts’ (Guess & Lyons, Reference Guess, Lyons, Persily and Tucker2020). As such, scientists and advocates can identify it through fact-checking, and respond by communicating and reinforcing scientific facts on a given topic (Schmid & Betsch, Reference Schmid and Betsch2019). In the logic-based approach, identification rather focuses on rhetorical techniques and strategies. Arguments are deconstructed into their basic premises and conclusions, which can then be interrogated for logical fallacies (Cook, Reference Cook, Holmes and Richardson2020; Cook et al., Reference Cook, Ellerton and Kinkead2018; Schmid & Betsch, Reference Schmid and Betsch2019; Vraga & van der Linden, Reference Vraga, van der Linden, Holmes and Richardson2020). If these fallacies are present, then arguments can be exposed as misleading. The logic-based approach calls for improvements to critical thinking as the basis for countering misinformation.

Compared to these two approaches, the focus on intent leads us to an analysis of the underlying actors, interests, and practices of misinformation production, in addition to the specific content claims that are made. Strictly speaking, the focus on intent already presupposes that misinformation is being spread and is rather concerned with whether it is intentional and can be classified as ‘disinformation’. Well-known examples include the study of propaganda campaigns and denial ‘playbooks’, and research into the spread of ideologically-motivated rhetoric through non-profit organisations and policy circles (Holder et al., Reference Holder, Mirza, Namson-Ngo-Lee, Carbone and McKie2023; Kukkonen & Malkamäki, Reference Kukkonen and Malkamäki2024; Lowes et al., Reference Lowes, Woodman and Speirs2020; McKie, Reference McKie2021; Moreno et al., Reference Moreno, Kinn and Naberhaus2022; Sassan et al., Reference Sassan, Mahat, Aronczyk and Brulle2023). As we have discussed, this approach suggests that mis/disinformation can be identified by observing the link between rhetoric and underlying interests, with a focus on interests, behaviour, and track record. It also implies that misinformation can be countered by rendering these links and activities transparent.

To what extent is the study of climate delay amenable to these different approaches? Arguably, all are relevant and complementary, particularly when one considers their individual weaknesses. For instance, fact-checking presents significant epistemic challenges when it comes to climate delay discourses. This is because policy discussions are often driven by values rather than verifiable facts. They deal with normative questions such as how responsible different parties are for addressing climate change, or they refer to hypothetical future scenarios about how we should reduce emissions. This is an area where different approaches ought to be legitimately discussed, not least because values and preferences for how to reduce greenhouse gas emissions vary markedly across society. Further, as in other fields of social policy, the ‘solutions’ to climate change often come with new challenges that themselves need to be deliberated (Grundmann, Reference Grundmann2016). Still, fact-checking may be useful in cases where the policy option space is constrained due to biophysical or techno-economic reasons. Not all policy positions are possible to hold, given the mounting evidence that certain mitigation options are simply too slow, expensive, or undeveloped compared to others to offer a viable pathway to ambitious climate targets (for instance, in the recurring debate on nuclear versus renewable energy for decarbonising the power sector).

Where claims cannot be straightforwardly fact-checked, the logic-based approach can offer more purchase. Using a taxonomy of misinformation techniques called Fake experts, Logical fallacies, Impossible expectations, Cherry picking, and Conspiracy theories, Cook et al. (Reference Cook, Ellerton and Kinkead2018) contend that a large number of climate denialist claims can be refuted using logical argumentation – an analysis that Flack et al. (Reference Flack, Cook, Ellerton, Kinkead, Coan, Boussalis and Dargaville2024) extend to focus on solution-oriented or climate delay claims. A strength of this approach is that it explicitly focuses on uncovering hidden premises, thereby exposing the ambiguity that is common in many claims. It also does not require specific domain knowledge on climate policies to apply. On the other hand, the logic approach struggles with inductive claims that do not feature definitive conclusions, that is, the difference between a definitive claim such as ‘carbon taxes do not work’ versus an inductive one such as ‘the evidence that carbon taxes work is limited and uncertain’ (Cook et al., Reference Cook, Ellerton and Kinkead2018).

Compared to fact and logic-based approaches, the focus on intent recognises that there are different positions to take on questions of climate policy, and that these positions will be occupied by actors with specific constellations of interests. In other words, it does not presuppose a given correct answer to climate policy, and it is not bound to certain types of content such as definitive claims. On the other hand, intent is by nature ambiguous and actor-based interests can be intransparent – often purposefully so. Judging intent is therefore a subjective and imprecise exercise, and carries strong normative implications for all those involved.

7. Conclusion

In this article, we aimed to expand on the concept of intent in climate policy discourse analysis and categorise how it has been evaluated in prior work. We find that intent is rather an implicit concept in most work. Nonetheless, we derive three categories to guide future actor-based analysis – interests, behaviour, and track record – while identifying three further categories that also deserve empirical attention – density, ambiguity, and professional context.

We believe these categories can assist the interdisciplinary and lay audience of this journal to critically reflect on climate delay discourses that they may observe in policy and media contexts (Box 1). In doing so, it should be noted that the overall weight of evidence can be considered. How many questions to judge intent have been answered in the affirmative? While a single affirmation may leave room for doubt, a multitude ought to constrain our credibility that someone is acting in faith and without intention to undermine climate policy.

To our knowledge, there has been no research that explores how actor analysis can support fact and logic-based techniques to identify and debunk climate misinformation. We perceive this as an important gap for future research, given that all three approaches have inherent strengths and weaknesses.

As climate policy implementation advances, discussions over how it takes place and who bears the costs or benefits will only become more frequent. Addressing climate change will require transformative and encompassing solutions across all sectors and segments of society. These solutions will likely give rise to new challenges, such as conflicts over land use, labour market shifts, or local impacts from the extraction of new minerals and resources. Such ‘wicked problems’ underscore the importance of listening to and engaging with affected parties, and doing so reflexively and repeatedly as new issues arise (Grundmann, Reference Grundmann2016; Levin et al., Reference Levin, Cashore, Bernstein and Auld2012). The notion that certain discourses constitute climate delay should not be used to stifle these conversations, as they are instrumental for reflecting the real concerns of citizens and stakeholders, and in the right setting can even strengthen support for climate policies (Dryzek et al., Reference Dryzek, Bächtiger, Chambers, Cohen, Druckman, Felicetti and Warren2019). Nonetheless, we should not be naive to the fact that a spectrum of powerful interests can benefit from casting doubt on the design and implementation of climate policy. The evaluation of intent can provide one entry point to shining a light on climate delay discourses and, therefore, on the broader phenomenon of climate obstruction.