Belfer et al. (2017) examined reporting on climate change and Indigenous peoples from four English-speaking countries between 1995 and 2015. They found several widespread trends across countries: First, news coverage largely treated Indigenous people as “intermediaries of urgency” (Roosvall and Tegelberg, 2013) who experience climate change most directly, rather than agents in their own right. Second, coverage was narrowly scoped; there was little to no context about colonialism and marginalization as core vectors of vulnerability that might themselves require a response. This absence frequently led to writing that treated Indigenous communities as responsible for choices that were made under duress, such as living in remote locations. Third, while many news reports mentioned traditional or Indigenous knowledge, it was nearly always treated as subordinate to scientific ways of knowing.

Belfer et al. (2017) called for an expansion of this research to compare different types of media, particularly outlets across the political spectrum. We respond to this call by comparing institutional U.S. media⁴ with different ideological leanings as well as two types of specialized outlets: those focusing on climate, and those focusing on Indigenous peoples. The modern media landscape is routinely described as high-choice, and the topic of climate is no exception. Climate change has become profoundly politically polarized in the US since the 1990s (Nisbet, 2009), and exposure to news sources with different political leanings has been linked to distinct climate attitudes (Newman et al., 2018). In addition, specialized sites that report primarily or exclusively on climate play an increasingly important role as a source of news (Painter et al., 2022; Russell et al., 2023). Meanwhile, Indigenous researchers have found that Indigenous publications are far more likely than non-Indigenous ones to frame COVID-19 stories thematically, grounding them in cultural context rather than focusing on communities’ lack of resources (LaPoe et al., 2022); we anticipated the same trend would hold for climate change. Moreover, there has been almost no research on how climate change is reported in news media by and for ethnically minoritized communities [but see Takahashi and Salas (2024)], despite racialized differences in how people understand environmental issues (Song et al., 2020).

Broadly speaking, the problems associated with climate change can be grouped into two larger categories: the causes of climate change, and its environmental effects. Similarly, solutions can be broadly grouped into mitigation and adaptation (cf. Swain, 2022).

We focus on solutions rather than problems because there is some evidence that hope-driven climate media may be more effective than media that focuses on fear (Johns and Jacquet, 2018; O’Neill and Nicholson-Cole, 2009), although the evidence remains mixed (Ettinger et al., 2021). In particular, hope-driven communications must be rooted in solutions rather than in the false hopes of denying or minimizing anthropogenic climate change (Geiger et al., 2023). One growing approach, which focuses on solutions rather than simply on problems, is known as constructive journalism (McIntyre and Gyldensted, 2018).⁵ Not only does this form of journalism show promise for raising support for collective climate action, perhaps because it expands feelings of control (Their and Lin, 2022), climate change educators are increasingly learning from its principles (Höhle and Bengtsson, 2023).

News stories often open with some sort of hook—a reason to care. Most research on moral framing leans heavily on moral foundations theory; a recent review argues in favor of including a broader range of moral theories (Brugman, 2024). We are particularly interested in framing that emphasizes duty and obligation either to oneself or others. As such, we focus on “the model of moral motives,” which Brugman (2024) specifically cites as a relevant alternative.

Psychologists use the term “moral motives” to refer to human needs to protect and nurture the people around ourselves (Janoff-Bulman and Carnes, 2013), which encompass the full social scale from the self to the societal. (Note that the term “moral” is descriptive rather than evaluative.) The framework assumes that human prosociality is foundational to being human [for theoretical syntheses of evidence see Tomasello (2014, 2019, 2020)] and for human survival (Tomasello et al., 2012). Human prosociality is automatic (Zaki and Mitchell, 2013) and emerges early in development (Hepach et al., 2013), ahead of norm learning and internalization (Warneken and Tomasello, 2015). Scholars have increasingly argued (Thomas et al., 2023) that prosocial motives for learning complement the intrinsic value of science, which was the focus of a landmark 2009 report (National Research Council, 2009) that connected the dots between identity, knowledge, attitudes, and motivations for STEM learning.

In the context of journalistic framing, studying moral motives means asking: Do news stories focus on providing benefits or preventing harms? And for whom? Do they focus on individuals or collectives?

We see climate solution stories as including up to four primary roles (Figure 1). There are causers of climate change, and solvers, both of whom are presented as active participants. Then there are those who are most harmed by climate change—and by its solutions. We refer to these roles as victims of climate change and sacrificers of solutions. None of these roles are mutually exclusive.

Focusing on these roles allows us to identify the types of people and groups that are represented as active participants and those treated primarily as passive recipients of others’ actions. Similarly, focusing on people who are quoted in news stories helps us identify the forms of expertise that journalists treat as relevant.

Expertise is not simply a trait of individuals but must be negotiated between parties in a conversation. Having the “right” credentials does not guarantee that one will be treated as an expert in the media (or, for that matter, that one will be interviewed or quoted at all); the interviewer or writer must also communicate the appropriate stance (Calsamiglia and López Ferrero, 2003; Lindwall and Lynch, 2021; Piirainen-Marsh and Jauni, 2012). All people have many different identities, which are made more or less salient in different contexts. Our relationships with these identities are shaped by widely shared narratives seen in news media and other pervasive cultural discourses (McAdams, 2019; McLean et al., 2018).

A single person might be a wife and mother, a citizen of the Navajo Nation, and a climate scientist who works for the state government. Choosing which of these to mention is one way in which journalists attribute the expertise of the people they quote, because of normative assumptions that all pertinent information, and only pertinent information, will be shared (cf. Grice, 1989). Focusing on these choices allows us to analyze discursive choices by which people (in this case, journalists) confer or withhold structural power from individuals and groups (Bucholtz and Hall, 2004).

Four researchers iteratively developed the codebook based on the considerations and theoretical framework outlined above. In each round, at least two of us would code a small sample of news stories from different types of outlets. Then all four of us would discuss and revise the instructions accordingly.

The full codebook is available as Supplement A.

We sampled stories published in 2023 from six outlets in each of the three categories: institutional U.S. media, climate-focused, and Indigenous-focused. The list of institutional outlets was adapted from a list of the most trusted US outlets as of February 2022 (Morning Consult, 2022), with attention to maximum variation in terms of medium, ideological leaning, and prestige.⁶ We initially attempted to include Fox News in our data set, but none of the stories met our inclusion criteria, as climate solutions were not substantively discussed in any story returned by our search procedure.

The lists of specialist climate- and Indigenous-focused outlets were developed in dialogue among the three collaborating organizations. The list of climate outlets was also informed by others’ research (especially Russell et al., 2023). The list of Indigenous outlets was led by the Indigenous journalists on our team and further informed by work in Indigenous Standpoint Theory (Azocar et al., 2021; Carter Olson et al., 2022; LaPoe et al., 2022).

No single database contained all selected outlets, and many of the smaller outlets were not represented in news databases at all.⁷ As such, we used a discrete sampling strategy for each category.

We searched each individual institutional outlet in Factiva for stories published in a consistent one-year interval (i.e., 01/01/2023–12/31/2023), using a query incorporating several climate solutions-related terms, which differed slightly for print and broadcast outlets (see Supplement B). Prior to sampling, we tested potential queries to ensure that a reasonable number of search results appeared to deal with solutions to climate problems even when not sorted by relevance. The query, date range, any other filters, and the sort order used for results were recorded for replicability (Supplement B). We then randomly sampled stories from each outlet and analyzed the first 8 which met our criteria for solutions-focused stories.

For Indigenous news outlets, the number of stories tagged as “environment” or “climate” (or appearing in a search for these terms, depending on the site) is low enough that we manually screened the full set of results from the one-year window for solutions-focused stories and sampled from those.

For climate news outlets, story volume is higher overall, and solutions-focused stories are also more frequent. In this case, we generated a list of 10 random dates and included the most recent five stories published on or before each date, then randomized the list to select stories.

In all cases, we excluded letters to the editor, opinion pieces, book or movie reviews, and audio or video over 20 min with no available transcript. We also excluded search results that were off-topic or did not discuss at least one concrete action which is being taken or proposed to proactively address harm or risk posed by the climate crisis.

Supplement B provides further detail about the sampling process, and Supplement C contains the final list of stories analyzed.

After the researchers independently coded several stories, one researcher coded all remaining stories. A second researcher provided triangulation by reviewing all coding and discussing any questions or discrepancies until both researchers were satisfied with the final codes applied.

We then calculated descriptive statistics. After a preliminary qualitative analysis, we conducted significance testing to confirm whether these observable differences between types of outlets could be attributed to chance. Wherever possible, we made two-way comparisons rather than three-way (pairwise) comparisons, particularly because the largest differences were between the Indigenous sources and the other two types.⁸ Even so, some comparisons had very small numbers of observations, meaning a χ² test would not be appropriate, so we used Fisher’s exact test. A commonly used measure of effect size for this test is the odds ratio (Rudas, 1998), which measures the association of a particular outcome with a particular condition. For example, an odds ratio of 3.0 means that the odds of an event occurring are three times higher in one condition than another. (Note that the odds ratio is not a measure of risk.)

Last but not least, we turned to close reading, narrative analysis (De Fina and Georgakopoulou, 2012; Ochs, 2005), and collaborative dialogue to shed further light on some of the compelling patterns we saw in our initial analysis. Narrative analysis provides systematic methods for identifying temporality, causal relationships, actors, and power dynamics in stories, both those told by individuals in interaction and more formal stories like news reports.

The most common appeals to moral motives across all types of outlets focused on harm to collectives (Table 1). Climate change “wreaks havoc across the globe” (Good Morning America, 2023a, 23 September 2023); it threatens “human life, tribal facilities, and overall infrastructure” (Celestine, 2023); it will even “increasingly exacerbate risks to U.S. national security interests” (Cho, 2023). Solutions are largely discussed in terms of the protection they offer against these harms, rather than additional benefits they might offer.

In institutional stories, benefit framing typically appeared in stories about consumer-facing solutions like carbon-friendly cruise ships (Diller, 2023) and ecological gardening (Roach, 2023). Meanwhile, only a single climate outlet, Inside Climate News, used benefit framing at all. These stories also focused primarily on consumer-facing solutions, including subsidies for rooftop solar (Gearino, 2023a) and cars built with low-emissions steel (Gearino, 2023b). In fact, climate specialist outlets were less likely (p = 0.013, odds ratio = 0.375) to appeal to moral motives at all, frequently opening with a dispassionate and technical introduction to the problem and solution at hand.

Indigenous outlets included more stories with a focus on collective benefits (p = 0.012, odds ratio = 3.91) than the other two types, particularly in stories focusing on the interdependence of people, animals, and land. In particular, a number of stories about Indigenous stewardship, food sovereignty, and land management—a common topic only in the Indigenous outlets—highlighted the benefits to people and animals alike.

Institutional and climate journalism appear to focus on similar problems (Table 2): these two sets of stories were roughly equally likely (p = 0.62) to focus on mitigating the causes of climate change and on adapting to a changed climate. Meanwhile, Indigenous outlets were much more likely (p = 0.006, odds ratio = 3.18) to tell stories about solutions that brought adaptation and mitigation together—and much less likely (p < 0.001, odds ratio = 0.262) to consider mitigation in isolation. The relative prevalence of stories about stewardship, described above, contributed to this difference. We also observed during the development of the data set that many of the institutional and climate outlets provided considerable episodic reporting on environmental issues like clean water and animal wellbeing, while the Indigenous outlets chose to contextualize these kinds of topics in broader climate issues.

All three types of outlets displayed a similar balance between “soft” (policy, administrative, or fiscal) solutions and “hard” (technological, including old technologies like planting trees and seeding oyster beds) ones (Table 3). However, Indigenous outlets were far more likely (p < 0.001, odds ratio = 54.00) to mention traditional or Indigenous ecological knowledge than the other two types: 38 of the 48 stories in Indigenous outlets did so, compared with just four stories in institutional outlets and two in climate outlets.

Two of the four institutional stories were USA Today pieces about Indigenous stewardship of the Klamath river, which were written by a single Indigenous journalist (Krol, 2023a,b); the third and fourth were New York Times pieces, one about a relocation agreement between Tuvalu and Australia (Frost N., 2023) and the other about Canadian approaches to wildfire prevention (Isai and Austin, 2023). Meanwhile, the two stories in climate outlets focused on silvopasture, or using the same land for tree planting and animal grazing (Chilukuri, 2023), and Colorado River water rights (James, 2023). Of these six, only the USA Today stories had a central focus on Indigenous practices and knowledge; in several of the others, it was mentioned only in passing. For example, the only reference to traditional knowledge in one story was a brief aside: “A USDA survey from 2017 showed that only 1.5% of U.S. farmers practice any kind of agroecology, the ancient regenerative farming movement that includes silvopasture” (Chilukuri, 2023).

Because of the types of stories we excluded from our data set, nearly every story from every outlet identified a problem-solver (Table 4). On the other hand, only half of stories across types of outlets identified actors who were primarily responsible for climate change. Climate outlets’ stories had fewer victims of climate change than institutional stories, and Indigenous outlets identified victims more often than the other two types, but both of these results could be attributed to chance (p = 0.41 and 0.074, respectively).

In general, victims were much more likely to appear in stories about adaptation solutions (p < 0.001, odds ratio = 7.30) whether they included a mitigation component or not, than stories that were exclusively about mitigation (Table 5). This is an expected result, since focusing on impacts rather than causes requires understanding who is impacted.

There were few differences between types of outlets in terms of most types of descriptors used for problem-solvers. Institutional outlets focus on individual problem-solvers more frequently than the other two types, but this finding could be attributed to chance (p = 0.08). A total of 14 institutional stories focused primarily on individual solvers, compared to seven stories in climate outlets and eight stories in Indigenous outlets. Individual politicians (PBS News Hour, 2023c, March 13, 2023; Davenport, 2023; Frost N., 2023; CNN Newsroom, 2023, September 20, 2023; Good Morning America, 2023b, November 7, 2023; Fareed Zakaria GPS, 2023, July 2, 2023) and CEOs (Dvorak, 2023; Good Morning America, May 8, 2023)—as well as the reader, indicated by the second-person pronoun “you” (Fenton, 2023; Frost A., 2023; PBS News Hour, 2023b, January 19, 2023; Good Morning America, 2023c, December 26, 2023)—comprised most of the individual solvers in the institutional stories, rounded out with two stories about individual activists. Second-person solvers did not make an appearance in the other two types of outlets, while researchers joined the types of leaders highlighted.

The authors found the relatively high number of individual solvers somewhat surprising. Since nobody can truly solve a problem of this magnitude alone, presenting problem-solvers in this way presupposes an individualistic approach to achievement [compare Echterling (2016)]. Even policies, a prototypical collective solution, were regularly framed as the accomplishment of a single named political leader, rather than ideas put forward and operationalized by coalitions.

There was one overarching narrative that appeared at least once in every single Indigenous outlet, as well as in the two Indigenous-authored USA today stories: “green colonialism.” When institutional or climate publications critiqued proposed solutions, it was typically for reasons of economic or practical feasibility. In contrast, the Indigenous outlets frequently reported on the colonial logics underpinning climate projects: building windmills on Sámi land without permission disrupted reindeer herders’ ability to do their work (Rasmussen, 2023); rare-metal mining operations did not take appropriate care of Indigenous sacred sites (Ricketts, 2023); cost–benefit analysis does not appropriately value relationships with land (Estus, 2023). Only the Indigenous outlets were unequivocal that just and ethical climate solutions require the partnership of Indigenous people who care for the land that provides resources. On the other hand, stories in our data set from institutional outlets rarely centered the perspective of Indigenous people, even on topics where they have been publicly vocal about their rights, such as the ongoing debate over Colorado River water use (PBS News Hour, February 3, 2023).

Indigenous outlets also told more stories where the sacrificers of solutions were victims rather than causers of climate change (Table 7). That is, the people being asked to give something up to respond to climate change were those who were disproportionately harmed by it, rather than those who were directly responsible for it. For example, a story about the Bureau of Land Management’s proposed changes to oil and gas leasing, noted that the changes violate treaty rights despite awareness of disproportionate climate impacts on Indigenous communities (NDN Collective, 2023). In contrast, institutional outlets were much more likely than Indigenous outlets to focus on sacrificers who were themselves causers of climate change (p = 0.032, odds ratio = 5.62), particularly in stories about the transition away from fossil fuels. In this case, fossil fuel companies would lose out—but their loss was largely presented as appropriate, even just.

Across institutional, climate-focused, and Indigenous news outlets, most stories that appealed to moral motives focused on how climate change harms groups of people. Out of the three outlet types, Indigenous outlets were most likely to highlight how a solution benefited others (for example, tribal stewardship of land benefiting the people and wildlife living there). Meanwhile, climate-focused outlets were less likely to use moral motives framing to draw in their audience.

Indigenous outlets were more likely to tell stories about solutions involving both mitigation of further climate change and adaptation to its existing effects, rather than one or the other. Across outlets, “soft” (policy and economic) and “hard” (technical) solutions occurred with roughly equal frequency. Traditional knowledge was rarely mentioned outside of Indigenous outlets.

While nearly all solutions stories identified an individual or group as responsible for the solution, fewer pointed to specific causers or victims of climate problems. Institutional outlets were more likely to identify individuals as problem-solvers than the other outlet types, and most of these individuals were politicians.

Indigenous outlets were far more likely to include Indigenous problem-solvers in climate stories. Outside of Indigenous outlets, Indigenous groups were rarely portrayed at all. Indigenous outlets also represented a much more diverse set of Indigenous people than other outlet types.

While all outlet types were equally likely to highlight how certain groups might be harmed by a proposed solution, institutional and climate outlets tended to identify companies and countries reliant on fossil fuels as sacrificers. Stories from Indigenous outlets identified how solutions examined uncritically in other outlets, like rare-earth mining for battery production, could still harm people who are already disproportionately affected by climate change.

Our data set was relatively small, comprising eight stories per outlet and six outlets per type. Moreover, we included stories that appeared in an outlet regardless of where they were produced, based on our knowledge of news users’ behavior. Partnerships and co-production and co-licensing agreements are common, however, and in practice, some stories that were produced by outlets in one category were analyzed as part of another category. Consider Grist as an illustrative example: although their overall focus is climate issues, and we categorized them as a climate outlet, they have a dedicated Indigenous affairs desk, and their stories regularly appear on Indigenous sites.

When it comes to climate solutions, climate journalism looks more like institutional journalism than either one looks like Indigenous journalism. Indigenous journalism is more likely to open with benefits, consider holistic solutions that address both adaptation and mitigation, and identify victims of climate change. Indigenous journalism is also much more likely to focus on Indigenous people in all roles, and to highlight traditional knowledge. These strategies are consistent with those Fine (2023) identifies for communicating about climate justice to social justice advocates who are not working on climate specifically: highlighting intersectional impacts and benefits of climate solutions to impacted communities.

The differences between institutional and climate outlets on the one hand, and their Indigenous counterparts on the other, illustrate some powerful stereotyped knowledge around climate problems, solutions, and problem solvers. Social stereotyping is both a strength and a weakness of human cognition; it has allowed us to function collectively by delegating some of our thinking to others, and it can both support and hinder learning. Stereotype violations likely function as “minimally counterintuitive stimuli” (cf., Upal, 2011). Minimally counterintuitive stimuli tend to be better remembered (Boyer and Ramble, 2001; Norenzayan et al., 2006) and elicit greater curiosity (Lewry et al., 2023), both of which may be key first steps for learning via updating false beliefs (Upal, 2010). The audiences of these outlets suggest parallels with climate advocates who are not engaged with social justice. Here, Fine (2023) argues for emphasizing local issues, promoting empathy, and advocating for systemic solutions.

Given the constraints specific to journalism, we suggest that institutional news can learn from Indigenous news in this area. Specificially, institutional outlets ought to consider: