Abstract
This article theorizes how social media images politicize. We argue that layering—digitally assembling diverse visual elements—is a key practice of visual politicization. Beyond just a technique, we conceptualize layering as a form of remix, a cultural tool that can be used to politicize: to render issues recognizable and actionable. To show how layering works, we analyze over 500 Instagram images related to climate activism. Adopting Annette Markham’s conceptual framework of remixing, we ask what elements are brought together in the images, what kinds of connections are produced, and what new meaning(s) emerge as a result of layering. Our analysis identifies four main types of layering: spatial, temporal, intertextual, and evaluative. We show how layered images create connections, positions, and scales between places, temporalities, arguments, and actions, prompting viewers to look at them in particular arrangements. Through layering, we argue, images name issues and actors relevant to the climate emergency, problematize the seemingly self-evident, and visualize alternatives to the status quo. Building a novel theoretical combination that puts together a pragmatist conception of political action and remix studies, this paper introduces layering as a concept to identify and analyze contemporary forms of doing politics online. It contributes to theories of politicization by positioning remixing as a key analytical lens to contemporary forms of politicization.
Introduction
In an increasingly visual public sphere (Meriluoto and Luhtakallio 2023), understanding visual and multimodal content and the particular ways in which they “demand attention” (Rovisco and Veneti, 2017) in the public has become vital. A prominent practice of building visual content on social media is combining or collaging different elements such as photographs, text, frames or drawings on top of each other or side by side. This type of social media content, which we propose to call “layered images,” is the result of digital editing—for example, assembling, meming, infographying, and curatorial designing. Visual elements in these images are often appropriated and re-purposed beyond their “original” use and meaning.
In this article, we argue that layering is a central contemporary tool of visual politicization. While sharing important aspects with existing concepts such as collage and bricolage, layering focuses on the relationship between different elements in social media content. As a concept, layering encapsulates the relational nature of meaning-making on social media content: by definition, a “layer” does not exist without other layers (think of, e.g., lasagna). By assembling different elements, layered images serve to transfer positionings or “stances” (Shifman, 2013) 1 ; to generate contrast or to reinforce the desired message by combining elements that support each other. By creating connections, positions, and scales between things, and by making the viewer look at them in particular arrangements, we argue that layered images can problematize the seemingly self-evident, visualize an alternative to the status quo, and interrupt the normal flow of everyday—to politicize (Eranti and Meriluoto, 2023; Luhtakallio 2012; Meriluoto, 2021). Moreover, by making the viewer see particular cultural units (actors, places, moments in time) side by side, different layers—and the actors they represent—are framed as part of the same whole, either as equal to each other or through an evaluative arrangement (e.g., comparison, contrast, moral binary). Alongside critique, layered images achieve an equally important element of politicization: rendering its basis recognizable to others (Eranti 2018; Luhtakallio 2012; Meriluoto 2021).
The conceptualization of layering as a tool of visual politicization is based on an empirical analysis of visual climate activism on Instagram, through which we also showcase how layering functions. The visualization of climate change and the activism around it presents a particularly interesting case for studying politicization in social media images, as the phenomenon has been gradual, complex, and in many ways still “invisible” (Frig and Penttilä, 2025) or “unseen” (Lester and Hutchins, 2012)—at least in the Global North. In this paper, we draw insights from over 500 layered Instagram images related to climate emergency and activism, asking what kinds of layers and layering practices can be found, and what these layers are doing to politicize climate emergency.
As a part of a wider research project on visual politicization among young Europeans (Luhtakallio, 2018), the data for this paper were collected during spring 2021 from Instagram with hashtags originating from online and offline ethnographic fieldwork in France, Finland, Portugal, and Germany. A novel image classification method (Maltezos et al., 2024) developed in the same project was used to decipher different forms of visual political action around climate emergency out of over 42 000 images. One of the forms of visual politicization that emerged from the analysis is “layered images”—that is, images that are “not just photos,” but where multiple visual elements or text and image are brought together, creating new meanings. For this article, we singled out 524 “layered images” as a category for a more detailed qualitative analysis.
To make sense of how politicization occurs through the different layered elements of the image, we follow the analytical foci of remix studies (Markham 2013, 2017; Navas et al., 2014). As the familiar music-inspired vernacular conveys, remixing refers to diverse practices of recombining and sampling existing elements into new versions. 2 Increasingly, remixing has been embraced to describe the novel kind of cultural production brought about by the proliferation and democratization of the means of digital visual culture through social media.
In this article, we turn to examine the critical potential of remixing (Conti, 2015), looking at how layering can be understood as a shared cultural tool of politicization (Luhtakallio et al., 2024; Swidler, 1986). We analyze the data through Annette Markham’s remix theory’s three core concepts: sampling, linkages and hybridity (translated in our analysis as “new meanings”), to understand what elements have been brought together in the layered image, what kind of connections are built between them, and what new meanings are constructed through the combination of elements. Based on the analysis, we identify four types of layerings as the most prevalent in climate-related social media images: spatial, temporal, intertextual, and evaluative.
The contribution of the paper is thus threefold: (1) we propose conceptualizing layering as a tool of politicization typical of social media, (2) we show how layering can be used to analyze social media images (3) we distinguish four types of layering typical to our empirical field of climate activism. We start with an overview of existing literature on visual social media politics: how the climate emergency is politicized visually, and how previous studies have approached the political work done in and through images. We then discuss politicization as a practice and connect this to the three elements of remixing: sampling, linkages, and hybridity. After outlining our data and methods, we present the four main types of layering we identified. We end by discussing the potential effects of the identified forms of politicization, and the future theoretical possibilities in combining remix studies to analyses of politicization.
Climate images on Instagram
Social media has transformed our relationship with images in a myriad of ways (Rose 2022); images are everywhere, they can be made by anyone, and the boundaries between production and consumption of (visual) media have long ago become blurred (Jenkins 2006). Scholars highlight the radically open-ended nature of social media content; it is “alive” (Hand, 2020), “open” and “malleable by design” (Frosh, 2003: p. 73), “designed as invitations for (creative) action” (Schifman, 2014), making it ripe for recycling, re-using and re-purposing on an unprecedented scale. As described by Saara Särmä (2016), social media content such as parody images or memes can very well be users’ first contact with current phenomena. In many ways, the composition and consumption of social media images has become interwoven into our common everyday practices (Schreiber, 2017) and “vernacular” epistemologies (Burgess, 2006; Conti, 2015).
The importance of multimodality and visuality is further highlighted when people are confronted with novel, complex, hard-to-grasp phenomena, such as the ongoing climate crisis. The complexity and partial “invisibility” of the climate change and wider environmental crisis has sparked lively debate and scholarly interest on how to best visualize it, initiating a growing body of research seeking to tackle the “image problem” of climate change (e.g., Chapman et al., 2016; León et al., 2022; Uusitalo, 2020). The literature unanimously recognizes the key role of imagery in influencing how people frame and conceptualize climate change (Lehman et al., 2019; Leiserowitz, 2006; O’Neill, 2013), as the different qualities of images hold great potential both for “witnessing” the climate emergency, as well as for shaping the social imagination around sustainable futures (O'Neill, 2017). As O’Neill and Smith argue (2014), by creating visualizations of climate change, “many actors—evoke climate change in particular ways, and make the issue meaningful in everyday discourse” (p. 73).
The climate-related visual analysis are often focused on the powerful and professional. While identified as a major forum for politicization, visually dominated social media platforms, such as Instagram, have long remained under-examined (see, however: Pearce et al., 2019; Koop-Monteiro et al., 2023; Schäfer, 2012; Mooseder et al., 2023). Extant research suggests that climate-related visual formats and contents vary widely across social media platforms (Pearce and Özkula, 2017) and that images of real people grappling with climate effects are more likely to be shared and liked (Corner et al., 2015). Taking and posting climate-related images on social media have also been analyzed as part of everyday political practices (Uusitalo, 2020). Such mundane visualities seem to create more online engagement than images of famous people and politicians (Metag et al., 2016). Social media visuality also emerges as an arena for the “personalization” and “celebritization” of climate politics (Haastrup, 2022; Stoddart et al., 2024). On the one hand, influencers, politicians, celebrities, and activists engage in visual genres of personal storytelling aimed at enhancing the relatability of climate issues (Haastrup, 2022). On the other hand, they are depicted on social media as heroes or villains as part of processes of critique concerning climate politics (Stoddart et al., 2024). Another prominent approach to climate images focuses on organized movements or networks, such as #fridaysforfuture (Herrman et al., 2022) or Extinction Rebellion (Jokela et al., 2024; Malafaia et al., 2025), or “climate celebrities” (Haastrup, 2022) such as the Swedish activist Greta Thunberg (Molder et al., 2022).
This article expands existing analyses of climate-related social media images on people’s everyday political visual practices. We are interested in the concrete practices through which social media users compose political arguments around the climate emergency, and the mechanisms through which political meaning is created. For this, we build on a novel theoretical combination that puts together a pragmatist conception of democracy and political action (Eranti and Meriluoto, 2023; Luhtakallio et al., 2024) with layering—a form of remix—as a cultural tool typical to digital media.
Remixing to politicize
In political and social theory, politicization is broadly understood as the opening of a debate—the process through which a certain state of affairs is interrogated, critiqued, and eventually deemed changeable and in need of disruption (e.g., Li, 2019; Palonen, 2003). The effects of politicization—putting into question something that has previously been presented and regarded as self-evident or private (e.g., Mills, 1959; Mouffe, 2013)—are well-established and fairly widely agreed-upon, but the practices through which this “opening” happens have been far less tackled (see, however, Carrel, 2015; Luhtakallio, 2012).
In this article, we build on recent scholarship theorizing politicization as a practice (Eranti and Meriluoto, 2023; Luhtakallio 2012; Meriluoto, 2021). Within this pragmatist democratic theorizing (Luhtakallio et al., 2024), politicization is understood as an act that interrupts the “normal flow of everyday” by voicing out an alternative to how things could be. This interruption needs to be substantiated by rendering the basis of the critique recognizable to others. Politicization, then, is verbal, visual or embodied argumentation that achieves two things: It (1) makes us notice something we have thus far taken for granted—often by showing us how this matter could also be otherwise, and (2) makes us care about, or connect to this matter by linking it to something already familiar to us (Eranti and Meriluoto, 2023; Luhtakallio, 2025).
Here, we investigate how the layered character of climate-related images posted on social media achieve the task of politicization. Crucially, we are reading the images in a context where the status quo is a perceived global inability to adequately respond to the climate crises (Browne, 2018) in a world that is continuously overshooting previously set critical thresholds (United Nations Environment Programme, 2025), and ignoring the steadily advancing ecological collapse in the form of biodiversity loss and mass extinction (Cowie et al., 2022). We are asking how climate-related posts on social media seek to interrupt this state of affairs. By observing them as tools of politicization, we are referring to their capacity to interrupt the seemingly self-evident trajectory of insufficient climate actions and deepening ecological crises, and inquire how they can be used to make the situation appear as something that can and should be acted upon. Here, we suggest layering as the key tool in achieving the two components of politicization: rendering the situation actionable and recognizable.
In this article, we use the concepts of remix theory to understand the practices through which these political actions take shape in the visual-digital environment. Inspired by the sampling and recombining practices of the remix culture rooted in the 1970s disco and hip-hop-scenes, scholars across disciplines have used the idea of remixing as regeneration, “the act of using pre-existing materials to create something new” (Navas et al., 2014), expanding it for analysis beyond music to conceptualize the kind of contemporary cultural production brought about by the proliferation and democratization of the means of digital visual culture.
Following Markham (2013, 2017), remixing culture is not only a technical quality of digital media, but primarily a very humane way of making sense of the world “out of the endless swirl of stimuli around us” (p. 225). To Markham, the novelty of social media as a site of remixing is not only about the speed or volume of the stimuli, but more importantly the way in which social media makes remixing practices visible and tangible.
Subsequently, social media images can be seen as a part of the aesthetic practices that arrange our sensible world (Niemelä-Nyrhinen and Uusitalo, 2021). Far from a side issue in democracy, in our increasingly hypervisual societies, the political significance of visibility and recognizability has only increased. As Rancière (2010) has argued, democracy is not a set of institutions, but an aesthetic regime that organizes and is constituted by the different hierarchical orderings of what we can sense (see Vihalem, 2018). The ability to be apprehended and recognized equals power and privilege, and major political struggles emerge over the right to be seen and taken into account as political subjects (Butler 2015): p. 5; Rancière 2010).
In this paper, we understand remixing as belonging to the cultural toolkit of being visible and recognized. 3 Defining remixing—and layering as a form of remix—as a cultural tool of visual politicization allows us to think about it, in Swidler’s (1986) terms, as a “repertoire” or a skill: culturally established and shared but not equally accessible for everyone. This way, remixing emerges as a form of politicization feeding on the vernacular: the ability to organize (or disrupt) our sensible world (see also Yusoff, 2010).
Layering draws on and extends artistic practices such as bricolage, collage, and montage, and should not be understood as separate from them. Layering of images is done through collaging tactics, that is, the “cutting and sticking” of different materials (Butler-Kisber, 2010: p. 102). It is also “bricolaging” in the sense that layering is composed of whatever materials are available (Lévi-Strauss, 1966). Furthermore, it shares with montage the logic of combining disparate or very diverse elements to produce new meaning (Manovich, 2001). While these existing concepts remain useful in describing artistic practice, methodology or pedagogy, they tend to foreground technique and materiality over meaning-making. With the term “layering,” we want to capture both the specific practical aspects—how images are created by adding layers—as well as the semantic function of layering, that is, how the creation of meaning happens in between the layered elements (Manovich, 2015).
Crucially, whereas collage and bricolage emphasize the act of assembling heterogeneous materials, and montage arranges elements temporally—one after another, creating meaning through sequence—layering places visual elements simultaneously on top of each other, creating meaning through their co-presence. This simultaneity is analytically significant: in social media images, political meaning often emerges through the co-presence of recognizable cultural elements. Furthermore, for us, the idea of layering always implies the possibility of adding another layer (either imaginatively in interpretation, or concretely by editing or “meming” the image). Thus, while layering shares the defining characters of remixing, it is, as a concept, more limited and precise: rather than describing a general cultural logic of recombination, it zooms in on the specific practice of bringing visual elements into simultaneous co-presence, and on the political meaning that this co-presence produces—allowing us to analyze how images are made and how they make meaning, and, in doing so, how they work as tools of politicization.
Data & methods
Data collection
The data for this study was scraped from Instagram, a visual social media platform owned by Meta. The original dataset consists of 42 708 publicly posted images on Instagram between March and April 2021, with climate-related hashtags. Instagram is well suited for the analysis of the visual repertoires in social media, as it is primarily a visual and personal platform (Jenkins et al., 2013). It is also one of the most used social media platforms in the world, boasting two billion monthly users in 2022 and ranked amongst the five most visited websites (Statista, 2025). While identified as a major forum for political activism and communication, Instagram has so far remained under-examined in comparison to, for example, Twitter and Facebook (Koop-Monteiro et al., 2023; Pearce et al., 2019).
Hashtags are the “anchor” of this dataset: they give the images a common context (Trillo et al., 2021). The list of hashtags used to collect this data was the result of a collective effort by a group of researchers conducting (n)ethnographic fieldwork in four European countries, Finland, Germany, Portugal and France, between years 2019 and 2023, as a part of a larger research project on visual politicization (Luhtakallio, 2018). In each country, relevant climate-related hashtags were identified based on both online observation and fieldwork with climate activists. Some of the traced hashtags were international, such as #xr or #fridaysforfuture, reflecting the cross-national nature and mobilization of climate movement(s). Others were more locally based, such as slogans in local languages.
Images with multiple layers or elements were filtered out by first using a novel, ethnography-informed machine-learning algorithm tailored to recognize and categorize political action in images (Maltezos et al., 2024). Using supervised deep learning-based AI and pre-trained neural networks, the model categorized a dataset of 42 708 climate-hashtagged images into nine categories (see Table A1 Appendix 1). The category that we are interested in this paper is the one that was initially labeled as “artificial,” since images in this category are the result of digital manufacturing, such as editing, collaging, filtering and infographying. The initial category −18 209 images in this dataset—included, for example, memes, cartoons, gifs, logos, infographics, statistics, screenshots, advertisements—anything more than a plain photograph.
For this analysis, we were interested in images that clearly exhibited multiple layers or elements by bringing together different visual and textual units, with each element “doing” something in the image, so that a new meaning would emerge from the combination of those various elements (Markham, 2013). Images produced explicitly for promoting candidates or marketing events—such as advertisements—were excluded, as we were primarily interested in non-commercial and non-promotional images. This produced a final dataset of 524 “layered” images, of which a limited number are reproduced here as representative examples of the analysis.
Methods of analysis
In our analysis, we used Markham’s (2017) three core elements of remixing as a framework to decipher how layered images construct political meaning: that is, how they render issues recognizable, and how they create commonality and/or conflict. Markham identifies these elements as sampling, linkages, and hybridity. Sampling refers to “the continual and experimental selection and subsequent recombination of cultural meaning” (p. 232), through selecting, extracting, and recombining cultural units. In our analysis, sampling focuses concretely on what elements have been selected and brought together in the image. Second, the notion of linkage allows us to zoom into the different kinds of connections that are produced through remixing of previously unlinked elements. These can be, for example, “a comparison, a juxtaposition, an interweaving, or some other type of relationship” (p. 235), created through different ways in which the elements are arranged with each other. Finally, hybridity, in Markham’s conceptualization, refers to the new meanings constructed through the layered image that can no longer be reduced to its composing parts (p. 233). The analysis of hybridity is here dubbed as “new meanings,” asking what are the new meaning(s) that emerge from the remixed images.
Dominant layers.
Analysis
Spatial layering
In what we call spatial layering, visual representations of the causes, consequences and responses to climate emergency—such as, pollution, wildfires, and demonstrations from different sites—are selected, brought together, and rearranged side by side (see Image 1, Image 2 and Image 3). In spatial layering, the process of sampling consists of selecting and combining images from different locations (or symbols of locations). These locations vary in their generality: we see visually unidentifiable flashes of burning forest, drone images of protest in big metropoles, as well as people taking selfies in their familiar environments. Sampling different scenes enables climate change to be visualized across different levels of action and abstraction. From global to local, scenes that are familiar and likely cherished by everyone—like forest, sea and beaches—can be harnessed to invoke collective action. A collage of wildfires in different geographical locations. Collage with pictures on the impact of climate change around the world. A collage of drone images from climate protest marches in different locations.


The linkages through which the different spatial layers are connected mainly rely on visual similarity, repetition, and concomitance, through which the particular meanings of images are de-territorialized and raised in generality; these are not isolated events, but part of a broader, global pattern.
In Image 1, visuals from different locations are positioned as equal. This highlights the effect of repetition: the climate emergency is happening and visible in seemingly unconnected places all around the world. In Image 2, repetition is combined with the grouping of visually different phenomena together to reveal the many faces of the emergency. Image 3, in turn, shows how people are brought together in various locations, visualizing the strength and scope of the global climate movement and its ability to coordinate collective action.
While images of wildfires or animals struggling for survival are, in themselves, powerful visual symbols of anthropogenic climate change, it is their assemblage that generates new meanings; for example, an argument emphasizing recurrence and highlighting the global scale of climate-related phenomena. Similarly, image collages of climate protests in diverse countries underscores the globalized and de-territorialized nature of social movements. By connecting places and actors that are physically distant and socially diverse, these collages highlight a common struggle around which people act and coordinate.
Comparative or juxtaposed arrangements of different spaces, in turn, create new meanings that foreground conflicting place-based notions of political action, climate justice, and moral values. For instance, the paired elements in Image 4 contrast two forms of doing politics: an assembly hall and a street demonstration, representing, formal, institutional modes of political decision-making and the grassroots, informal forms of political mobilization. The disparity between the stillness of institutional politics and the dynamism of street protests incites the viewer to weigh these forms of political action within the context of the “climate emergency,” thereby articulating a critique of governmental inaction. A photograph from a Parliamentary/Assembly hall contrasted with a photo from climate protest. The caption states: “Climate emergency? It does not exist in Piemonte.” Two samples of the same landscape from the polar sea in two points in time, showing how the glacier has melted.

Temporal layering
The gradual and intangible development of climate change is often considered one of the main challenges in portraying it as a matter of urgency. This “image problem” (e.g., Chapman et al., 2016) has spurred a visual social media practice we term temporal layering where images are arranged to construct a temporal narrative. In practice, different moments in time—often from the same location (Images 5–7)—are sampled together. These layers frequently feature natural landscapes and locations where the effects of environmental change are visually evident. The most common format are pairs of photographs, although in some images the contrasting moment is implied or imagined (Images 8 and 9). Often, textual cues such as dates mark the different time points, that is, the year of the photograph or the moment in the future imagined scenario. In some cases, instead of dates, textual cues signal different rhythms or speed, as in Image 6, which depicts how the forest that takes 100 years to grow is destroyed in 6 days. Two photographs representing different timelines, former referring to the 100 years it takes to grow a forest and latter indicating the relatively short time it takes to destroy the landscape. Time comparisons showing the positive impact of environmental action in Mumbai beach. A nostalgic image with a 50’s-style housewife grocery shopping plastic-free, with a text asking “can we make it that way again?” A dystopic image of a city covered in yellow dust, sampled with the text “this is not your social media feed. Yet.” A painting by Sir John Everett Millais of Ophelia from Shakespeare’s Hamlet, with trash edited on top of the original painting. Original edited version credited to Raquela Aparicio (2020).




The linkages created through temporal layering entail comparing, juxtaposing and connecting different points in time. These images visually witness and narrate environmental change by bringing different moments in dialogue with each other, also breaking down and sequencing the passage of time to make the change visible (Images 7–9). Temporal contrast may also emerge between human action and natural rhythms.
While much of the temporal layering serves the purpose of documenting change, some examples promote the potential of human agency. The image pair in Image 7 shows Mumbai Beach across a timespan: a filthy beach 20 years ago, and a clean one 20 years later. Unlike temporal layering that alerts us to the speed and disastrous direction of climate change, this temporal layering visualizes a hopeful future narrative and highlights human impact. In the images linking the present to an imagined future, the visual plea of agency is forward-looking, projecting present fears or hopes onto what is to come.
Temporal layering situates environmental change within a historical continuum, creating meaning that one image cannot convey, inviting us to witness a change, or to imagine a future. This demonstrates how layering infuses its end product with an argumentative potential that is absent when the photos are viewed alone: a sense of immediacy concerning the speed and scale of environmental degradation. Climate change as a political issue is deeply time-sensitive and social media images embody this sense of urgency. By making the effects of climate change visible, temporally layered images challenge narratives that downplay climate change as part of natural fluctuation and variability. On the other hand, temporal comparisons are also used to showcase human agency’s disruptive force, in both negative and positive senses, thus making room for an alternative course of events, contra the inevitability of climate change.
Beyond witnessing past change, temporally layered images also invite future imaginaries, functioning as “prospective” (Shifman, 2014) or “forecasted photos” (Jänkälä et al., 2019). Such visual configurations create new meanings attached to a world that does not yet exist. Examples range from hopeful futures—sometimes evoking nostalgic references to more sustainable pasts (Image 8)—to dystopian warnings. Here, images from the past are brought to the viewers’ attention in a new light, contextualized by the present climate emergency, alluding to the possibility of reclaiming more sustainable cultures and lifestyles. On the dystopic side, images such as Image 9 encourage audiences to tangibly imagine imminent futures if the status quo remains: what will the next image in this sequence look like, if we continue along the same path?
Intertextual layering
While intertextuality, broadly understood, is typical to all remixing and memes, Intertextual layering here highlights the practice of using popular cultural symbols and references in the context of politicizing climate change. In the analyzed images, the sampled units often originate from iconic cultural productions—such as films, TV series, and well-known artwork—that are altered by superimposing text and/or other visual elements, leading to reinterpretations of the imagery components. They also use other iconic visuals, sometimes by mimicking, tweaking or editing a well-known visual theme. Image 10, for example, features the 1851–52 painting by British artist Sir John Everett Millais, depicting Ophelia from Shakespeare’s Hamlet, with garbage edited on top of it.
4
Other examples include references to popular media, celebrities or famous actors and characters, such as Image 11, which adapts an iconic scene from the Harry Potter movies. A defining feature of intertextual layering is the recognizability of the “main” sampled visual component—a culturally significant image—which ensures resonance and elicits engagement. A classic scene from Harry Potter and the Philosopher’s stone, remixed with Hermione correcting a misunderstanding about climate change. Two expressions of Greta Thunberg with a text (in French): “Going to school”/Saving the planet by taking part in an inter-generational ecological movement.

Unlike temporal or spatial layering, where comparisons are explicit and visually apparent, the linkage process in intertextual layering relies on subtler connections rooted in shared cultural knowledge. These connections invite viewers to decode layers of meaning by linking visual cues and references not typically linked, such as Ophelia’s tragic death in Hamlet with polluted waters. Shifman’s (2013) concept of “stance,” developed for analyzing internet memes, elucidates how these connections position creators and audiences in relation to the images. For instance, Image 11, from Harry Potter movies, portrays Hermione correcting Ron’s mispronunciation of a spell. This well-known scene is re-captioned to challenge a common climate-skeptic argument, borrowing Hermione’s characteristic aura of rigorous intellect and stubborn determination. As discussed extensively in meme literature, the depth of meaning in such images depends on viewers’ ability to bridge visual and cultural references, making intertextual connections more interpretive than literal, and creating distinctions based on cultural literacy and “subcultural knowledge” (Nissenbaum and Shifman, 2017).
The dominant remix process in practices of intertextual layering is the creation of new meanings, since entirely new meanings emerge from assembled visual elements. This includes the re-appropriations of known visualities, subverting their meanings and creating new ones, leveraging pre-existing cultural associations to make climate arguments comprehensible. As Markham (2017) explains, “the product that emerges is wholly dependent on the use of existing elements, but the meaning is unique” (p. 234). By hybridizing climate-related and culturally shared visuals, these images emotionally engage audiences, fostering arguments’ recognition by indexing their meaning to the cultural context being evoked. Such indexicality can be exemplified by the newly memetic Ophelia image that, for those familiar with Shakespeare’s Hamlet, invokes notions of despair and fragility, now metaphorically tied to the climate crisis. Similarly, for Harry Potter fans, Hermione symbolically encapsulates intelligence and morality, setting the argumentative tone for how climate change should be addressed. Thus, there is a tacit appeal to a community (e.g., fans, knowledgeable individuals) with shared experiences of the same cultural artifacts, characters, and scenes, which are not merely mobilized as “symbols or signs, but as vehicles for deeply personal attachments” (Thévenot, 2014, 20).
Intertextual layering leverages characters and situations with which people already have emotional relationships, redirecting these emotions—both positive or negative. This familiarity builds in-group feelings (“we”), while also drawing distinctions from out-groups (“the others”). The duality enables intertextual layering to build commonality and create space for critique and conflict. Its ability to align with or distance from dominant discourses underpins its politicization power (Conti, 2015). By tapping into the emotions and meanings tied to icons and scenes that belong to collective imaginaries, intertextually layered images stimulate cognitive and affective engagements. They turn familiarity into a tool for transforming personal and culturally shared experiences into frameworks for meaning-making around climate challenges.
Evaluative layering
While all the layering types we have identified can be understood as (e)valuative, that is, assigning and organizing worth and passing judgment in one way or another (Cefai et al., 2015; Lamont, 2012), in some images, these evaluative arrangements are more at the forefront. The fourth type of layering we propose is therefore called evaluative layering. Sampled elements are often famous and powerful persons, mobilized to draw attention and create visibility (Luhtakallio and Meriluoto, 2023, 2026), while their representational qualities are used to substantiate arguments about either the status quo or alternatives to it. Just as Greta Thunberg (Image 12) is not merely a young activist, but also represents the contemporary generation of climate activists, a German politician (Image 13) potentially alludes to the broader political class. Visual arrangements emphasize these dualities, using actors’ representational significance to unpack conflicting valuations at play in specific situations. Layered images can sample duplicated visuals of the same person (e.g., Images 12 and 13), or combine two different individuals (e.g., Image 14). Beyond depicting persons per se, these individuals are connected to choices, which are presented in an evaluative arrangement. Emotional cues—such as facial expressions—often signal the moral fitness of a course of action. Two expressions of Germany’s former federal minister for economic affairs and energy, Peter Altmaier with a text (in German): “110 000 jobs in renewable energy”/20 000 jobs in coal industry.” A comparison between climate policy decisions by the governments of Denmark and the UK prime ministers at the time, Mette Frederiksen and Boris Johnson. A climate policy variation of the classic Left Exit 12 Off Ramp meme.


The use of common signifiers is central to evaluative layering, including a selection of publicly known actors—activists versus politicians, for example—and reliance on famous visual formats to build moral arguments. Popular meme templates, such as the Highway Exit Turn Meme (Image 15)—indicating the ethical and morally “right” path—is used to critique political inertia by contrasting politicians with scientists, a binary often employed in climate activism.
Regarding the linkages built in evaluative layering, connections are primarily presented as moral binaries, embodied by specific actors, often through memetic genres. Themes like “good” versus “bad,” casting “heroes” and “villains,” and judging “responsible” versus “irresponsible” action are visually emphasized, leaving no doubt as to which side is more legitimate. For example, in Image 12, Greta Thunberg’s contrasting expressions illustrate two options: “going to school” with a discontented frown and “saving the planet” with an approving smile. Similarly, Image 13 contrasts two political paths for German minister Peter Altmaier, using emotional cues to highlight the chosen course of action.
Images where evaluative layering is dominant create new meanings by framing climate change as a matter of morally hierarchical choices, often cast on individual actors. These images compare courses of action, often in a binary, polarized manner. The moral hierarchy is clear: there is a right and a wrong thing to do (Walker, 2024). Evaluative layering politicizes by creating sharp divisions between actors and actions deemed to belong on the “right side”—aligned with how the world should be and the necessary paths to build it—and those critiqued for their opposing values. In this sense, evaluative layered images “open up” the conflict and prompt viewers to take a stance, ultimately raising what Walzer (2002, p. 629) would call “the unavoidable political question: which side are you on?”
Conclusions
Four types of layering explained through three elements of remixing.
Each form of layering is characterized by the sampling of specific visual themes, the building of certain types of linkages and the production of novel meaning(s) that are irreducible to the parts they are composed of. These differences are not only technical but also illustrative of the different logics through which layered images can do political work. While differences in sampling are probably the most evident, a closer look at the topology of linkages and new meanings reveal that it is at least as important to look at the way these elements are put in relation to each other to unpack the overall stance of the image.
Multiple practices of layering embody what we have elsewhere (Eranti and Meriluoto, 2023) identified as the plurality of ways to politicize: different types of layering can be seen as creating both commonalities, that is, “the different logics by which a ‘we’ can be formed, and action coordinated,” and conflicts, or “ways to identify injustices and formulate and justify claims in public struggles” (p. 9). Recognizing the plurality of both commonalities and conflicts in the politicization of climate emergency on visual social media expands the understanding of the where, the who and the how of political action, allowing the analysis of the political beyond traditional arenas of politics. Through sampling, certain subjects, spaces, temporalities and objects become identified as relevant for climate politics, and their interrelations are arranged and evaluated. Acknowledging this, the approach developed in this paper helps to grasp how layering names and defines climate-related phenomena and activism, as well as its visible “ecology” of actors and actions.
Recognizing political action in visual social media can be difficult, as images are polysemous and harder to hold accountable than words. Our analysis shows that while social media images can feel like “seamless” experiences (Markham, 2013), images can also disrupt and challenge status quo, opening the field of visual politicization. As argued by Markham, remix is inherently “an invitation and provocation,” implying ethical responsibility (2017, p. 238). Layered images are not neutral, but interpretative and argumentative; they create equations, comparisons and juxtapositions that can be normative, arbitrary, or disproportionate. Layering can render complex issues like climate change recognizable and interrupt the status quo, making space for political and societal change. However, as images tend to create an illusion of “seamless” wholeness, layered images also pose a risk of being misleading and polarizing in ways that are more difficult to break down than verbal arguments. Furthermore, layering as a form of participation can also be differently accessible for different users, as demonstrated with the example of Intertextual layering. Layering can thus potentially contribute to novel types of inequalities emerging in online spheres, where new skills are needed for impactful participation.
By demonstrating the potential of using remix studies to analyze the politicization embedded in layering images, this paper establishes layering—as a type of remix—as a cultural tool of politicization, contributing to the understanding of contemporary ways of “doing society” (Luhtakallio et al., 2024), as well as expanding the existing applications of remix studies. While the analysis in this paper was conducted with climate activist images, the approach developed here can be applied to systematic analyses of layering in other visual online content, however, with sensitivity to the empirical context. Future research could expand our framework by identifying new types of layering across different forms of online activism and comparing how diverse local cultures shape those visual practices of politicization. Comparative research across platforms like TikTok or YouTube could also study the possibilities and limits of the analytical framework in analyzing audio-visual content, exploring the possibility of, for example, sonic layers, as well as the impact of generative AI tools on layering practices.
Footnotes
Acknowledgments
The authors would like to thank the PI and members of the Imagidem research team for the collective effort of developing the research method used in this article: Eeva Luhtakallio, Vasileios Maltezos, Karine Clément, Jenni Kettunen, and Jyrki Rasku. They would also like to thank the anonymous reviewers for their comments and insight, as well as the entire research community at the Centre for the Sociology of Democracy at the University of Helsinki for their support and unwavering enthusiasm in driving forward discussions on politicization and democracy.
Ethical considerations
The project (see above) has been approved by extensive ethical reviews by the European Research Council and the Ethics Committee of Tampere University.
Consent to participate
Based on this approval, no consent on the republication of publicly shared images needs to be obtained, provided that no metadata on the posts is collected or shared.
Author contributions
Juulia Heikkinen: Project lead, conceptualization, data collection, data analysis, literature review (climate images, remix studies), conclusions, editing
Taina Meriluoto: Conceptualization, data analysis, literature review (politicization), editing
Carla Malafaia: Conceptualization, data analysis, literature review (climate images), editing
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project has received funding from the following sources: European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (Grant agreement No. 804024); and the Kone Foundation; Arvonannon ekologia algoritmiyhteiskunnassa, PALJAS –Bare activism: Politics of authenticity and exposure in the margins.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
All data analyzed are publicly available online.
Notes
Appendix
Results of the ethnography-informed computational categorization. Categories by ImagiDem research team 2019–2024 (for detailed description of the method, see Maltezos et al., 2024).
Category
n
Category
n
Category
n
1. Artificial/layered
18209
4. Protest materials
3991
7. Threat
126
2. Selfies
9131
5. Crowds
2092
8. Performance
115
3. Groupies
4862
6. Meeting/Deliberation
415
9. Ambiguous/Neutral
210/3556
Total 42 708
