Atmospheric Practices: On Affecting and Being Affected

Introduction

In his book Encountering Affect, Ben Anderson (2016) notes that “there is no such thing as affect ‘itself’” (p. 13). ¹ This elucidation needs to be emphasized at a time when affect theory has swept through the social sciences for more than a decade and has done so to such a degree that it qualifies for the pronouncement of an “affective turn” (Clough & Halley, 2007). It may at first appear to be a peculiar statement, judging from the many fruitful analyses stemming from the so-called affective turn that encompasses pivotal aspects of contemporary and past society; see, for example, Anderson and Holden (2008), Anderson and Harrison (2010), McCormack (2008a, 2008b), Stewart (2007) and a reader edited by Gregg and Seigworth (2010). Yet Anderson has an important point here, we argue, since the turn has also had its side-effects.

First and foremost, in the effort to de-psychologize the concept by approaching affect as a noun (rather than an adjective or verb), “affect” has been verging on an ontological dumping. This happens despite the best intentions of scholars who stress the relationality of affect as the capacity to affect and be affected. For instance, affect becomes either “an active force, a fluid and dynamic process that is continually made and remade” (Buser, 2014, p. 232) or inherent properties of artifacts working “as the bearer of positive or negative affects” and “affect generators” (Reckwitz, 2017, pp. 120, 123). Avoiding this, Anderson instead connects his statement to another important clarification, saying that affective life is always already mediated. In this way, he sees affects as a kind of by-product located in the midst of other processes, structures, apparatuses, and collectives.

Yet in the wake of such efforts to detach “affect” from a subject-oriented psychology and engaging with the spatial implications, there have also been other side effects, for instance, a usurpation of related terms most clearly illustrated through Anderson’s own notion of “affective atmospheres” (Anderson, 2009, 2016). Another consequence of these side effects is that some of the philosophical nuances that make “affect” relevant in the social sciences are bypassed, leaving inconsistency and unfruitful vagueness in its wake. How does affect, as a noun, actually work? Are affect and emotion the same? Are affect and atmosphere the same?

It is these shifts and lack of clarity in meaning between different concepts that have inspired this article. Here, we review some of the recent social science literature on affect, particularly in geography, and show how there has been a remarkable shift from affect to affective atmospheres, while missing out the role of practices in shaping affect, which other areas of social science have highlighted (although see Duff, 2010; Reckwitz, 2017). We argue that affect as a noun or an inherent property of artifacts as “affect generators” actually falls short of an understanding of the circularity of human bodies’ capabilities of moving and being moved. To get to this, we argue that affect, as a verb and adjective, needs to be situated in practices but that these affective practices are spatially embedded and felt phenomena. In this lies a necessity to turn to the atmospheric aspects of human practices in a phenomenological perspective. Rather than viewing atmosphere and affect (verb, adjective) as something in the relation between people, place, and things, they unfold as the relation, in what we call atmospheric practices. The implication of this is that atmospheric practices highlight the constant attunement to atmospheres through the human body.

Outlining the article, we first turn to how the notion of affect has spread rapidly in social science but leaving (at least) two problematic limitations, one of distinction and one of transmission. We then turn to practice theory, to highlight the role of practices in affective phenomena but also arguing that to come to grips with the affective sphere of human life, practice theory needs a more considerate understanding of the lived body. Following this, we scrutinize another parallel genealogical turn within geography, where affect becomes affective atmospheres, and discuss the implications. Finally, we merge the perspectives of practices and atmospheres into a synthesis of atmospheric practices to show how spatially embedded practices affect and are affected by the emerging atmospheres.

Affect and Its Conceptual Problems

Affect has indeed become an inherent part of contemporary geographical vocabulary—theoretically and empirically. In much of the empirical work, it appears in the form of the conceptual pair affect and emotion, used together, interchangeably, or as one simply, unconsciously denoting the other to avoid having to define clear distinctions. When more elaborate definitions are offered, they are, for the most part, connected to the posthumanist thinking in Gilles Deleuze’s biophilosophy, often translated/interpreted by Brian Massumi (2002). This understanding rests on a distinction between the notions of affect, feelings, and emotions. Affect is seen as field of intensity that is “pre- and postcontextual as well as pre- and postpersonal, an excess of continuity that is invested only in the ongoing of its own” (p. 217). Affect in this sense is an autonomous and self-generating movement, primarily thought of as processes of circulation, flow, transmission, or contagion—most notably seen in Teresa Brennan’s The Transmission of Affect ( 2004). Drawing on Massumi, Derek McCormack (2008a) constructs the distinction as between “affect (as a prepersonal field of intensity), feeling (as that intensity registered in a sensing body), and emotion (as that felt intensity expressed in a socio-culturally recognizable form)” (p. 426). Despite flirting with divergent approaches such as phenomenology, psychobiology, and Darwinism, Nigel Thrift (2008) is also inclined to this pre and post line of thinking, as demonstrated in his synoptic statement “Affect is understood as a set of flows moving through the bodies of human and other beings” (p. 236)—whereby affect becomes an autonomous phenomenon. While this body of literature has been persuasive because of its neatness in definition and its emphasis on bodies and materialities, it has its limitations in its posthumanist ontological basis tending toward the old problem of naturalism (with its ontological and methodological equalization of nature and human beings). The first limitation is what we can call the problem of distinction, in particular the distinction between affect and emotion. Explicit or implicit critique of the conceptual division of affect and emotion has been forwarded, for instance, by Simonsen (2007), who suggests an understanding of emotion that attempts to transcend the boundary between the two (see also Anderson & Harrison, 2013). Pile (2010) has also addressed the division into discrete emotional and affective geographies, which he considers unproductive and tends to reify both concepts. Similarly, both Bondi and Davidson (2011) and Edensor (2012) have suggested that the conceptual boundaries are unproductive and should remain amorphous and elusive.

In line with the above critique of distinction, the points we want to pursue here are more internal to the conceptual formulation—arguing that it tends to reinstall the distinctions between body and mind, or between object and subject, that many of the authors set out to overcome initially. The distinction tends to reduce the two concepts (i.e., affect and emotion) to a pure physical/biological process and a cognitive representation, respectively. On the one hand, this dichotomous model does not recognize how that which is emotionally experienced is not necessarily conscious and cognitive; experiencing is, to a large degree, a bodily process, and past experiences become incorporated into bodily memories that influence present reactions and might—or might not—be linguistically expressed. On the other hand, the distinction severs emotions from the lived experience of being and having a body. Emotions involve sensation and bodily impressions and expressions (see also Ahmed, 2004, 2010; Harris & Sørensen, 2010; Leys, 2011). Consequently, bodies and sociolinguistic expressions are inseparable and intertwined. What we need is an understanding that can grasp the blurred borders between the two concepts, including how and when they come to matter and are expressed in their empirical contexts (Navaro, 2016), and the wholeness of the emotional–affective field with the internal relations of its parts. Anderson, initially one of the main proponents of the distinction (see Anderson, 2006), also later attempted to unsettle it. When first discussing his idea of affective atmosphere (Anderson, 2009) as a solution to the problematic distinction, he acknowledges that it is caught up in the subjective/objective issue that it originally sought to overcome. Later on again, Anderson (2018) even questions the more naturalist side of the approach: “I’ve been cautious about the authority claims vested in science and how those authority claims bolstered a particular kind of ontological treatment of affect” (p. 231). The distinction is unsettled but only by turning to another concept of affective atmospheres, discussed later.

If we turn to Thrift’s (2008) previous statement on affect as “a set of flows moving through the bodies of human and other beings” (p. 236), it directs our attention to the second problem we want to discuss: the problem of transmission. The statement depicts affect as a kind of autonomous social and physical/biological energy, flowing through humans and nonhumans, like waves in the sea. It suggests “that a self-contained packet of emotional stuff is being transferred wholesale from one body to another” (Wetherell, 2012, p. 141). An alternative term for transmission has been circulation, but it is no more helpful in answering the basic questions that arise from the theories and definitions of affect: How does the transmission of affect proceed? How can a physical state move from one body to another? And, as also raised by Wetherell (2015), what are the limits of affective transmission? How and why does it stop? And why does it seize some bodies and not others?

Some authors seek to overcome this problem of transmission by way of naturalistic metaphors such as “affective contagion” spreading like a wildfire (Thrift, 2009, p. 88). For example, Brennan (2004) turns to processes of “chemical entrainment” to explain how affect become distributed across bodies. The chemical agent is “pheromones,” chemicals emitted by one body that cause the release of hormones in the blood of the other body, thus leading to a change in the somatic state. However, as a solution to the problem of transmission, the idea of affective contagion leaves much to be desired. It is based in a biological concept of the body as a “machinic assemblage,” where the human, according to Ansell-Pearson (1999), is implicated in a “postbiological evolution” as part of its very definition (p. 216). This naturalistic account, based on biological and chemical understandings of human bodies, again reifies the bodies and tends to remove them from their social and political contexts. It also disregards their social practices, since practical human meaning making as well as cultural processes and differences disappear from the analysis. What might be intersubjective reactions, such as a reunion of a family or football fans getting worked up, is reduced to a physical/biological effect, much like a contagious yawn, and the question of transmission remains somewhat mysterious.

Practice Theory and Emotion/Affect

To avoid these problems, some authors turn to practice theory. Prominent examples are Reckwitz (2012, 2017), who aims to develop what he calls a praxeological perspective on affect, and Wetherell (2012, 2015) who presents the concept of affective practices.

Practice theory characterizes a bundle of writings that pin down practices as the primary or generic element of social analysis, and some authors even talk about a practice turn in contemporary theory (Schatzki, Knorr Cetina, & von Savigny, 2001). Underlying this has often been the desire to push social theory beyond current problematic dualisms such as objectivism/subjectivism or structure/agency (e.g., Bourdieu, 1977, 1990; Dreyfus, 1991; Giddens, 1984). The first wave of practice theories within human geography followed this line of thinking—–see, for example, Thrift (1983), Pred (1984, 1986), and Simonsen (1991). Remnants of this wave still inform practice thinking, but they have developed in radically different directions; for instance, in his books Spatial Formations (1996) and Non-Representational Theory (2008), Thrift demonstrates his roots in theories of practice, stating that “non-representational theory concentrates, therefore, on practices, understood as the material bodies of work and style that have gained enough stability over time” (Thrift, 2008, p. 8). But on the other hand, he simultaneously moves toward the more objectivist ontologies of posthumanism and the physical/biological concept of affect as discussed above.

Another direction can be found in a new strand of practice theory mostly espoused by the American philosopher Theodore Schatzki (Everts, Lahr-Kurten, & Watson, 2011; Gram-Hanssen, 2010, Reckwitz, 2012; Shove, Pantzar, & Watson, 2012). Since some of the emergent discussions arguing for practice theory as a starting point for understanding affect/emotion put their faith in this approach, we will dwell a moment here (see, e.g., Everts et al., 2011, Reckwick, 2012; Wetherell, 2012, 2015). Like his predecessors, Schatzki tries to develop a social ontology that “transcends rigid action-structure oppositions” (Schatzki et al., 2001, p. 1) and treats practices as the key element for this development. However, he moves a step further, formulating what he calls a site ontology (Schatzki, 2002), seeking reconciliation between “theories of arrangements” (forwarding poststructuralist and posthumanist ontologies) and practice theories while, at the same time, clarifying their differences. For Schatzki, practices are nexuses of human activity, open-ended sets of doings and sayings organized by understandings, rules, and “teleoaffectivities,” which aim to link the doings and sayings of a practice. The fundamental units to analyze social phenomena consist accordingly of bundles of practices and arrangements. To clarify his diverging from theories of arrangements, he states,

Latour’s associations bear an obvious resemblance to my arrangements. His account, however, recognizes no pendant to what I call practices. On my account, social affairs consist, not just in connected associations as Latour holds, but in linked arrangements and practices. (Schatzki, 2015, p. 3)

Particular emphasis in the exercise of building affect or emotion into practice theory has been placed on Schatzki’s (2002) concept of teleoaffective structures, defined as

a range of normativized and hierarchical ordered ends, projects and tasks, to varying degrees allied with normativized emotions and even moods. (p. 80)

This definition indicates that coordinated within the teleoaffective structuring are emotions and moods that participants should, or might, enjoy. This idea, however, is not developed in Schatzki’s writings, and the notion of teleoaffective structures is more likely intended to analyze normative connections between doings and sayings rather than elements of people’s emotional life. A quick inference would be that practice theory is not very helpful when it comes to the analysis of emotional life, and there are scant examples from empirical work.

But examples do exist. One notable case is conducted by Jackson and Everts (2010), who argue that geographies of emotion could benefit from theories of practice by adding the ways in which emotions are practiced to their agenda—how being emotional is learned and unlearned and how affect resonates with practical understandings of knowing how to do things or how to proceed (Everts et al., 2011). The analysis by Jackson and Everts (2010) is about anxiety as social practice. They start from practice theory but choose to go behind the theory, to its ontological foundation in Heidegger’s (1996) existential phenomenology. Here, they can draw on his basic theory of “Angst” in everyday life and his idea of the “thrownness” of the human being into the material world and our practical ways of dealing with it. For them, “anxiety is not some free-floating mental activity; it is embedded in specific (often complex) doings and sayings. Anxieties are embodied and social, practical and practiced. Like other social practices, they are routinised, collective, and conventional in character” (Jackson & Everts, 2010, p. 2801).

Another example can be taken from the work of Wetherell (2012, 2015) and her idea of “affective practice.” Drawing on Schatzki and Reckwitz, she characterizes affective activity as a form of social practice and highlights how individuals—as bodily and mental agents—act as carriers of practices. In relation to affect, she argues,

The unit of analysis for social and cultural research on affect, is not some kind of inarticulable, momentary, spurious, hard-to-detect, pre-conscious judder. It is affective-discursive practice, or that domain of social practice which bears on and formulates the conduct of activities we conventionally recognise as making psychological and emotional sense, and, in the process, making psychological subjects and emotional events. (Wetherell, 2015, p. 152).

Empirically, this approach is used to analyze what she and her colleagues have called the “affective-discursive canon,” found, for instance, in print media coverage of Aotearoa, New Zealand’s national day. They are here investigating how affective-discursive practices are mobilized to “cover the nation” and “settle space” (Wetherell, McCreanor, McConville, Barnes, & le Grice, 2015, pp. 59-62). With the notion of “affective practice,” she, like Jackson and Everts (2010), in this way grounds affect (and/or emotion) in practice. However, in her move toward affective-discursive practices, and in the empirical display of that move, she twists the emphasis toward “concepts derived from discourse studies in social psychology” (Wetherell et al., 2015) such as “affective-discursive canon,” “affective-discursive positions,” and “repertoires.” In this way, she turns to a more social-constructionist account and subsequently downgrades the emphasis on bodies and materiality otherwise prioritized in the aforementioned approaches.

We appreciate the emphasis on practice formulated in these examples, and our argument would be that even if practice theories have been largely silent around issues of emotion and affect, they do have the potential to make a contribution. They open up the possibility of seeing humans not as affective dopes—as “schools of fish briefly stabilised by particular spaces, temporary solidifications which pulse with particular affects” (Thrift, 2008, p. 236)—but as practitioners of emotions, embedded in specific social, spatial, and temporal contexts. We would, however, argue that to fulfil these potentials, it is necessary that the theories are developed with a stronger understanding of the lived body. Many practice theories refer to the role of bodies, but most of them do not develop a theoretical understanding of them. Schatzki (2015) himself is aware of this deficit when he writes,

I do not claim that objective configurations and interwoven path-place arrays exhaust the spatial features of practice-arrangement bundles. Neither phenomenon, for instance does justice to the spatiality of the lived body. (p. 2)

To confront the paucity of detailed dealing with the lived body, it is necessary, once more, to go behind practice theories and dig into their phenomenological inheritance—this time to Merleau-Ponty’s (1962, 1968) sensuous phenomenology of the body. ²

Lived Bodies and Emotions

Merleau-Ponty’s phenomenology is based on a nondualist ontology of the body and its environment. The body is not just an object in the world as studied by science, and neither is it a mere assemblage or juxtaposition of its parts; it is a sensuous, lived body that changes through interacting with an environment that it both responds to and actively structures. This body is always in process; it emerges in its perceptual engagement with the world and reaches out to, and questions, its environment. Perception, then, is not an inner representation of an outer world; it is an opening into that world. This emphasis on engagement and participation indicates a strong affinity with the theories of practice. Meaning and significance are created in practice. The practically oriented body continuously weaves meaning throughout its life course, and its own capacities materialize through its interactions with others and with its environment. This understanding of the body underlines its relational character; it is radically intercorporeal and concerned with coexistence—the integration of the body into the order of things and the unfolding of collective life.

Merleau-Ponty (1968) further highlights the intertwining of the body and the world—he calls this common element the flesh. The body as self-sentient flesh becomes part of the flesh of the world but is not reducible to it. Probably the most important aspect of the flesh is its reversibility, a “double sensation” by which the body-subject’s practices and perceptions are connected in an interworld or “intermundane space.” Because of this reversibility, the body is simultaneously passive and active. It is at the same time seeing and seen, creative and receptive, affecting and affected. Although this latter pairing immediately sounds similar to how affect theory sees affect as “the capacity to affect and to be affected” (e.g., Barnfield, 2016, p. 3), we argue that there is a prominent imbalance toward a passive subject being exposed to the flows moving through people, places, and things.

The reversibility is repeated in Merleau-Ponty’s ideas on emotion, and in our opinion, it provides an input that can advance the current debate about emotion and affect. An elaboration of Merleau-Ponty’s nondichotomous approach cuts through the distinction between emotion as subjective experiences or significations, and affect as an impersonal “set of flows moving through the bodies of human and other beings.” It starts from the general phenomenological insight that we are never “un-touched” by the world around us—we are always already attuned to it. Situatedness, mood, and the collapse of the distinction between “inner” and “outer” are crucial aspects of Merleau-Ponty’s idea of emotions. Emotions are neither purely mental nor purely physical but are ways of relating and interacting with the surrounding world. The problem of distinction, elucidated above, is with Merleau-Ponty, thus not even a distinction to begin with.

This relational account gives rise to two sides of an emotional spatiality (Simonsen, 2007): an expressive space and an affective space. The expressive space can be seen as a performative element of emotion. Emotions are something practiced and shown, enacted by the expressive and communicative body. The body, Merleau-Ponty (1962) argues, is comparable to an expressive work of art but one that expresses emotions in the form of living meaning. Emotional meanings are “secreted” in bodily gestures in the same way that musical/poetic meaning is “secreted” in a phrase of a sonata/poem. These meanings are communicated and “blindly” apprehended through corporeal intentions and gestures that reciprocally link one body to another. Emotional experience is something public and “in-between,” situated in the perceptibility of bodily gestures—as when Merleau-Ponty says,

Faced with an angry or threatening gesture, I have no need, in order to understand it, to recall the feelings, which I myself experienced when I used these gestures on my own account. . . . The gesture does not make me think of anger, it is anger itself. (p. 184)

The other emotional spatiality is affective space, which is the space in which we are also emotionally in touch—open to the world and its capacity to affect us. This means that emotions are not just actions, something that our bodies express or articulate. Another aspect of them is how we are possessed by them or swept into their grasp—for example, when experiencing or appreciating a work of art or being moved by a beautiful landscape. It is related to the experience of the world around us, to things, rooms, and architecture but also feelings elicited—for example, being gripped with fear, overcome by shame, or filled with joy—which is imprinted in the sense of the body as contracting, light and uplifted, or tense. In that way, it is the felt sense of having been moved emotionally, the more passive side of emotional experience.

Together, these two sides of emotional spatiality are related in active/passive circularity—in the form of mutual implication (Cataldi, 1993). Emotions are neither an expressive nor an affective phenomenon—they are both simultaneously (see also Harris & Sørensen, 2010). Furthermore, emotions are essentially public and relational; they are formed in the intertwining of our own bodily flesh with the flesh of the world and the flesh of others. Merleau-Ponty’s (1962) stand on emotions, then, is related to his ideas of bodily doing and perception as well as his interest in aesthetics and, later, his flesh ontology. They take the form of the whole register of different emotions—such as love, desire, hate, fear—and in their reversibility they relate to human, as well as nonhuman, parts of the flesh of the world.

Together, the emphasis on practices and the understanding of the lived body provide us with a nondualist ontology of the body and its environment, and the active/passive circularity in the understanding of emotions unsettles the distinction between affect and emotion. The reversibility and the expressive/affective duality of embodied emotions render the two concepts interconnected and indivisible. This provides us with the double concept of emotions connecting expressive space and affective space and involves practices of active, living bodies intertwining with their social and material environment. This is also seen in the later development of affect theory, which highlights the felt space—the atmosphere—to which we now turn.

From Affect to Affective Atmospheres

Recent theories of affect have, to a greater degree than before, highlighted the role of the material world. The context has, for instance, been “a vital element in the constitution of affect” (Thrift, 2004, p. 60), yet nonetheless, to emphasize the material elements even more, there has been a turn from the concept of affect to atmospheres—or, more precisely, a conceptual reorientation achieved by making a noun into an adjective: affective atmospheres. This can be seen as an attempt to deal with the issues discussed above surrounding the conceptual problems in the notion of affect.

The turn to atmospheres in current geography largely finds its philosophical base in the writings of the German philosopher Gernot Böhme. In the mid-1990s, Böhme drew attention to how the material environment is felt (Böhme, 1993, 1995; see also Bille, Bjerregaard, & Sørensen, 2015; Bollnow, 1941; Hauskeller, 1995; Riedel, 2019; Runkel, 2016; Tellenbach, 1968). In this line of thinking, atmospheres have emerged as a particular lens for understanding people’s affective engagements with the environment. Böhme’s (1998) starting point is phenomenological (taking inspiration from Schmitz, 2014; Schmitz, Müllan, & Slaby, 2011); he conceptualizes atmospheres as the copresence of subject(s) and object(s). Their ontological status is vague—belonging not in the human psyche alone or in the material surroundings but having to be felt by a sentient being as a “subjective fact” or “half-things” (Böhme, 1998, p. 114, 2001, p. 59; Schmitz, 2014, p. 39)—different from a table, for example, but no less real. As Tonino Griffero (2014) writes, “Why on earth, in fact, should solid and contoured bodies be more real than vague entities, which we experience without referring them to solidity, such as fluids, gas processes or even quasi-things like atmospheres?” (p. 10). The challenge then is to describe “a vague entity in a precise way” (p. 7).

Like Merleau-Ponty, Böhme (1993) is highlighting the issue of perception, and according to him, the primary “object” of perception is atmospheres. What is first and foremost perceived is neither sensations nor shapes, or objects or their constellations, but atmospheres (see also Thibaud, 2003). Perception thus “involves not only discerning objects in the environment, but experiencing the state of the medium at a given time” (Thibaud, 2011, p. 212). The consequence is that atmospheres are not in or properties of things but with them. Atmospheres are thus a “spatial carrier of attunement” (Böhme, 1995, p. 29, our translation), and “without the sentient subject they are nothing” (Böhme, 2013, p. 103, our translation).

Although these discussions emerged throughout the 1990s and 2000s in geography, particularly with Jürgen Hasse (2002, 2008, 2014, 2016), the term affective atmosphere has become particularly prominent in Anglophone geography with Anderson’s (2009) article of the same name. Here, he discusses how this term enables us to reflect on affective experiences by unsettling the distinction between emotions as personal and subjective, and affect as impersonal and objective. Through Böhme and Mikel Dufrenne, he argues that “atmospheres are spatially discharged affective qualities that are autonomous from the bodies [italics added] that they emerge from, enable and perish with” (p. 80).

The term affective atmosphere allows, Anderson (2009) argues, for more focus on the collective than on the personal, in that they are impersonal by belonging to “collective situations, and yet can be felt as intensely personal” (p. 80). Anderson also draws on McCormack’s work (2008a), where the term affective is used to distinguish atmosphere in its meteorological sense from its more experiential sense. In essence, it is argued, affective atmospheres entail a copresence of objects and subjects in places, stressing the collective rather than individual experience. There has since been a widespread use of the term affective atmospheres (discussed in Gandy, 2017) to highlight everything from cities at nighttime (Shaw, 2014), to constructing homes (Pink & Mackley, 2016), to conflict (Fregonese, 2017), to nationalism (Stephens, 2016). The latter study highlights, for instance, how the notion of affective atmospheres allows us to reflect on the “unpredictable affective encounters that cannot be traced back to the feelings or emotions of an individual or (national) group” (p. 185). Stephens further argues how foregrounding affective atmospheres places “affective feelings centre stage and invites us to address how they matter politically” (p. 182). In this way, the term helps us understand “the transmissions between the singular and the collective” (p. 185). It becomes clear that in the process of turning toward atmosphere, à la Böhme, the noun affect has turned into the adjective affective.

As is evident, there are at least two different sets of arguments involved in the turn from affect (as a noun) to affective (as an adjective qualifying atmosphere). First, there is the distinction it makes from the supposedly original notion of atmosphere as a meteorological concept (but see a critique of this in Riedel, 2019). Second, it highlights the collective, rather than individual, aspects of experiences and thereby seeks to unsettle the distinction between emotion and affect.

In both cases, atmospheres are seen as relational yet autonomous (Anderson, 2009; Buser, 2014), emerging from practices, bodies, and materials. As Anderson (2009) argues, “Affective qualities emanate from the assembling of the human bodies, discursive bodies, non-human bodies, and all the other bodies that make up everyday situation” (p. 80). Ash (2013) takes the notion of affective atmospheres even further when exploring how interobject relations shape affective atmospheres as equally important as those shaped through relationships between subjects and objects. Whereas in the recent spate of geography literature on affective atmospheres, atmospheres are seen as foregrounding the body in space (Duff, 2016), Ash (2013) radically departs from this “clause of subjectivity” (Sørensen, 2015), in which there has to be a corporeal subject for an atmosphere to exist. To Ash, the focus is more profoundly on the affective qualities of objects to shape atmospheres.

The question is, however, whether an affective aspect is not inherent in the very notion of atmosphere, which, according to Böhme (1993), is a premise for being-in-the-world as the first object of perception. Are we not simply talking about the same thing in different ways—a reformulated version of affect as atmosphere? For instance, Cameron Duff (2010) writes that “affects are autonomous in this sense, in that they reside neither in individual places nor in individual bodies but rather in the dynamic and relational interaction of places and bodies” (p. 886). Similarly, Brennan’s (2004) work on the transmission of affect actually starts off with the rhetorical question “Is there anyone who has not, at least once, walked into a room and ‘felt the atmosphere’?” (p. 1). And Navaro-Yashin, drawing on Guattari in her work on affect, where she highlights the necessity of situating affect in concrete ethnographic settings (Navaro, 2016), also coins terms such as “melancholic objects,” for “things which exude an affect of melancholy,” and “spatial melancholia, an environment or atmosphere which discharges such an affect” (Navaro-Yashin, 2009, p. 16). Here affect, following Guattari, is defined as “hazy and atmospheric” (Navaro-Yashin, 2009, p. 13). In the above, we thus see how the reconceptualized version of affect as a flow, as intensity, and as a collective, rather than an individual phenomenon, is defined by reference to, or even more precisely by, the term atmosphere. “Affect” has turned to “atmospheres” in half a decade.

Atmospheric Practices

Many of the above scholars draw on Böhme in particular to achieve the spatialization of affect through the term affective atmosphere as the copresence of subject and object. Yet, unlike the phenomenological tradition that Böhme draws on, they insist on adding the adjective affective to atmosphere, as if there are atmospheres of this existential kind that are not affective or simply to distinguish it from a meteorological one. In reading the work of Böhme and earlier scholars, such as Tellenbach (1968), or those working with the related notion of Stimmung (Bollnow, 1941; Heidegger, 1996), ³ the adjective mostly appears in the form of a pleonasm, since atmospheres are by definition, in this line of work, affective (see also Riedel, 2019). We may not be able to verbalize what kind of affect is experienced through an atmosphere, but it is affective nonetheless (Böhme, 2006). Similarly, it has also been questioned to what extent it makes sense to claim that atmospheres are relational, because the premise, at least in the phenomenological tradition, is that atmospheres are the first object of perception (Böhme, 1993, p. 125) and even actually condition perception (Thibaud, 2003).

We will therefore suggest even more pragmatically that atmosphere, at least in this existential sense with a lived body à la Merleau-Ponty (1962), is not in the relation; atmosphere is the relation (Bille, 2015a). Atmospheres are the conditions for perceiving the world, in the sense that people are always already attuned to them (Heidegger, 1996, p. 126). By being embedded in material worlds, with or without other people present, atmospheres also emerge in socio-material contexts with affective qualities, potentially surpassed by people’s ability to affect their environment. Hence, one can also ask what the attention to the “affective” in Anglophone geography has contributed to, that the phenomenological notion of atmospheres did not have the potential to answer, and doing this without leaving us with the problem of transmission and distinction. For example, the problem of transmission is bypassed by atmospheres, as Griffero (2014) notes, because “atmospheres appear and disappear, but we cannot sensibly ask ourselves where they have gone and how they have existed in the meantime” (p. 119).

The atmosphere literature following Böhme very much centers on perception of the world—that is, how atmospheres shape the experience of situations and spaces. Seen in relation to the emotional spatiality we developed above, this corresponds to the affective space, or the passive side of emotional experience. The other aspect was the expressive space, seeing emotions as something shown, made, and practiced by an active human body. This other, active side is associated with the turn toward practice theory and the lived body—in this context drawing attention to the issue of affective practices.

This is not to say that all the research on atmosphere has ignored such practices. Bille (2015b, 2017, 2019) has, for instance, concentrated on practices of attuning spaces through light, and Riedel (2015, 2019) and Sørensen (2015, 2016) have focused on atmosphere and movement; Rob Shaw (2014) similarly notes that

as atmospheres emerge from bodies, they are connected to the activities of those bodies: what Edensor labels the “reiterative practices” which “coproduce” atmosphere (Edensor, 2012). Perhaps contrarily, the atmosphere helps “ground” affect, by showing how experiences which may appear ephemeral, loose and floating are often associated with material practices, bodies and places. (p. 89)

And Duff (2010), in his work on youth practices in Vancouver, Canada, drawing on Edvard Casey’s notion of thick places as the imbrications of affect, habit, and meaning, states that

practices like walking, socialising, skateboarding, and breakdancing provided participants with various means of producing place. However, it was the affective force of these practices that contributed most directly to the production of thick places. (p. 882)

Yet Duff is also in a sense fixating the atmosphere, contrary to Griffero’s (2014) promotion of vagueness, when he states that “one is affected by the atmosphere, the potential in place, before one investigates its value as a space of and for practice” (p. 891) and “one is affected by place before one might be affected in place through one’s practices and habits” (p. 892). In other words, atmospheres become the property of a place rather than emerging from it.

While largely in line with Duff, we want, however, to promote a more active person who is affecting and being affected through the practices of the lived body taking part in atmospheres. It is a theoretical synthesis between the two lines of thought of affective practices and affective atmospheres that allows us to add a more active bodily element, seeing atmosphere as—and not just in—the relation constituted in the affective duality between materiality and somatic and social practice. In this way, atmosphere is not only something humans feel, or that conditions perception, but it also simultaneously positions the felt space as something humans do. There is a continuum of bodies in architecture, space, and embracing sensuous elements that is shaped through the emotional spatiality of practices that enables us to study the affective aspects of human lives through a decentering of the human being—as also seen in practice theory’s emphasis on practices rather than individual motivations—without the posthumanist biological ontology of affect. Merleau-Ponty’s (1962) reversibility helps us underline the importance of dealing with this lived, more than a biological, body in the understanding of atmosphere.

Turning to atmospheric practices is also to highlight another element: Although most of the literature on affective atmospheres focuses on how “they are impersonal in that they belong to collective situations and yet can be felt as intensely personal” (Anderson, 2009, p. 80), and highlights how crowds are shaped by and shape atmospheres (Borch, 2009, 2014; Runkel, 2018), practices of cooking, running, reading, and so on, are also atmospheric practices of affecting and being affected performed in solitude within a world of material objects, without any direct affective relation to other people.

The implication of attending to atmospheric practices beyond atmospheric perception is to see human bodies and materiality as attuned with affective potentials, where the resulting atmospheres are affective forces. It is attending to mundane as well as spectacular situations, collective or in solitude, that has an impact on our lives. The practice of running, for instance, becomes different atmospheric practices qualifying the act of moving your legs at a certain pace. Running in a competition solicits a different atmosphere from that evoked by jogging, interval training, or social running. That is, not only does the atmosphere affect the mood and intensity of the running, so too does the runner affect the feeling of a space depending on outfit, speed, time, and numbers. By attending to atmospheric practices and atmospheres as practiced, the cultural logics of a sensing, lived body as well as sensuous norms and normalization become a starting point for analysis. What atmosphere is sought in everyday domestic life? How do practices fail or succeed? How does that affect people in that space? What objects and services are consumed to shape diverse atmospheres? What atmospheric norms require what types of practices? And how can different practices compete in shaping atmospheres?

Concluding Remarks

We began this article by reference to Anderson’s perspective that there is no such thing as affect itself. Definitions of affect have shifted, being problematic and apparently difficult to sustain. Yet from our perspective, the most problematic aspect is the lack of human agency and practices of the lived body in the formation of affect as a noun with an autonomous ontological status. That does not, however, disqualify the idea of highlighting the capacity of individual and collective bodies to affect and to be affected. To develop this idea with a focus on the spatiality of the lived body, we have followed two lines of literature: first, the posthumanist theories of affect, in particular the way in which they have developed within geography, and, second, the emergent interest in taking up affect and emotion within practice theory.

In relation to posthumanist affect theories, we have experienced different problems related to the autonomous, free-floating status ascribed to the concept. This is probably also part of the background for a rather conspicuous shift that can be traced in the conceptual discussion. The central premise of the original theory, emphasizing flows, circulation, and transmission, seems to have shifted within geography—particularly visible in Anderson’s (2009, 2016) work—toward turning affect into an adjective qualifying atmosphere. Seeing the problems with the concept of affect, this collusion tends to render superfluous the concept as an autonomous “thing”. We do, however, appreciate the introduction of, and focus on, the concept of atmosphere. This development has ascribed a much stronger role to the material world and collective, sentient beings. While acknowledging this move, however, we also think that considerable parts of the literature on affective atmospheres share another problem with the literature on affect: They tend to exaggerate the passive side of emotional spatiality and undervalue how both emotions and atmospheres are also something people do. That is why we turn toward practice theories to achieve a more performative dimension of the emotional/affective field.

In relation to practice theories, we are particularly appreciative of the contributions offering concepts such as “emotion as social practice” (Jackson & Everts, 2010), “affective practices” (Wetherell, 2012), and “affective habitus” (Reckwitz, 2012). We do, however, still argue that to fulfil these potentials, the theories need to have a better understanding of the lived body entangled in space in order to understand how perceiving atmosphere involves feeling and sentient bodies. Such a theory will offer a fuller understanding of shifts in how places are felt, the politics that this entails, and the kinds of geography evolving through lived bodies. This is what we obtained by considering the phenomenal body of Merleau-Ponty’s (1962, 1968) work and deriving from that the active/passive circularity of emotions. This combination of practice theories, sensuous phenomenology, and the theories of atmosphere highlights how atmospheres are created by both materiality and the presence and practices of people. Here, there are bodily practices of affecting and being affected, as well as material qualities that might be affective. In this version, practice theory has the potential to add to the active side of the constitution of atmospheres as a condition for perception.