Genres as Social Affect: Cultivating Moods and Emotions through Playlists on Spotify

Abstract

This article bridges work on media technologies and affect theories through an analysis of how users appropriate playlists on Spotify. Our study draws on 30 interviews with users of music streaming services in Costa Rica and an analysis of their accounts on these platforms. We discuss how users create playlists as a means to cultivate affect. The notion of cultivation stresses the dynamic and ritual work involved in producing, capturing, and exploring moods and emotions. We argue that playlists work as “genres”—fusions of musical substance, sociotechnological assemblages, and sociomaterial practices—to respond to the exigencies of affect. This helps turning the platform into an obligatory intermediary in the establishment of a utilitarian relationship between users and music. We draw on Berlant’s notion of “intimate public” to analyze how playlists form the basis of collective experiences that serve Spotify’s political-economic project. As material embodiments of cultivated affect, playlists offer a promise of identification and belonging to “intimate publics” formed by strangers through the specific bonds between music, technology, and affect they enact (as genres). Our analysis concludes with a discussion of the implications of our study for rethinking the relationship between technology, affect, and genre.

“Gabriela” is a 21-year-old sound engineer based in San José, Costa Rica. Music occupies a good part of her day, both for professional and personal reasons. In 2014, she created her own account on Spotify. Although she also uses a wide range of social media platforms, Spotify is, by her own admission, the only app that she is willing to pay for. During an interview we conducted for this project, “Gabriela” was hard pressed to find something she did not like about the platform. She discussed extensively what made the service so appealing to her. The following quote summarizes what she mainly uses Spotify for:

I made a playlist that was called “Nostalgic Jams,” which is a playlist that is simply like nostalgic vibes. I made a little cover for it and it is a playlist similar to R&B [rhythm and blues], but slow and nostalgic. For me, that’s like a genre or subgenre that I have to conceptualize, which I named and made public.

“Gabriela’s” comments provide us with an opportunity to present four issues discussed in this article. First, we elaborate on how music “is a cultural form that has strong connections to emotions, feelings, and moods: the domain of affect” (Hesmondhalgh, 2013, p. 11). Specifically, this article discusses how users turn to Spotify as a means to cultivate moods and emotions. The notion of cultivation stresses the dynamic and ritual work involved in producing, capturing, and exploring moods and emotions. It is also meant to stress how music and affect mutually constitute each other (DeNora, 2000).

Second, “Gabriela” reveals the centrality of playlists in how users experience music streaming services nowadays. On its support website, Spotify (2019) promotes playlists as “collection[s] of music. You can make them for yourself, you can share them, and you can enjoy the millions of other playlists created by Spotify, artists, and fans.” Dias, Gonçalves, and Fonseca (2017) define playlists as “ordered sequence[s] of songs meant to be listened to as a group” (p. 14379). We show that creating playlists requires a set of practices and technologies to materialize affect into an artifact and thus cultivate moods and emotions (Orlikowski, 2007).

Third, like “Gabriela,” we argue that playlists on music streaming services can be theorized as “affective genres” or fusions of musical substance, sociotechnical assemblages, and sociomaterial practices that respond to the exigencies of affect. We depart from standard views of genres that center on the properties of (musical) texts and extend work in various fields that defines genres as cultural categories and actions (Lena, 2012; Miller, 1984; Mittell, 2004).

Finally, we examine how playlists can form the basis of collective experiences. We contend that when playlists are shared for public consumption, they can offer a promise of identification and belonging to “intimate publics” (Berlant, 2008) formed by strangers, through the specific bonds between music, technology, and affect they enact. The formation of “intimate publics” favors Spotify’s political and economic project that promotes playlists as a means to build a utilitarian relationship with music, that is, one that conceives of users as self-governing subjects who control themselves through moods and of music as “situational and functional for certain activities” (Eriksson, Fleischer, Johansson, Snickars, & Vonderau, 2019, p. 123).

Music consumption has shifted to digital services and platforms. Accordingly, scholarship on streaming services has flourished over the past years (Johansson, Werner, Åker, & Goldenzwaig, 2017; Nowak, 2016; Prior, 2018). Many studies tend to privilege a “top-down” perspective that focuses on how content is provided to audiences to guide their consumption behaviors (Anderson, 2015). As a supplement, we examine how users build genres from the “bottom-up” to cultivate affect. In this way, this article contributes to scholarly dialogues between social media and technology research and affect theories (Döveling, Harju, & Sommer, 2018; Hillis, Paasonen, & Petit Michael, 2015; Papacharissi, 2015).

In what follows, we elaborate on the four issues introduced above. Our study draws on interviews with 30 users of music streaming services in Costa Rica and an inductive analysis of their accounts and profiles on these platforms. We conclude with a discussion of the implications of our study for rethinking the relationship between affect, genre, and technology. We show how users experience the development of playlists as a need and thus incorporate them as an obligatory means of making sense of their moods and emotions.

Theorizing Playlists as Genres

Playlists have gathered increasing scholarly attention (Dias et al., 2017; Eriksson & Johansson, 2017; Hagen, 2015). As Dhaenens and Burgess (2018) note, “the playlist has become a primary means to discover, order and share music” (p. 2). Playlists are a central component of music streaming services’ business model. Eriksson and colleagues (2019) contend that playlists “occupy a central role in Spotify’s strategy for attracting advertisers [. . .] [They ensure] that advertisements can be delivered to particular groups of users at particular points in time” (p. 137). Playlists can also operate as means of self-performance, building culture, expressing political opinions, forming social relations, and fostering certain kind of (govern)mentality (Dhaenens and Burgess, 2018; Gorzelany-Mostak, 2015; Hagen & Lüders, 2017; Johansson et al., 2017).

To contribute to this body of work, we argue that playlists are best defined as (affective) genres. We depart from traditional theoretical approaches to genres that center exclusively on the properties of (musical) texts and build instead on work that emphasizes how they operate as cultural categories (Mittell, 2004). Lena (2012), for example, distinguishes between music styles, that is, “musical idioms (e.g. polka or techno)” and music genres or “systems of orientations, expectations, and conventions that bind together industry, performers, and fans in making what they identify as a distinctive sort of music” (p. 6). This approach emphasizes the cultural work that genres accomplish: they help to define practices and interpret their cultural significance.

In a similar manner, Miller (1984) contends that definitions of genre must center not on the “substance”—or “the semantic aspects of discourse” (p. 159)—or form—how substance is symbolized—but on the action that it achieves, “on the action it is used to accomplish” (p. 151). A genre typifies social actions through repetition by blending substance and form. These actions emerge as a response to what Miller (1984) describes as exigencies or “objectified social need[s]” (p. 157). Miller envisions context as a level of meaning that enables these exigencies. In her words, “Genre refers to a conventional category of discourse based in large-scale typification of rhetorical action; as action, it acquires meaning from situation and from the social context in which that situation arose” (Miller, 1984, p. 163).

We draw on genre theories to conceptualize playlists but expand them in various ways. Building on Miller (1984), we define genres as fusions of substance, sociotechnical assemblages, and sociomaterial practices to respond to the exigencies of affect. First, genres involve specific kinds of substance, that is, the features, symbols, and aesthetic particularities of content or media texts. In the case of music, most scholarship defines genres as a pre-established collections of textual properties, such as style of composition, arranging, performance, and programming, which are typically associated with a number of listening situations (Stockfelt, 1997). Instead, and concurring with Nowak (2016), we argue that musical features are assigned specific roles to “[suit] the settings of the particular context within which [music] is listened to” (p. 82). These roles emerge out of patterns of practice and, we here contend, affect.

Genres also consist of technology and not only “form” (cf. Siles, 2011). An increasing number of scholars have analyzed the significance of digital media in how people relate to music (Nowak, 2016; Prey, 2018). This has resulted in a more thorough attention to two specific notions: affordances—how users perceive certain features of technologies as possibilities or constraints for action—and sociotechnical assemblages—how technology is always embedded within a large network of human and non-human actors from which the agency of users and artifacts derive. The notion of assemblage comes from work in Science and Technology Studies (STS) and helps to make visible the “clusters of material, non-material, human and non-human that are constantly ‘becoming’—that is, constantly forming through active processes, rather than fixed or static” (Prior, 2018, p. 20). This approach invites recognizing the ecologies of practices, actors, and technologies (“old” and “new”)—devices, software programs, algorithms, standards, media, and computational tools—through which Spotify is constantly “becoming.”

Third, key in Miller’s (1984) definition of genre are typified social actions. As a supplement, we suggest that the notion of sociomaterial practice allows to better account for the constitutive nature of actions and the role they play in the entanglement of substance and technology (Orlikowski, 2007). As Orlikowski (2007) puts it, “the social and the material are considered to be inextricably related—there is no social that is not also material, and no material that is not also social” (p. 1437). Sociomaterial practices thus shape the contours of genre in specific situations.

We also expand genre theory by arguing that substance, sociotechnical assemblages, and sociomaterial practices are fused to respond not only to the exigencies of context but also to affect. We draw on a body of work that defines affect as intensities of feeling and attachment (Massumi, 2002, p. 27). Affect refers to that which “contains a particular energy, mood, or movement that may lead to a particular feeling, and possibly the subsequent expression of emotion, it both precedes and sustains or possibly annuls feeling and emotion” (Papacharissi, 2015, p. 21). We argue that genres (such as playlists) are a means to cultivate affect to produce, capture, and explore moods and emotions. Mood may be understood as

an enduring affective state, characterized by being global and not clearly elicited by an external event. [. . .] Emotion is [. . .] the awareness of situational demands, personal concerns, action readiness, and often physiological change, along with hedonic quality. (Konijn, 2008, p. 124)

Media technologies have historically played a key role in the cultivation of affect. Mood management theory, for example, shows there is a direct relationship between media users’ mood and the content they typically choose (Knobloch-Westerwick, 2008; Zillmann, 2000). Papacharissi (2015) demonstrates that media works to “sustain affective feedback loops that generate and reproduce affective patterns of relating to others that are further reproduced as affect” (p. 23). Despite their many contributions, these approaches have focused primarily on the symbolic dimension of media texts, rather than the technological or practical dimensions (as defined above through work in STS). This study makes visible the analytical advantages of further considering these three dimensions and their interplay.

Research Design

This study examined how users in Costa Rica appropriated music streaming services. The focus on a Central American country is meant to further understand the significance of platforms such as Spotify as a transnational cultural phenomenon. Costa Rica has a relatively large middle class and is among Latin America’s heaviest Internet users (Gao, 2015). The country has the highest rate of mobile Internet access in the region (Economic Commission for Latin America and the Caribbean, 2017). Spotify was launched in Costa Rica in late 2013 and has found a steady user base since. A recent study considered Spotify the most-used entertainment app on mobile phones in the country (Red 506, 2018).

We implemented a criterion sampling strategy to find “heavy users” of music streaming services to identify individuals who have a highly reflexive relationship with these platforms. We shared a call for participants on social media profiles associated with the University of Costa Rica. Thirty individuals who responded to our call were selected for interviews. Our final sample privileged sociodemographic diversity and thus included 15 men and 15 women, aged 19 to 52 years. Our informants were mostly educated people with different professional backgrounds.

Interviews were conducted in person between August and November 2018 and lasted for an average of 40 min. We recorded these interviews and transcribed them in their entirety. We used pseudonyms to protect the identity of our interviewees. (All interviews were conducted in Spanish. Translations are our own.)

We used an adapted version of the “think aloud protocol” (Fonteyn, Kuipers, & Grobe, 1993). During the interviews, we asked informants to open their Spotify accounts on a computer, which was projected on a screen so that the research team could see contents available and interactions with the platform. Informants who access Spotify primarily on mobile phones opened their accounts on these devices instead. We asked interviewees to describe their history of music consumption and typical appropriation practices. We discussed specific instances of content available on their accounts and requested explanations of their accounts’ configurations. We focused on playlists created or followed by our interviewees and discussed extensively their history, significance, and contours. We also inquired into “folk theories” of how recommendation algorithms work and how users experience and domesticate them. We triangulated data sources by capturing screenshots (of both computer and mobile accounts) for analytical purposes and by considering Spotify’s discourse about its own services in official outlets.

We coded the data inductively in a grounded theory manner (Corbin & Strauss, 2008). Coding was conducted by mixing individual and collective work phases. First, members of the research team conducted open and axial coding individually to identify data patterns and relationships between these patterns. Second, selective coding was carried out collectively to develop general analytical categories that combined the categories, patterns, and relationships identified in the first rounds of coding.

Cultivating Affect Through Music Streaming Services

Users appropriate music streaming services in large part to cultivate affect. We employ the notion of “cultivation” to make visible several issues. First, it highlights the work involved in producing moods as outcomes. Capturing, eliciting, exploring, or maintaining moods and emotions requires significant time and involves specific practices. Second, cultivation stresses the ritualistic nature of this work. By rituals, we refer to “structured mechanisms of repetition that work by resonating with, and reproducing patterned—and patterning—relations” (Singleton & Law, 2013, p. 265). Obtaining moods and emotions with and through music is thus a constant endeavor, reproduced every day and repeated throughout the day. Third, cultivation brings to the fore the dynamic nature of users’ relation to affect. It posits affect as both source and outcome; it situates affect and music as relative products of one another. For these reasons, we prefer the notion of cultivation to others traditionally used, such as mood management (Zillmann, 2000).

One key way of cultivating affect is the production of specific moods and emotions through music. In the words of “Mateo,” a 19-year-old college student, music in this sense is “the engine of how you want to feel.” To be in a mood thus means “to think and feel at length through that mood about everything” (Anderson, 2015, p. 817, emphasis in original). Users typically experience the production of specific moods and emotions as a need. For example, “Ricardo,” a 31-year-old software developer, links this need to the particular demands of his job:

I do choose music based on how I want to feel, not so much the other way around. This is not only because I need to concentrate but also because I need to concentrate on a certain kind of job, so I put on that kind music. [This helps to] enter the atmosphere we want for the product we are developing. (Emphasis added)

He illustrated this dynamic by putting on music created in the 1980s to inspire himself to work on a video game that had what he described as an “80s look.” Similar accounts were given for many other practices, most notably exercising and studying. The basic premise here is that music allows people to enter into a particular kind of affective state where moods and emotions flourish that allow them to perform professional or personal tasks in desired conditions.

Users turn to music to create not only individual but also collective moods and emotions. This is done in both work environments and domestic settings. “Eugenia,” a 22-year-old journalism student, thus, narrates a common situation that involves other members of her household:

I’m in charge of music at home; my brother supports it and my parents don’t know how the sound equipment works. [I choose music] that helps to transport people from one place to another. That’s what I try: to create an environment so [that everyone] is tuned, on the same page.

In this way, music is a key in the production of a collective mood through rituals. Mastery of technology—that is, the sound equipment—is what allows control of this ritual.

Another dimension of affect cultivation involves the use of music not to create but to respond to moods and emotions derived from specific experiences and activities. “Valentina,” a 29-year-old specialist in natural resources management, notes, “When I’m energetic, I look for very strong music, like heavy metal. Maybe it helps me express what I’m feeling in those moments.” For her, music provides a language to articulate affect. Some users even suggest that moods and emotions are incomplete without a musical expression. For “Mateo,” music is what allows him to sustain moods and emotions over time. He plans his music selections accordingly. In his words, “If I listen to something, it’s because I will be in that mood for some time.” A common expression of these ideas is the use of music as a soundtrack for daily life activities. Music is typically seen as the perfect companion for “whatever I’m doing,” as “Enrique,” a 22-year-old public administration student, put it (cf. DeNora, 2000).

Finally, users also turn to music to explore the temporality of their life. In this way, music becomes a way to revisit and restore specific moods and emotions associated with events in the life of users. “Valentina” explains,

There are songs I remember because I used to go to a bar with some friends and these songs were played. When I listen to them, I say, “That’s from when I used to go to that bar!” They connect me with that part of my life.

To crystallize these moods and emotions, users turn to technology.

Playlists as Affective Genres

Playlists are the quintessential way to cultivate affect associated with music streaming services. In what follows, we examine how playlists function as genres—and thus fuse specific kinds of content, sociotechnical assemblages, and sociomaterial practices—by discussing three issues: (1) the affective exigencies that motivate the creation of playlists, (2) the dynamics that sustain their transformation into specific kinds of artifacts, and (3) the process through which they become an expression of user identity and a shortcut to moods and emotions.

The Affective Origins of Playlists

Playlists begin with an affective exigence. In other words, users turn to playlists to produce, capture, and explore moods and emotions associated with a variety of events and experiences in their life. A common source of exigence comes from ordinary activities. “Adrián,” a 22-year-old public relations specialist, provides a telling example. During our interview, he explained how a mundane activity demanded the creation of a new playlist: “This [playlist] is called ‘Embers’ (Brasadas). I was with my friends smoking and it was like ‘I have to make a playlist for this,’ and only songs along this line started to come to my mind” (emphasis added). In this way, “Adrián” sought to create an appropriate affective state for this ritual in his life.

Like “Adrián,” interviewees described many other mundane activities that became opportunities for creating playlists: playing a video game, reading a book, watching a movie or a series, studying, thinking of a person or concept, or simply having an idea. These situations operate as exigencies in that users experience the possibility to produce, capture, and explore the moods and emotions linked to these situations as an objectified social need (Miller, 1984). In this way, users begin incorporating the playlist as an obligatory means to make sense of moods and emotions.

Another source of affective exigence comes from extraordinary events. “Julián,” a 20-year-old audiovisual producer and musician, explains,

The common trigger for all the playlists I have are very important moments in my life. On November 27 of last year, an online friend died. It was a very sad day so the first thing I did was to create this playlist. [He] was from Brazil so all songs are in Portuguese or are from Brazilian artists.

The experience demanded a response from the user, who envisioned the playlist as a means to produce an affective state that could work as a tribute to his friend. What counts as an extraordinary event is relative. During interviews, users mentioned examples that included the first day of college, Christmas, the end of the academic semester, the month before the end of the semester, a party, an evening out with friends, a particular class, or a concert attended or planned, among others.

Whether about mundane events or something extraordinary, playlists are all about capturing the proliferation of moods and emotions. Accordingly, for many users, playlists are never an individual or unique creation; they are one instance of a larger series. This is what “Julián” expresses when he claims that, “On Spotify, making a playlist is something like a ritual.” Playlist creation is an affective reaction integrated into everyday situations. It is thus not surprising that, when asked to describe her playlists, “Carla,” a 22-years-old graduate student in Linguistics, stated, “I’ve got everything in here.” By this, she meant both that playlists responded to every experience one could think of and that all that was important for her was represented by the ensemble of playlists she had created over the years.

From Exigencies to Artifacts

Media technologies have historically afforded ways to materialize affective exigencies. Many interviewees, particularly the eldest in our sample, envisioned CDs and mixtapes as antecedents to their playlists. The features afforded by Spotify to automate this process are what drove some of them to the platform in the first place. “Emma,” a 52-year-old audit specialist, explains the convenience offered by Spotify when compared with other technologies she used in the past:

You had to begin classifying and identifying cassettes. Now you don’t have those complications. [. . .] If you wanted to make a playlist, you needed to call a radio station and ask the person: “Could you play that song so I can record it?” [. . .] It used to take time. And you had to carry around cassettes and CDs. Now you just show up at a party with your phone and [ask]: “Does the speaker have bluetooth?”

Creating and maintaining a playlist is now codified in a few automated actions on Spotify’s platform. A video available on Spotify’s (2019) support website instructs users on how to create playlists by following five steps: “1. Play song then tap . . . 2. Tap add to playlist. 3. Tap create. 4. Enter playlist name. Tap create. 5. Tap [to] add more songs.”

But building and maintaining genres still takes time and effort. A key in these processes is reconciling the exigencies of affect with specific musical substance. As Lena (2012) notes, defining and building a genre requires establishing some sense of consistency. Users look for songs that seem coherent to them with the mood and emotions intended and then “tap to add” them to the playlist. Several interviewees called this process finding “the line” of the playlist; others used terms such as “thematizing” it. By these, users refer to the search for a pattern of similarity through which they articulate songs that could work to produce, respond to, or explore specific moods and emotions. In turn, this selection of songs begins to bound and give an expression to the genre.

The use of genres as a relatively pre-defined set of musical properties—or what Lena (2012) calls styles—is flexible. However, users draw on this conception of genres to establish the sense of similarity between songs that will be included in playlists. During our interviews, it was common to find references to playlists that had a “rock” or “jazz” feel. These definitions honor a pre-existing symbolic contract and usually meet the expectations about the kind of music they contain. But, on the contrary, users are quick to note the limitations that these labels have in defining music, playlists, and themselves. The most indicative answer in this sense came from “Rubén,” a 39-year-old psychologist who stated, “I am multi-genre.” “Javier,” who is 20 years old, put it differently: “The ‘genre’ does not matter as much as what the song evokes in me.” In this sense, playlists exceed the boundaries of traditional genres because affect affords opportunities for finding new roles for music. These opportunities are not unlimited but rather culturally and historically situated (Nowak, 2016).

Selected songs need to be ordered to provoke the appropriate affective response. “Ricardo” explains his approach as follows: “What I do is to grab similar music and arrange a playlist; I do spend some time to put it in the order I like” (emphasis added). Users employ notions such “rhythm” to describe the goal of this practice. (Spotify’s “Shuffle Play” feature allows them to experiment with this “rhythm” when the playlist is reproduced.) The notion that users are engaging in storytelling through the order of songs is not uncommon (cf. Dhaenens & Burgess, 2018). As one interviewee puts it, “No order is fortuitous. It tells a message.” “Gabriela,” the sound engineer, similarly maintains: “I have a concept, a story, or something I want to tell to somebody, so I do it through the songs [of the playlist].” To crystallize the affective exigencies imposed by these abstract “concepts,” she experiments with the grammar of the playlist (both lyrically and sequentially). She defines the result as “narrative playlists.”

Users approach the selection of names carefully and reflexively. “Julián,” the musician, notes, “All the names mean something. They are all like a book: they have a title. Choosing the title of the playlist is a moment because that is how I am going to reproduce it subsequently.” Naming playlists means naming a chapter in the affective life of the user and a way to bring the genre to life. Although the platform automatically creates an image or thumbnail for each playlist, some users choose images themselves. “María” (20 years old) asserts,

I like to be as creative as I can because I believe music represents a lot of one’s personality. I try to use [images that] are related to what I’m going to hear, but at the same time it’s an issue of aesthetics. I do believe that [images] give each playlist its personality.

Playlists thus get a specific name and look to reflect the user’s own personality. In this way, the genre and the user become a reflection of each other.

Users typically create and develop playlists individually. However, when they want to transcend their range of musical knowledge, they can open the playlist for external intervention. The case of “Rubén” is fairly typical. He explains, “There’s a Spotify feature [called] ‘Similar to . . .’ [From] there you build the sequence, iterating and iterating, and you get to know a lot of [artists]. You hear the song and then include it in the playlist.” “Rubén” thus envisions Spotify’s algorithms as a means to exploit his own knowledge of music. Users also think of algorithms as a means to incorporate “surprise” into the playlist (i.e., not necessarily knowing what song comes next). Another link is thus established between music streaming services and their antecedents, most notably radio.

This link is further promoted by Spotify, which uses the term “radio” to launch algorithmic recommendations associated with specific songs, albums, or artists. This use of “radio” also brings to the fore the importance of playlists in Spotify’s political and economic project. Current discussions between labels and artists for inclusion on the company’s playlists are reminiscent of negotiations for radio programming. In the presentation of a new tool launched in mid-2018 to help labels and artists submit music to the company for playlist consideration, Spotify emphasized affect as a main guide for making algorithmic recommendations:

It’s important to give us as much information about the track as possible—genre, mood, and other data points all help us make decisions about where it may fit. [. . .] The data you share will be complemented by what we already know about you. (Spotify, 2018)

But the limitations of this comparison with radio (and of algorithms) are constantly reached. For “Rubén,” Spotify’s algorithmic radio metaphor quickly falls short:

Radio has something that these applications don’t have: social interaction. I actually need the real live interaction and entertainment. Spotify, YouTube and all those [apps] let me adapt things to my liking, but [they] became so individualistic that you get bored.

Thus, when users want recommendations from someone (rather than something), they turn to people rather than algorithms. Referring to how he maintained a playlist with a friend, “Augusto” (19 years old) says, “I made the playlist and we named it ‘Listen to this, man’ (Oiga esto, mae). The playlist is meant to see what each other is finding and is fed by both. By sharing the playlist, it became social.” (We elaborate on the significance of sharing playlists below).

The process of creating a playlist can be tied not only to antecedent media technologies, as noted above, but also to other cultural practices. For example, playlist creation holds evocative parallels with the practices of DJs. Like DJs, playlist creators take finished products and transform them into a performance (Greasley & Prior, 2013, p. 25); they carefully design transitions and develop sequences of music “sets”; they seek to bring potential audiences to particular affective states (Katz, 2012). It would not be an overstatement to suggest that, through playlists, Spotify users turn themselves into the DJs of their everyday life.

The Endings and Beginnings of Playlists

On some occasions, playlists find an ending. Sometimes they are purposefully finished; sometimes they are abandoned or ignored. The deliberate end of a playlist is usually met with emotion. “Eugenia,” the journalism student, recalls,

I made this playlist for my birthday. I am very proud of it because I have been working on it [for a long time]. When I finished [it], it was 44 hours long. I am [so] proud that I wrote on Twitter: “Hey, I have a very cool (tuanis) playlist, if someone wants access to it, let me know.”

“Eugenia’s” words capture the emotional attachment that users have with playlists: they are seen as valuable possessions. Spotify then helps transform these possessions into commodities in a market of moods and emotions.

Reasons for explaining why users stop including songs to playlists range from practical (they take time and effort) to affective (new moods and emotions demand attention). They can also be cabalistic: “Julián,” for example, claims that playlists are finished once they include 47 songs. More generally, the potential of playlists for affect cultivation is not unlimited. At some point, music fails to produce or capture the desired moods and emotions. Users typically employ local slang to refer to this issue: songs and playlists get “burned” (están quemadas). This expression suggests that, each time a song or playlist is reproduced, its affective potential diminishes until they are turned to ashes and cannot ignite an emotional fire.

Not only do playlists have symbolic worth but, as noted previously, they are seen as a prime instance of self-expression. The fact that these playlists are hosted in users’ “personal” profile on the platform encourages this process. According to a user interviewed, “Playlists [have to] be under my name, my username, because they are like my identity.” For this reason, playlists are almost never deleted, even if abandoned or forgotten. “Carla,” the Linguistics graduate student, captured a common reaction when asked whether she had deleted any playlist: “No, no, no, I never delete them! Because I say to myself: ‘I might come back,’ even if I never do.” This assertion reveals how playlists come to play a practical role: they automate access to both the songs and the moods and emotions enacted by their genres.

In many ways, the end of playlists is also a starting point. When one playlist is (relatively) finished, new opportunities arise to restart the process. Moreover, once they are concluded, they become the place where the subsequent use of streaming services actually begin. Many interviewees reported going directly to their playlists when they start using the platform. In this way, playlists provide a shortcut to the cultivation of affect. Spotify’s sociomaterial assemblage is once again crucial in making this possible. Among the services it provides to paying customers, Spotify Premium allows users to download their own playlists to mobile devices and skip as many songs they want. This is of importance for users who manage their mobile data plan carefully. “Agustina,” a 30-year-old computer scientist, explains why she decided to pay for the Spotify Premium:

When I travel to visit my parents, there are areas where I don’t have coverage with the cellphone, and it disconnects me. [If] I download the playlist, I know that I will be listening to it appropriately on the road.

In this way, users connect playlists to broader infrastructural issues that are part of the sociomaterial assemblage within which listening to music takes place.

From Playlists to “Intimate Publics”

Music has a social nature. Research has consistently demonstrated the role of music as a “device of collective ordering, [ . . .] a means of organizing potentially disparate individuals such that their actions may appear to be intersubjective, mutually oriented, coordinated, entrained and aligned” (DeNora, 2000, p. 109). For the most part, studies have focused on the consumption of music in public spaces to make sense of how music enables community building and performance.

In a similar manner, playlists can become the basis of a shared affective experience. To make sense of this process, we draw on Berlant’s (2008) notion of “intimate public.” According to this author, “A public is intimate when it foregrounds affective and emotional attachment located in fantasies of the common, the everyday, and a sense of ordinariness” (Berlant, 2008, p. 10). As affective genres, playlists can enable the formation of “intimate publics” when a fantasy of social belonging emerges through the moods and emotions they evoke. When this happens, a playlist “flourishes as a porous, affective scene of identification among strangers that promises a certain experience of belonging and provides a complex consolation, confirmation, discipline, and discussion” (Berlant, 2008, p. 25). The formation of “intimate publics” is thus an achievement that serves a political purpose. In the case of Spotify, this helps turning the platform into an obligatory intermediary in the establishment of a utilitarian relationship between users and music (Eriksson et al., 2019). This requires establishing the playlist as a privileged object of affective control in daily life.

As noted previously, Spotify allows users to “share” their playlists and make them publicly available for others to “follow” and thus incorporate them into “their” set of playlists (or “library”). This allows genres to circulate and enables users to find a representation of their affective interests. “Emma,” the audit specialist, explains how she deals with the availability of public playlists:

The app “boxes” things together; within the “drawers,” there is a game of chance and probability that I might like something or not. I start to see if there is something that catches my attention visually when I start navigating, something that sounds interesting to me. The title of playlists facilitates the trial and error.

Through their titles, playlists seek to evoke a sense of common, ordinary affect available to users. The metaphor of “drawers” denotes a sense of unity and coherence in the musical substance of each playlist, but also a lack of openness. In this way, Spotify enables a market of moods and emotions that can be consumed through the act of “following” playlists. This market is naturalized in the name of convenience. “Eugenia” says,

There is nothing more comfortable than going to Spotify and [see playlists] already made to play them while you study, or to type: “I had a bad day, what should I listen to?” Since there is so much variety, it is easy to find something that fits your needs so you don’t have to go through that work and just listen.

The consolation for the ordinary experience of having a “bad day” comes in the musical form of ready-made playlists.

Spotify also commodifies “intimate publics” by showing the number of “followers” each playlist has. Although many denied to be interested in the success of their playlists to attract public attention, most interviewees revealed to be very aware of the number of followers they had gained. When “Felipe,” a 24-year-old computer scientist who works for a scientific laboratory, was explaining to us the playlist he made for playing a video game, he interrupted himself to declare, with no small pride, “This is my playlist with more subscribers. It has 782.” This kind of statement was relatively common during our interviews.

As the DJs of everyday life, playlist creators need an audience for their “sets.” Many users work explicitly to find these “followers.” Some share their playlists on social media or messaging applications with comments that convey their emotional singularity. The goal of sharing these lists is to allow others to experience the affective genre they enact. “Carlos,” a 20-year-old electric engineering student, explains,

If [a playlist] made me feel good, I would like it to make others feel good in a certain way. I would like people to listen to new and different things and to explore the different sensations that [playlists] give.

In this sense, the affective genres enacted by playlists operate as a “sentimental intervention” that “mobiliz[es] a fantasy scene of collective desire, instruction, and identification that endures within the contingencies of the everyday” (Berlant, 2008, p. 21).

“Intimate publics” can form through playlists created not only by users but also by Spotify. Some of these playlists have titles that reveal the commercial interest of affective genres for Spotify. There is a special section devoted to “Genres & Moods” on the platform’s browse interface. The “Moods” section recommends titles such as, “Coffee, Books” (Café, Libros), “On the Road” (De Camino), or “FrienDeSemana,” a play of words on weekend (in Spanish) and friend (in English). Each playlist has a short description and a thumbnail. This introduces important tensions in the market of moods and emotions enacted by the platform. For many interviewees, Spotify’s curated playlists signal the success of its algorithmic model of recommendation. “Gabriela” shares this view: “[Spotify] definitely knows what I like. They already know what the formula is to make a playlist for me” (emphasis added). For this user, the limitations of algorithms are surpassed when something becomes like someone who has come to know her preferences. She attributes this outcome to a right interpretation of the input she has given to algorithms through her use practices (cf. Siles, Espinoza, Naranjo & Tristán, in press).

Other users seem warier of Spotify and its algorithms. “Adrián,” the PR specialist, claims to never listen to Spotify’s own playlists. In his opinion, “What I feel is ‘On the Bus going Home’ is not what Spotify says it is.” Here, the platform is criticized for producing affect that is perceived as artificial rather than organic. The evidence “Adrián” puts forth is to have personally experienced what being “On the Bus going Home” actually means.

Concluding Remarks

Despite profound transformations in communications industries, genres persist as a fundamental component of the digital ecology. Like the case of Spotify shows, genres remain a crucial mechanism of navigating digital content. To further understand the meaning of genres in the digital era, this article argued for conceptualizing them as fusions of substance, sociotechnical assemblages, and sociomaterial practices that respond to and crystallize affect. We argue that this definition is much better suited than some traditional approaches to genre to help understand present reconfigurations of media industries. This approach to genres involved three important issues: (1) integrating affect into the study of media technologies; (2) examining the interplay of content, technology, and practice; and (3) considering how affect becomes the basis of collective experiences. To conclude, we elaborate on the implications of these issues by discussing the main findings from our study.

Scholarship from a variety of fields has worked to further integrate the study of affect and media technologies. This has helped to better understand the emotional investments of audiences when they encounter media texts and how these investments are a key to account for deep social and political transformations associated with the media. To supplement this body of work, we argued that affect is constitutive of genre. We put forth the notion of cultivation to suggest that when considering the case of music streaming services, users seek to produce, respond to, and explore moods and emotions in particular ways. This requires the performance of rituals. Cultivation also posits both affect and music as a product of one another. It is not only that moods and emotions reflect the music choices of users, but that music also reflects their moods and emotions.

Bringing affect into the analytical equation also helps to illuminate important aspects of Spotify’s political and economic project. Affect has significant monetary value. Music is a fundamental component of the contemporary market of moods and emotions. As Eriksson and Johansson (2017) aptly put it: “As the Spotify machinery gears more towards the provision of playlists that evoke intimate moods, the service becomes (financially) dependent on users’ willingness to disclose their feelings by selecting a playlist that suits them” (p. 77; emphasis added). This article showed how this process is achieved not only through the selection of playlists but primarily through their creation and public circulation. We shed light on how the cultivation of affect through playlists is actually incorporated into daily life as a necessity, carried out by users as an obligatory means of interpreting moods and emotions. It is perhaps a telling illustration of how, as discourses on user agency continue to gain currency, contemporary audiences are partly disciplined on how to materialize their moods and emotions through platforms. Integrated as an obligation into the daily life of self-governing subjects, a functional relationship with music ends up infusing self-performance dynamics.

Examining the case of Spotify required moving beyond definitions of genre that center on pre-established and relatively fixed media texts. When the actual practices of users are examined, genres become much more fluid entities. To account for this analytical shift, we argued that three dimensions are necessary: substance, sociotechnical assemblages, and sociomaterial practices. Regarding substance, playlists involve the search for patterns of similarity between songs that seem compatible with the exigencies of affect. This involves finding the “line” or “theme” that links these songs together, and negotiating the singularities of musical texts and styles (Lena, 2012). There is a practical dimension to this process: these songs are organized in a particular order that is meant to say something about the person who chose them; they are named evocatively to convey their affective singularity; they are given a cover that expresses their aesthetic value; they are shared for others to enjoy the “atmosphere” they enable; and they are promoted in the sociotechnical assemblage in which music listening takes place today.

To become genres, these playlists also require a sociomaterial basis: Spotify has coded into a few affordances the process of creating, developing, and sharing them. Its algorithms are a key in exploiting the musical knowledge of the user and enacting the singularities of previous media technologies (such as radio, CDs, and mixtapes). Features such as “Shuffle Play” add new dimensions to the “rhythm” intended for songs. The Spotify sociomaterial assemblage is also critical in creating a market of moods and emotions that can be experienced by simply “following” ready-made playlists.

Affective genres not only require the study of these three dimensions but demand an account of how they interact. Thus, although concepts such as “affective scripts” (Tomkins, 1962) or “resilient reception” (Cavalcante, 2018) would help stress the significance of certain practices, they would run the risk of overlooking how technology shapes the process of creating, using, or sharing them. On the contrary, Anderson’s (2015) notion of “moodscape” highlights the role of “technology for the delivery and consumption of musical content from any genre” (p. 833), but is less attentive to the practices involved in producing, responding to, and exploring moods and emotions. In a similar manner, theories of genre that emphasize the properties of media texts do not necessarily account for the work involved in making genres emerge, evolve, and operate as cultural categories that are assigned roles for specific moments (ordinary, extraordinary, and everything in between). Our approach to genres is useful beyond the specific case of Spotify. It could also be applied to theorize dynamics that are constitutive of the contemporary digital ecology, as the case of Netflix’s “alternative genres” suggests.

Finally, a growing concern in scholarship on media technologies is how affect becomes the basis for collective experiences that have cultural and political implications. Papacharissi (2015) thus speaks of “networked publics” or “networked public formations that are mobilized and connected or disconnected through expressions of sentiment” (p. 125). Although clearly related, we argue that Berlant’s notion of “intimate publics” is better suited to account for the collective experience enabled by affective genres and their economic value. Unlike “networked publics,” “intimate publics” are not oriented toward collective action or power disruption but rather coalesce around emotional attachments to certain discourses of how to make sense of ordinary and extraordinary experiences, which then serves political-economic projects such as Spotify’s. We also prefer Berlant’s notion to the concept of “affect cultures” (Döveling et al., 2018) because it highlights the sense of intimacy that playlists seek to evoke by articulating sequences of songs, titles, and images. We posit “intimate publics” as one possible means to better understand how “affect cultures” actually come into being around the world and why they matter. In this sense, by discussing the links between affect, technology, and music, we hope to contribute to further problematize the rise and significance of platforms such as Spotify as a transnational cultural phenomenon.