Streaming’s Effects on Music Culture: Old Anxieties and New Simplifications

Legacies of Critique

Academic research has paid considerable attention to music streaming. As in much of the online commentary just mentioned, there has often been a focus on questions of business and/or technology, topics that have generally dominated research on popular music in the social sciences and humanities since 2000. The crisis that gripped the recording industry during the 2000s and early 2010s made it an object of great interest for researchers concerned with the disruptive role of new technologies (e.g. Prior, 2010, Rogers, 2013, Arditi, 2015). During a period when many were speculating and hoping that digitalisation might democratise culture (see Hesmondhalgh, 2019), the recording industry seemed to offer an early test case, with some analysts seeing it in relatively optimistic terms (Suhr, 2012), while others discerned new forms of control and resistance (Burkart, 2010).

Music streaming hardly featured in published research on music before 2015, but since it became clear, around that time, that streaming was emerging as the new centre of musical production and consumption, a new wave of research on digital music has emerged. While it is universally recognised that legal streaming has, for the time being, ended the recording industry crisis of the noughties, there continues to be an emphasis on business and/or technological change, though now it is increasingly focused on how disruption has been overcome (e.g. Sun, 2019). There has been much attention to the methods used by now-dominant music platforms to convert user activities into data (Drott, 2018; Meier and Manzerolle, 2018; Negus, 2019; Prey, 2016), and how opaque technological systems might be researched (Eriksson et al., 2018). A smaller body of research has investigated changing audience musical practices and values in the era of streaming (Hagen, 2015; Nowak, 2015). Yet relatively little academic attention has been paid to issues of quality, diversity and aesthetic experience raised by claims about the new musical system. I address some of the more substantial efforts to do so here, but relate these academic critiques to public debate and online commentary. While academic sources have drawn on online commentary, this has often been done in passing and with little sense of caution, scepticism or reflexivity (with some recent exceptions such as Morris, 2020).

Here are five key themes, or sets of criticisms, that have emerged in online commentary on the new musical system centred on streaming, and in related academic research:

In what follows, I address each of the themes in turn, examining the degree to which criticisms of streaming’s effects on musical experience along these lines might be considered valid, and the degree to which they might genuinely enhance understanding of musical experience in the 21st century. In actual discussion, these themes often blur and even merge. For example, critique of music’s functionalism is often connected to critique of its blandness, as I show in Section 2; I separate them here for analytical clarity. The third and fourth categories overlap in terms of a concern with distraction and inattentiveness. Importantly, I also consider these themes in relation to older forms of evaluation, particularly those that developed in the 20th century in response to the industrialisation of music. In doing so, I build on sociological and cultural studies research, such as that of academic and critic Simon Frith, who in his classic book Sound Effects, traced a history of 20th-century concerns about music and mass culture, pointing out how popular music was ‘frequently censured by rock fans and critics according to the values and practices of both “high” and “folk” art’ (Frith, 1981: 40). Frith also showed how rock fans and critics alike often valued records and songs ‘for their artistic intensity, for their truth to experience’, but ‘condemned [them] for commercialism’ (1981: 40). I explore some ways in which the terms of censure apparent in 20th-century discourse about mass culture might still be at work in 21st-century digital music culture, and what this might tell us about how critique may at times inadvertently limit, rather than promote, critical and emancipatory aesthetics. But I also seek to identify forms of critique that might provide greater insight into the present state of musical culture. My claim is that recent critiques of streaming often draw upon long-standing aesthetic anxieties about the effects of technologies on music, about the supposedly debasing influence of commerce, and about perceived decline in the standards of musical taste and practice in society, in ways that sometimes come close to elitism and snobbery, and certainly risk simplification. I also discuss these matters in the context of academic research in music sociology, musicology, popular music studies, psychology of music, and other related fields, identifying cases where academic commentators have pursued similar criticisms (sometimes reproducing the problems just mentioned) and cases where their perspectives might cast light on such debate. The aim is decidedly not to assert the superiority of academic analysis over the supposed naivety of online commentary. It is to suggest some potential lines of dialogue between academic researchers and online commentators, which might provide a basis for more robust forms of analysis of emerging musical systems centred on streaming than those which currently prevail.

Streaming Encourages Functional Musical Experience

For one very prominent critic of MSPs, the US journalist Liz Pelly (2017), Spotify offers a mere ‘illusion of choice’. The music available ‘appears limitless’ but for Pelly, in reality the ambition of MSPs such as Spotify (or Spotify in particular, because of its intensive focus on function-based playlists) is to ‘turn all music into emotional wallpaper’. Just as clickbait content farms attack ‘a reader’s basest human mood and instincts’, Pelly writes, Spotify makes music conform to ‘an emotional regulation capsule optimized for maximum clicks’. Pelly is particularly troubled by the prevalence of ‘chill’ playlists, and the prevalence of music intended to be heard ‘in the background’. Spotify’s intense attention to mood and activity-based playlists has contributed, Pelly (2017) claims, to ‘all music becoming more like Muzak’. ² Academic researchers have taken similar views. Paul Allen Anderson sees contemporary MSPs as a form of ‘neo-Muzak’, which ‘works as an aurally administered affective mood stimulant and mood elevator by forestalling exhaustion and boredom’ (Anderson, 2015: 832). This can involve using music to increase stimulation and mental acuity to keep up with ‘accelerated work’ (2015: 834) or to decrease stimulation to counteract ‘perceptive surplus, hypervigilance and stress levels’. For political theorist Paul Rekret (2019: 62), streaming offers a more individualised, personalised version of muzak, embodying ‘a movement away from music’s embedded social meaning towards its identification with social function’ (2019: 64). While he does not use the term ‘muzak’, A.T. Chodos (2019: 51) argues that Spotify embodies a view of music that understands its meaning as residing in its relationship to other activities or feelings, and that listening on Spotify is not about ‘attending to music but using music to evoke a desired feeling or achieve some other secondary effect’.

Some critics have gone further, to argue that, in their focus on function, MSPs promote forms of musical subjectivity that resonate with problematic forms of contemporary power, especially those associated with ‘neoliberalism’. Maria Eriksson and Anna Johansson (2017: 74), for example, argue that Spotify’s featured playlists, through their emphasis on using music for a purpose ‘may serve as a disciplinary technology that promotes neoliberal subjectivities and attitudes’ and in particular they argue that Spotify’s playlists promote ‘a type of subjectivity informed by positive thinking as a contemporary ideology’, encouraging fantasies of an unattainable good life. Writing in the context of the COVID-19 pandemic, Marie Thompson and Eric Drott (2020) observe that music is increasingly being ‘put to work’ to enable ‘social reproduction’: the maintenance of daily and intergenerational life, at the levels of family, community and society.

These concerns about music serving functional purposes, rather than as something to be ‘attended’ to, have a long history, which we can begin to explore here by examining the muzak metaphor. Practically everyone who has written seriously about muzak points out that it is a misnomer. The Muzak company operated from the 1930s to 2011, when it was bought by Canadian company Mood Music, and was only one of a number of businesses that have supplied ‘programmed music’ (Sterne, 1997: 24). As well as public spaces, programmed music was very widely used in American factories and retail spaces from the 1940s to the 1980s (Jones and Schumacher, 1992). Liberals and radicals alike feared that programmed music represented a new phase, driven by technologies associated with industrialisation, whereby music more than ever would serve non-aesthetic functions. Muzak and programmed music were thus condemned by radicals not only for having functional rather than aesthetic goals, but also for the particular functions they served. Programmed and background music were understood to serve business, or capital, by manipulating workers to work harder, or consumers to spend more. In a widely read book on the politics of music and sound published in the 1970s, the French polymath Jacques Attali argued that muzak in the workplace was not just a way of drowning out ‘tedious noises’ but more generally heralded ‘the general silence of men [sic] before the spectacle of commodities . . . and the end of the isolable musical work, which may have been only a brief footnote in history’ (Attali, 1985 [1977]: 112). For Cold War liberal critics of ‘mass culture’, meanwhile, the greater concern was musical banality or blandness (issues I return to in Section 2). Building on long-standing anxieties about the introduction of ‘canned music’ into public places (Jordan, 2014), programmed music’s many 20th-century critics ‘seemed particularly incensed by its technological component, which they feared would force a form of sonic banality upon the public’ (Radano, 1989: 448).

There were also liberal fears about a loss of control over the sonic environment, encapsulated in the notions of ‘forced listening’ and ‘captive audiences’. Such ‘forced listening . . . was a threat to individual liberty’, especially in public places such as railway stations (Jordan, 2014: 296). Sometimes, such fears led to important questioning of the expansion and intrusion of commerce into public space, as in a campaign to stop music and advertisements from being played in New York’s Grand Central Station in 1950. But often the level of anxiety about the fate of music itself seems disproportionate, as when, in 2000, a British Conservative politician introduced an unsuccessful but well-backed parliamentary bill intended to protect the public from involuntary exposure to music by banning the playing of ‘piped music’ in places such as bus stations and swimming pools, remarking that all music ‘is devalued if it is treated as acoustic wallpaper’ (quoted by Frith, 2002: 35). As we shall see, this rather clichéd metaphor of music as wallpaper recurs throughout much mass culture criticism of music – including recent critiques of streaming.

How might we think about these concerns about music’s functionalism (mingled with concerns about banality), often expressed via the muzak metaphor? Music has of course often been tied to function, both in terms of social ritual (funeral music, music to mark religious festivals or changes of season) or at a more intimate level (singing lullabies to get children to sleep) and all points in between (Gregory, 1996). The recent concerns about the use of music to accompany other activities can seem rather odd when seen in this larger historical context. After all, if the objection is to music serving a function beyond the music itself, even dance music would fall foul of such complaints. At work in some expressions of concern about musical functionalism is surely a certain version of the notion of the autonomous artwork – the idea that art should stand separately from social function in order to achieve its full aesthetic and/or political power. Janet Wolff has defined ‘the ideology of autonomous art’ as the idea that ‘Art, or at least Great Art, transcends the social, the political and the everyday’ (Wolff, 1987: 1). Few people interested in the conditions of musical production and consumption would defend such a view today, at least not by invoking the idea of transcendence, with all its religious and spiritual connotations, as though art might soar free of racism, colonialism, sexism, and social suffering. Nevertheless, a version of this ideology seems to underlie many of the criticisms being made of streaming’s musical functionalism.

Music sociologists and ethnomusicologists, often more distanced from romanticism and modernism than cultural critics, have tended to approach the use of music to serve functions for individuals and institutions, and everyday musical engagement in general, with somewhat greater interest and sympathy than cultural critics – and with greater regard to what people actually do with music. A prominent example is Tia DeNora’s Music in Everyday Life, one of the key works of music sociology of the last half-century (DeNora, 2000). DeNora sought to understand music’s powers by examining neglected cases of music’s ‘agency’, including its use for making sense of social situations, or as a way of ‘creating and sustaining ontological security’ and managing mood (DeNora, 2000: 16) – such as getting ready to go to work or to go out for leisure (2000: 53–61). Her work – and that of other researchers concerned with music and everyday life, including music psychologists (Sloboda, 2010) – raises the possibility that recent expressions of concern about the proliferation of playlists oriented towards mood and function on Spotify and other streaming services risk a certain unconscious snobbery towards the ways in which people use music to get themselves through the day, or to prepare themselves for action of some kind, whether going out with friends or working out in a gym.

I want to suggest that we need a more sociological but also a more dialectical understanding of musical value that would allow for a more nuanced and complex understanding of problems associated with streaming. Eriksson and Johansson’s identification of positive thinking ideology in the framing of some Spotify playlists is significant and suggests the importance of analysing MSPs’ ways of curating and framing music. But it is a leap to imply that ‘neoliberal or radical individualist ideologies’ are at work when Spotify users click on a ‘Wake up to good vibes’ playlist. We know little yet about the kinds of experiences that result when people click on such lists, and what conscious and less conscious responses might be involved. Other critical claims may, conversely, require better grounding in political-economic explanation and theorisation (rather than sociological analysis of experience) such as Thompson and Drott’s idea that MSP playlists are an important means by which music is being ‘put to work’ to enable ‘social reproduction’ and are thereby ‘inextricably tied to the reproduction of capital’ (Thompson and Drott, 2020). Is there a danger here of overstating the place of music’s functional roles within contemporary society, or the capitalist system? Even if not, which agents (people and/or institutions) are putting music to such work, and in whose interests?

It is to the credit of those academic analysts concerned with music and everyday life that they strive to capture the nuance and complexity of people’s ordinary interactions with music. A problem with this approach is that the emotional and psychic power of music can sometimes be obscured, and so can the ideological work of music that the recent critics of streaming are at least seeking to comprehend. But such critique needs to avoid simplistically posing music’s embedded social meaning against its functional uses in everyday life, as though these were always incompatible.

Streaming Encourages Bland, Unchallenging Music

The muzak and wallpaper metaphors/metonyms discussed earlier invoke banality as well as anxieties about music serving everyday functions. For a number of commentators, the mood-driven quality encouraged by streaming leads to a pervasive blandness. Liz Pelly (2018a) has discussed the prevalence on Spotify of ‘streambait pop’: ‘muted, mid-tempo, melancholy pop, a sound that has practically become synonymous with the platform’. A distinct but related term bemoans the prevalence on Spotify of ‘Spotify-core’, a term often attributed to New York Times pop critics (see also Morris, 2020). Joe Coscarelli defines this type of music as follows: ‘It’s like vaguely electronic, usually female vocals, glitchy effects’, and his fellow NYT critic Jon Caramanica elaborates: ‘Slow, galloping BPM, sort of dance music alluding, but not actual dance music’ (New York Times, 2018). In a piece entitled ‘Has 10 years of Spotify ruined music?’, two Guardian writers define Spotify-core as ‘homogenous mid-tempo pop drawing from rap and EDM’ (Beaumont-Thomas and Snapes, 2018). According to one independent label owner quoted by Pelly (2017), artists and independent record companies struggle to gain access to prime playlists, because the ‘more vanilla the release, the better it works for Spotify. If it’s challenging music? Nah’ (Pelly, 2017). ‘Vanilla’ here draws on a different discursive repertoire from muzak and wallpaper, but with similar implications of blandness: Spotify offers the musical equivalent of dull, conventional sex. What’s more, this blandness supposedly makes streaming the perfect fit for the neoliberal subjectivities produced by contemporary capitalism: algorithmically designed playlists, Pelly writes, ‘have seized on an audience of distracted, perhaps overworked, or anxious listeners whose stress-filled clicks’ then in turn ‘generate anesthetized, algorithmically designed playlists’. Washington Post critic Chris Richards (2017) writes that streaming is consistent with the prominence of the drug Xanax in culture, and refers to a pervasive ‘21st century pill-pop’, marked by ‘a smoothness, a softness, a steadiness’; ‘the sound washes over you, smooth and steady’, and streaming services promise ‘fluid, frictionless listening’. Paul Rekret notes how ‘what seems to sell today is sound that takes its audience neither too high nor too low’ (Rekret, 2019: 59) and like many others, comments on the omnipresence on streaming platforms of ‘chill’ playlists.

Prominent UK music industry consultant Mark Mulligan (2019) offers an account of how industry dynamics might serve to reinforce blandness. The supposedly casual, passive users of MSPs (see below) are likely to become increasingly important to MSPs’ expansion, as services seek to expand beyond the current base of subscriptions based on committed listeners. Such listeners already subscribe to MSPs in large numbers, and so the main place to expand is into the realm of casual users. Mulligan points to the creation of background playlists stuffed full of short tracks, such as a Sony UK playlist called ‘Sleep and mindfulness thunderstorms’ consisting of 330 tracks of just over one minute each. For Mulligan, such efforts to game the playlist system might make commercial sense but are ‘creatively bankrupt’. ³

The sense of a vast culture of bland music which in some way dominates musical culture, or some prominent component of it, has been a persistent theme in vernacular criticisms of mass culture over many decades. One example was the frequent criticism, on the part of rock critics and fans, of ‘middle of the road’ and ‘easy listening’ music. Keir Keightley (2008) has provided a definitive account of ‘the easy listening musical formation’ that arose in the immediate aftermath of the Second World War in the USA and that ‘dominated the mainstream of popular music, particularly on popular LPs, for a generation’, from the demise of swing until the rise of rock in the late 1960s. Easy listening (under its different guises, including ‘sweet’ music, ‘mood’ music and later Middle of the Road or ‘MOR’ – this range of terms is why Keightley uses the term ‘formation’ to label this loose alliance) had been the dominant and most commercially successful music of the preceding years, enjoyed by and associated with many of the parents of the rock generation. Keightley shows how as rock rose to prominence, rock critics and fans criticised such music, borrowing from various values characteristic of jazz and swing in a previous era, positioning their own music as masculine and superior, and other musics as feminine and/or feminising, and therefore inferior. Rock critics and fans pointed to various characteristics of easy listening that, in their view, marked it as sub-standard as a result of its unchallenging commercialism – ‘its pronounced melody, its easily recognizable/singable/hummable qualities, its nostalgic tendencies’ (Keightley, 2008: 326). Among the adjectives regularly used from the 1940s to the 1960s by the pre-rock generation to describe the music of the easy-listening musical formation in positive terms were ‘soft, smooth, soothing, gentle, quiet, muted, lush, romantic, dreamy, sugary, civilized’ (Keightley, 2008: 327). Keightley shows how the valence of these terms changed – they were used by rock generation as criticisms, rather than as the basis of praise. Similar terms were also applied to pop music from the 1960s onwards, shaping the rock–pop divisions that structured popular music cultural politics for decades to come. It is striking that terms such as softness, smoothness and mutedness also occur in the critiques of streaming cited earlier. And, as Keightley notes, these are terms strongly associated with femininity.

Keightley’s account also makes conspicuously clear how rock culture’s criticisms of the easy listening musical formation (including ‘MOR’), like recent critiques of MSPs, focused on its ‘explicit functionality’. Keightley points out that the ‘mood music’ variant of easy listening was frequently conceived as background accompaniment, highlighted and elaborated upon in album titles, artwork and liner notes, and he lists titles such as Music for Dreaming, Music for Dining, Music for Reading, Music for Peace and Inner Calm (along with hundreds more). The similarity to titles of streaming playlists are notable. So too are the parallels between rock’s rejection of music oriented for such purposes and similar attacks on ‘functional’ mood music in the streaming era.

Blandness and functionality are closely aligned within discourses of musical taste and value that were formed in the jazz, swing and rock eras and that have been resilient beyond it. In retrospect, rockist dismissals of easy listening and MOR, and of mood-oriented music more generally, can feel crude, masculinist, and at times unthinkingly adolescent in their loathing of ‘conventional’ adulthood (there is nothing intrinsically wrong with adolescence, but nor is there anything intrinsically wrong with being an adult). The contempt expressed for ‘Spotify-core’ and ‘chill’ playlists in recent commentary may be in danger of reductions and simplifications of a similar order to those apparent in the rock era. Critiques of streaming would benefit from greater attention to ways in which their terms might unwittingly reproduce an earlier era of dismissal of other people’s music and from greater understanding of people’s varied musical practices, without resorting to a debilitating aesthetic relativism. Brian Eno expressed an interest in musical function, evident in album titles such as Ambient 1: Music for Airports (1978), and he praised ambient music for its ability ‘to induce calm’ – anathema to rock and punk’s equation of good music with rebellion and transgression, and therefore with strong emotions of frustration, anger and fury. The 1990s saw an intensification of efforts, sometimes in kitsch form, to reclaim elevator music, lounge music, new age music and other musics aiming to evoke mood, or fulfil a function, other than the generation of intense engagement and emotion. ⁴ The creativity, inventiveness and sophistication of some easy listening and MOR has increasingly been recognised (see Weisbard, 2014). These are just some potential sources for a richer and more varied critical aesthetic that would not dismiss entire swathes of music on masculinist grounds, and that would recognise that the use of music to serve functions does not necessarily or entirely negate its cultural value.

Streaming Makes Musical Experience Distracted and Passive, and Music Recedes into the Background

Another major theme, closely connected to the concerns about functional music discussed in section 1, is the idea that streaming encourages passive musical experience. Liz Pelly (2018b) uses the term ‘lean back listener’ to refer to ‘a new type of music listener, one who thinks less about the artists or album they are seeking out, and instead connects with emotions, moods and activities, where they just pick a playlist and let it roll’. Blogger Michael Donaldson (2019) concurs: ‘the seductive nature of playlists and algorithmic recommendations is turning fans into more passive listeners’, as opposed to ‘intentional’ ones (see also Bainbridge, 2019). We can distinguish two different notions of passivity in the criticisms of streaming. One is that streaming encourages and/or entrenches a more general culture, and specifically a musical culture based on distractedness and lack of attention. The other is that streaming discourages musical discovery and exploration; I address the former set of criticisms here, the latter in Section 5.

The same muzak and wallpaper metaphors used to criticise the way that streaming supposedly leads to an increased functionality and blandness of music (Sections 1 and 2) are also frequently used to criticise passivity. But in this context the muzak metaphor is used by critics to evoke the dangers of music that does not demand, require or merit sustained attention. The metaphor of music as wallpaper draws on similar concerns. Wallpaper is intended to be pleasing but ultimately merely decorative. The implicit underlying idea is that music should demand more of us, that it should force its way into our attention, requiring that we be more actively engaged, and that streaming is preventing this from happening. As with concerns about music’s functionality and blandness, an understanding of the historical roots of such ideas might place them into perspective and allow a more dialectical understanding of MSPs. I also want to consider whether concepts of ‘ubiquitous music’ and ‘ubiquitous listening’, which have sometimes been used by academics in efforts to understand recent developments in music technologies, including streaming, offer much potential in this regard.

A sense of the value of auditory attention is built into everyday distinctions between listening and hearing, with the former configured as ‘active’ and ‘focused’ and the latter as ‘passive’ or ‘background’, and into distinctions between different kinds and intensities of listening (Clarke et al., 2010: 65). Jonathan Sterne has shown how a set of ‘audile techniques’ (a set of values attached to auditory attention), initially developed in medicine and telegraphy, proliferated in the late 19th and early 20th centuries, and how close listening became ‘a symbol of modernity, sophistication, skill, and engagement’ (Sterne, 2003: 137). David Suisman (2009: 240–272) recounts how a new musical soundscape emerged at around the same period, including the new music-related technologies of phonography and radio, embraced by some as cultural democracy, but accompanied by new anxieties about the effects of business and technology on sound and music, including ‘a culture of degraded listening’ (2009: 255). This is the context in which metaphors about muzak and wallpaper emerge. An essay from the Musical Quarterly in 1917 provided a typology of musical listening, lambasting ‘the listener who is no listener at all’, and who waits ‘like a cabbage or stone, for something to happen to him’; by contrast, ‘the ideal listener’ is one who prepares for concerts and is ‘ready to merge completely with the music’ (Sophie Gibling, quoted by Supper and Bijsterveld, 2015: 124). William Glock, a key figure in the development of classical music radio in Britain as BBC Controller of Music from 1959 to 1972, expressed concern that the introduction of less demanding classical music as part of the BBC’s new station, The Music Programme, in 1965 would lead to the ‘danger of musical wallpaper’ so that ‘by the next day even the greatest things had been forgotten’ (quoted by Stoller, 2018: 107). Public service paternalism, intended to induce citizen engagement with culture, sought to challenge the passivity associated with commercial consumption.

The most relentless critic of music in capitalist modernity, Theodor Adorno, offered various versions of a typology of music listeners (Paddison, 1993: 210–212) but the most often cited one in English is to be found in Adorno (1976 [1962]). Adequate listeners could be professional experts or committed amateurs – but the latter were under threat in an era of ‘mechanical reproduction’ (i.e. the rise of the media). Regressive listeners of various kinds treated music ‘as a comfortable distraction’ (quoted by Paddison, 1993: 212). Adorno’s friend Walter Benjamin famously saw things differently. For him, mechanical reproduction of culture for mass audiences, exemplified by cinema, might democratise art by undermining the authority attached to contemplative appreciation such as gazing at paintings (Benjamin 1968 [1935]). From this point of view, the cinema-going public could be distracted experts. ⁵ As we have seen, critical commentary has often tended to go in a pessimistic Adornian direction rather than towards Benjaminian optimism, but for all the differences between visual art and music, Benjamin’s perspective suggests the possibility of a more dialectical reading that might find critical possibilities in digital capitalism – an understanding that is not to be found in the vast majority of recent treatments of music streaming.

At first sight, or on first hearing, the concept of ‘ubiquitous listening’, developed by Anahid Kassabian, may seem to offer a route towards such a dialectical reading. Referring to the idea of ‘ubiquitous computing’, Kassabian (2013) developed the notion of ‘ubiquitous musics’ to refer to the way that music is everywhere, some through our own choices, some without our sanction or control’ (2013: xii). Her focus was on the experience of such music, in particular the fact that these musics ‘are listened to without the kind of attention assumed by most scholarship to date’, and so although her ideas were developed before the consolidation of the new musical system, and Kassabian hardly mentioned streaming, the way that MSPs arguably involve an intensification of the constant, background nature of music makes her work of potential interest for analysts of streaming. For example, Rasmus Pedersen has argued that ‘data-driven curatorial practices’ on MSPs perform music recommendation ‘in ways that amplify and encourage user practices of ubiquitous listening’ (Pedersen, 2020: 72), and thereby create ‘a bias towards inattentive listening’ (2020: 87). Such ubiquitous listening, according to Pedersen, also involves valuing music as a resource, ‘rather than music’s aesthetic value and the depth of the emotions it produces’ (Pedersen, 2020: 87), echoing critiques of functionalism discussed earlier.

Of course, the idea that music is all around us for much of the time is not new. The concept of ‘ubiquitous listening’ is primarily an intervention in musicology. Musicologist Nicholas Cook wrote in 1990 of ‘a widespread consensus of opinion among twentieth-century aestheticians and critics that listening to music is, or at any rate should be, a higher-order mental activity which combines sensory perception with a rational understanding based on some kind of knowledge of musical structure’ (Cook, 1990: 21). However, in recent years, some music scholars have moved to dethrone such attentive and structural musical listening from musicology, and from music criticism and aesthetics, including Cook himself, and Rose Subotnik (1995), who deconstructed structural listening’s devaluation of music and musical experiences not based on such western notions of reason. Kassabian went further, by demanding much greater focus on the kinds of musical experience that most people engage in. In this respect, she sounded like a sociologist, but her book showed little interest in sociology (or psychology) of music, including work on music in everyday life. More importantly, perhaps, the analytical and political implications of Kassabian’s account were left unclear. For Kassabian, recognition of ubiquitous music and its associated listening patterns meant that music analysts should focus on ‘affect’ rather than emotion, and on sensory and physiological responses rather than meaning – a fashionable position in recent humanities in the wake of the ‘affective turn’. But there was a sense in Kassabian’s account that the ubiquity of music and of distracted listening should be accepted as part of a new digital reality. Kassabian helpfully clarifies that we shouldn’t assume the superiority of close listening, but her book is not concerned with the implications of shifting regimes of attention/inattention.

Recent critics of music streaming’s effects on attentiveness are much more concerned with such issues, and in so doing share anxieties with Adorno and certain modes of academic musicology. While we need to question universalising assumptions about close and structural listening, and understand the particular historical and cultural conditions that produced them, we might still want to value attentiveness to music in everyday life as part of a dialectical account of streaming. After all, it does seem that mindful, attentive listening might often enhance musical experience. Certainly, it seems that the most powerful musical experiences often involve situations in which ‘the music completely dominates one’s attention and shuts out everything else’, as some moving accounts suggest (Gabrielsson, 2011: 67). Another reason to value attentiveness is related to the general tendency for things often to become more interesting, absorbing and enjoyable the more people know about them, and listeners are more likely to learn more about music and its various possibilities if they pay attention, no matter what the genre. And given that musicians may well often have put considerable effort into writing, arranging and performing well, again regardless of genre, it is arguably more respectful to their work to listen closely. Expressions of anxieties about musical passivity and distractedness are helpful in providing reminders of this value.

The problem though, here as with the concerns about musical functionalism discussed in Section 1, is that critics of music streaming often seem to posit overly dualistic sets of values in assessing questions of passivity and attention. As music psychologist Ruth Herbert points out, simplistic contrasts between concentrated listening versus inattentive and everyday listening are often misleading; at best they might act as crude ways of thinking about behaviour but they do not capture the variety of internal experience involved (Herbert, 2011: 187). Herbert’s research shows that the same musical experience may feature close attention and ‘moments of multiply directed attention’ (Herbert, 2011: 187).

Beyond academia, musicians have long questioned the post-romantic fetishisation of musical attentiveness for some time – in ways that also question the excessive critique of functionalism discussed in section 1. Both Erik Satie and John Cage wished to create ‘a music which is like furniture, a music, that is, which will be part of the noises of the environment’ (Satie, quoted by Cage, 1961: 76). It’s often forgotten that when Eno built on this legacy to advocate ambient music, he did so not on the basis that music might merge into the background, but because it offered flexibility, the ability to ‘accommodate many levels of listening attention’ (Eno, 1978).

A dualism of foreground good, background bad, overly simplifies people’s ability to find beauty and revelation in the everyday. After all, for a very long time, music has been used as a ‘background’ to other activities, and not just in the form of programmed music. Perhaps the most striking example is radio, which from the early 20th century ‘established the possibility of music as an ever-playing soundtrack to our lives’ (Frith, 2002: 41). Radio’s history is riddled with commercialism, formulaic homogenisation and corporate control (Barnes, 1988), but the medium has also offered experiences where sociality, community and even joy are woven into everyday life (Douglas, 1999; Fairchild, 2012; Scannell, 1996). Rather than blocking such experiences of musical beauty and wonder, radio’s ‘background-ness’ arguably produces these qualities in the everyday, even if corporate commercialism can nullify its potential to do so. Might streaming be marked by similar contradictions? Research on such matters is still in its infancy. ⁶

Streaming Makes Tracks and Songs Shorter, and Musical Experience More Fragmented

Many commentators point to ways in which the abundance offered by MSPs reinforces a tendency towards distracted and inattentive musical experience, which then has ‘textual’ ramifications, beyond the blandness noted in Section 2. In particular, there has been a great deal of emphasis on the idea that musical tracks are getting shorter, musical attention more fragmented, and that the album form, integral to popular music for many decades, is under threat. To quote Verge’s ‘Switched on pop’ podcast, ‘we had songs that were coming in at [an average of] four minutes and 30 seconds’ in 1995, but by 2019, ‘songs are down to three minutes and 42 seconds’ (Mack, 2019; see also Kabash and Araújo, 2019). Short rap tracks are often the basis of such claims, such as a piece in Quartz that remarks on the very short length of recent hits such as Lil Pump’s ‘Gucci gang’, or his collaboration with Kanye West, ‘I love it’ (Hassan and Kopf, 2018). Shortening of tracks is widely attributed to MSPs, or to the musical system in which they now play a central part, including the now widely known fact that a stream counts as a stream only if it is played for 30 seconds or more, and does not count at all if it is played for less than 30 seconds. A later Quartz headline, for example, proclaims that ‘the economics of streaming is making songs shorter’ (Kopf, 2019). One blogger is unusually clear in delineating the causal processes supposedly involved:

When digital music services started offering listeners millions of songs on demand, it became really easy to get lost in the chaos induced by this abundance of choice. Labels realized this, and molded the creative process to better suit the nature of the medium, reducing pop music to short, easily digestible bits of information, offering listeners instant gratification with no complexity, enabling an ‘easy listening experience’. (Bagal, 2019)

Relatedly, there is a surprising amount of commentary on the idea that song introductions are getting shorter, and are changing in form, and this reflects (and affects) short attention spans. A BBC news story from 2017, for example, reported the view of record producer Mark Ralph that ‘[g]reat song intros, where a tune builds up before the vocal kicks in, are becoming an endangered species as fickle music fans skip tracks if they don’t get immediate gratification’ (BBC, 2017). The same story also reported an academic study (Léveillé Gauvin, 2018) which found that ‘intro lengths had dropped by 78% between 1986 and 2015’. During this same period, there was much coverage of the mooted death of the album as a musical form in the wake of streaming (among them Ingham, 2018). The immense success of short video app TikTok might only seem to confirm such ideas. It has been described in the Guardian as ‘the perfect medium for the splintered attention spans of [pandemic] lockdown’, where ‘everything is so fast that it doesn’t feel like it’s stealing your time’ (Haigney, 2020). Commentators point out how TikTok’s new centrality to music marketing (Leight, 2019) means that rap verses become widely disseminated before they’re even songs (Leight, 2020).

Distraction and a lack of attention have of course been widespread themes in more general social analysis and commentary, in economics and management studies treatments of ‘the attention economy’ (Davenport and Beck, 2001) or in social theory addressing social ‘acceleration’ (Noys, 2014; Rosa, 2013). However, some sociologists have questioned generalisations about technology-fuelled acceleration and time-shortage of a kind that underpin some of the concerns about musical experience in the age of streaming. Judy Wajcman (2015), for example, has shown that time poverty affects some groups much more than others – for example, single parents are far more time-poor than couples without children. She also argues that information and communication technologies (ICTs) have positive as well as negative effects, in leisure time as well as working time, and that such technologies ‘make possible new and multiple temporalities’ (Wajcman, 2015: 9) rather than homogenising the experience of time in the direction of short attention spans and distraction.

While it seems undeniable that there is increasing competition for our attention, whether this is having effects on musical texts is another matter. The very sources cited by commentators claiming that MSPs are driving shorter tracks (e.g. Tauberg, 2018) in fact provide evidence that the trend actually began around the turn of the century – much earlier than the rise of MSPs. Nor should we assume that brevity is without interest or merit. Kieran Press-Reynolds (2020) adroitly delineates the musical features of the bass-driven ‘no melody’ rap that provides the score for many of the 15-second videos on TikTok, an almost comically distorted sound that refuses the sentimentality of rap’s earlier turn to song structures. Press-Reynolds points to the condensed musical creativity involved. Concerns about shorter songs can plausibly be seen as an echo of earlier anxieties about the three-minute pop single in the wake of the rise of the recording and radio industries (Katz, 2010). A critic Dave Marsh put it in his epic book on singles, ‘there’s no way two to six minutes of music can compete with thirty to fifty’ but short songs ‘can distil an essence, seize a moment, sum up a movement, galvanize an audience’ (Marsh, 1989: xv).

There are good reasons to value close musical attention, but we should be careful not to assume that people really do listen less attentively than before or to draw conclusions about attentiveness from dubious statistics about shorter average length of tracks. After all, the headphones used by many people to access MSPs arguably enable a great deal of attentive listening. There is an abundance of long tracks on MSPs as well as the many shorter ones that lower the average length. Meanwhile, there are almost as many articles about the album’s surprising resilience as a cultural form (Blake, 2020) as about its predicted demise (Hall, 2018). In addition, anxieties about TikTok need to be placed in context. Although TikTok has grown rapidly in popularity, there is no plausible reason to think that it will somehow displace or eliminate the other musical experiences available to contemporary music listeners. In line with Wajcman’s point about the multiple temporalities potentially made available (to some) in modernity, TikTok exists alongside many other modes of musical experience, including, it would seem, a great deal of attentive and ‘self-curated’ listening.

Streaming Discourages and/or Limits Exploration and Discovery

I now turn to another dimension of the passivity encouraged by streaming, according to its critics: the way it discourages musical exploration and discovery. For Chris Richards, the ‘fluid, frictionless listening’ offered by streaming leads to ‘an experience that can be entirely predictable, even when you don’t know exactly what’s coming next’ with ‘most of the major platforms’ recommendation algorithms . . . designed to suggest music that’s similar to what you’re already listening to. Instead of going on a “trip,” streaming allows you stay put [sic]’ (Richards, 2017). New York Times critic Ben Ratliff (2016: 6) writes that ‘[i]nfinite access, unused or misused, can lead to an atrophy of the desire to seek out new songs, and a hardening of taste, such that all you want to do is confirm what you already know’. Many academic researchers have developed perspectives consistent with these views. For Rekret (2019: 63), the use of automated and personalised curation and selection by streaming services, by contrast with preceding media forms based on purchase, means that streaming ‘is a model that incites passivity in the user’. Such views are found across many sources (e.g. Bainbridge, 2019; Pelly 2018b).

What is striking about many of these claims (and those addressed in previous sections) is how uninterested many of the authors who make them seem in actual evidence concerning what people do with streaming. Whether streaming is damaging musical culture through encouraging greater levels of distractedness or passivity are surely best addressed by examining what listeners actually do with streaming, rather than merely supposing that listeners are more passive and distracted than in some previous era. Anja Nylund Hagen’s research, which drew on user diaries to avoid ‘the potential pitfalls associated with retrospective inquiries’ (Hagen, 2015: 126), identified a wide range of practices on the part of users, some of them strongly shaped by streaming services, and some independent and even idiosyncratic, exercising skill and creativity in searching and browsing, and engaging in substantial curation, often via the self-created playlist features common to many MSPs. Similarly, though some of the young Swedish and Russian streaming users interviewed by Sofia Johansson and her colleagues admitted to a new laziness or ‘indolence’, involving less knowledge of artists, and less willingness to explore (in the words of one user, ‘you expect things to come to you’ (quoted by Johansson et al., 2017: 38)), they also found evidence of great care being taken over the construction of playlists. Drawing on interviews from 2010 to 2013, Marika Lüders (2019) found that Spotify users are relentless explorers, but that they also express concern over their own restlessness. Their very awareness of this problem, and their ability to reflect on it, surely somewhat qualifies the passivity attributed to them by some critics. Quantitative evidence also challenges some of the speculations surrounding streaming. In an analysis of the music usage of a random sample of 5000 users of a music recommendation service over a one-year period, Datta et al. (2018) found that new users of Spotify significantly increased their consumption of artists, tracks and genres that they had not previously encountered, compared with the period before they started using the service.

Such evidence suggests a messy and complex picture that belies simplified and overly generalised claims. Nevertheless, we should not discount the possibility that music streaming might be harming music’s capacity to enhance people’s lives (Hesmondhalgh, 2013). Nancy Hanrahan (2018, 2019) analyses how possibilities for musical discovery, including a meaningful encounter with difference, might be foreclosed by various ways in which MSPs structure musical diversity, including dynamics of personalisation. Crucial here is Hanrahan’s conviction, quoting Susan Buck-Morss, that aesthetic experience potentially involves ‘not only critical judgement of cultural forms, but social forms of our being-in-the-world’ (Buck-Morss, quoted by Hanrahan, 2018: 289). Based on focus groups she conducted with students and young professionals, Hanrahan notes that ‘while everyone claims to have eclectic tastes’, her respondents ‘were at a loss to describe anything they heard as different, nor could they describe having an experience of difference’ (Hanrahan, 2019: 167). Indeed, she found evidence of deep fear and anxiety about using music as a means of exploring intersubjectivity, including a reliance on quantified metrics, abundantly available on MSPs, to make judgements about what to listen to (Hanrahan, 2018: 295–296). This in turn leads to situations in which judgement about music becomes increasingly private or tacit, rather than ‘sociable, deliberative, performative or public’ (Hanrahan, 2018: 296). Hanrahan also found that ‘discovery is oriented not toward what might be surprising, unsettling, new or different but toward the self – our moods, activities and memories’ (2018: 296). In a situation where users expect music to be tailored to their needs, via dynamics of personalisation, based on algorithmic and human recommendation, she detects an impatience and disappointment with music that does not fit.

Although Hanrahan’s research was exploratory, it offers a potential model for exploring how digital technologies are combining with socio-cultural forces to create a climate of consumption in which various emancipatory possibilities offered by music might be in danger of erosion, and how some potentially meaningful aspects of musical experience might be under threat. It does so by combining interview material with theory that bravely posits a positive theory of the value of aesthetic judgement and deliberation against which the interview material can be assessed – as indicated by the Buck-Morss quotation given earlier (though Hanrahan uses a range of other sources too). The relevant point for the current article is that rather than make generalisations about passivity, Hanrahan bothers to listen to actual users, and consider their experiences sympathetically but critically, in the context of a positive vision of what music might offer, rather than a projection of some previous time, when music was somehow better.