Eric Drott

Streaming Music, Streaming Capital surveys the political economy of music streaming. Attentive to the way streaming has reordered music’s social, technical, aesthetic, and economic bases, the book at the same time treats the transformations heralded by the rise of digital music platforms as an opportunity to rethink music’s complex relation to capitalism more broadly. Drawing on Marxist-Feminist and ecosocialist frameworks, Streaming Music, Streaming Capital highlights how music’s significance for capital isn’t just a product of its incorporation into the formal economy, resulting from its commodification or assetization at the hands of the music industries. Just as significant is music’s excorporation from capitalism, as its economic exceptionalism fuels processes of accumulation in an indirect albeit profound manner.

Across its five chapters, Streaming Music, Streaming Capital considers a number of key features of this new political economy of music: the various mediations that platforms perform in making music fit to be streamed; the contradictory economic status music assumes by virtue of the multi-sided architecture of streaming platforms; the important role played by data collection, data sharing, and corporate surveillance in the streaming economy; the changes in musical production and style prompted by the peculiar incentive structure of platforms; the kinds of mundane deception enabled by platforms’ attempts to quantify user engagement; and the way streaming helps recast music not as entertainment, but as a necessity, a cheap fix that is called on to help mitigate the deepening crises of social reproduction.

This article reviews recent literature on music, protest, and social movements. Its principal focus is on English-language research being conducted in North America and the United Kingdom, dispersed across such disciplines as music studies, social movement studies, anthropology, political science, sociology, and area studies, among others. Four recent trends are highlighted: work that stresses the importance of affect to music’s political efficacy; studies addressing the soundscapes of protest events, including the tactical use of noise and silence by activists; research on media ecosystems, with a particular emphasis on online and social media’s impact on protest movements; and work that throws into relief the contradictory and ambivalent effects of protest musicking. By drawing attention to these areas of common concern, the article aims to foster dialogue among scholars working in different disciplinary spaces, as a way of mapping the terrain where a future protest music studies might take root and flourish.

Few intellectual movements have been as influential as cybernetics was in the 1950s and ’60s. Fewer still have seen their stock fall so precipitously in the years since. Despite the growing body of literature that has reassessed this postwar “cybernetics moment” (Hayles, Kline, Pickering, Medina, et al.), its far-reaching impact remains curiously underappreciated, especially as regards music. This article seeks to redress this neglect, by focusing not on works and practices that spectacularize cybernetics (the “cybernetic sublime”), but instead on those activities, discourses, and projects that so thoroughly internalized and normalized the cybernetic ethos that it eludes notice (the “cybernetic mundane”). A first case study considers the little-known role played by information theory and cybernetics in the design of the RCA Synthesizer, one of the first instruments of its ilk to be developed. Among other things, I contend that cybernetic thinking pervaded the instrument’s conception to such an extent that it paradoxically contributed to the subsequent erasure of its influence from accounts of the instrument’s development and subsequent implementation as part of the Columbia-Princeton Electronic Music Center. The second case concerns more recent applications of cybernetic ideals to digital music distribution, exemplified by the platform Spotify, whose routinization of these ideals has ensured not just their persistence, but their persistent misrecognition.

Since 2015 a number of startups have emerged seeking to commercialize music AI. Two trends stand out: one involves firms marketing services directly to consumers, in the form of adaptive music that responds to contextual and/or activity-related cues; the other involves companies marketing AI-generated music to other cultural producers, in the form of algorithmically-generated, royalty-free production music. Initiatives like these have generated debate among legal scholars about notions of copyright and authorship. But until recently discussion has focused on who (or what) should be awarded rights over the products of so-called “expressive AI”: Its programmers? Its users? Or the AI itself? Largely overlooked in such debates is the status of another repertoire: not the music put out by an AI, but that which is put into it, the music that constitutes the training set necessary for machine learners to learn.

The dependence of machine-learning systems upon the human labor they threaten to displace raises important questions of distributive justice. To address them, this paper examines precedents for music AI that also relied on the “dead labor” of musicians (specifically, the “Music Composing Machine” developed by RCA in the 1950s), as well as proposals that have been advanced in response to more recent developments. I argue in particular that existing copyright regimes are ill-equipped to remedy the potential economic harms of music AI. Commercial practices premised on the extraction of value from a special kind of common-pool resource—the shared knowledge of a given music community—demand remedies grounded not in the methodological individualism of copyright law, but commons-based responses instead. Here, too, history provides precedents. Specifically, the Performance Trust Fund instituted by the American Federation of Musicians offers one model of how to respond to technology-induced unemployment, one that resulted from an earlier innovation in music technology: sound recording.

A specter is haunting the world of cloud-based music streaming—the specter of fake streams. Recent years have witnessed the growth of an underground market for various tokens of user engagement (plays, followers, reposts, etc.), generated by a range of human and nonhuman actors (independent contractors, clickfarm workers, automated scripts, etc.). Typically marketed to unknown and/or struggling artists eager to bring their music to wider attention, rumors nonetheless abound of major labels and artists availing themselves of such services in order to boost the popularity of tracks—and hence their royalty shares. How far such counterfeit signals of user activity have permeated the streaming ecosystem was made abundantly clear by the revelations in early 2018 that Tidal had consistently inflated both its subscriber numbers and the playcounts of headline artists like Kanye West.

This article examines the market for fake streams, attending to the way the architecture of streaming platforms have encouraged the growth of this market, even as platform operators strain to root out the counterfeit digital goods thus circulated. Two facets of the streaming economy are noteworthy in this connection. One is its formalization of the principle whereby attention begets attention, as the more streams a track garners, the more likely it is to appear in search results, playlists, recommendations, and other outputs of algorithmic filtering. The other is the role played by play counts in determining how much of the royalty pool is distributed to rights holders. Taken together, these two characteristics reinforce the sort of winner-take-all dynamic that digital music was supposed to attenuate. But they also incentivize the trade in fake streams, insofar as these promise to help musicians stave off consignment to what, playing on Marx, one might refer to as an artistic surplus population—a group rendered superfluous not just for listeners, but for the attention-extractive industries that monetizes their listening practices.

This essay stages an encounter between two different ways of thinking music’s relation to the social. One of these is historical materialism. The other is what might best be described as a neopragmatist current in music sociology, exemplified by Tia DeNora’s influential studies of music’s use as a “technology of the self,” as a means for regulating mood, modifying conduct, and inducing changes to one’s physiological state.

Motivating the dialogue fostered between these frameworks are three objectives. First, this dialogue extends the insights of neopragmatist approaches beyond the microsociological level to which they are typically confined, elucidating how the work music performs in mundane social interactions is shaped by shifts taking place within the broader political economy. Second, taking seriously the proposition that music is itself a technology offers a way past some of the impasses of older Marxian accounts of music. Rather than treat music as part of the ideological superstructure, rather than ruminate on the processes whereby music has been progressively commodified (or resisted commodification), the argument advanced here highlights music’s productivity. This points to the essay’s third and most important aim, which is to use the insights generated from the encounter of neopragmatist and materialist approaches as a way of shedding light on music’s changing functions within contemporary neoliberalism. Above all, music’s utility as a “technology of the self” has made it ripe for conscription as a technology of social reproduction, especially as new technologies of music distribution promote (and normalize) its therapeutic, prosthetic, and self-regulatory uses through the activity- and mood-based playlists that popular music streaming services. Coupled with the downward pressure that digitization exerts on music’s price (to say nothing of artists’ compensation), these developments have transformed music into a cheap resource that can be harnessed to replenish the cognitive, affective, and/or communicative energies increasingly strained by neoliberalism’s “crisis of care.”

This article explores questions of music use, commodification, and online surveillance in cloud-based music streaming services. Key catalysts in the transition from ownership- to access-based models of music distribution, services like Pandora, Spotify, Deezer, and others have positioned themselves as a means whereby listeners may be reintegrated into a “digital enclosure,” a space over which rights holders can exercise greater control. Yet online streaming’s promise of remonetizing musical commodities previously demonetized by practices of file sharing has been called into doubt by difficulties in converting users of advertising-based “freemium” services into paying subscribers. This has impelled Pandora, Spotify and others to develop alternative means of extracting value from users. Streaming sites have thus transformed into enterprises whose business is not limited to the sale of music-related services, but relies increasingly upon the collection, aggregation, and exchange of user data.

A key issue this article pursues concerns the changing status of music within the commercial strategies of online streaming. While previous research has indicated how various features, functionalities, and interfaces serve to distinguish competing services, less attention has been paid to the way they position themselves vis-à-vis other new media companies also trading in user data and user-commodities. Notable in this respect is how music figures into marketing campaigns directed not at consumers, but at prospective advertisers and investors. Close examination of music’s representation in such marketing discourse underlines how it too has been transformed by the logic of user monitoring and commodification. Such discourse casts music as a medium that offers streaming services, advertisers, and data brokers privileged access to listeners’ innermost selves. But it also casts music as an ideal tracking device, pervading our everyday lives, insinuating itself into any and every activity, and accompanying individuals across the social, physical and geographical spaces they traverse. In this way, the very attributes that make music so powerful as a “technology of the self” facilitate its transformation into an equally powerful technology of surveillance.

This article explores how human curation and algorithmic recommendation are figured in cloud-based streaming platforms. In promoting their services as alternatives to illicit file-sharing, platforms such as Spotify, Deezer, and Apple Music have long touted the access they provide to a massive database of music. Yet the effectiveness of appeals to musical plenitude have been thrown into doubt, as high rates of user turnover threaten streaming's economic viability. Curation and recommendation have thus been posited as solutions to this problem, as means of producing and reproducing consumer desire. By attending to the fantasies woven around streaming and music recommendation more specifically, this article highlights the peculiar form of subjectivation at work in the way recommendation hails listeners. The normative listener constructed through such modes of hyper-personalized address is ideally one that is as dynamic and adaptive as the algorithmic systems that adjust to their fluctuating needs, dispositions, and desires.

This article asks how we might map the complex relations between music and socialism. If an abiding concern of socialism is the collective appropriation of the means of production, what are the points at which music might participate in this project? Three moments from the history of music's entanglements with socialism shed light on this question. The first considers music's role in workers’ struggles in Europe c. 1900. Here what is at stake is how music and musical associations (most notably workers’ choruses) were mobilized as a tool in the struggle to socialize the means of production. A second vignette turns to France after May 1968, when music itself became a site of political intervention. Here what is at stake is the struggle to socialize the means of musical production, part of the struggle to institute a truly democratic culture. The third vignette turns its attentions to the proximate future, considering music's place in emerging forms of digital capitalism. Crucial in this regard is the way music is being transformed into a means of production in its own right – a means, that is, of producing the kinds of subjects required by contemporary capitalism – as well as the political constraints and possibilities these shifts present for a socialism of the future.

Writing in the wake of 1968, critic Louis-Jean Calvet observed that during periods of revolutionary upheaval people seldom sing songs born of the moment. “One always sings the previous revolution” he declared, his remark echoing Marx’s claim (in the Eighteenth Brumaire of Louis Bonaparte) that in “epochs of revolutionary crisis [the living] anxiously conjure up the spirits of the past to their service.” The event that prompted Calvet’s reflections, the uprising of May-June 1968 in France, appeared to bear out his claim. If any music can be said to have provided the soundtrack to the street protests and strike actions of May ’68, it was not so much contemporary chanson or jazz, but songs drawn from the established canon of late nineteenth- and early twentieth-century chanson révolutionnaire, most notably the “Internationale,” ritualistically intoned at the meetings, marches, and rallies that animated the movement.

This paper considers music’s perceived belatedness with regard to May-June 1968. More specifically, it considers the anxieties that this perception aroused in musicians, and the strategies that were developed to mitigate the troubling sense that the time of musical production was out of step with that of revolutionary action—what Peter Schmelz has described as the sentiment that music in general and art music in particular “tends to lag behind history.” Of note is the incorporation of field recordings made during the uprising of May-June ’68 into music produced after the fact. An examination of two works created in the months following May ’68—Guy Reibel’s and François Bayle’s mixed media composition Rumeurs and Colette Magny’s song “Nous sommes le pouvoir” (“We are the power”)—sheds light on how “sound documents” (to borrow Andrea Bohlman’s term) were deployed to confer sociopolitical authority upon music and musician alike. Construed as a way of collapsing the distance separating musical action from political action, recourse to documentary recordings in these pieces also sheds light on a different kind of politics at play around May ’68: the politics of music’s presumed separation from politics.

This paper addresses music's role in social movements, using work in mediation, assemblage, and actor-network theories to address the complex, contradictory, and contentious status that music often assumes within political protest. Three premises guide the discussion. First, if music is multiply mediated, then it follows that music's mediation of other activities (including protest) is likewise multiple. Second, this constellation of mediations may be productively conceptualized as an assemblage, with music forming part of movement-assemblages, and movements forming part of musical assemblages. Third, this mutual imbrication of musical and political assemblages— and the reversibility of positions it entails—introduces a source of potential friction. To illustrate these points, the second part of the article examines the controversies surrounding the drum circle that animated Liberty Park during the Occupy Wall Street protests of 2011.

This paper considers how the importance attached to psychoacoustics and music perception within the French new music community since the 1970s has enabled stylistic elements evocative of tonality to be simultaneously rehabilitated and disavowed by composers. Beginning with the spectral movement in the 1970s and institutionally enshrined at IRCAM in the 1980s was a tendency for composers to legitimize their work through what I refer to as the ideology of psychoacoustic realism—namely the notion that musical communication could be vouchsafed by paying greater heed to fundamental mechanisms of human perception and the constraints these imposed. Among other things, this stance allowed composers to frame their use of consonance, periodicity, polarities of tension and release, and directed harmonic motion—all taboo according to the prevailing modernist doxa—in a way that accommodated the demands of this doxa. Stylistic traits that might otherwise have been criticized as marking a retreat could instead be characterized as an advance, in part because of their alleged grounding in the empirical realities of human perception, in part because of the aura of scientific rationality surrounding the discourse of psychoacoustics.

To illustrate these themes I examine the influence that research in timbre perception exercised on the stylistic development of Kaija Saariaho, a Finnish composer who has resided in Paris for the past thirty years. In a series of essays published during and shortly after her tenure at IRCAM in the mid-1980s, Saariaho outlined a compositional approach that would substitute timbre for harmony as the main determinant of musical form. The models of timbral space formulated by Saariaho and realized in her compositions not only bore the imprint of spectralist thought, but also—and more significantly—the tonal frameworks they were intended to supplant. Of note is her evident discomfort in drawing such parallels with tonality. What this ambivalence indicates is how Saariaho’s desire to develop a musical language capable of performing the same narrative functions to which tonality is so well-adapted ran up against the aesthetic norms of the new music community in France, norms she had clearly internalized and which made too overt a reconciliation with tonal means professionally perilous.

In the three decades since it first appeared in English translation, Jacques Attali’s Bruits has come to occupy an important place in Anglo-American music studies, a standard point of reference in discussions concerning music’s social mediations. Much of the book’s appeal derives from its inversion of the relationship conventionally held to exist between music and the social. In Attali’s account, music functions not as a medium that passively registers the influence of extrinsic social forces, but as an augur, its sonic patterns providing a presentiment of some future socio-economic order. Bruits thus turns the traditional Marxian understanding of the relation between base and superstructure on its head: music, long seen as standing at a remove from political economy, is instead placed squarely at its center.

Yet for all of its suasive power, the central claim of Bruits—that musical change prefigures socio-political change—is also its least substantiated. This article seeks to make sense of the lacuna at the center of Bruits by situating it in the context of the political debates taking place in France at the time of its initial publication in 1977. Attali’s position as one of the leading intellectuals of the Parti socialiste, his role in the ideological reorientation of French socialism, and his involvement in disputes pitting the party against various rivals—above all the Parti communiste—find expression in key concepts used to support his claims regarding music’s prognostic power. But they also find expression in the model of socio-economic change sketched in Bruits, a model that has remained an enduring feature of Attali’s thought to the present day—a fact borne out by the heavily revised version of the book published in 2001.

Taking heed of the particular context out of which Bruits was born compels a reassessment of Attali’s arguments, as well as the influence they have exercised on Anglo-American music scholarship. This reassessment reveals the degree to which the brand of libertarian socialism sketched in Bruits laid the intellectual groundwork for the policies of deregulation and economic liberalization that Attali now champions in his capacity as a prominent public intellectual and advisor to successive governments of both the left and the right. Beyond this, it indicates the need to rethink certain concepts central to Bruits, most notably that of noise itself, whose political valence can no longer be assumed within the reigning neoliberal order.

This article presents a critique of the commonplace trope that holds genre to have declined in relevance under modernism. Contrary to the widespread notion that composers’ repudiation of received tradition rendered the very idea of genre categories obsolete, this article argues that such categories have never ceased playing a decisive role in the production, circulation, and reception of post-1945 art music. In interrogating the assumptions that underpin the “decline-of-genre” thesis, this article underlines the utility that renewed attention to genre and its framing effects may have for the analysis of this repertoire. To this end, an alternative to standard theories of genre is advanced, one that draws on actor-network theory to destabilize categories too often conceived as fixed, solid, and binding. This revised theory of genre is applied to Gérard Grisey’s six-part cycle, Les Espaces acoustiques (1974-1985). Habitually regarded as an exemplar of spectral music, Grisey’s cycle may be understood as participating in a number of additional generic contexts at the same time. Taking such generic overdetermination into account not only sheds light on the range of conflicting interpretations that Les Espaces acoustiques affords, but also suggests how music analysis might better address the heterogeneous contexts and multiple listener competences that this and other musics engage.

This paper reconsiders what is signified by the term “protest music,” taking as its point of departure an ephemeral performance of sorts that took place during the 2016 protests against the loi travail in France. During a march in Paris on May 26, a bottleneck resulting from the confrontation of protesters and police led a handful of activists who found themselves pressed against a corrugated partition to begin beating out a rhythm on the aluminum sheets, imitating the scansion of one of the more popular chants of the protest movement. This incident—which was over less than a minute after it begun—raises a number of questions. Does this fleeting, impromptu performance count as protest music? If it does not, why not? If it does, what are the conditions that would have to be satisfied to recognize it as such? And what, if anything, is gained by revising our notion of protest music so that we might hear it as such?
In pursuing these questions, this paper enters into dialogue with a number of scholars who have drawn attention to the limitations that terms “protest music” and “protest anthem” have historically imposed on musicking and politicking alike (Brooks 2015; Bohlman 2017; Tausig forthcoming). The paper begins by examining some of the unspoken ontological commitments embedded in received notions of protest music, before suggesting an alternative way of conceiving the interaction of protest and music, one premised on their reciprocal mediation. It then turns to the conditions whereby forms of sonic dissent can be recognized as such, considering the way in which genre (in the case of music) and tactical repertoires (in the case of protest) furnish normative horizons of action and expectation, and how the interaction of the two may reshape our conceptions of what counts as music, what counts as protest, and what counts as musical protest. The final portion of this paper explores the question of how protest tactics may be evolving at present, including sound-based ones, and considers what the responsibility of music scholars may be in tracking such changes.

This paper explores transformations that notions of musical genre have undergone in response to developments in the field of Music Information Retrieval (MIR). With the growth of online streaming services, algorithms developed to identify, classify, and search large libraries of music files have become integral to music distribution infrastructures, their effects registered, if indirectly, in the playlists generated by Spotify, iTunes and Rdio, among others. As has been noted with regard to search engines and face recognition software, the power wielded by these algorithms is inversely proportional to their transparency. Because their procedural logics generally operate below the threshold of users’ awareness, the results they return are liable to be taken at face value; and because the assumptions guiding these algorithms are equally obscured—assumptions governing how they go about relating songs, artists, and genres, and how they define the categories they present as given—it is often difficult to discern to what extent these assumptions act upon and perhaps even create the musical realities they purport to describe.

A key question this paper examines concerns how everyday musical categories are being reimagined in and through the genre recognition algorithms developed by MIR researchers. A number of mediations are required to make music amenable to the kind of measurement, comparison, and categorization that MIR systems entail. By treating classification as a computational problem, genre recognition engines transform songs into feature vectors and genres into statistical clusters. In the process, the relation between genre and similarity is altered in ways that have significant implications for the ontology of genre. Furthermore, the criteria by which genre categories are enacted shift as new types of user information are collected, aggregated, and analyzed by online music distributors, perhaps heralding a future where the articulation of genre to the social no longer centers on identity but on function.