Air Flows

Breath, Voice, and Authenticity in Three Recordings

Note: This article was originally published in IASPM@Journal, the open-access journal of the International Association for the Study of Popular Music. You can find the article in the journal here.


Miley Cyrus inhales audibly before launching into the final chorus of “Wrecking Ball”. Cellist David Soyer grunts as he tears through the Beethoven’s Grosse Fuge. Saxophonist Colin Stetson’s loud circle breathing on “Hunted” pushes the limits of the physically possible. These recordings all contain traces of the musicking bodies that they purport to represent. This article will treat the breath not as excess, but rather as part of the core musical content of recordings. The breath, the grunt are sounds are often considered excess—not properly musical—yet they are crucial to the affects of these tracks. They validate the recorded performance by pointing to the lingering aura of the musicians who produced them, even while they reveal the ways in which the recording studio can reconfigure, extend, and naturalize these mediatized musicking bodies.  On recordings breath becomes a tool through which musicians and recordists can construct the embodied authenticity of the recording. It argues that the sound of breath is essential to the perception of embodied authenticity on recordings—evidenced by the development of genre-specific conventions for treating breath and constructing the body on recordings.

KEYWORDS: recording, breath, embodiment.

I’m still breathing,
I’m alive.
—Sia, “Alive” (2015)

Introduction: From Voice to Breath

Scholars of music are very used to thinking about the voice as an embodied musical sound (for example, see Brooks 2000; Feldman 2015; Rahaim 2008). Singers produce sound from within the body itself. That musical sound emanates from the chest, throat, and mouth, and is projected into a physical space where it resounds and is perhaps perceived by listeners. Once the voice departs from the singer’s body it enters a network of complex social and sonic relationships with other sounds, other forms of “musicking” (Small 1998),[i] and other types of mediation that characterize contemporary musical cultures. The perception of the singer’s voice through these networks and media constitutes an embodied act. Such embodied acts are called “ubiquitous listening” by Anahid Kassabian in order to account for the vast array of listening modes and contexts that comprise one’s affective sonic world (Kassabian 2013). Listening of this sort makes the singer’s body a central figure for analysis, via the voice, and in this regard, singing is conceived as the form of musicking that is innately human. As Richard Middleton has observed, the voice is “commonly understood…as the profoundest mark of the human” (1990: 262).

The direct connection of voice to body is suggested by Carolyn Abbate, who remarks that “singers, unlike instrumentalists, are not mutely clutching dead objects of wood or metal that ‘make music’… And singers are not media. They seem to ‘speak for themselves’” (Abbate, 1999: 483). The word “seem” is critical to Abbate’s meaning. Listeners perceive the singing voice to be the authentic voice of the person who is singing, yet the voice of the singer is simultaneously strange and familiar—uncanny, perhaps—because of listeners’ knowledge of their own bodies and their own voices. The singing voice is not only communicative; it is also the result of production techniques that are mysterious to or hidden from the listener. In psychoanalytical terms, the voice is prelinguistic, precultural, and pure. It is “the link which ties the signifier [i.e., language] to the body,” according to Mladen Dolar. “It indicates that the signifier, however purely logical and differential, must have a point of origin and emission in the body. There must be a body to support it” (Dolar, 2006: 59).

The body is figured as a precondition for the voice and all of its expressive possibilities, and the voice is presumed to emanate from a coherent and vocally capable body. Coherence—the perception of a voice emanating naturally from a body—is an important assumption for scholars who have positioned the voice as uniquely human. According to Simon Frith,

The voice as a direct expression of the body…is as important for the way we listen as for the way we interpret what we hear: we can sing along, reconstruct in fantasy our own sung versions of songs, in ways we can’t even fantasize instrumental technique…because with singing, we feel we know what to do. We have bodies, too, throats and stomachs and lungs. (Frith, 1996: 192, emphasis original).

Yet in arguing for the embodied perception of the voice, scholars like Frith do not account for the particular physiological processes that produce the voice—one of which is breath. Scholars seem much less comfortable thinking about breath as a sonic signifier in music; few do more than nod to its presence.[ii] Roland Barthes acknowledges breath, but he does so only to lament that “the whole of musical pedagogy teaches not the culture of the ‘grain’ of the voice”—which might convey something of the performer, and therefore constitutes the proper object of musical study—“but the emotive modes of its delivery—the myth of respiration” (Barthes 1977: 183, emphasis added). For Barthes, the lung is “a stupid organ”. Rather, “it is in the throat, place where the phonic metal hardens and is segmented, in the mask that significance explodes” (ibid., emphasis original). Steve Savage suggests that it is not “the actual breath that [Barthes is] referring to but rather the reliance on breath for expression” (Savage 2011: 55). Savage’s observation aligns with the ideas of renowned vocal teacher ands scholar Richard Miller, who argued that singers should strive for “immediate and silent renewal of the breath” (1997: 37, emphasis in original). Miller, like the vocal pedagogues criticized by Barthes, relegates breath to a hidden role in musical expression. However, following Savage, I see breath as more than a necessary but inexpressive complement to voice; rather I will show how the sonic presence of breath on a recording can contribute to the perception of bodily coherence.

This article aims to take a small step towards considering breath as a sonic element of embodied music production and perception. Its purpose is to move beyond the idea that breath is merely an index of the voice and to propose that it is perhaps better considered an expressive device in its own right. Just as Barthes suggests that a voice has a “grain”, we should follow film scholar Davina Quinlivan’s proposal that breath has “grain,” too (Quinlivan 2012: 137). As Serge Lacasse (2010) has recognized in his discussion of “paralinguistic features” of the recorded voice, the sonic properties of breath are tremendously affective, conveying effort, intensity, and even intimacy to listeners. Breath functions similarly to voice in that it is perceived in an embodied way by listeners, who understand the sound of breath with reference to their own bodily function.

However, breath, like voice, can be mediated and manipulated—subjected to all of the same musical and technological processes as voice. Within modern recording studios, a singer’s voice and breath can be processed and spliced to contribute to a sonic persona and a continuous performance that may exist nowhere other than in the mediatized space of the studio. The recording studio problematizes the notion of voice as coherent and, therefore, the correlating idea that a coherent voice necessarily corresponds with a coherent “musicking body”, a term developed by Matthew Rahaim with regard to Hindustani music to describe “a trained body in action, engaged mindfully in singing and/or playing an instrument” (Rahaim 2012: 2). Further, as I have already suggested about recorded musicians, Rahaim observes that the “musicking body” is at once continuous with and marked as distinct from the musician’s quotidian body: “When a singer beings to sing…it is as though…some entirely different person has replaced the person who was there a moment before” (ibid.).

The second and primary purpose of this article is to argue that the sound of breath on music recordings should be understood as a part of the expressive sonic apparatus of the recording medium. In this regard, I dispute a longstanding assumption of scholarship on music and recording—namely, that breath does not have any aesthetic meaning or value in its own right. Simon Zagorski-Thomas argues that

In any “real life” performance situation there will be both desirable sound, the noise the musicians make, and undesirable sound, any unintended noise that nevertheless happens during the performance… Close microphone placement can reduce some of the sounds but can also make others worse: the musicians turning the pages of their music, their chairs creaking, the rustle of their clothing as they move or the sound of their breathing. (2014: 59, emphasis added)

Zagorski-Thomas elaborates that the presence of an “undesirable sound” such as breathing contributes to an “unwanted realism” (ibid.) of a recording, and therefore, the removal of the sound of breath from recordings is done “to render the sound less human and more abstract” (ibid.: 60). There are at least two terms that might refer to the removal of breath from a recording: “de-essing…which can be used to reduce the volume of sibilance or breath noises” (ibid.); and “de-breathing”, a term that recording engineers sometimes use to describe the painstaking process of cutting (manually or digitally) the sound of breathing from a recording.[iii] Savage describes a piece of software called DeBreath, which is used for the “manipulation or elimination of breaths” in the recording process (2011: 53).

At the same time, the complete removal of breath from a recording is not always desirable. For instance, in addition to the removal of breath, Savage details the painstaking process of deciding which breaths best fit in which positions in a recording (2011:53). Guided by genre conventions in the recording studio, recordists must determine how much breath to leave in so that listeners will perceive the recording as ‘real’. In an earlier meditation on the topic, Zagorski-Thomas observed one question that might be troubling to listeners of a recording: “How much room for breath needs to be left in a spliced performance before we perceive it as an activity that is impossible to generate, and therefore artificial?” (Zagorski-Thomas 2007: 197). Paradoxically, those “less human” recordings with fewer (and more tightly controlled) breath sounds may be perceived by listeners as more human than an untouched recording.

The trick of producing breath on a recording lies in the ability of the musicians and recordists (a general term to refer to producers and engineers) to mediate the desire for a relatively pure rendering of a musical work with the need for a recorded performance to be perceived as human. The precise nature of this balancing is dependent on a large number of factors, primarily revolving around the particular conventions of musical genres and forms. Rock recordings conventionally permit different sorts of interventions than do classical recordings. For example, and listeners’ expectations about the ‘authenticity’ of the recorded performance varies according to genre cultural context.[iv] This article does not assert any hard-and-fast rules for how breath should appear on recordings, however, its three case studies reveal the breadth of possibilities for breath in a recording context. These dramatically contrasting recordings highlight the points of contact and overlap between the musical practices and recording conventions of apparently distinct musical forms. For it may seem that the case of the Guarneri Quartet presents the most natural and unmediated representation of breath, insofar as the sound of breath in their recording is less of an artistic intention than a by-product of their other performance practices and habits. However, I will show that the authenticity of the Guarneri recording, as embodied by the sound of breath, has much in common with the far more deliberate breathing and recording techniques of saxophonist Colin Stetson and pop singer Miley Cyrus. Each of these recordings makes particular claims about the “liveness”—the mediatized, embodied authenticity (see: Auslander 2008)—of the recorded musical performance. By attending to the sound of breath, we can understand something of the processes and conventions through which the body is produced within the collective and collaborative acts of recording music.