I am a music-maker who likes to explore the limits of what I think music is (Iyer, Reference Iyer2017, Reference Iyer2021; Iyer & Smith, Reference Iyer and Smith2016; Iyer & Taborn, Reference Iyer and Taborn2019). I am also a music scholar, and I have offered critical perspectives on the music cognition since the 1990s (Iyer, Reference Iyer1998, Reference Iyer2002, Reference Iyer2004, Reference Iyer, Lewis and Piekut2016, Reference Iyer, Rehding and Rings2020a, Reference Iyer2020b). I have long argued for an embodied understanding of music as, first and foremost, human action: the sounds of bodies in motion. As composer-pianist Cecil Taylor once put it, music is “everything that you do” (Mann, Reference Mann1981).

I'm not here to assess the relative scientific merits of these papers. I am not an evolutionary scientist, nor am I invested in the research question of why music might have evolved. Instead, I wish to scrutinize some assumptions underlying both articles. I submit that (1) the category “music” as used in this area of science is inconsistent and unstable, (2) its logical relationship to the word “musicality” – used by scientists to denote the “human capacity for music” – is circular (Iyer & Born, Reference Iyer and Born2020), and (3) this circularity conceals a problem at the heart of this scientific enterprise, namely that its very object of study does not strictly exist, but rather is constructed and managed through scientific discourse.

An influential study by Mehr et al. (Reference Mehr, Singh, Knox, Ketter, Pickens-Jones, Atwood and Glowacki2019), cited in both papers, drew from a massive corpus of “human song”: field recordings described with tags such as “song,” “voice,” “vocal,” and so forth. Although its purview is immense, it expressly does not include purely instrumental music. Yet, the domain of this study has been routinely conflated with the larger category of “music,” as is done in Table 1 of the current study by Savage et al. This category called music, however, apparently does not include poetic or rhythmic speech: in their current study, for example, Mehr et al. assert that a coxswain (whose rhythmic chanting coordinates rowers) “does not sing” and should, therefore, be taken as an example of “an alternative to music,” namely, language. Meanwhile, Savage et al. posit that we should include dance as “a core part of music-making.”

The pattern here is that common human behaviors are freely added to or subtracted from the category of music by scientists as a matter of course. These unpredictable inclusions and exclusions are presented as if normal, even commonsensical. Such arbitrary assertions cause a humanist's ears to perk up, for they are a telltale sign of what Michel Foucault (Reference Foucault1972) called a discursive formation: an accumulation of statements insistently held together, less by facts than by language, desire, and power. By extension, if musicality is defined as the species-wide capacity for making music, and music is then defined as (an arbitrarily delimited subset of) the diverse range of outcomes of that capacity known as musicality, then these two definitions orbit, reinforce, and validate each other, but do not resolve onto a definable, consistent, or stable object of study.

We can historicize our contemporary use of the word “music,” analyzing it as a contingent, historically specific concept. Our understanding of music today is bound up with our technologically mediated, market-dependent listening experiences, which condition us to think of music as an object or substance that is nameable and consumable (see, e.g., Sterne, Reference Sterne2003). Our conception of music additionally depends on western epistemologies of aurality (Erlmann, Reference Erlmann2010; Ochoa Gautier, Reference Ochoa Gautier2014), instrumentality (Dolan, Reference Dolan2013), and humanity/animality (Jackson, Reference Jackson2020; Mundy, Reference Mundy2018; Wynter, Reference Wynter2003) that emerged in the last half-millennium of European imperialism, “exploration,” “enlightenment,” enslavement, genocide, and extraction. As European “explorers” encountered their racialized others, they believed indigenous people to be at the edge of the human category, and took their musical behaviors as noise; they, like today's scientists, discursively delimited what human music must and must not be (Mundy, Reference Mundy2018; Ochoa Gautier, Reference Ochoa Gautier2014).

Instead, we must listen beyond our last century of sound mediation, beyond the last five centuries of racial capitalism that undergird western science and reason (Robinson, Reference Robinson1983/2000), and reexamine what music might have been on the timescale of 100,000 years – the era of human sociality, as evidenced in the remains of Blombos Cave (Henshilwood et al., Reference Henshilwood, d'Errico, van Niekerk, Coquinot, Jacobs, Lauritzen, Menu and García-Moreno2011; Wynter & McKittrick, Reference Wynter, McKittrick and McKittrick2015).

Recently, I posed a question on Twitter (Iyer, Reference Iyer2020b): “What's not music, but feels like music to you?” This post received an avalanche of responses: “Cooking.” “A really good conversation.” “Longing.” “Dancing, weather, grandparents.” “People being insulted in Yoruba.” “Construction sites.” “Water droplets, door creaks, astrological planet alignments, white noise, difference tones.” “My youngest son's gentle snore, my older son's footsteps, my daughter's laughter.” “The wind through desiccating leaves.” “Dogs barking.” “The unfolding of an exquisite meal.” “Shafts of sunlight through slow-moving clouds.” “Astrology, Astrophysics, Poetry, the crunch of leaves on a crisp Chicago autumn day, looking deeply in the eyes of a lover, traffic.” “The groove a washing machine or dishwasher gets into.” “The ocean.” “Love.” “Memories.” “Rollercoasters.” “Touch.” “Thought.” “Everything.” Hundreds of people from around the planet offered examples of what “feels like music,” illuminating the contested edges of the category, tracing out a larger space of musical mattering: affectively charged attending to phenomenological experience. These included human and nonhuman actions, complex sensations, emotions, thoughts, and meaningful social relations among people. The category of music encompasses many behaviors, but it is also surrounded by a vast sea of experiences that “feel like” they belong in the category too.

Therefore, rather than pursue the question, “Why did music(-ality) evolve?” let us ask more inclusively, “What experiences in humankind's deep past might have felt like music, and how, and why?” Feelings are not scientifically trivial; they are ongoing bodily activation at the intersection of biology and culture. As the tweets revealed, virtually any experience can “feel like music” to somebody; and they cannot be proven wrong, for music is felt into being. We must, therefore, treat the category of music(ality) not as one that coheres as the direct effect of a specific cause (be it social bonding, credible signaling, or anything else), but instead as a sphere of experience, opening out endlessly to human possibility.