Globally, men and women’s singing and speaking voices are not exactly one octave apart: Commentary on Bannan et al., “The evolution of gender dimorphism in the human voice: The role of octave equivalence”

Abstract

I was pleased to read Bannan et al.’s (In press) interesting and timely target article arguing for an evolutionary hypothesis for the under-explored phenomenon of octave equivalence. The article nicely extends and synthesizes previous proposals (e.g., Savage et al. 2021; Wagner and Hoeschele 2022; Dunbar 2012), including our own “Music and social bonding hypothesis”. As such, it may not be surprising that I agree with their basic argument that octave equivalence facilitates social bonding. But I particularly welcome the way in which they outline the potential trade-offs between social bonding and sexual selection (which we only discussed in passing) to arrive at a “compromise that allows male voices to be lower in frequency, while still acoustically complementing female and juvenile voices”. Given that the presence or absence of sexual selection for human musicality has been historically controversial (Miller 2000; Fitch 2006; Mehr et al. 2021; Bertolo et al. 2023), it is nice to see a nuanced proposal for how sexual selection and social bonding may have jointly shaped the evolution of human musicality. That said...comparison of previously published audio recordings of singing and speaking around the world suggests that the adult male:female vocal fundamental frequency (F0) ratio is greater than 0.55 (~1,000 cents), not exactly 0.5 (1200 cents / one octave) as claimed by Bannan et al....None of this detracts from the value of Bannan et al.’s hypothesis. Instead, it highlights the complex nature of the evolution of vocal communication, and the need for more research....