“Music” refers to and is studied as a set of cultural practices. As such, understanding its nature and origins requires some specification of how social interaction shapes, channels, and scaffolds cultural practices. Yet, both articles only emphasize the role of global, psychological factors in the cultural evolution of music – and in doing so they risk obscuring key aspects of the problem. Savage et al. highlight stress reduction, cooperation, group identification, and other such factors that contribute to social bonding. Mehr et al. remark that features of music “will interact with evolved capacities for nonmusical traits… increasing variability in music.” Here, “evolved” means (roughly) global, psychological factors, and indeed such factors clearly help to explain why, for instance, some different song types (dance, lullabies, healing, and love) exhibit cross-cultural similarity and robust form–function relations, discernible by listeners of other cultural backgrounds (Mehr, Singh, York, Glowacki, & Krasnow, Reference Mehr, Singh, York, Glowacki and Krasnow2018). Nevertheless, factors relevant to cultural evolution can also be local rather than global, and ecological rather than psychological. In Table 1, we list a few specific examples of each type relevant to the cultural evolution of music.

Table 1. Factors relevant to cultural evolution, with examples relevant to the cultural evolution of music

The terminology of local and global, ecological and psychological, is taken from Cultural Attraction Theory (Sperber, Reference Sperber1996; Scott-Phillips, Blancke, & Heintz, Reference Scott-Phillips, Blancke and Heintz2018).

As a specific case, consider the evolution of violin design (Tai, Shen, Lin, & Chung, Reference Tai, Shen, Lin and Chung2018). Early violins, in particular those developed by Andrea Amati (1505–1577), had acoustic properties characteristic of male singers. However, as female voices became increasingly common and popular in Baroque music in the early 1600s, violin designs changed in ways that complemented their sound. In particular, the designs of Antonio Stradivari (1644–1737), widely considered to be the gold standard of violin making, have acoustic properties that closely resemble those of female singers (ibid.). It seems, then, that the acoustic properties of the female voice were an important ecological factor in the evolution of violin design (see also Miton, Wolf, Vesper, Knoblich, & Sperber, Reference Miton, Wolf, Vesper, Knoblich and Sperber2020, for experimentally induced demonstration of how ecological factors can influence the cultural evolution of rhythms).

We expect the authors of both target articles would agree that existing musical behaviours, and associated mental representations, interact not only with universal features of the human mind, but also with non-universal features of minds, and features of the world beyond the mind. Moreover, the sheer number and range of such interactions is how massive cultural variation can emerge from human psychic unity. However, the absence of any clear exposition of these points from their papers is, we believe, an oversight that risks, unnecessarily, downplaying important aspects of the problem.

In any case, it is revealing to ask in what ways, if any, these cultural evolutionary considerations might inform our understanding of the origins of music. Here, the two target articles seem to differ. At one extreme, it may be that music's origins are largely independent of its cultural evolution. This is (more or less) the approach adopted by Mehr et al., who distinguish between the “proper domain” and the “actual domain” of musicality and its associated cognitive capacities, and attempt to distinguish the contents of each (see also e.g., Honing, ten Cate, Peretz, & Trehub, Reference Honing, ten Cate, Peretz and Trehub2015). After all, cultural items frequently fall inside the actual domain but outside the proper domain of the relevant cognitive capacities, such as, for instance, portraits, caricatures, and masks, all of which trigger cognitive processes for facial recognition, but none of which fall within the proper domain of those cognitive processes (Sperber & Hirschfeld, Reference Sperber and Hirschfeld2004). At the other extreme, it may be that music's origins are irreducibly intertwined with its cultural evolution. By placing significant weight on gene-culture coevolution, in which musical practices impact biological evolution because of their impact on social bonding, which in turn feeds back on culture, Savage et al. tend more towards this pole.

The different empirical claims developed in the two target articles thus seem to reflect, in part, different tacit and prior assumptions about the relationship between cognition and culture. To what extent is any particular aspect of the human cognitive phenotype (e.g., musicality) shaped, by natural selection, in order to acquire specific cultural phenomena (e.g., music)? This is a major question for the evolutionary human sciences, relevant not just to music but numerous cultural domains (language, religion, and politics, inter alia). A range of answers are possible. In the specific case of music, we do not believe that any answer will achieve strong scientific consensus unless and until there is a more detailed analysis of the factors that shape its cultural evolution, as it proliferates from its proper domain to its (larger) actual domain. It is certain that these factors are not exclusively global and psychological.