Singh’s explanation of near-universal “super-attractors” is the most persuasive case I’ve read yet, generalising cultural attractor theory to phenomena widespread across cultures. Singh frames his approach as addressing what he calls “laws that apply across time and throughout human history” in order to solve what Boas (Reference Boas1896) called “the most difficult problem of anthropology.” This makes it sound like Singh will address what are known as “absolute universals” found in all societies without exception. But in fact, Singh focuses on “near-universal” traditions (also known as “statistical universals”; Bickel, Reference Bickel and Hogan2011; Brown, Reference Brown1991; Savage, Reference Savage and Sturman2019)

At the top of most lists of absolute universals is the uniquely human capacity for language, which has been described as “what makes us human” (Fitch, Reference Fitch2010). Arguably, next on the list might be music (for which Singh has helped provide some of the strongest evidence for absolute universality; Mehr et al., Reference Mehr, Singh, Knox, Ketter, Pickens-Jones, Atwood, Lucas, Jacoby, Egner, Hopkins, Howard, Hartshorne, Jennings, Simson, Bainbridge, Pinker, O’Donnell, Krasnow and Glowacki2019), along with things like religion, tool use, and control of fire (Brown, Reference Brown1991).

The examples of super-attractors Singh provides include multiple examples that can be seen as restricted cases of broader absolute universals such as music (e.g., lullabies, dance songs), language (e.g., hero stories, happy endings), or religion (e.g., shamanism, prosocial religions). Yet the broader, most universal domains of music, language, and religion themselves are curiously absent from the target article (Fig. 1).

Figure 1.

An adapted version of Figure 1 from Singh’s target article with an added red box containing some of the most complex and universal cultural domains. These “absolute universals” found in all known human societies without exception (e.g., language, music, religion) appear beyond the explanatory framework of super-attractor theory, and instead imply at least some genetic component.

Super-attractor theory is proposed as a phenomenon relying purely on cultural evolutionary mechanisms, without requiring any genetic component. This contrasts with alternative theories that invoke gene-culture coevolutionary mechanisms (e.g., Patel, Reference Patel and Honing2018, Reference Patel, Margulis, Loui and Loughridge2023; Savage et al., Reference Savage, Loui, Tarr, Schachner, Glowacki, Mithen and Fitch2021; Savage, Reference SavageIn press). Singh argues that the existence of societies such as the Northern Aché without lullabies (Aubinet, Reference Aubinet2024; Singh & Hill, Reference Singh and Hill2025) is a key challenge to gene-culture coevolutionary theories of the evolution of musicality. Yet this conflates the broader, universal category of music with the narrower, less universal sub-phenomenon of lullabies. Even the Northern Aché still sing other, non-lullaby songs (Singh & Hill, Reference Singh and Hill2025), as do every other known human society throughout history (Brown, Reference Brown1991; Mehr et al., Reference Mehr, Singh, Knox, Ketter, Pickens-Jones, Atwood, Lucas, Jacoby, Egner, Hopkins, Howard, Hartshorne, Jennings, Simson, Bainbridge, Pinker, O’Donnell, Krasnow and Glowacki2019; Savage, Reference SavageIn press).

The analogy of “cheesecake” recurs repeatedly to explain super-attractor theory. Pinker (Reference Pinker1997) famously used the cheesecake metaphor to argue that music (“auditory cheesecake”) and other aspects of arts and culture were non-adaptive byproducts of other adaptations, such as the capacity for language. But even Pinker acknowledges that the language instinct, at least, suggests biological adaptive explanations beyond a purely cultural evolutionary one (although cf. Reference KirbyKirby, 2017, for potential cultural evolutionary mechanisms in language evolution). Further, I have argued that this “cheesecake” metaphor is fundamentally flawed because it again conflates a non-universal sub-phenomenon (cheesecake) with the universal category it is a part of (food; Savage, Reference SavageIn press).

It is true that our capacity for synchronised dancing to a beat may have originated as a byproduct of learning, as it appears to have done in other vocal learners such as cockatoos (Patel, Reference Patel2024; Patel et al., Reference Patel, Iversen, Bregman and Schulz2009; Schachner et al., Reference Schachner, Brady, Pepperberg and Hauser2009). But among those non-human animals that have the capacity to synchronise to a beat, they do not universally employ this capacity. Humans do, which I have argued was a result of a gene-culture coevolutionary selection following its initial origins as a byproduct of vocal learning (Savage, Reference SavageIn press).

Singh correctly notes that one of the major challenges for biological explanations of music (and other cultural universals) is the “absence of evidence of adaptation.” But there is relatively strong evidence for the adaptive value of language, at least (Patel, Reference Patel2008), and the absence of evidence for music is not necessarily the same as evidence of absence. Global experiments reveal strong and cross-culturally consistent acoustic differences between singing and speaking (Albouy et al., Reference Albouy, Mehr, Hoyer, Ginzburg, Du and Zatorre2024; Anikin et al., Reference Anikin, Canessa-Pollard, Pisanski, Massenet and Reby2023; Ozaki et al., Reference Ozaki, Tierney, Pfordresher, McBride, Benetos, Proutskouva, Chiba, Liu, Jacoby, Purdy, Opondo, Fitch, Hegde, Rocamora, Thorne, Nweke, Sadaphal, Sadaphal, Hadavi and Savage2024), and follow-ups are underway to directly test the potentially adaptive social bonding functions of singing vs. speaking (Savage et al., Reference Savage, Ampiah-Bonney, Arabadjiev, Arhine, Ariza, Arnese, Bamford, Barbosa, Beck, Belyk, Benetos, Bulbulia, Cabildo, Calhoun, Chiba, Dabaghi Varnosfaderani, Dias, Duran, Echim, Færøvik and ZariquieyIn Principle Accepted).

Studies of the genetic bases of musicality and language are still in their infancy, but already a number of potential candidates have begun to emerge, such as the FOXP2 gene that could be implicated in either song and/or speech evolution (Enard et al., Reference Enard, Przeworski, Fisher, Lai, Wiebe, Kitano, Monaco and Pääbo2002; Fisher, Reference Fisher2019; Haesler et al., Reference Haesler, Rochefort, Georgi, Licznerski, Osten and Scharff2007). Genome-wide association studies have started identifying dozens of other genetic variants associated with musicality and language (Alagöz et al., Reference Alagöz, Eising, Mekki, Bignardi, Fontanillas, Nivard, Luciano, Fisher and Gordon2025; Niarchou et al., Reference Niarchou, Gustavson, Sathirapongsasuti, Anglada-Tort, Eising, Bell, McArthur, Straub, McAuley, Capra, Ullén, Creanza, Mosing, Hinds, Davis, Jacoby and Gordon2022; Wesseldijk, Ullén, & Mosing, Reference Wesseldijk, Ullén and Mosing2023). More will likely come as the quantity and quality of data and methods continue to improve.

Such new data will provide an important opportunity to test Singh’s cultural evolutionary predictions against predictions of alternative hypotheses that invoke genetic mechanisms. However, as currently framed, Singh’s predictions do not mention genetics. Future studies should aim to formulate testable hypotheses, making specific predictions for purely cultural evolutionary theories vs. those that include genetic components. I will admit, however, that this is easier said than done. We did make some genetic predictions for our gene-culture coevolution hypothesis (e.g. “genetic differences (e.g., in DRD2) may predict variation in bonding experienced through musical activities.”; Savage et al., Reference Savage, Loui, Tarr, Schachner, Glowacki, Mithen and Fitch2021). However, it is not immediately clear how to conclusively test such predictions against alternative predictions of a purely cultural evolutionary theory, such as Singh’s.

On that note, I’d like to emphasise that it is much easier to criticise than to propose constructive new theories, and my criticisms of Singh’s proposal should not detract from the value he has provided by proposing an impressive and broad new theory. I look forward to future tests of his and competing theories to help us understand the complex cultural and biological bases of human evolution.