Whitehouse proposes that altruistic self-sacrifice stems from identity fusion, “a visceral feeling of oneness with the group,” which stems from shared, traumatic, or at least life-altering experience and “perceptions of shared essence” resulting in an all-or-nothing relationship, where you are either included or not.

Relational models theory (Fiske Reference Fiske1991; Reference Fiske1992) posits that communal sharing (CS) is a relationship of equivalence and unity, in which people feel they have a common essence. They communally coordinate resources, decisions, responsibilities, tasks, or whatever matters. Underpinning this relationship of altruistic solidarity – prototypically implemented among close kin – is the mathematical structure of an equivalence relation, homologous to a nominal scale of measurement that results in categorical in–out distinctions. The validity of the CS construct has been established in hundreds of studies by hundreds of authors using all kinds of methods (http://www.sscnet.ucla.edu/anthro/faculty/fiske/RM_PDFs/RM_bibliography.htm). Fiske and Rai (Reference Fiske and Rai2014) have shown that CS and the other three relational models are moral frameworks that motivate most violence across cultures.

Although identity fusion (IF) is remarkably similar to the CS construct, we fundamentally differ from Whitehouse in our theory of the origins and functional scope of this sense of equivalence.

In Whitehouse's proposal, it is unclear why traumatic experience with present others should result specifically in “visceral oneness” with them, rather than any other relationship, such as relative ranking (e.g., Who is most brave, most loyal, or even dominant?). In contrast, relational models theory explains the range and specificity of experiences of consubstantial assimilation that bond persons in CS relationships by making their bodies equivalent or contiguous (Fiske Reference Fiske and Haslam2004). CS results from giving birth, nursing, feeding, commensalism, sharing bodily substances (e.g., blood brotherhood rituals), caressing, cuddling and sleeping together, intimate sex, synchronous rhythmic movement, and marking or modifying the body (e.g., circumcision or clitoridectomy), thereby creating the impression of one merged, social body containing the same essence. These bonding experiences are rarely painful, fearful, or in any way traumatic: Shared traumatic experience is not necessary for even the most intense CS, nor is reflection on episodic memory. Furthermore, CS relationships are adaptively essential to, and universally used for, the coordination of labor, consumption, use of resources and land, and everything else social: The feeling of oneness is not an adaptive specialization primarily for killing or dying for the group.

In contrast to our view that CS relationships implement and instantiate an evolved, innate, intuitive relational form, Whitehouse posits that identity fusion results from individual cognitive deliberation. It arises from “internal processes of reflection and individual learning” (sect. 6, para. 7) about dysphoric experiences, where episodic memory of traumatic or life-altering events binds the self to the particular people who participated and to their common social identity. This view predicts that infants could not feel or understand the altruistic implications of social oneness/equivalence, insofar as they are not yet capable of extended cognitive reflection, they have not yet solidified a self-concept, and their explicit episodic memories are limited. But CS and its accompanying altruism are readily intuitive to infants. Indeed, we posit this must be so if they are to solve the fundamental learning problem of figuring out who relates to whom and how (Fiske Reference Fiske1991; Thomsen & Carey Reference Thomsen, Carey, Banaji and Gelman2013).

For example, reflecting the role played by synchronous motion, spatial closeness, and looking and acting alike in constituting CS relations of equivalence, infants expect individuals who move in synchrony and close to each other to later act alike. They infer that imitation, as well as shared ritualistic actions, marks social affiliation motives and group membership (Liberman et al. Reference Liberman, Kinzler and Woodward2018; Powell & Spelke Reference Powell and Spelke2013; Reference Powell and Spelke2018). They prefer those who help similar and hinder dissimilar others (Hamlin et al. Reference Hamlin, Mahajan, Liberman and Wynn2013) and copy the food preferences of helpers over hinderers (Hamlin & Wynn Reference Hamlin and Wynn2012). They expect altruistic support to be selectively directed to in-group members who are marked to look alike (Jin & Baillargeon Reference Jin and Baillargeon2017), overriding fairness considerations (Bian et al. Reference Bian, Sloane and Baillargeon2018). They also respond with increased helping to primes of closeness and affectionate touch between dolls (Over & Carpenter Reference Over and Carpenter2009).

Whitehouse cites experimental evidence that episodic recall of dysphoric experience correlates with altruistic self-sacrifice and that this is mediated by measures of fusion. From this, he infers that traumatic (or at least life-altering) experiences are somehow the essential root of feeling one. However, these effects need not imply that the basic representation of unity and the communal motivations and expectations it entails are created by shared dysphoric, life-altering experiences. Only a small proportion of the people who have had a shared dysphoric experience kill or die because of it. On the contrary, collective trauma typically leads to extraordinary kindness and compassion to everyone around (James Reference James1906; Lessa Reference Lessa1964; Oliver-Smith Reference Oliver-Smith1986; Solnit Reference Solnit2009), and only a small proportion of those who kill or die for others have ever met most of the people they kill or die for.

Whitehouse proposes that fusion to imagined communities results from projecting local bonds from personally shared experiences to entire social categories such as nation and religion. However, rather than scaling up a local bond to the whole category, recent empirical evidence demonstrates that people may also fuse to ethnic out-groups across the globe with whom they share political ideology, but no dysphoric experience, let alone shared participation in a face-to-face event (Kunst et al. Reference Kunst, Boos, Kimel, Obaidi, Shani and Thomsen2018). This suggests that what is “projected” cannot be concrete personal experience, but must be an innate, intuitive, social relationship of CS. Indeed, intense CS relationships are ubiquitous among people who have never had a traumatic experience or ever met, yet will kill or die for each other (Fiske Reference Fiske1991; Fiske & Rai Reference Fiske and Rai2014; Ginges et al. Reference Ginges, Atran, Medin and Shikaki2007).

Personally shared experiences of trauma are an important bonding mechanism, and people surely reflect on such experiences. However, few, if any survivors of school shootings, for example, are thereby motivated to kill and die for their schoolmates. Feeling and understanding communal sharing are fundamental to human living but do not require or typically stem from cognitive reflection on episodic memory of shared traumatic experiences.