We appreciate Whitehouse's efforts to construct a comprehensive theory of self-sacrifice. However, there currently appear to be inconsistencies between the predictions of the model regarding which people theoretically should be expected to engage in self-sacrifice and the well-documented patterns concerning which people actually do sacrifice themselves for their groups. One notable disconnect involves the relationship between age and self-sacrifice. Here, we outline this issue and make concrete suggestions for how age – and its psychological correlates – might be effectively incorporated into the generative theoretical framework that Whitehouse has proposed.

A central assumption of Whitehouse's model is that fusion is accumulative in nature: The more shared experiences that a person has with fellow group members, the more intense and/or opaque these experiences are, the more a person reflects on these experiences, the tighter are the bonds of kinship, and so on. All of these factors are said to heighten fusion with the group. This accumulative account would suggest that fusion – and therefore the propensity to self-sacrifice – should be higher among individuals who have spent more time as members of their groups. Therefore, the model would seem to predict that self-sacrifice, on average, should be more common among older individuals. These predictions, however, are at odds with empirical data concerning real-world self-sacrifice: The majority of individuals who self-sacrifice for their group are young, and have therefore typically spent less time as members of their groups. For example, the average age of Palestinian suicide bombers was 21 years (Dickey Reference Dickey2009), the average age of the 9/11 hijackers was 24.2 years (Benmelech & Berrebi Reference Benmelech and Berrebi2007), and the average age for all suicide attackers (with known ages) between 1982 and September 2015 was 23.2 years (Chicago Project on Security and Terrorism 2015). Similarly, other work has suggested that most suicide bombers spend little time as members of terrorist groups before engaging in their first (and usually only) attack (Lankford Reference Lankford2015; Pedahzur Reference Pedahzur2005).

Why is it the case that those who die for their groups tend to be young? One possibility is that this trend relates to the general tendency for younger people to commit more violent acts (Farrington Reference Farrington1986) and is unrelated to identity fusion. Indeed, research has identified a wide range of biological and social factors that contribute to the association between age and violence, including physical strength, impulsivity, cognitive competence, freedom from supervision, and greater access to resources (Casey et al. Reference Casey, Jones and Hare2008; Fonagy Reference Fonagy2003; Klausen et al. Reference Klausen, Morrill and Libretti2016; Ulmer & Steffensmeier Reference Ulmer, Steffensmeier, Beaver, Boutwell and Barnes2014). However, we propose that age may in fact interact with identity fusion effects through its relationship with identity development. Late adolescence and early adulthood are periods in which individuals strive to understand their place in the world and develop a coherent sense of self (Arnett Reference Arnett2000; Kroger Reference Kroger, McLean and Syed2015). In doing so, they typically rely on social referents, such as friends, peers, and others in their social networks, to help navigate the development of a formative identity (McLean Reference McLean2005). Often, young adults undergo intense and frequent changes in their personal and group identities until they find one that most readily “fits” within a given time and place (Grotevant Reference Grotevant, Adams, Gullotta and Montemayor1992; Newman & Newman Reference Newman and Newman2001; Tanti et al. Reference Tanti, Stukas, Halloran and Foddy2011). During this process, the groups with which individuals identify are highly salient, and the roles related to their group memberships are extremely accessible (Ashmore et al. Reference Ashmore, Deaux and McLaughlin-Volpe2004; Ethier & Deaux Reference Ethier and Deaux1994). It is plausible that this saliency of group membership among young adults is conducive to development of high levels of fusion and, therefore, leads to subsequent self-sacrifice for one's group.

Specifically, individuals who are undergoing or have recently undergone identity changes may be most likely to define themselves by their group membership, to fuse quickly and tightly with their group, and thus to be more likely to self-sacrifice, compared with others whose identities are less in flux. Supporting this possibility, some scholars have contended that a drive to find meaning and reduce uncertainty through identity and group membership can galvanize engagement in terrorist activity and violence on behalf of the group (Hogg & Adelman Reference Hogg and Adelman2013; Kruglanski & Fishman Reference Kruglanski and Fishman2009; Kruglanski et al. Reference Kruglanski, Bélanger, Gelfand, Gunaratna, Hettiarachchi, Reinares, Orehek, Sasota and Sharvit2013). To the extent that young people are especially inclined to search for meaning and a sense of purpose through their identities (Arnett Reference Arnett2000; Kroger et al. Reference Kroger, Martinussen and Marcia2010; Nelson & Barry Reference Nelson and Barry2005), identity salience could help explain why young individuals, in particular, would be inclined to resort to violence on behalf of the group.

To translate the above into concrete predictions for Whitehouse's model, we propose that there are two specific ways in which age and associated group identity salience could be incorporated. First, group identity salience could function as a moderator of the degree to which shared group experiences predict identity fusion, such that shared experiences lead to higher levels of fusion when group identity is salient. Second, group identity salience could moderate the relationship between identity fusion and self-sacrifice for the group, such that the salience of group identities among already fused individuals leads to increased willingness to die for the group. The latter explanation is in line with research by Swann et al. (Reference Swann, Gómez, Seyle, Morales and Huici2009), who provide evidence that “fused persons were more willing to fight or die for the group … when their personal or social identities had been activated” (p. 995).

Additional research will be needed to determine how age can be properly situated in Whitehouse's model. Nonetheless, we suggest that incorporating the link between age and identity salience into this model might help resolve the apparent inconsistencies discussed here, whereby younger people who have had less time to develop group fusion are more likely to engage in suicidal self-sacrifice for their groups. Although the parsimony of Whitehouse's model is appealing, it is likely that a theoretical model that can fully account for a behavior as complex as self-sacrifice will require additional refinement and nuance. We hope that the argument that we have presented here might help facilitate this process.