Baumeister et al. provide a great intellectual service by revisiting social psychology's “master problem”: explaining the connection between the individual and the collective, including groups, organizations, communities, and society. Vestiges of Allport's (Reference Allport1924) antigroup orientation continue to influence theorists' and researchers' willingness to consider group-level concepts and processes, and Baumeister et al.'s analysis is a reminder that anyone who wants to understand something about the human condition – creativity, decision-making, productivity, or even violence – needs to understand group processes because nearly all humans are members of groups and those groups have a dramatic impact on them.

Baumeister et al. get many things right. Allport did champion a psychogenic analysis in his 1924 paper (although he amended his position in his later writings, e.g., Allport Reference Allport1962). Although coordination of action is more difficult in larger groups, most organizations cope by creating and sustaining small subgroups (e.g., Kozlowski & Bell Reference Kozlowski, Bell, Schmitt and Highhouse2013). Many individuals are motivated by self-serving aims that prompt them to maximize self-interest (unless their self-system is a collectivistic one that rejects the marked distinction between self and other that characterizes Western thought, e.g., Brewer & Chen Reference Brewer and Chen2007). Groups with complex social systems do tend to triumph over ones that lack such systems (as confirmed by anthropological studies of the shift from hunter-gatherer societies to agrarian and pastoral ones, e.g., Mulder et al. Reference Mulder, Bowles, Hertz, Bell, Beise, Clark, Fazzio, Gurven, Hill, Hooper, Irons, Kaplan, Leonetti, Low, Marlowe, McElreath, Naidu, Nolin, Piraino, Quinlan, Schniter, Sear, Shenk, Smith, von Rueden and Wiessner2009). And threat of evaluation does increase the likelihood that group members will work harder, loaf less, communicate more carefully, offer more creative ideas, and resist conformity pressures, and it decreases the likelihood that they will exploit shared resources, riot, or act violently (but then again, evaluative pressures do that same thing to individuals who are not in groups).

They also raise a number of questions. Did the study of the self and groups proceed along independent lines? Social identity and symbolic interactionists would likely say no (e.g., Hogg et al. Reference Hogg, Terry and White1995). Do group researchers continually claim that groups are more than the sum of their parts? No, for the evidence of synergy in groups is thin (Larson Reference Larson2010). Is role differentiation unique to larger groups? Roles are what distinguish any group from such aggregates as crowds and audiences and are the hallmark of one of the most common of all human groups: teams (e.g., Bunderson & Boumgarden Reference Bunderson and Boumgarden2010). Does social facilitation require the threat of social evaluation by others? Social evaluation certainly enhances the effect (e.g., Harkins Reference Harkins1987), but it is not a necessary condition, at least according to studies that have shown facilitation in species that likely worry little about evaluation (e.g., cockroaches; Zajonc et al. Reference Zajonc, Heingartner and Herman1969). Do groups pass through two identifiable stages as they develop? Tuckman (Reference Tuckman1965) identified four stages, including one characterized by the formation of stable status hierarchies that promote both cohesion and individuality, but many theorists believe groups cycle repeatedly through periods of increased and decreased cohesion, productivity, and conflict (Forsyth Reference Forsyth2014).

But what of their most provocative claim, that group membership can trigger a cascade of psychological processes that results in a loss of individuality such that “the individual self is lost or forgotten as identity is submerged in the group” (sect. 4, para. 1)? Le Bon invoked this idea back in the nineteenth century when he suggested people lose touch with their individual sensibilities when caught up in a crowd. The crowd is “anonymous, and in consequence irresponsible” (Le Bon Reference Le Bon1896/1960, p. 30). Freud (Reference Freud and Strachey1922) trotted out this theory to suggest that the group, as a kind of primal horde, satisfies latent aggressive and libidinal tendencies by allowing members to abandon the regulations of the superego and follow the lead of others. In Reference Cantril1940, Cantril invoked it again to explain why a handful of people panicked when listening to the Orson Welles broadcast of The War of the Worlds. In 1969, Zimbardo, in his input–process–output model of deindividuation, theorized that in anonymous groups individuals lose their sense of responsibility and individuality and so engage in irrational, emotional, and impulsive actions.

Individuals certainly do sometimes act in unusual ways when in groups, but researchers have yet to definitively document any of these hypothesized psychological transformations. McPhail's (Reference McPhail1991) field studies of individuals submerged in mobs and crowds concluded most members of such groups act rationally. Postmes and Spears's (Reference Postmes and Spears1998) meta-analytic review of studies of deindividuation concluded anonymity and group membership rarely trigger psychological changes or that these changes mediate the relationship between situational factors and aberrant actions. Bromley (Reference Bromley, Zablocki and Robbins2001), investigating the most extreme types of group identifications (e.g., cult indoctrination, religious conversions), concluded that these processes can be explained by such quotidian mechanisms as persuasion and social networking rather than by dramatic shifts in identity.

Multilevel, multicomponent models of the self, such as sociology's identity theories (e.g., Stets & Burke Reference Stets and Burke2000), social identity theory (e.g., Hogg Reference Hogg and Wheelan2005b), and Brewer's (Reference Brewer, Van Lange, Kruglanski and Higgins2012) optimal distinctiveness theory, consider the self to be an ongoing process that integrates individualistic, collective, and relational aspects of identity in an associative network. Rather than privileging the self comprising primarily individual qualities as the “true self,” and the self that derives from group memberships as inauthentic, they suggest situational factors influence the spread of activation across this network so that some aspects of the self may be more accessible cognitively and, in consequence, regulatorily (e.g., intergroup settings, competitive contexts), but rarely will the activation of one subset of closely related elements in this set fully suppress other components of identity. Humans have sufficient cognitive resources to develop and maintain an elaborate self-system that continuously integrates multiple self-defining constituents, so only in highly unusual circumstances does the self becomes narrowly defined by only one category of self qualities (e.g., Sedikides et al. Reference Sedikides, Gaertner, Luke, O'Mara and Gebauer2013; Swann et al. Reference Swann, Gómez, Huici, Morales and Hixon2010). In most cases, personal and collective identities coexist amicably, making concepts such as mob mentality and deindividuation theoretically and empirically suspect. Humans are social, but not so social that when they join a group they risk losing touch with their individual, self-defining qualities. We are not the Borg.