There is much to like in De Dreu and Gross (D&G)’s article, which develops a unique perspective on the differing psychological and neural processes of attack and defense. In a sweeping synthesis of findings from behavioral economics, neuroscience, cognitive science, and social psychology, the authors distinguish the unique properties of attack and defense. These differences generate several novel hypotheses and explain why it is easier to defend the status quo than bring about societal change.

Things get more complex, however, when delving further into two interrelated aspects of the theory: (1) the reality of real-world conflicts in which both sides construe their position as one of defense, not attack, and (2) the central role of anger in defensive intergroup conflict, which blurs the distinction between active attack and vigilant defense. The galvanizing and cohesive effects of anger help explain why political leaders justify attack as defense.

Consider first what may be the bigger problem for D&G, the tendency of leaders to “sell” attack as defense. The study of public opinion during war time provides clear evidence that Americans are most willing to engage in war when faced with a serious threat to national security, but, generally, reluctant absent immediate danger (Aldrich et al. Reference Aldrich, Gelpi, Feaver, Reifler and Sharp2006; Herrmann et al. Reference Herrmann, Tetlock and Visser1999; Jentleson Reference Jentleson1992; Jentleson & Britton Reference Jentleson and Britton1998). National leaders may understand instinctively the psychology of attack and defense and reframe conflicts as defense even when they constitute attack. The Iraq War provides a compelling example. In the leadup to the war, the Bush administration advanced several specious or erroneous claims concerning Saddam Hussein's attempts to build nuclear weapons and connections to Al Qaeda that linked him to the 9/11 terror attacks (Althaus & Largio Reference Althaus and Largio2004; Feldman et al. Reference Feldman, Huddy and Marcus2015; Kull et al. Reference Kull, Ramsay and Lewis2003; Liberman & Skitka Reference Liberman and Skitka2017). The administration persuaded a majority, if not all, Americans of Hussein's villainy and the necessity of a defensive attack on Iraq.

Collective action, which fits D&G's definition of attack as a challenge to the status quo, is also grounded in defensive psychology. In D&G's Attack-Defender Game, attackers and defenders have equal resources. But most political conflicts involve unequal power, complicating the subjective perception of attack and defense. For the less powerful, collective action can be viewed as a defense against existing ills such as sexual harassment (#MeToo), systemic racism (Black Lives Matter), or the erosion of living standards for the French working class (Yellow Vests). The status quo involves continued threat and loss for the less powerful consistent with evidence that institutions, norms, and leadership help structure the status quo and the value of its alternatives (Aldrich et al. Reference Aldrich, Gelpi, Feaver, Reifler and Sharp2006; Gelpi Reference Gelpi1997). Furthermore, successful collective action requires strong group identities, a clear sense of efficacy, and a shared sense of grievance and deprivation, features described by D&G as emblematic of defense not attack (Klandermans & van Stekelenburg Reference Klandermans, van Stekelenburg, Huddy, Sears and Levy2013; van Zomeren et al. Reference Van Zomeren, Postmes and Spears2008).

Our second concern is that D&G oversimplify the neuropsychology of attack and defense. They deserve credit for the development of a parsimonious model grounded in a basic neuropsychological dichotomy. In this scheme, attack is linked to reward, behavioral activation, approach, and positive emotions such as enthusiasm and hope, whereas defense is associated with punishment, behavioral inhibition, vigilance to threat, avoidance, and the negative emotions of fear and anxiety. This dichotomy is taken to its extreme when attack is equated with trait psychopathy and defense with trait paranoia. The binary distinction may aptly characterize reactions to potential gains and losses among individuals. Both animal studies (Crofoot & Gilby Reference Crofoot and Gilby2012; Kitchen & Beehner Reference Kitchen and Beehner2007; Rusch Reference Rusch2014b) and historical conflicts (Glowacki & Wrangham Reference Glowacki and Wrangham2013; Rusch Reference Rusch2013), for example, Allied Europe versus the United States in World War II, suggest differences in the cost/benefit structure of intergroup attack versus defense. D&G wish to extrapolate to broader societal conflicts, however, that are inherently intergroup, changing defensive emotional and behavioral responses from anxiety to anger, a shift that is especially pronounced among strong group identifiers who believe the group can prevail (Mackie et al. Reference Mackie, Devos and Smith2000).

The link between anger and group defense poses an obvious challenge to D&G's model. Anger is a common response to group threat, especially among highly identified group members, and, in that sense, is integral to defense and collective action (Huddy et al. Reference Huddy, Mason and Aarøe2015; Van Zomeren et al. Reference Van Zomeren, Postmes and Spears2008). But problematically for D&G, anger is a defensive emotion that is linked to behavioral approach, occurs in regions of the brain associated with approach, and is linked to other positive (not negative) emotions (Berkowitz & Harmon-Jones Reference Berkowitz and Harmon-Jones2004; Carver Reference Carver2004; Harmon-Jones & Allen Reference Harmon-Jones and Allen1998; Harmon-Jones & Sigelman Reference Harmon-Jones and Sigelman2001). D&G acknowledge the prevalence of hostile attributions in defense, but they do not take the extra step to link hostility and anger. The absence of anger from their model is jarring and poses clear problems. They repeatedly characterize defense as avoidant and linked to behavioral inhibition, but anger does not fit that profile. It is an approach emotion that is associated with reward and behavioral activation, changing the essential psychological nature of defense. In that sense, D&G's characterization of defense as vigilant avoidance is overly simplistic.

The fundamentally different nature of individual and intergroup defense needs better explication in D&G's model. As they note, group cohesion intensifies under conditions of group defense, consistent with the well-documented effects of threat. This helps explain why group leaders transform attack into defense. The creation of a threatening enemy is an effective way to foster group solidarity, generate anger, decrease risk aversion, and promote action. Leaders would not adopt this strategy if it led to risk aversion, avoidance, and inaction. As noted in several places in the manuscript, leaders specialize in the creation of threatening out-groups and enemies, which is difficult to explain without a more complex portrayal of the psychology of intergroup defense.