Kemp on conflict and in-group cooperation

Refs: Kemp and the rise of the state as collective action failure Bauer et al on war and cooperation Churchill on war as the human default Cultural evolution

Summary

In Goliath's Curse (2025), Luke Kemp makes a short argument that conflict is not purely destructive. External threat raises cooperation inside a group and raises hostility toward outsiders. He rests the first claim on the review in Bauer et al on war and cooperation (sixteen studies, forty countries), and adds three supports:

• Symbols and stories. Attacked groups develop shared symbols and behaviors, such as patriotism. His example is US flag sales after 9/11: Walmart sold 250,000 flags the next day, against 10,000 on the same day a year earlier.
• Tightness and prejudice. Across societies, exposure to threats such as war, disease and scarcity is the best predictor of prejudice and 'cultural tightness' (stricter norms, harsher punishments, less tolerance of deviance).
• Neural and organizational evidence. In one study, the brainwaves of Chinese students synchronized while they read about a threat from Japan, and they coordinated better on group tasks. Kemp links this to the bonding seen in terrorist cells, revolutionary groups and military units, and to military training that rebuilds recruits into a single team.

The claim in the last line of the passage is that the same mechanism raises both in-group cooperation and the capacity for violence against outsiders. This is the parochial pattern Bauer et al flag as preliminary.

Excerpts

Conflict isn’t a completely destructive force. It can spur cooperation, at least within a group. Data from sixteen studies covering forty countries all suggest that exposure to violence from war increases local cooperation within groups.47 Groups who have suffered an external attack tend to develop symbols, stories, and behaviours that bond them together, such as a shared sense of patriotism. We saw that in the wake of the 9/11 terrorist attacks in the US. The day after the attacks, Walmart sold 250,000 US flags; it had only sold 10,000 the same day the previous year. The supermarket chain was almost running out of flags by the end of the week.48 Unfortunately, that increased coordination within a group does seem to come at a cost. It requires having an ‘other’ group or enemy to rally against. Across the world the best predictor of levels of prejudice and ‘cultural tightness’ (tighter cultures have stricter norms, harsher punishments, and less tolerance for deviant behaviour) globally is how much a society has been exposed to threats such as war, disease, and scarcity. A people under threat stand together but apart from others.49

Conflict, or even just the threat of an external enemy, changes our brains. One study scanned the brains of ninety Chinese students while they read about a potential threat from Japan. Remarkably, their brainwaves started to synchronize. This led to better coordination and completing group tasks more quickly.50 It makes sense: if your group is in trouble then the ability to cooperate intensely and survive is critical. It’s a finding backed up by decades of research on terrorist cells, revolutionary groups, and squads of soldiers, all of whom develop strong emotional ties and a willingness to sacrifice themselves for each other (hence phrases like ‘esprit de corps’ and ‘brothers in arms’). It’s a phenomenon that militaries capitalize on, by training small units under duress to break down the individuals and rebuild them into a single team that is both more organized and more willing to kill. An external threat supercharges cooperation within groups and the capacity for violence against outsiders.51

— Luke Kemp, Goliath's Curse: The History and Future of Societal Collapse (2025); endnote 47 cites Bauer et al (page number not recorded)

So what

The same threat that makes a group cooperate also defines who is outside it. For coordination problems that need a large or global in-group, such as those the metacrisis literature describes, threat-driven bonding is an unreliable tool, since it works by drawing a boundary. The open design question is whether the cooperative effect can be had without the 'other', for example with a shared threat that no group is cast as the source of.