Start a long way back, because Greene does, and the chapter only works from there.
Our ancestors were badly equipped. Animals had speed, teeth, claws, fur against the cold. Humans had none of it, and must have felt terrifyingly exposed and alone.1
The only way to compensate for such weakness was to form groups.1
The group gave defence against predators, effectiveness in the hunt, and enough people to watch your back. Larger groups allowed the division of labour, which freed individuals from immediate survival for higher tasks, and the roles reinforced each other. The net result was an increase in human strength.1
Then settlements, then towns, and eventually people forgot the fear that had made them form tribes at all.
But in one group—the army—that primal terror remained as strong as ever.1
Ancient warfare was hand-to-hand: individuals exposed to death from behind and on both sides at every moment. Commanders learned early to form soldiers into tight, cohesive ranks, because a man who trusts the men beside him not to run can fight the man in front of him with more spirit.1
The Romans went further, placing the youngest and most impetuous in the front ranks, the most experienced in the rear, and everyone else in the centre — so that the weakest soldiers, the ones most prone to panic, were surrounded by the braver and steadier.1
No army went into battle with more cohesion.
The nineteenth-century military writer Ardant du Picq noticed something odd about the most celebrated ancient battles: the losses were fantastically disproportionate. A few hundred for the victors at Cannae and Pharsalus; thousands upon thousands among the defeated.1
His explanation is the mechanism this whole chapter runs on.
Through maneuver, the winning army had surprised the enemy and splintered its lines into parts. Seeing their ranks break up, losing their sense of solidarity and support, feeling isolated, soldiers panicked, dropped their weapons and fled — and a soldier who turns his back is an easy soldier to kill.1
These great victories, then, were essentially psychological.1
Hannibal was vastly outnumbered at Cannae. He made the Romans feel vulnerable and isolated, they overreacted and retreated in confusion, and the slaughter followed. The full treatment is at Du Picq and the Terror of the Isolated Soldier.
The claim Greene extracts is that nothing about this has changed:
Lurking deep in even the most civilized among us is the same basic fear of being alone, unsupported, and exposed to danger. People today are more dispersed and society is less cohesive than ever before, but that only increases our need to belong to a group.1
cut people off from their group—make them feel alienated, alone, and unprotected—and you weaken them enormously.1
The chapter contains two genuinely different strategies under one name, and separating them is the most useful thing a reader can do with it.
Divide to attack. Splitting an enemy so the pieces can be defeated one at a time — "in detail," as the military say. Genghis Khan was a master, using Mongol cavalry mobility to cut communications and isolate parts of enemy armies so they felt alone and unprotected. He worked consciously to instil terror. Napoleon and Mao's guerrillas used it too.1
Divide to rule. Governing a group by preventing any faction inside it from consolidating. That has its own page — Divide and Rule and the Central Position — because the mechanism, the target and the ethics are all different.
The first method has a precondition, and it is the chapter's cleanest tactical idea.
by allowing the larger force to come to you, at full strength and unified, you increase the odds against you; a large and powerful army on the move will gain an irresistible momentum if left unchecked.1
So you go to meet it — and the way you make a large force divide is to occupy the centre.
Napoleon's system, quoted at length from Chandler, is the model: accumulate intelligence, plot the enemy armies on a map, find the point where their boundaries converge — the "hinge" or "joint" of their dispositions — and fall on it in full strength. Once you hold that point you are the central position, interposed between them, operating on interior lines while they operate on exterior ones and have further to march.1
Greene generalises the board:
The chessboard's center can be physical—an actual place like Marathon—or more subtle and psychological: the levers of power within a group, the support of a critical ally, a troublemaker at the eye of the storm.1
And once something has been split, splitting gets easier:
once something large is divided, it is prone to further division, to being splintered into nothingness.1
Marathon is the case — Miltiades marching an outnumbered Athenian army toward a vastly larger Persian force to hold a pass, and forcing the division that made the battle winnable. See Marathon and the Central Position.
The second half of the chapter is about a subtler target: not the parts, but what holds them together.
A joint is the weakest part of any structure. Break it and you divide people internally, making them vulnerable to suggestion and change. Divide their minds in order to conquer them.1
Samuel Adams is the case, and the reason it is the chapter's best is that the joint was emotional rather than organisational. See Samuel Adams and the Severed Bond.
Greene's extraction from it is the most transferable claim in the chapter:
rational arguments go in one ear and out the other. No one is changed; you are preaching to the converted. In the war to win people's attention and influence them, you must first separate them from whatever ties them to the past and makes them resist change. You must realize that these ties are generally not rational but emotional.1
Appeal to the emotions and you can make people see the past in a new light — tyrannical, boring, ugly, immoral. Then there is room to insinuate new ideas.
When you size up your targets, look for what connects them to the past, the source of their resistance to the new.1
Two placements are named, and both are practical.
Between the leadership and the people. Leaders function poorly once they lose support among their own, so work to make them look authoritarian or out of touch.1
Or steal their foundation. Nixon in 1972 wooed the blue-collar voters who had traditionally voted Democrat and split the Democrats' base — and Greene's parenthesis is that the Republicans have been doing it ever since.1
Mao's dismantling of Lin Biao is the chapter's most surgical case: public support for the man paired with quiet demotion of his military allies. See Mao and the Isolation of Lin Biao.
And Greene adds a note on momentum:
once your enemy begins to splinter in any way, the rupture will tend to gain momentum. Division usually leads to more division.1
The chapter's last move takes the strategy off other people entirely, and it is the part most worth keeping.
Miyamoto Musashi repeatedly faced bands of warriors determined to kill him. Most people would hesitate — fatal for a samurai — or lash out violently, trying to kill as many as possible at once and risking loss of control.
He did neither. He would place himself so that the men had to come at him in a line or at an angle, then focus on killing the first man and move swiftly down the line.1
Instead of being overwhelmed or trying too hard, he broke the band into parts. He had only to kill opponent number one while staying positioned for number two, and he kept his mind from being clouded by the others waiting. The effect compounded: as he proceeded down the line, they became the ones intimidated and flustered.1
Greene's transposition:
Whether you are beset by many small problems or by one giant problem, make Musashi the model for your mental process. If you let the complexity of the situation confuse you and either hesitate or lash out without thought, you will lose mental control, which will only add momentum to the negative force coming at you.1
The procedure: divide the issue, place yourself in a central position, and proceed down the line. And a sequencing note that is not obvious — begin with the smallest problem while keeping the most dangerous one at bay, because solving it creates physical and psychological momentum that helps overwhelm the rest.1
Greene's Reversal is unusually substantial here and contains the strategy's real limit.
Your own divisions must be temporary and controlled. Napoleon's corps system and guerrilla dispersal both work by splitting forces — and both require precise coordination and the ability to reassemble. A guerrilla army whose parts cannot communicate fails.1
And a blow that is too strong unites the target. That was Hitler's miscalculation in the London Blitz: a bombing campaign intended to demoralise the British public made them more determined, willing to suffer short-term danger to beat him in the long run.1
The last paragraph is the defence, and it is not what you would expect:
The best way to maintain unity may seem to be the creation of enthusiasm and high morale, but while enthusiasm is important, in time it will naturally wane, and if you have come to depend on it, you will fail. Far greater defenses against the forces of division are knowledge and strategic thinking. No army or group can be divided if it is aware of the enemy's intentions and makes an intelligent response.1
You are facing something too big to take on whole — an opposing coalition, a hostile department, a pile of problems arriving at once.
Do not wait for it to arrive unified. That is the chapter's first and least intuitive instruction, and it's Marathon: an outnumbered force marching toward a larger one, because a large body allowed to come at you at full strength picks up momentum you cannot stop.
Then find the joint. Not the strongest member and not the weakest — the connection. Where do two parts of this coalition meet? Whose interests only partly overlap? Which alliance is recent, or transactional, or held together by a shared irritation with you? A structure fails at its seams and the seams are usually visible if you stop looking at the components.
Then occupy the centre. Physically that's a location; usually it's an information position — being the point through which the parts have to coordinate. Once you hold the centre, they have further to travel to reach each other than you have to reach either of them.
Where the resistance is a belief rather than an organisation, go for the emotional tie instead. This is the Adams move and it inverts what most people do. You will want to argue that the new thing is better. Adams spent years doing that and got nowhere. What worked was making the old attachment feel different — not wrong, but domineering, exploitative, no longer what people had thought it was. Ask what your audience is loyal to, and what would make that loyalty feel like a cost.
And run the Musashi version on your own week. Get the problems into a line rather than a crowd. Take the smallest one first, deliberately, while holding the most dangerous at bay — not because it matters most but because finishing it changes what the rest feel like.
Then the two guards from the Reversal. Keep your own divisions temporary and coordinated, and check the force of the blow — a pressure hard enough to threaten everyone at once will consolidate the thing you were trying to split. If your action gives the other side a shared grievance, you have built their unity for them.
Du Picq's observation about disproportionate ancient casualties is a real and influential piece of military analysis, and the Roman ordering of ranks is well attested. [POPULAR SOURCE] for the framing.
The evolutionary opening is doing rhetorical work rather than evidentiary work. "Thousands of years ago our ancestors felt terrifyingly exposed and alone, so they formed groups" is a plausible story with no evidence attached, and it is used to license a strong claim — that isolation triggers a specific primal terror that can be strategically induced. The battlefield panic du Picq describes is real; the deep-time explanation for it is decoration, and the chapter would lose nothing by dropping it. 🚩 It is also the kind of just-so evolutionary framing that this vault flags elsewhere.
The chapter conflates two strategies with different ethics and never separates them. Dividing an enemy coalition in war and dividing your own subordinates to prevent them consolidating are grouped under one heading. The second is aimed at people who depend on you, and the chapter recommends it warmly. The Elizabeth and Hitchcock material sits three paragraphs from the Genghis Khan material with no seam.
And the strategy has no stated ethical floor for the emotional version. "Separate them from whatever ties them to the past" is offered as a general technique for winning attention and influence, illustrated with a revolution. The same procedure describes how a movement recruits someone away from their family, and Greene supplies no criterion distinguishing the two beyond who is doing it.
Against Strategy 5: The Command-and-Control Strategy there is a contradiction the book creates twice over.
Strategy 5's whole argument is that divided leadership causes history's greatest defeats — unity of command, one voice, one plan. Strategy 17 teaches that divided leadership is the seam you exploit, and that dividing your own subordinates is how you rule.
Which means Greene is simultaneously arguing that you must have unified command and that you must prevent your own people from consolidating. Both are argued at full strength twelve chapters apart. The reconciliation available and unstated: unity is required in decision-making and division is sought in power bases — a leader wants one plan and many rivals. That is coherent, and it is also a fair description of a paranoid organisation, which the chapter never acknowledges.
Against Strategy 6: The Controlled-Chaos Strategy the Reversal does the reconciling for once: segmenting your forces is division for mobility, and it works only with the coordination Strategy 6 supplies. Greene is explicit here that any division of your own forces must be temporary, strategic and controlled — which is the first time in this build that he has closed one of his own seams.
Against Strategy 19: The Annihilation Strategy, the du Picq mechanism is the same one: what kills armies is the feeling of being surrounded and unsupported, not the casualties. Division and envelopment are two ways of producing one psychological state.
Psychology → Social Support as a Load-Bearing Structure. The chapter's claim that isolation makes people weak and suggestible is not merely a battlefield observation. Perceived social support consistently predicts resilience under stress, and its removal is the common feature of the environments where people become most influenceable — interrogation, high-control groups, and the manufactured isolation of an abusive relationship.
The insight the pairing produces: isolation is the universal precondition of coercive influence, which means it is also the most reliable early-warning signal. Persuasion techniques vary enormously and are hard to spot in the moment. Isolation is structural, slow, and visible from outside — the friends who drop away, the second opinion that becomes inconvenient, the channel that becomes the only channel. This reframes the defence entirely: you cannot reliably detect manipulation by examining the arguments being made to you, because that examination happens inside the compromised system. You can detect it by counting your independent connections, which is a question with an answer. Du Picq's soldier does not need to assess whether the retreat is wise; he needs to know whether anyone is still on his flank.
Business → Coalitions Fail at the Seams. Enterprise deals, standards bodies, joint ventures and buying committees are all coalitions, and they are never defeated as units. They come apart where two members' interests only partly overlap — the partner who joined for a different reason, the department that signed up to avoid a worse outcome.
The insight neither field gives alone: the seam is almost always where an alliance was cheapest to form, and cheapness is legible from outside. Nobody has to tell you which member joined reluctantly; you can infer it from what each one is getting. Which means the highest-value competitive intelligence about any coalition is not what it wants but why each member is in it — and the standard practice of pitching to the coalition's stated shared objective is precisely the move that treats it as a unit and keeps the seams closed. The corollary for anyone building a coalition is uncomfortable: an alliance assembled quickly around a shared irritation is the easiest kind to break, and the ease of assembly is the tell.
The great slaughters were psychological, not physical — men died in thousands because their line broke and they felt alone, and a man who turns his back is easy to kill. Which means the thing that actually protects people in any contested situation is not their strength or even their position, but whether they can still see someone on their flank. Take that away and everything else they have stops working.