Picture the version of you that never says who's actually in the way. Something's been off for months. A project keeps stalling. A relationship keeps costing you more than it gives back. A room keeps going quiet when you walk in. You feel it. You just can't say who or what it is.
So you keep being "fine." You keep being flexible, generous with the benefit of the doubt, endlessly willing to see it from the other side. And you stay exactly where you are — because a fight you won't name is a fight you can't win.
Robert Greene's first strategy is blunt about why: you need an enemy.1 Not because conflict is good for its own sake, but because a mind without a clear opposite has no direction to move in. Watch a compass with no magnetic north. The needle doesn't rest. It just spins. Naming your enemy is what gives the needle somewhere to point.
Here's the mechanism underneath the advice, and it's older than strategy — it's developmental. A toddler doesn't discover who they are by being told. They discover it by pushing against a parent's "no" and finding out where their own edges are. Differentiation runs on friction. You find out what you actually want partly by finding out, hard, what you refuse.
Adults do the same thing at scale, and most adults are trained out of doing it cleanly. The culture Greene is writing against prizes fairness, cooperation, seeing every side — genuinely useful values, until they get used as a way to avoid ever drawing a line.
"I don't really have enemies" sounds like maturity. Often it's just fear of the discomfort a clear opposition brings. Greene's claim is that the discomfort is the point. Without a "not this," your "this" stays vague. And a vague identity produces vague strategy, which loses to anyone whose sense of direction is sharper than yours.
Think about it from the outside, watching two people negotiate. One knows exactly what they're protecting and exactly who's trying to take it. The other is still privately hoping the whole thing resolves itself amicably, still framing the other side as "probably reasonable, deep down." The first person makes fewer, sharper moves. The second hedges every move, because every move risks confirming a conflict they haven't let themselves name yet.
The asymmetry isn't about aggression. It's about clock speed. A named enemy lets you stop re-litigating whether this is actually a fight every time a new data point arrives. An unnamed one forces you to re-run that judgment call constantly, which is exhausting, and exhaustion is where bad decisions live.
This is where the strategy gets its teeth, and where it's easiest to misread. Greene widens the word past "person who wants to hurt me." An enemy can be:
Salvador Dalí ran his whole career this way. He couldn't stand conformity, sentimentality, piety — so at every stage of his life he found someone who embodied exactly that and built an opposition against them. First the poet Federico García Lorca, whose romanticism he needed to reject. Later André Breton, the surrealist movement's heavy-handed enforcer.1 Having a named target to push off from didn't limit Dalí. It's part of what made him move at all — each rejection was also a definition, a public statement of what he was not, which sharpened what he was.
Notice that none of Dalí's targets were trying to destroy him. Lorca and Breton weren't villains in any conventional sense; one was a friend whose sensibility Dalí simply found stifling, the other a movement leader whose control he needed to escape. This is the part of the strategy people resist most: the enemy doesn't have to be malicious. It only has to be genuinely, structurally opposed to what you're trying to become.
The failure mode is obvious enough that Greene builds it into the strategy's own Reversal: see an enemy in everyone and you've stopped being a strategist. You've become a tyrant losing his grip.1 The line between the two isn't about how many enemies you name. It's about whether the naming produces clarity or produces noise.
A genuine enemy sharpens your priorities. You know what you're for because you now know, concretely, what you're against, and the two definitions reinforce each other. Paranoia does the opposite. It multiplies threats faster than you can act on any of them, until every relationship reads as a potential betrayal and you can no longer tell a real signal from your own anxiety.
Margaret Thatcher, whose career is this strategy's central case study, eventually crossed exactly that line. The polarizing instinct that won her two elections stopped discriminating between genuine political opponents and colleagues who simply disagreed with her on a Tuesday, and it cost her the premiership.1
Franklin Roosevelt ran the same play with more discipline. Draw the line sharply, then — once the line has done its work of defining you — back off, and let yourself look like the reasonable one.1 Same tactic, opposite management of when to stop.
The difference isn't intensity. Thatcher and Roosevelt both polarized hard when it mattered. The difference is that Roosevelt treated polarization as a tool with an off switch, deployed for a specific purpose and then set down, while Thatcher let it become the only tool in the kit — reached for automatically, in situations that actually called for something else.
Most enemies in adult life don't announce themselves. They're indirect, because direct hostility carries a social cost nobody wants to pay anymore. Greene gives four concrete ways to smoke out someone who's working against you while performing friendliness.
Mao noticed that his lieutenant Lin Biao had become embarrassingly, excessively warm toward him right around the time Lin was positioning for a takeover.1 Overpraise is data. Genuine allies don't usually need to perform their loyalty that loudly — the performance itself is the tell, because it's aimed at an audience, not at you.
King David suspected his father-in-law Saul wanted him dead, so instead of asking directly, he sent an excuse for missing a feast — an excuse readable as either innocent or as a small insult. A friend reads it as innocent and shrugs. Saul read it as insult and exploded, confirming exactly what David suspected.1
The test works because it gives no ground to react to except the other person's own interpretation. You haven't accused anyone of anything. You've just left a small, deliberately ambiguous gap, and watched which way the other person's mind fills it.
The studio boss Harry Cohn would deliberately attack a director's work or take an extreme position just to see who dropped their diplomatic mask and showed their real position.1 People under mild pressure reveal more than people offered a safe, open-ended question. Comfort produces performance. A little friction produces truth.
Cortés, suspicious of soldiers he thought disloyal, never confronted them. He pretended to approve of what they did. Thinking him weak or complicit, they pushed further, and the further they pushed the more clearly they identified themselves as enemies he could now move against with justification.1 Where Cohn provokes to expose, Cortés withholds to expose — same diagnostic goal, opposite mechanism. Cohn applies pressure and watches the crack. Cortés removes resistance and watches how far someone runs once nothing is stopping them.
Not every enemy has a face. Greene's widest category — an idea or value you find intolerable — is also the easiest to get wrong, because an abstraction can't be tested the way a person can. You can't send David's ambiguous excuse to "smugness" and watch how it reacts.
The fix is to keep the abstraction pinned to something concrete. Dalí didn't declare war on "romanticism" in the abstract — he declared it on Lorca, a specific person whose specific poems he could point to. The abstraction gave the opposition its meaning; the person gave it a location. Without the person, "I hate conformity" is a mood. With one, it's a strategy.
You're three months into a project that keeps losing momentum for reasons nobody names out loud. In the weekly meeting, one colleague keeps finding a reason the timeline should slip. Never hostile. Always reasonable-sounding. Always someone else's fault this time. You've been telling yourself it's just caution. Tonight you sit with the pattern instead of explaining it away.
You test it the way David tested Saul. Next week, you float a plan with a real deadline attached — not confrontational, just concrete — and you watch. If the colleague engages, raises real logistics, helps solve them: that's information, not hostility. Cross it off.
But the colleague finds a reason it can't work, again, phrased as concern for the team, and you notice your own gut going tight in the same specific place it goes tight every time this happens. That's the signal. Not proof of malice — proof of a real, locatable opposition, whatever its motive actually is underneath.
Now you have somewhere to point the compass. You don't announce a feud. You stop bringing this person into decisions where their default is "not yet." You route around them, quietly, the way Cortés let his suspects show themselves before he moved.
And notice the side effect. Once you've named this one clear obstacle, the rest of the project's ambiguity thins out with it. You know what you're pushing against, so you finally know, in sharp focus, what you're actually building toward. The vague dread you'd been carrying about the whole project turns out to have had a single, locatable source — and most of it lifts the moment you stop pretending you hadn't already found it.
Greene's strongest evidence is structural, not statistical. Xenophon's Ten Thousand1 go from a demoralized, arguing mob to a disciplined force capable of a 1,500-mile fighting retreat the moment they stop debating and declare unambiguous war on the Persians who betrayed them. The behavioral shift is dramatic and immediate, which is persuasive — but it's also a single dramatic case, chosen because it illustrates the thesis cleanly.
The real tension the strategy doesn't resolve: naming an enemy and being fair to them are not automatically compatible, and Greene says so outright — "victory is your goal, not fairness and balance."1 That's honest, but it leaves open exactly how far the distortion is supposed to go before it curdles into the paranoia the Reversal warns against.
The book gives you Thatcher's failure and Roosevelt's success as bookends without giving you the dial in between. It never specifies how you'd know, in the moment, which one you currently resemble — the discipline seems to live entirely in Roosevelt's move (polarize to define, then deliberately step back) rather than in any fixed rule about who counts as a legitimate target in the first place.
Put this next to The Self-Opinion, also Greene, from Laws of Human Nature. That page argues the self-image gets defended at nearly any cost to reality — it's a structure that forms mostly by accident, in childhood, and then resists correction for the rest of a life.
The Polarity Strategy is close kin, but inverted in direction. Instead of describing a distortion people fall into by accident, it's a distortion Greene recommends deliberately, on the theory that a slightly exaggerated enemy is an acceptable price for a sharply defined self.
Read together, the two pages suggest the self-image isn't only defended reactively — it can be actively constructed by choosing, on purpose, what to stand against. That's a more agentic reading of identity than the LOHN material usually allows, and Greene never quite flags that he's shifted registers between the two books. One treats the self-image as something that happens to you. The other treats it as something you can build, on purpose, using an enemy as the load-bearing wall.
Psychology → Developmental Differentiation. Object-relations theory treats a child's early "no" — pushing the parent away, testing the boundary — as a necessary step toward a coherent self, not a failure of attachment. The Polarity Strategy is that same mechanism redeployed on purpose, in adulthood, at a much larger scale. What neither the child nor the parent could plan consciously (the friction that actually builds the boundary) Greene turns into deliberate technique: identify friction, then use it.
The insight neither domain reaches alone: differentiation isn't a childhood phase you graduate out of. It's a structural requirement of any bounded identity, at any age, and it has to be periodically re-run against a live opponent or the boundary starts to dissolve back into the crowd. The toddler who successfully differentiates once doesn't stay differentiated forever without more friction later — and neither does the adult who declared one enemy a decade ago and hasn't updated since.
History → Carl Schmitt's Friend-Enemy Distinction. Schmitt argued that the defining act of the political — not a side effect of it, the actual definition — is the ability to distinguish friend from enemy with concrete clarity.1 Greene quotes him directly. The handshake runs deeper than the citation: Schmitt is describing a state's survival logic, and Greene is claiming the identical logic scales down to a single person managing a career.
The insight the pairing produces is uncomfortable but real. The reason "we should just get along" feels so hollow as political advice, and so hollow as personal advice, might be the same reason: neither a nation nor a person can locate its own boundary without something concrete on the other side of it. A state with no enemies isn't at peace in any stable sense — it's a state that hasn't yet discovered where it ends. Same for a person.
The people who tell you conflict is primitive and beneath you are very often disguising their own fear of it — and if you listen to them, you inherit their vagueness along with their peace. A self with no named opposite isn't at peace. It's just undefined, and undefined things get pushed around by anyone who has bothered to define themselves.