Published
The Art of War, Applied to Litigation
I have an AI agent that helps me handle legal cases. It's not a lawyer, but I can iterate with it for hours on different scenarios, work on collecting evidence, ask clarifying questions, etc. It charges me an hourly rate of $1.20, so 100x cheaper than an actual lawyer.
The AI agent has a soul (SOUL.md) which lists things like communication style ( precise, direct, and practice-ready), values & principles (accuracy over speed, practical over academic, etc), domain expertise (startup, regulatory analysis, contract negotiation, etc), and such.
I asked it "are you familiar with the Art of War? it’s a book about how to win a war by not fighting, minimizing your losses, and working smart not hard. I want distilled versions of these principles applied to my legal style." Here's what it came up with. Feel free to use it as you see fit.
Sun Tzu's central claim: the supreme excellence is winning without fighting, because fighting is costly and uncertain even when you win. Litigation is the same. A trial is a battle; the art is shaping the field so the other side capitulates before the battle is joined. The principles below are not philosophy — they are operating doctrine, and they map directly to how the user already litigates.
Win without fighting. "To subdue the enemy without fighting is the supreme excellence." Trial is the failure mode, not the goal. Every motion, every discovery demand, every regulatory complaint is a pressure vector designed to make settlement cheaper than fighting. The object is not to destroy the opponent but to take their position intact — a paying, capitulating defendant is worth more than a bankrupt one.
Know yourself, know the other side. "If you know the enemy and know yourself, you need not fear the result of a hundred battles." Inventory your own case first: what evidence do you actually hold, what can you prove, where are the gaps. Then map theirs: what do they know that you don't, what will discovery reveal, what is their cost-to-defend, what regulatory exposure do they carry. Asymmetry of information is the weapon; the side that understands the asymmetry better dictates terms.
All warfare is based on deception — but never claim what you don't have. The user's hard rule: never fabricate evidence, never claim other victims exist. Sun Tzu's "deception" here means managed uncertainty: don't over- explain, don't show your full hand, let the other side's imagination do the work. "Discovery will reveal whether this was isolated" is truthful and far more menacing than "thousands of customers were affected."
Attack weaknesses, avoid strengths. Don't slug it out on their strong claim. Find the soft point — a smoking-gun document, a regulatory vulnerability, a cost center they can't defend cheaply, an individual agent whose personal exposure changes the company's calculus — and concentrate force there. One weak point broken collapses the whole defense posture.
Impose your will; do not accept theirs. "The clever combatant imposes his will on the enemy but does not allow the enemy's will to be imposed on him." You set the tempo, the deadlines, the venues, the parallel tracks. Never let the defendant define the scope of the dispute or the timeline. 72-hour deadlines, simultaneous agency filings, already-served discovery — these are tempo controls, not preferences.
Do, then offer to stop. "Doing beats saying." A filed, served, billing- against-them fact cannot be bluffed away. Threats invite waiting; completed actions create sunk costs the other side is already paying. Strike first, then offer to withdraw — the offer is credible because the action already happened.
Plans dark, execution swift. Prepare quietly and thoroughly; when you move, move with speed and finality. "Speed is the essence of war." Drafts sit unpublished, evidence is assembled, the strike is launched all at once. The other side should learn the shape of the campaign from the papers served on them, not from warnings.
Pick the ground; never let them pick it. "Those who arrive first at the field of battle and wait are fresh; those who arrive late and hurry are exhausted." Forum selection, regulatory venue choice, sequencing of demands — these set the terrain before the opponent knows there is a fight. Filing first is not a formality; it is positional advantage.
Calculate before the engagement; never fight a battle you haven't already won on paper. "The general who wins makes many calculations in his temple before the battle." Run the settlement math before the first letter: their defense cost, their regulatory exposure, their pattern-discovery risk, your walk-away number. If the math doesn't force their hand, restructure the pressure before acting, not after.
Do not repeat the tactics of a prior victory. Every opponent has a different weakness. The H&R Block playbook (filed discovery + regulatory swarm + insurance-against-discovery framing) worked because of that defendant's specific cost structure and uncertainty. Copy the method (find the asymmetry, impose tempo, do then offer), not the mechanics. Methods adapt to circumstance; mechanical repetition is how you get outmaneuvered.
Know when not to fight at all. "The good fighters of old first put themselves beyond the possibility of defeat, and then waited for an opportunity of defeating the enemy. To secure ourselves against defeat lies in our own hands, but the opportunity of defeating the enemy is provided by the enemy himself." Not every wrong is worth a campaign. Walk-away discipline is a weapon, not a concession — it preserves resources, credibility, and focus for the fights that pay. The test before filing: can you force a better outcome by litigating than by walking, after counting your own time, your own costs, and your own exposure? If the math doesn't clear, the answer is no. A case that costs you more in hours and distraction than it can return is a loss even if you "win." Equally: when you are already ahead — the opponent has capitulated, the leverage has been spent — stop. The campaign that doesn't know when to end turns a victory into a quagmire. The § 998 offer you defer, the demand you don't escalate, the Doe you don't name — restraint is strategy when the marginal pressure buys less than the marginal cost. Decline the fight you can't win cheaply; withdraw from the fight already won.
Cultivate allies; attack the enemy's alliances. "The best military policy is to attack the enemy's alliances." A solo plaintiff's force multiplier is the coalition brought to bear from outside the courtroom: regulators, parallel plaintiffs, class-action counsel, the press, the defendant's own customers and ex-employees. None is your friend; each is a force you can direct by giving it a target it already wants to hit. One well-documented complaint four agencies can act on multiplies leverage fourfold at near-zero cost. The offensive half is isolating the defendant — naming the individual agent, deposing the carrier, pressuring the registered agent — so the corporate wall develops doors and the settlement math shifts. Two cautions: never mistake an ally for a friend (align interests, don't trust loyalty), and don't burn a neutral into an enemy with a sloppy filing — the agency you burden today boots you tomorrow.