In 1944 the flying bombs came down on London at a rate nobody had seen before. Over two thousand V-1s, more than five thousand people killed, many more wounded.1
And they kept missing.
Bombs aimed at Tower Bridge or Piccadilly fell short of the city, into the less populated suburbs. Then further short. Greene's closing line on the campaign: "By the end of the campaign they were landing on cows in the country."2
The Germans were fixing their targets using secret agents planted in England. What they did not know was that those agents had been found, and that English-controlled agents had taken their place — feeding back "subtly deceptive information."3
The word doing all the work in that sentence is subtly.
A false report that the bombs are landing five miles north gets checked, disbelieved, and burns the agent. A report that they are landing slightly long does not get checked, because it is exactly the kind of thing a real report says.
So the deception did not relocate the aim point. It nudged it, then nudged it again from the new position, and the new position was the one the Germans now believed. Each correction was small enough to be plausible against the last one, and none of them was ever measured against the original.
Greene registers the shape without naming it: "The bombs would hit farther and farther from their targets every time they fell."4
That is the whole technique, and it is not really about lying. It is about being the only party who can see the result. The Germans had a weapon, a target, and a firing solution — and no independent way to observe where anything landed. Every fact they held about their own accuracy arrived through a channel the enemy owned.
Greene's image for it is the best compression in the law: "While spying gives you a third eye, disinformation puts out one of your enemy's eyes. A cyclops, he always misses his target."5
Depth perception is the right metaphor and it is more precise than he probably intended. A cyclops is not blind. He sees everything, in detail, confidently — and he cannot judge distance, which is the one thing you need to hit something.
That is the exact condition of a German targeting officer in 1944. He is not short of information. He has a steady flow of specific, timely, granular reports from trusted sources. His problem is that he has only one source of depth, and the enemy is holding it.
Set the two available strategies side by side.
Shoot the bombs down. Requires guns, fighters, radar, barrage balloons, crews, and continuous operation across the whole approach. Costs enormous resources, works imperfectly, and every bomb that gets through still lands where it was aimed.
Move the target. Requires a handful of turned agents already in custody and some carefully worded radio traffic. Costs almost nothing per week, and it does not intercept a single bomb — every V-1 launched still flies, still arrives, still detonates. What changes is only where.
The second strategy is superior on every dimension except one, and the exception is the whole moral content of the case: it does not reduce the number of bombs. It relocates them.
Greene's sentence — bombs "landing in the less populated suburbs" — carries that relocation as though it were a natural consequence of error. It was not. Someone chose the direction of the drift, and the choice determined which people were under it. The account contains no agent making that decision, and no acknowledgement that a decision existed.
Read Greene's construction again: the bombs "would fall well short of the city, landing in the less populated suburbs."6
Nobody in that sentence does anything. The bombs fall. The location is a fact about geography rather than a product of a choice made by named people who knew what they were choosing.
This is not a small stylistic point, because the entire ethical weight of the case sits precisely where the grammar has removed the actor. The deception's success is measured in the story by German inaccuracy. It could equally be measured by who was living in the less populated suburbs, and the chapter has no vocabulary for that measurement at all — the law it illustrates is Pose as a Friend, Work as a Spy, and the frame admits only two parties.
Flagged, per this build's standing practice: documented, not editorialised against. Greene's framing is what it is, and the omission is the finding.
The law's Reversal opens on a principle rather than a case: "Information is critical to power, but just as you spy on other people, you must be prepared for them to spy on you."7 The counter is not secrecy but supply — "one of the most potent weapons in the battle for information, then, is giving out false information."8
Churchill supplies the line: "Truth is so precious that she should always be attended by a bodyguard of lies."9 Greene's gloss: "You must surround yourself with such a bodyguard, so that your truth cannot be penetrated. By planting the information of your choice, you control the game."10
Note what that concedes. The bodyguard metaphor assumes the enemy will get in — you are not sealing the perimeter, you are ensuring that what they find is mostly yours. Secrecy tries to reduce the volume of information leaving; this accepts the volume and controls the composition.
Which is why it pairs with spying rather than opposing it. Both are claims about owning a channel. The spy owns the channel coming in; the disinformer owns the channel going out. Greene puts them in one law and only half-explains why they belong together.
You are running a project and every piece of information you have about how it is going comes from the same person.
That sentence should feel worse than it does. Nothing they have told you is wrong. The reports are timely, granular, specific — a bit behind here, recovered there, a small slip in one workstream. Each one is entirely plausible against the last one you received.
Now do the thing nobody does: compare the current position against the original plan rather than against last week's. Not the revised plan, not the re-baselined one. The first one.
Sometimes the gap is enormous and you have never seen it, because you have only ever been shown one increment at a time, and no single increment was ever implausible. That is not necessarily deception. Drift produces the same reading as disinformation, and from inside the channel they are indistinguishable — which is exactly the point.
So the fix is not to distrust the person. It is to get a second eye. One measurement from anywhere else — a customer, a log, somebody two levels down who has no idea what the official position is — restores depth perception. You do not need it to be a good measurement. You only need it to be independent, because a cyclops with a bad second eye can still judge distance.
The strongest evidence for the mechanism is the shape of the failure: a progressive walk rather than a single displacement. That pattern is the signature of incremental correction from a moving reference point, and it is hard to produce any other way.
Tension: Greene's account has no actor at the moment of choice. The bombs "fall short" and "land in the less populated suburbs" — the redirection is narrated as German error, when the entire premise is that the English were choosing the correction. The chapter's frame cannot register that a third party was under the new impact point.
Tension: the Reversal is filed as a reversal and is not one. Greene's own text makes disinformation the complement of spying — own the inbound channel, own the outbound channel — rather than its inversion. This is the second Reversal in Cluster E that completes its law instead of inverting it, after Law 7's in Cluster C.
🚩 SINGLE SOURCE · 🚩 SECONDARY WITHOUT PRIMARY — no source is named. The casualty figures ("over two thousand" V-1s, "more than five thousand people" killed) are given without citation, and the "landing on cows" line is presented as reportage without attribution. [POPULAR SOURCE] · [PLAUSIBLE — needs corroboration] on all figures. The existence of a British programme running turned German agents is well established in the general literature; the specific claims in this passage have not been checked against a scholarly source in this build, and should be before any of it is used in output.
Open questions. Who chose the direction of the drift, on what authority, and was the trade-off argued at the time or simply executed? And the operational question the case raises and cannot answer: how long can an incremental deception run before the accumulated displacement becomes visible as a displacement — is there a natural limit, or does it only end when the campaign does?
Set this against Duveen and the Elevator at Claridge's in the same chapter and the two halves of Law 14 come apart cleanly.
Duveen buys a channel into Mellon's household and uses what comes out. The V-1 operation captures a channel out of England and controls what goes in. Greene treats the second as the reversal of the first, but they are the same asset viewed from opposite ends — a communication line that one party owns and the other believes is neutral.
The difference that matters, and neither page states alone: the inbound version degrades if you are detected, and the outbound version degrades if you are believed too much. Duveen's spying fails the moment Mellon's staff are caught. The V-1 deception fails if the corrections are too large to be plausible — the constraint is not secrecy but credibility, and credibility caps how much you can move the target per week. That is why the bombs walked. It was not caution; it was the maximum rate the channel could carry without breaking.
The convergence within Greene's own corpus is with the formlessness material of The 33 Strategies of War — Sun-Tzu, Deception and Formlessness — where the aim is to present the enemy with nothing to read. This is the opposite move and arguably the stronger one: rather than being unreadable, be readable and wrong. An enemy who can read nothing knows he is blind and will go looking. An enemy who reads a clean, consistent, detailed picture has no reason to look anywhere, and will defend the picture against contrary evidence. Greene has both doctrines in his corpus and never sets the cost of one against the other.
Psychology — Anchoring and Adjustment
Spin a wheel, get a random 65, then estimate what percentage of UN countries are African. People adjust away from the irrelevant number — and adjust insufficiently. The anchor keeps a gravitational pull on the final answer even when everyone knows it was random.
That is the German targeting officer's condition, run as a weapon. Each week's firing solution is an anchor. Each new report produces an adjustment from that anchor, not from the true impact point, and each adjustment is deliberately sized to be insufficient.
The insight neither generates alone: anchoring is usually described as a bug in individual judgement, and the V-1 case shows it is exploitable as an ongoing process rather than a one-shot trick. The laboratory version is a single anchor and a single estimate. The operational version chains them — every insufficient adjustment becomes the anchor for the next one, so the bias compounds instead of averaging out. Which reverses the usual reassurance about anchoring, that repeated exposure and feedback should wash it out. Feedback does not help if the feedback arrives through the same channel that set the anchor. The condition for anchoring to compound rather than correct is not gullibility; it is a closed loop, and the loop is what an adversary supplies.
Behavioral mechanics — Information Control and Opacity Networks
That page's claim is positional: in any organisation, whoever receives information first can act before others know it exists, prepare responses before questions are asked, and shape the narrative before the true story spreads. Information asymmetry as a durable source of power.
The V-1 case is a stronger version of the same structure, and the difference is worth being precise about. Being first is a timing advantage — it decays, because the others eventually find out. Being the only source is a standing advantage that does not decay at all, because there is no eventual.
What the pair produces: the org-politics page's remedy is implicitly speed — get on the channel, be earlier. The V-1 case shows that remedy is worthless against the stronger form. If one party owns the only channel, arriving earlier on it changes nothing; you receive the same constructed picture sooner. The only countermeasure to sole-source control is independence, not priority — a second channel of any quality beats a faster seat on the first one. That reframes what is actually dangerous inside an organisation. It is not the person who knows things before you do. It is the topic on which everybody's knowledge traces back to one mouth, which is a structural property nobody owns and therefore nobody watches.
Sharpest implication. A channel you cannot cross-check is not a source of information. It is a source of someone's information, and the more detailed, timely and internally consistent it is, the less able you are to notice which. Confidence in a picture scales with the picture's coherence, and coherence is the cheapest thing for a single source to manufacture.
Generative questions.