During the Blitz, psychologists watched what happened to a city being bombed and found something counterintuitive:
when the bombing was frequent and somewhat regular, the people of the city became numb to it; they grew accustomed to the noise, discomfort, and carnage.1
But when the bombing was irregular and sporadic, fear became terror. It was much harder to deal with the uncertainty of when the next one would land.1
Read that again with the quantities held constant. More bombing, on a schedule, produced less terror than less bombing, at random.
Which means the damage is not being done by the explosions. It is being done by the interval — and the interval is free.
Greene supplies the mechanism a few paragraphs earlier, and it is the chapter's real content.
We manage fear by putting ground under ourselves: the distractions of daily life: work, social routines, activities with friends, and belief systems that offer hope. These become the ground beneath us, keeping us upright and able to walk on without the paralysis that fear can bring.1
And the crucial claim about what we can and cannot absorb:
What troubles us most in these situations is not any specific dreadful event that happened in the recent past; we have a tremendous capacity to overcome and adapt to anything horrible.1
It is the uncertain future, the fear that more terrible things are coming and that we might soon suffer some unpredictable tragedy—that is what unnerves us.1
Regularity is absorbable because it can be built into a routine. A raid every night at nine becomes a fact of life — you plan around it, you sleep in the shelter, and the fear attaches to a known thing and stays there. The ground holds.
Irregularity cannot be built into anything. There is no arrangement you can make with an unpredictable event, so the fear does not attach and does not resolve. Greene's description of the resulting state:
We cannot crowd out these thoughts with routines or religion. Fear becomes chronic and intense, our minds besieged by all kinds of irrational thoughts. The specific fears become more general.1
The specific fears become more general. That is the whole failure. A specific fear is finite and can be managed; a general one has no edges and consumes everything.
Hasan-i-Sabah's campaign ran on exactly this, six centuries earlier, and Greene records the tempo without connecting it to the Blitz material:
the killings went on—but they were highly sporadic. Months would pass without one, and then suddenly there would be two within a week. There was no real rhyme or reason to when it happened or which high administrator was singled out.1
And the effect:
Officials would talk endlessly about a pattern, analyzing every Ismaili move. Without their realizing it, this little sect had come to dominate their thoughts.1
Fifty assassinations across a lifetime. A regular one-a-month campaign would have cost more lives and produced less panic, because a rate is something an administration can plan against.
Greene's own summary of Hasan's design: he struck leaders in a somewhat random way: it was impossible to see any pattern in his choices, and the possibility of being the next victim was more disturbing than many could bear.1
There is a second-order effect the chapter records and does not name, and it may be the more important one.
Officials talked endlessly about a pattern — and there was no pattern. A search for structure in randomness cannot conclude, because absence of evidence is indistinguishable from insufficient analysis. So the search continues indefinitely, absorbing the attention of exactly the people whose attention the institution depends on.
That is a resource transfer, not a side effect. Hasan spent fifty killings; the empire spent its senior administrators' capacity to think about anything else.
And it generalises well past terror campaigns. Any adversary who behaves unpredictably converts your analytical function into a permanent, unresolvable task — and the more capable your analysts, the more thoroughly they will pursue it.
Greene's other example inverts the tempo and reaches the same place. The Mongols would level a few cities here and there, in as horrible a manner as possible — and the point was not the destruction but what preceded the army:
The terrifying legend of the Mongol Horde spread quickly. At its very approach to a city, panic would ensue as the inhabitants could only imagine the worst.1
More often than not, the city would surrender without a fight—the Mongols' goal all along.1
The inhabitants could only imagine the worst. A few unpredictable, disproportionate acts do the work of an army, because what arrives at the next city is not a force but an open question about what these people will do.
The economics are stated plainly: a relatively small army far from home, they could not afford long sieges or protracted wars.1 The irregularity was not cruelty for its own sake — it was the substitute for a capability they did not have.
There is a limit to the principle that Greene does not state here, though the vault holds it from two other chapters.
Unpredictability produces terror only while the target is still trying to predict. Past a certain density, people stop modelling the source and start treating it as a permanent condition — weather rather than an adversary — and at that point the fear flattens into something duller and more durable, and the leverage is gone.
Strategy 24's dosage rule says this for an individual: less is more—do this too much and you may be locked up. And the Windigokan case shows the terminal state, where enemies simply refused any encounter at all.
Which means the principle has a working range rather than a slope. Too regular and people adapt; too constant and people disengage. The effect Greene describes lives in a band between the two, and the practitioner has to stay inside it — which is a considerably harder operational requirement than the chapter's presentation suggests.
Note what this page is for. The mechanism below is documented so it can be recognised and countered; the offensive application is the deliberate manufacture of chronic fear in a population, and this vault's treatment of it stops at description.
You are on the receiving end of something sporadic and unpredictable — sabotage, a hostile actor, a pattern of incidents nobody can explain.
Start by measuring the actual rate and writing it down. This sounds trivial and it is the entire correction. Your sense of frequency will be wildly wrong, because unpredictable events are recalled as more frequent than they were, and the count is the only thing that contradicts the feeling. Fifty assassinations across a lifetime is roughly one a year, and it read as omnipresence.
Then stop the pattern search, explicitly, and say so. If there is no pattern the search cannot conclude, and it will silently consume your best analytical capacity indefinitely. Replace it with a bounded question: what is the rate, what are the actual targets, what defences failed. Those have answers.
Then impose regularity where you can, since that is what the mechanism says will help. Scheduled briefings rather than alerts as things occur. A fixed time when people learn what has happened. You cannot make the events predictable; you can make the information about them predictable, which restores some of the ground the irregularity removed.
And watch what your response is telling everyone. Precautions calibrated to a felt threat rather than a measured one broadcast your state, and Greene's Sanjar is the case — the harshness of the security revealed a man already inside the panic he thought he was outside.
The Blitz observation is real and is a well-known finding in the literature on civilian morale under bombing — the wider result being that sustained bombing generally failed to break morale, and that predictability aided adaptation. [POPULAR SOURCE] for Greene's rendering, which cites no study.
Greene states it as a clean finding and it is a summary of a contested literature. The British and American strategic bombing surveys, and the psychiatric expectations that preceded them, are a much-argued body of work; the broad conclusion that populations adapt better than planners expected is well supported, and the specific regular-versus-irregular contrast as Greene frames it is a simplification of a more complex picture that includes shelter availability, community structure, and prior warning systems. The mechanism is sound; the crispness is Greene's.
And the principle has an obvious dark application that the chapter does not flag. If irregularity is what produces chronic fear, then the finding is an instruction manual, and Greene presents it as one — the way to paralyze their will and destroy their capacity to think straight is to consciously create uncertainty, confusion, and an unmanageable fear.1 This is the point in the book where the analytical register becomes hardest to justify, because the technique's stated output is a traumatised population and the surrounding prose is unchanged from the chapter on winning arguments with colleagues.
One genuinely valuable corollary Greene does not draw. If regularity permits adaptation, then the defensive use of the principle is to make everything you control predictable — which is a cheap, available and largely unexploited intervention for any organisation under sustained uncertainty, and it is what Churchill's steady public rhythm actually supplied.
Against Strategy 33: The Chain-Reaction Strategy this is the chapter's operative mechanism, separated out because Greene states it in one paragraph and then relies on it for twenty.
Against Hasan-i-Sabah and the Anatomy of Panic the case supplies the historical demonstration and the vault now holds both halves: an eleventh-century campaign and a twentieth-century bombing survey arriving at the same finding, which Greene prints eighty lines apart without connecting.
Against Strategy 24: The Ordinary-Extraordinary Strategy and The Windigokan and Strategic Madness there is a direct line. Strategy 24's act crazy like a fox and its dosage rule — less is more, an occasional flash rather than constant unpredictability — is this principle stated for an individual. And the Windigokan page's finding holds here too: past a threshold the target stops modelling you and starts routing around you, at which point the unpredictability has stopped being leverage.
Against Time and Space as Offensive Weapons the convergence is exact and Greene does not mark it. That chapter's calibration rule — keep your enemies just barely going, avoid the sharp setback that would let them see the trap and pull out — is the same instruction: manage the tempo so the target can neither adapt nor conclude.
Psychology → Habituation requires predictability, and terror is its failure. The Blitz finding is a habituation result: a repeated stimulus at a stable rate produces a declining response, while a variable-interval stimulus does not extinguish and remains salient indefinitely. The vault's Autonomic Oscillation holds the physiological side of a system that cannot settle.
The insight the pairing produces: the schedule is a more powerful variable than the magnitude, and the two are commonly confused — which means an organisation or a person is often far more damaged by small unpredictable stressors than by larger scheduled ones, and will misattribute the damage to the size of what happened. This has a directly usable consequence for anyone managing people under strain: the intervention with the best ratio is almost never reducing the load, it is regularising it. A hard deadline every Friday costs less than an unpredictable request that might arrive any day and usually doesn't, even when the Friday work is greater.
History → Reputation substitutes for capability, and unpredictability is what makes it convertible. The Mongols could not afford sieges, so they made a few cities into a legend and collected surrenders. Set that beside Strategy 10: Deterrence Strategies and its engineered reputations, and John Boyd and the Porcupine, where a reputation for being expensive to fight buys a career's worth of room.
Together they produce something sharper than "reputation matters": a reputation for unpredictability is worth more than a reputation for strength, because strength can be estimated and planned against while unpredictability cannot be. A city facing a known force can calculate whether to resist; a city facing an unknown one can only imagine the worst, and imagination is more expensive than any actual assessment. The corollary is the defence — the counter to a fearsome reputation is not courage but specification: establish what the actor has actually done, at what rate, against what defences. That converts an open question into an estimate, and an estimate can be resisted.
More bombing on a schedule produced less terror than less bombing at random — which means the damage is done by the interval rather than the event, and the interval costs nothing. The corollary is the cheapest available defence and almost nobody uses it: you cannot make the events predictable, but you can make the information about them predictable, and that restores most of the ground.