Across his entire lifetime, Hasan-i-Sabah's followers carried out about fifty assassinations.
Only fifty or so assassinations are recorded in his entire lifetime, and yet he won as much political power through them as if he had an enormous army.1
That ratio is the whole subject. A sect Greene describes as quite weak and vulnerable except for the castles it held, extracting from the Seljuk empire what a field army could not.
Late eleventh century, Isfahan. Nizam al-Mulk, vizier to Sultan Malik Shah of the great Islamic empire of the period, becomes aware of a small problem in northern Persia: the Nizari Ismailis, followers of a religion combining mysticism with the Koran, whose charismatic leader Hasan-i-Sabah has recruited thousands of converts alienated by the tight control the empire exercised over religious and political practices.1
And the specific thing that disturbed him:
what was most disturbing to Nizam al-Mulk was the utter secrecy in which they operated: it was impossible to know who had converted to the sect, for its members did so in private and kept their allegiance hidden.1
Then the news that forced action: over years, thousands of these secret converts had infiltrated key castles, and now taken them over in Hasan's name — a kind of independent state within the empire.1
Nizam al-Mulk was a benevolent administrator, but he knew the danger of letting such sects flourish. Better to snuff them out early on than face revolution. In 1092 he convinced the sultan to send two armies against the castles.1
The castles were strongly defended, and the countryside around them teemed with sympathizers. The war turned into a stalemate, and eventually the sultan's armies were forced to come home.1
Nizam al-Mulk began considering an occupying force. And then:
a few months later, as he was travelling from Isfahan to Baghdad, a Sufi monk approached the litter on which he was carried, pulled out a dagger from under his clothes, and stabbed the vizier to death.1
The killer was revealed to be an Ismaili dressed as a peaceful Sufi, and he confessed to his captors that Hasan himself had assigned him to do the job.1
Then the compounding accident:
The death of Nizam al-Mulk was followed within weeks by the death, from natural causes, of Malik Shah. His loss would have been a blow at any time, but without his crafty vizier to oversee the succession, the empire fell into a period of chaos that lasted several years.1
With one murder they had managed to make the entire empire tremble.1
That last claim needs the sentence before it to be honest — the chaos required both deaths, and only one of them was Hasan's doing.
By 1105 stability had returned and attention swung back to the Ismailis. A new campaign was launched. And then the empire learned what it was actually facing:
soon it was revealed that the assassination of Nizam al-Mulk was not a single act of revenge, as it had seemed at the time, but an Ismaili policy, a strange and frightening new way of waging war.1
The form:
Over the next few years, key members of the administration of the new sultan, Muhammad Tapar, were assassinated in the same ritualistic fashion: a killer would emerge from a crowd to deliver a deadly blow with a dagger.1
The deed was most often done in public and in broad daylight; sometimes, though, it took place while the victim was in bed, a secret Ismaili having infiltrated his household staff.1
Two variants, and each attacks a different assumption. In public, in daylight — you are not safe among people. In your own bed, by household staff — you are not safe among your own.
A wave of fear fanned out among the empire's hierarchy. It was impossible to tell who was an Ismaili: the sect's adherents were patient, disciplined, and had mastered the art of keeping their beliefs to themselves and fitting in anywhere.1
And a refinement that turned every interrogation into a weapon:
It did not help that when the assassins were captured and tortured, they would accuse various people within the sultan's inner circle of being either paid spies for the Ismailis or secret converts. No one could know for certain if they were telling the truth, but suspicion was cast on everyone.1
A captured assassin was still working. The confession under torture — which the empire believed it had extracted — was the delivery mechanism for the sect's most damaging output.
The visible consequences:
Now viziers, judges, and local officials had to travel surrounded by bodyguards. Many of them began to wear thick, uncomfortable shirts of mail.1
In certain cities no one could move from house to house without a permit, which spread disaffection among the citizenry and made it easier for the Ismailis to recruit converts.1
Note that second one carefully. The security measures generated the sect's recruits. The empire's response was the sect's supply line.
Many found it hard to sleep at night or to trust their closest friends. All kinds of wild rumors were spread by those who had grown delirious with paranoia.1
Bitter divisions sprang up within the hierarchy, as some argued for a hard-line approach to Hasan, while others preached accommodation as the only answer.1
The tempo is the part worth studying:
the killings went on—but they were highly sporadic. Months would pass without one, and then suddenly there would be two within a week.1
There was no real rhyme or reason to when it happened or which high administrator was singled out.1
And then:
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
They were searching for a structure that did not exist, which is an infinite task, and the searching itself was the occupation of the empire's administrative attention.
He took precautions — changing his sleeping arrangements and allowing only those he knew well to approach him.
By making himself personally secure, he believed he could stay free of the panic around him.1
Hasan sent ambassador after ambassador offering to negotiate an end to the killings. All turned away.
The tables seemed to have turned: now it was the Ismailis who were frightened.1
Then, shortly before the campaign launched:
the sultan awoke one morning to find a dagger thrust neatly in the ground a few feet away from the position where his breast lay on the bed.1
How did it get there? What did it mean? The more he thought about it, the more he began to literally tremble with fear—it was clearly a message. He told no one about this, for whom could he trust? Even his wives were suspect. By the end of the day, he was an emotional wreck.1
That evening, from Hasan himself:
"Did I not wish the sultan well, that dagger which was struck into the hard ground would have been planted in his soft breast."1
Sanjar had had enough. He could not spend another day like this. He was not willing to live in constant fear, his mind deranged by uncertainty and suspicion. It was better, he thought, to negotiate with this demon. He called off his campaign and made peace with Hasan.1
An army was cancelled by a piece of metal in a floor. And the negotiating offers that preceded it — which looked like weakness — were what made the dagger unmistakable when it arrived.
Over the years, as the Ismailis' political power grew and the sect expanded into Syria, its killers became almost mythic. The assassins had never tried to escape; their killing done, they were caught, tortured, and executed to a man, but new ones kept on coming, and nothing seemed to deter them from completing their task.1
They seemed possessed, utterly devoted to their cause. Some called them hashshashin, from the Arabic word hashish, because they acted as if they were drugged.1
European crusaders carried the stories home, the word slowly transforming into "assassins," passing forever into the language.1
The objective was modest and the method was not:
Hasan-i-Sabah had one goal: to carve out a state for his sect in northern Persia, allowing it to survive and thrive within the Islamic empire. Given his relatively small numbers and the powers arrayed against him, he could not hope for more, so he devised what was surely history's first organized terrorist campaign for political power.1
The mechanism:
In the Islamic world, a leader who had won respect was invested with considerable authority, and to the extent that he had authority, his death could sow chaos.1
Accordingly Hasan chose to strike these leaders, but 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
And the illusion that did the real work:
In truth, except for the castles they held, the Ismailis were quite weak and vulnerable, but by patiently infiltrating his men deep into the heart of the sultan's administration, Hasan was able to create the illusion they were everywhere.1
Then the social mechanism, which is the chapter's actual contribution:
This power could not come by merely making individuals feel afraid. It depended on the effect the killings would have on the entire social group.1
The weakest officials in the hierarchy were the ones who would succumb to paranoia and voice doubts and rumors that would spread and infect those who were less weak. The result was a ripple effect—wild swings of emotion, from anger to surrender, up and down the line.1
A group caught by this kind of panic cannot find its balance and can fall with the slightest push.1
And Sanjar as the proof:
Even the strongest and most determined will be infected in the end, as Sultan Sanjar was: his attempts at security, and the harsh life to which he subjected himself as protection, revealed that he was under the influence of this panic. One simple dagger in the ground was enough to push him over the edge.1
The general claim:
we are all extremely susceptible to the emotions of those around us. It is often hard for us to perceive how deeply we are affected by the moods that can pass through a group.1
The strong may respond angrily and violently to the terror campaign, but that only shows how influenced they are by the panic; they are reacting rather than strategizing—a sign of weakness, not strength.1
And the sentence that makes the whole case operational:
In normal circumstances individuals who become frightened in some way can often regain their mental balance over time, especially when they are around others who are calm. But this is almost impossible within a panicked group.1
You are inside an organisation where something bad has happened more than once, unpredictably, and everyone has started behaving strangely.
Start by recognising that your own security measures are data about your state rather than solutions to the problem. Sanjar changed his sleeping arrangements and restricted access, believed this kept him free of the panic, and was destroyed by a piece of metal in a floor. Greene's reading is exact: the harshness of the precautions revealed that he was already inside the thing he thought he was outside. If you cannot describe what threat a measure actually reduces, it is managing your feelings.
Then check whether your response is generating the problem. The movement permits spread disaffection among the citizenry and made it easier for the Ismailis to recruit converts. This is the most common and expensive failure available, and the test is simple: does this measure fall on the people you are protecting, and what does it cost you in their goodwill?
Then stop looking for the pattern. The officials talked endlessly about a pattern, analyzing every Ismaili move — and there was none, which meant the analysis could never conclude and consumed the administration's attention indefinitely. An unfalsifiable search is not diligence; it is the mechanism. Establish the rate and the actual count instead.
Then treat the weakest people in the hierarchy as the transmission channel, not as a nuisance. Greene's mechanism is specific: the least resilient voice the doubts, the doubts spread, and the more resilient are infected second-hand. The intervention is not silencing them — it is being visibly calm near them, because individuals can regain their balance around others who are calm, and cannot inside a panicked group.
And know the two things that ended it. Sanjar did not lose to a capability; he lost to a message, delivered at the moment his precautions had proved worthless. And the empire's armies had already failed against the castles, so the campaign he cancelled might well have failed anyway. The decision that looks like collapse was made with better information than the one that looked like resolve.
The assassination of Nizam al-Mulk in 1092, the seizure of Alamut and other fortresses, the Nizari campaign of targeted killings, Sanjar's accommodation with the sect, and the etymology of "assassin" are all established. [POPULAR SOURCE]
The hashshashin etymology Greene reports is the popular one and is disputed. The drug story descends largely from Marco Polo and European crusader accounts, and most modern scholarship treats it as hostile propaganda rather than description — the term more plausibly functioning as an insult meaning something like "rabble" rather than a report of actual hashish use. Greene reproduces the legend and its explanation without qualification, which matters because the drugged-fanatic image was itself part of how the sect's enemies made sense of behaviour they could not otherwise explain. 🚩 SECONDARY WITHOUT PRIMARY.
And the "one murder made an empire tremble" framing overstates a coincidence. Greene's own narrative supplies the correction: the chaos followed the assassination and Malik Shah's death from natural causes weeks later, without which the succession would have been managed. The killing was consequential; the collapse required luck, and the chapter does not separate them.
The sect's own account is entirely absent, as is the political context. The Nizaris were a persecuted religious minority within a hostile empire, their converts drawn from people alienated by the tight control the empire exercised over religious and political practices — Greene's own words — and the campaign began after two armies were sent to destroy them. None of that appears in the strategic reading, which treats the sect as a case study in method. [CONTESTED] as a complete account.
One thing the case establishes cleanly and unusually well: the group-transmission mechanism. That panic propagates from the least resilient to the most, that a frightened person recovers among calm people and cannot inside a panicked group, and that security measures can be read as evidence of the panic they were meant to address — all of this is well demonstrated here and is the chapter's durable content.
Against Strategy 33: The Chain-Reaction Strategy this is the chapter's founding case and the source of its central claim — that the power comes from the effect on the group rather than on individuals.
Against Nizari Ismaili Assassins: Extermination by Hulegu this page is the other end of the same history and the vault now holds both. That page covers how the sect was destroyed in 1256, from a different source; this covers how it was built. Read together they close a loop neither closes alone: a strategy that made the sect unassailable for a century and a half, and the specific method — systematic reduction of the stronghold network, with no negotiated accommodation available — that finally ended it. Sanjar's peace and Hulegu's campaign are the two possible responses, and the vault holds the outcome of each.
Against The Irregularity Principle the tempo here is the chapter's own best evidence: months of nothing, then two in a week, no discernible logic to the targets. The unpredictability is not operational sloppiness — it is the payload.
Against Strategy 31: The Inner-Front Strategy and The Blooming Lotus the infiltration here is that doctrine's purest historical form — secret converts inside castles and inside household staffs, over years, with the visible act coming last. Greene tells the Alamut infiltration story in Strategy 31 and the assassination campaign in Strategy 33 without noting they are one operation.
Psychology → Panic is transmitted, not experienced. Greene's mechanism is precise: the weakest officials succumb first, voice doubts and rumours, and infect those who were less susceptible, producing swings from anger to surrender up and down the line. The vault's The Crowd Morality Paradox holds the general case of a group state overriding individual judgement without the individual noticing the override.
The insight the pairing produces: the unit of vulnerability is the group rather than the person, which means individual resilience is not additive and cannot be relied on — Sanjar was the strongest and most determined figure in the story and he broke, because he was inside a hierarchy that had already broken. This inverts how institutions usually approach the problem, which is to harden individuals through training, briefing and precaution. The operative variable is the ambient state, and the only intervention that touches it is visible calm from people others are watching — which is why Greene's throwaway line about recovering among calm people is the most actionable sentence in the chapter.
Behavioral mechanics → The response is a supply line. Movement permits spread disaffection and made it easier for the Ismailis to recruit converts. Set that beside Strategy 31's Reversal — that only unhealthy and decaying bodies let cancerous cells take root, and that the real safeguard is satisfied people rather than surveillance.
Together they produce a claim that generalises well beyond terror campaigns: a defensive measure that imposes costs on your own population is a recruitment subsidy for whoever you are defending against, and the subsidy is largest exactly when fear makes the measure feel most necessary. The empire's permits, mail shirts and bodyguards were all rational responses to a real threat, and collectively they converted an alienated population into an available one. The test is not whether a measure is justified but who pays for it — and in a panic that question stops being asked, which is precisely when it most needs to be.
Fifty killings across a lifetime bought a state, because the power was never in the killings — it was in the group's inability to find a pattern, and the searching was the occupation. Sanjar's precautions did not protect him; they were evidence he was already inside the panic, and a dagger in a floor was enough to finish what his own security had begun.