There is a fundamental problem with Western intelligence in Iran that has received remarkably little attention. It is not primarily a problem of satellites, intercepted communications, artificial intelligence, drones, cyberwarfare or the ability to locate individual commanders. In many of these areas, American and Israeli intelligence capabilities are extraordinary. The problem is much more basic.
Who, exactly, is telling them what Iranians think?
This question matters because there is an enormous difference between knowing where an Iranian general sleeps and knowing what ninety million Iranians will do when their country is attacked. One is tactical intelligence. The other is strategic intelligence. It is entirely possible to be spectacularly good at the first and disastrously bad at the second.
That may be precisely what has happened with Iran.
The Intelligence Prostitution Problem
Put yourself in the position of a serious Iranian living inside Iran.
You have a family. You have a profession. You have a reputation. You have relationships extending back decades. You understand your country, its history, its contradictions and its people. You may despise the Islamic Republic. You may desperately want political change. You may have marched against the government yourself.
Now somebody asks you to become an agent of the CIA or Mossad.
Why would you?
The personal risks are enormous. More importantly, there is no reason to assume that the foreign government recruiting you will protect you indefinitely. Administrations change. Governments change. priorities change. Intelligence officers rotate. Political alliances disappear. Today’s indispensable asset can become tomorrow’s inconvenient liability.
A visa can be offered. Money can be offered. Protection can be offered. Political influence can be offered.
And people can certainly be purchased.
But purchasing somebody and earning somebody’s allegiance are two very different things.
This creates what I call the intelligence prostitution problem. If the primary basis of the relationship is transactional, the intelligence relationship begins to resemble every other transactional market. The person providing the service has an incentive to keep the customer satisfied.
And that is precisely what an intelligence service should not want.
The best intelligence source is not the person who tells you what you want to hear. It is the person who repeatedly tells you that you are wrong.
Adverse Selection
There is an economic concept called adverse selection. Sometimes the structure of a market causes the people most eager to participate in it to be precisely the people you should be most suspicious of.
Iran presents an extraordinary version of this problem.
The Iranian most willing to work with a foreign intelligence organization may be someone motivated by money, immigration, revenge, ideology, ethnic separatism, personal ambition or the expectation of a position in a future government.
Some of these people may provide excellent information. The problem is that, collectively, they need not constitute anything resembling a representative sample of Iranian society.
Indeed, they may constitute its opposite.
Consider an Iranian exile activist who has spent twenty years telling Western officials that the Islamic Republic is about to collapse. His political relevance, fundraising ability and access to Western institutions may depend upon that proposition remaining believable.
What incentive does he have to walk into Washington and say:
“The regime is unpopular, but most Iranians don’t support us either. If you bomb the country, nationalism may overwhelm opposition to the government. If you arm separatists, Iranians who hate the Islamic Republic may nevertheless rally to defend Iran. And if you expect millions of people to come into the streets to install your preferred exile leadership, you are dreaming.”
That might be extremely valuable intelligence.
It is also a terrible sales pitch.
Tell the Customer What He Wants to Hear
Now consider the information chain.
An Iranian source speaks to a handler. The handler reports upward through an intelligence bureaucracy. Intelligence is incorporated into assessments. Assessments reach political officials. Political officials make decisions.
At every stage there are incentives.
The source wants to remain valuable. The handler wants productive sources. The intelligence organization wants actionable intelligence. Political leaders want evidence supporting policies they have already decided to pursue.
The result can become an extraordinary feedback mechanism:
Poor access to Iranian society → dependence upon highly motivated dissidents → distorted picture of Iranian opinion → optimistic assessments → failed policy → demand for additional intelligence → even greater dependence upon the same networks.
Eventually everybody is talking to everybody else and mistaking repetition for corroboration.
Ten Iranian opposition activists telling Washington the same thing does not necessarily constitute ten independent pieces of intelligence if all ten belong to the same political ecosystem.
It may simply constitute one mistaken assumption repeated ten times.
The Ghost of the Shah
There is another reason why genuinely valuable Iranian assets may hesitate to entrust their futures to Washington: the historical memory of Mohammad Reza Shah.
The Shah was about as useful a regional partner as the United States could reasonably have hoped to possess. Iran was strategically located, fiercely opposed to Soviet expansion, enormously important to Western energy security, militarily powerful, relatively secular and closely aligned with Washington.
Yet when the revolutionary crisis arrived, America’s ability and willingness to preserve that political order proved finite. Historians can debate endlessly whether Washington abandoned the Shah, mishandled the crisis, had already concluded that his position was unsustainable, or simply lacked the ability to reverse events. The important issue for intelligence recruitment is not which interpretation ultimately wins the historical argument.
It is the lesson a prospective asset might draw from it.
If America’s relationship with the Shah could end that way, what exactly is America’s promise to you worth?
That question becomes even more serious in a political system in which presidents can be replaced every four years.
A Republican administration embraces one faction. A Democratic administration rejects it. Another president reverses course again. Political appointees disappear. Priorities move from the Middle East to Asia and back again.
Iran remains.
That asymmetry is enormous.
Iranian institutions can think in decades. American administrations frequently think in election cycles. Yet successful human intelligence networks require precisely the opposite: patience, continuity, trust and the credible expectation that relationships will survive political changes.
Without that credibility, the people easiest to recruit become the people whose loyalty is easiest to purchase.
And purchased loyalty has a terrible habit of remaining available to the highest bidder.
Great Tactical Intelligence, Terrible Strategic Intelligence
This helps explain one of the strangest features of Western intelligence operations against Iran.
How can intelligence services demonstrate astonishing knowledge about individual Iranian officials, military facilities, nuclear scientists, communications networks and weapons systems while repeatedly appearing surprised by Iran’s political behavior?
Because these are different forms of intelligence.
Finding a missile launcher is an engineering problem.
Understanding Iranian nationalism is a human problem.
Identifying a commander is a surveillance problem.
Predicting whether Iranians will revolt during a foreign attack is a sociological problem.
Intercepting communications requires technology.
Understanding why an Iranian who despises the Islamic Republic might nevertheless fight fiercely against Israel or the United States requires understanding Iran.
That distinction is absolutely critical.
You can penetrate a government without penetrating a society.
Indeed, an intelligence organization can know almost everything about a state’s internal workings while understanding remarkably little about the civilization surrounding it.
The Regime Is Not Iran
This is where the consequences become enormous.
One of the easiest mistakes outsiders make about Iran is assuming that opposition to the Islamic Republic translates automatically into support for whatever harms the Islamic Republic.
It doesn’t.
An Iranian can hate the government and oppose bombing Iran.
An Iranian can want regime change and oppose ethnic separatism.
An Iranian can despise the Revolutionary Guards and still want Iran to possess a powerful military.
An Iranian can oppose compulsory religious rule and simultaneously resent Israel.
An Iranian can admire America culturally while deeply distrusting American foreign policy.
None of these positions is contradictory.
They become contradictory only if Iran is viewed through a simplistic binary framework in which everyone is either “pro-regime” or “pro-Western.”
Iranian nationalism is thousands of years older than the Islamic Republic. The state, the government, the nation, and the civilization are not interchangeable concepts.
Any intelligence assessment that fails to understand this distinction is almost guaranteed eventually to produce catastrophic policy recommendations.
The Exile Bubble
The problem becomes even worse when policymakers confuse the Iranian diaspora with Iran itself.
Exiles are disproportionately people who left because they were unhappy with conditions inside the country. Political activists within that population are an even more highly selected subgroup. Opposition organizations represent another level of selection beyond that.
By the time an intelligence service reaches somebody who actively seeks meetings with foreign governments and promises that his movement can govern Iran after regime change, it may be dealing with a person several layers removed from representative Iranian public opinion.
Yet these are precisely the people most accessible to Western policymakers.
They speak English. They live in Washington, London, Paris or Los Angeles. They understand Western media. They know congressional staffers. They appear on television. They know what Western audiences want to hear.
Meanwhile, the shopkeeper in Mashhad, the engineer in Isfahan, the student in Shiraz, the factory worker in Tabriz, or the middle-class family in Tehran whose political behavior ultimately determines Iran’s future remain largely invisible.
The result is an enormous sampling error masquerading as intelligence.
When Bad Intelligence Becomes War
This would merely be an interesting bureaucratic problem if the consequences were confined to intelligence reports.
They are not.
Imagine building a strategy around assumptions such as these:
The government is about to collapse.
The population will rise after military strikes.
The armed forces will defect.
Ethnic minorities will rebel.
Opposition groups abroad have substantial domestic support.
Military pressure will weaken nationalism rather than strengthen it.
The population will distinguish perfectly between attacks against “the regime” and attacks against their country.
Now imagine that most of those assumptions ultimately came from people whose careers, incomes, immigration status, political ambitions, or institutional relevance depended on Western governments believing them.
Suddenly the quality of human intelligence becomes one of the most important variables in the entire war.
The problem is no longer simply that intelligence produced a mistaken forecast.
Bad intelligence can produce the strategy that produces the war that proves the intelligence was bad.
Intelligence Needs People Who Say No
There is a simple test for any intelligence organization operating in a difficult country.
Ask how many of its most valued sources regularly tell it things it does not want to hear.
If every Iranian source says the regime is six months from collapse, there is probably a problem.
If every opposition figure says his organization is enormously popular, there is probably a problem.
If every exile says Iranians will welcome foreign military intervention, there is definitely a problem.
Real societies are messy. Real intelligence should therefore be messy too.
The most valuable Iranian source may be the person saying:
“You don’t understand this country.”
“You are confusing hatred of the government with hatred of Iran.”
“These opposition leaders aren’t popular.”
“Bombing will strengthen nationalism.”
“Arming separatists will destroy your credibility.”
“Your assumptions are wrong.”
An intelligence service that systematically filters such voices out will eventually create an exquisitely detailed map of a country that does not actually exist.
The Ultimate Irony
America and Israel may possess some of the world’s most sophisticated intelligence capabilities. They can see extraordinary things. They can intercept extraordinary things. They can find people who believed themselves impossible to find.
But technological superiority creates its own temptation: the belief that because you can see almost everything, you therefore understand everything.
You don’t.
A satellite can photograph Iran.
It cannot understand Iran.
A phone intercept can reveal what an Iranian commander said yesterday.
It cannot tell you what millions of Iranians will decide tomorrow.
And an agent recruited because he desperately wants your money, your visa, your protection, or your political sponsorship may know exactly what his handler wants to hear.
That is the intelligence prostitution problem.
If you build an intelligence network primarily from people whose loyalty is for sale, you should not be surprised when the intelligence they provide is also for sale.
Perhaps this explains one of the great paradoxes of Western policy toward Iran: extraordinary intelligence capabilities accompanied by extraordinary strategic miscalculations.
The problem may never have been that America couldn’t see Iran.
The problem may be that it kept paying the wrong people to explain what it was seeing.
Leave a Reply