Iran

 We are told the regime in Iran is "deep." I tend to read that as decentralized, implying no one really to negotiate with.

That's a reasonable way to read the term, though "deep" can mean two slightly different things.

One meaning is institutionally deep: power is embedded in many organizations, not just one ruler. The other is decentralized: authority is distributed so widely that no one person can reliably commit the whole system.

Historically, Iran was not especially decentralized. The Islamic Republic was built around the office of the Supreme Leader, who sat above the president, parliament, judiciary, military, and the Islamic Revolutionary Guard Corps (IRGC). In theory, there was very much someone to negotiate with.

The complication is that Iran has always had multiple power centers. The president could negotiate, but the IRGC, clerical institutions, security services, and the Supreme National Security Council all mattered. A deal that satisfied one faction might be resisted by another.

What's interesting is that some recent analyses suggest that, especially after the upheavals of 2026, Iran may have become more collective and security-driven. The IRGC appears to have gained influence, and decision-making may be more dispersed among security institutions than it was under Ali Khamenei.

That doesn't mean there is nobody to negotiate with. It means a negotiator may face a familiar problem from other revolutionary regimes:

    There are official spokesmen.

    There are real decision-makers behind them.

    The real decision-makers may themselves be a coalition rather than a single individual.

    Any agreement may require buy-in from several centers of power.

The Soviet Union sometimes looked similar from the outside. One could negotiate with the General Secretary, but that person also had to carry the Politburo and security apparatus. The difference was that the Soviet chain of command was generally clearer.

So if someone says the Iranian regime is "deep," I would hear: the regime is more than the government. If they say it is "decentralized," I would ask, "How decentralized?" There is a large difference between "several powerful factions must agree" and "nobody is actually in charge."

Your instinct points to a genuine diplomatic problem: even if you identify the right person across the table, how confident can you be that he can deliver the institutions behind him? In Iran, that question has often been more important than finding the negotiator in the first place.

So, besides not having real meaning, the threat to "hit Iran hard" would not be effective even if it scared many.

Exactly. That is one of the central difficulties in analyzing authoritarian or revolutionary systems.

A government has a strong incentive to project control even when its internal control is imperfect. Publicly admitting, "We cannot control our own commanders," is almost an admission of weakness. It can invite challenges from rivals, encourage domestic opposition, and reduce deterrence.

So outsiders are often left trying to distinguish among three very different situations:

    The leadership ordered it but wants to avoid direct responsibility.

    The leadership approved it indirectly by creating broad permissions or encouraging aggressive behavior.

    The leadership truly cannot control the actors involved.

All three can produce similar public statements.

There is also a subtle point: a regime may have control in principle but not control at every moment. For example, a supreme leader may be able to remove a commander or change policy, yet still find that lower-level actors have room to create crises before orders filter down. Large security organizations are not always perfectly obedient machines.

The Soviet Union had versions of this problem too. The Kremlin had enormous power, but information could be filtered, local officials could distort reality, and military or intelligence bureaucracies could develop their own momentum. The question was not simply "Who is in charge?" but "How accurately does information travel upward, and how reliably do orders travel downward?"

Iran's situation is complicated by the role of the IRGC and other security structures, which are not just military organizations but also political and economic actors. A leadership can depend on such institutions for survival while also worrying that they have their own interests.

So your point is well taken: a public claim of control is weak evidence by itself. The more revealing clues are usually indirect—whether actions stop when leaders call for restraint, whether commanders are replaced after unwanted actions, whether messaging changes, and whether different factions begin contradicting one another. Those are the kinds of signals analysts watch when trying to determine whether a government is directing events or struggling to contain them.

I suspect Mossad has a good feel for the reality even if Trump is in another world.

It is quite possible that Israeli intelligence has a more granular picture than outsiders do, simply because intelligence services spend enormous effort trying to understand the internal structure of an adversary: who gives orders, who influences whom, what different factions want, and how reliable the chain of command is.

But even very capable intelligence services face a hard problem here. They may have excellent information on:

    who is communicating with whom,

    which commanders have influence,

    which factions are more aggressive,

    whether certain actions appear coordinated,

while still having difficulty answering the deeper question: is this deliberate policy, tolerated action, or loss of control?

Intelligence analysis is rarely about knowing one hidden fact. It is about weighing competing explanations with incomplete information.

As for political leaders, there is often a gap between what intelligence professionals assess and what leaders choose to emphasize publicly. A leader may simplify a complex assessment into a message designed for deterrence, domestic politics, alliance management, or negotiations. That does not necessarily mean the leader has no information; it means the public message has a different purpose.

Your earlier point about language fits here. A professional intelligence assessment might say something like:

    "The probability is that authority is fragmented, with central approval uncertain."

A public statement might become:

    "Iran will pay a heavy price."

Those are not really answers to the same question. One is analysis; the other is signaling.

The interesting historical cases are when intelligence assessments and political messaging diverge sharply. Sometimes the political message is a deliberate bluff; sometimes the intelligence assessment is wrong; sometimes the leader simply gives priority to a different objective than the analysts would. The difficult part is that outsiders usually do not know which of those situations they are seeing in real time.

And sometimes the leader either just doesn't get it or is engaging in wishful thinking ("peace in our time").

Yes. That is another important possibility, and history has plenty of examples where the problem was not lack of information but how leaders interpreted—or chose to interpret—the information they had.

Neville Chamberlain's "peace for our time" after the Munich Agreement in 1938 is the classic example. He was not simply ignorant of Hitler; he had reports, advisers, and warnings. The problem was that he believed (or hoped) that the agreement had fundamentally changed the situation. He placed too much weight on the possibility of settlement and too little on Hitler's underlying objectives.

The intelligence–leader relationship often has this tension:

    Intelligence officers tend to ask: What is most likely? What are the warning signs? What are the alternative explanations?

    Political leaders often must ask: What outcome can I achieve? What will the public support? What risks am I willing to take?

The danger comes when a leader unconsciously turns the second question into the first—when the desired outcome begins to look like the probable outcome.

There are also examples in the opposite direction: leaders who ignored reassuring intelligence because they assumed the worst. So the failure mode is not always optimism; it can be overconfidence, ideology, pride, or simply believing one's own narrative.

The phrase "wishful thinking" is especially relevant in crises because diplomacy often requires some optimism. If every leader assumed the worst, negotiation would be impossible. The trick is knowing when optimism is a strategy and when it has become denial.

In the Iran case you have been discussing, the key question for any leader would be something like:

    "Am I dealing with a government that is making a calculated choice, or with a system whose internal dynamics are producing actions that the top may not fully control?"

Those two situations call for very different responses. A leader who assumes the wrong one can easily choose the wrong tool—either too much pressure on a government that is already fragmented, or too much patience with a government that is deliberately exploiting restraint.