In the aftermath of Iran’s massive missile attack on Israel this week, it has become clear that Israeli missile defenses are robust. Of the estimated 180 ballistic missiles that Iran launched, only a small percentage evaded Israel’s anti-missile defenses, causing limited damage at or near some Israeli intelligence and military sites and apparently having little impact on Israeli military operations. But the attack marks a major escalation in the Israel-Iran conflict and has led to widespread speculation about when and where Israel will respond. Much of that speculation has centered on the question of whether Israel will attack facilities related to Iran’s nuclear program.

Late this week, I asked James Acton, a physicist and wide-ranging nuclear policy expert who co-directs the Nuclear Policy Program at the Carnegie Endowment for International Peace, for his assessment of the Israel-Iran situation, especially as regards the possibility of an Israeli attack on Iran’s nuclear facilities. His answers follow in a lightly edited and condensed Q&A format.

John Mecklin: I gather you think it would be a bad idea for Israel to attack Iran’s nuclear facilities. Can you explain why for our readers?

James Acton: Sure. If Israel or the United States tries to attack Iran’s nuclear facilities, my belief is that that will harden Iranian resolve to acquire nuclear weapons without eliminating Iran’s capability to do so. Israel would be motivated, in part, to punish Iran for its recent attack on Israel, using that as an opportunity to try and destroy Iran’s nuclear program, so the Israelis didn’t have to worry about it in the future. I think if they decide to attack Iran’s nuclear program, they will find themselves worrying much more about Iran’s nuclear program in the future. We’ll elaborate on this, but an attack would, I believe, simultaneously harden Iranian resolve to acquire nuclear weapons while also not destroying permanently their capability to achieve that goal.

Mecklin: In a sense, that’s essentially impossible, right? People in Iran have the knowledge to pursue nuclear weapons, right?

Acton: Yes. The situation is quite different from the Israeli attack on Osirak in Iraq in 1981. The Iraqis did have a nuclear weapons program that was centered around a reactor designed to produce plutonium. And I think the Israeli attack on Osirak in ‘81 did slow the Iraqi program very dramatically by many years. If the Iranian program today comprised a single reactor that had not been turned on, I think you could make a fair argument that it could be in Israel’s interests to attack it. But that’s nothing like what the Iranian program actually looks like. And I think the Osirak example, which many experts have at the back of their mind as a successful example of military nonproliferation. And also, the attack on the Syrian reactor in 2007.

But the Iranian program today is based around centrifuges, which are very small and can be manufactured quickly and placed almost anywhere. So even if an Israeli attack destroys Iran’s current centrifuge plants at Fordow and Natanz—and it’s not obvious to me that Iran has the capability to destroy Fordow, which is buried inside a mountain—but even if Israel can destroy Iran’s existing centrifuge plants, Iran is almost certainly going to reconstruct centrifuge facilities. In fact, it may already have clandestine centrifuge facilities. We don’t know. But even if it doesn’t, it will construct, I feel very certain, more centrifuge facilities, potentially at multiple sites, some of them potentially hidden in plain sight inside normal industrial buildings, and some of them buried even deeper than Fordow, so they’re definitely out of the range of the Israelis.

And I think Iran will likely kick out [International Atomic Energy Agency] inspectors and try to manufacture highly enriched uranium for weapons in the centrifuge plants. So even in the most optimistic case that this attack is highly successful, a centrifuge program can be reconstituted in a more survivable form relatively quickly.

One option then, is what euphemistically called mowing the grass, i.e., conducting repeat strikes on Iran. But the idea that the attacks are going to work indefinitely, that you’re always going to know where everything is, you’re always going to be able to destroy everything—that is, I think, the height of wishful thinking.

Or you invade Iran, and you occupy the country. Fundamentally, the only way I think you can absolutely guarantee the Iranians never get a nuclear weapon is by occupying the country indefinitely. Israel clearly doesn’t have the will or the capabilities to do that. The United States, I think, has very clearly learned its lesson in terms of occupying other countries, and I don’t think there is any appetite in the United States to do that. And quite rightly; I think it would be a catastrophe trying to invade and occupy Iran.

So people tend to say the Israelis can destroy Iran’s nuclear weapons program. Even if that is true in the short term, the question they have to answer is: Then what?

Mecklin: Okay, the second question is: How likely do you think it is that Israel is actually contemplating attacking the nuclear facilities?

Acton: Let me distinguish between two ideas. Are they contemplating doing so? And will they do so?

I think there is an extremely high probability that there is a serious discussion going on right now in the Israeli military cabinet about whether to attack Iran’s nuclear facilities. Many Israeli leaders have openly called for that at this juncture. And you know, Netanyahu has been publicly mentioning this possibility on and off for many years now. So I would be staggered if there was not a serious discussion within Israel right now about attacking Iranian nuclear facilities.

Would Israel actually go ahead and do that? I think it would be tough without a lot of US support. And Biden has come out and said unequivocally, no. And doing it without US support would do enormous damage to the US Israeli relationship. And I think the Israelis understand that.

I think the Israelis fully understand that if they attack Iran’s nuclear program, Iran then attacks Israel in a much larger way than we’ve seen before. The Israelis are going to want America’s help in defending against those attacks, and there must be at least some uncertainty in their mind, if they just point blank defy an American president, whether that help would be forthcoming. So for all of those reasons, if the US is being as clear and private as it is in public, I do think it’s substantially less than 50/50 that the Israelis are going to attack Iranian nuclear facilities. I think it’s higher than 10 percent, but it’s not, I think, 50/50. Which I find somewhat reassuring.

We’ve discussed the effect of an attack on Iranian capabilities. Can we also talk a bit about the effect of an attack on Iranian intentions?

Mecklin: Explain what you mean by that.

Acton: For a long time, my assumption has been that Iran wanted the capability to build a nuclear weapon at short notice but had not made the decision to actually build one. Whether Iran has started to change its thinking in the last few weeks in its escalation sequence with Israel, and given that it’s kind of proved incapable of protecting all of its proxies, one thing that I feel pretty confident in saying is that if Iran has not yet made a decision to build a nuclear weapon, an Israeli strike makes it much, much more likely that It will make that decision to do so—both for reasons of defending the state and for reasons of domestic politics.

A strike against Iran’s nuclear sites, which are arguably the most sensitive places in the entire country, would very strongly empower those, I think, who want to have and want to build nuclear weapons. This is why I emphasize you have to look at both the effect on intent and capabilities. Following an Israeli attack, Iran’s intentions, I think, will become much more focused on building nuclear weapons. And while an Israeli strike can dent its capabilities in the short term, it can’t remove those capabilities in the long term. And indeed, as I say, Iran can redevelop its nuclear weapons program in a way that’s more survivable.

So, you know, from both the capability and intent perspectives, when you combine them together, an Israeli strike looks like a pretty bad idea to me right now.

Mecklin: Israel is going to respond somehow. That’s clear, right?

Acton: Yes.

Mecklin: And if the United States’ advice is followed, the Israelis won’t go after anything to do with the nuclear program. Not to put you in the position of telling them what to attack, but what are the next likely possibilities?

Acton: This is kind of starting to move some way outside my area of expertise. But in the array of targets the Israelis have been talking about, one that has been mentioned publicly is Iran’s oil infrastructure, which would, I think, clearly be somewhat less escalatory than attacking Iran’s nuclear facilities. But it’s still going to be immensely escalatory. And it actually has quite significant domestic implications for the US. If there is a strike against Iranian oil facilities, that increases the global price of oil. The US does not import oil from Iran, but there is a global market for oil. So if the supply of oil goes down, then the price of oil will go up.

Beyond that, there’s then any number of military, industrial, and government targets that the Israelis could attack. When I say government, I’m not suggesting that they would try and assassinate key Iranian leaders. I don’t think they would do that. But you know, one could imagine a strike intended to significantly weaken the Iranian state through attacks against various different government and military targets.

I’m assuming at this point that an Israeli response is highly likely and that it will be pretty big scale; there’s a lot of uncertainty from my perspective about what it will look like, what targets will be hit.

Mecklin: A final question. In your view, how likely is it that superpowers are going to become directly involved as this continues? Is there a significant possibility of escalation to include US and Russian forces.

Acton: I think that’s very unlikely. That’s the good news. Let me be clear: If we’re talking about a nuclear crisis between the US and Russia, I think that’s pretty unlikely. I can absolutely see the possibility of the US getting dragged further into this crisis. One can imagine a massive Israeli attack on Iran, a massive counterstrike, and the US getting increasingly sucked into helping Israel defend itself against these counter strikes.

As you know, the US has already played a role in defending Israel against strikes from Iran. By the way, I think that’s a good idea, because I think it does lessen the possibility of further escalation. The US can bring very useful capabilities to the region. If you help save Israeli lives, then you help prevent Israel from going as far as it might.

But how long will that restraint last? As we’ve discussed, if Israel feels obliged to respond in a major way, one can easily imagine the US getting sucked in. I kind of find it harder to see the Russians getting sucked in right now. They do back Iran, and they do have this growing relationship of closeness with Iran. They’re also enmeshed in this quagmire in Ukraine, and I don’t think they’re going to want to get involved militarily on Iran’s behalf. So yes, I can see the US getting much more deeply involved right now. The good news is, I find it pretty difficult to see how this escalates into a US-Russia nuclear crisis.

Mecklin: I told you that was the last question. But I’m a journalist, so I’m going to ask one more: Are there any lessons that you take in regard to missile defense from what’s gone on?

Acton: It’s a great question. In any missile-defense discussion, it’s necessary to distinguish between different kinds of missile defenses. And the most basic distinction is between point and area defenses. Point defenses are where you’re trying to protect a relatively small area, like a military base or a city or a carrier group. To use a football analogy, these are your cornerbacks. Hopefully, from their perspective, they intercept the ball just before it reaches the wide receiver’s hands, so they only protect a very small area. The other form of missile defense, wide-area missile defense, or homeland missile defense, where you’re trying to defend wide areas—that’s what the US system in Alaska and California is about, trying to defend the whole continental United States against North Korea. That’s the equivalent of your defensive lineman in football, you know, trying to knock down the ball just after it leaves the quarterback’s hands and defend the whole of the downfield area.

Israel is basically a point defense situation. It’s a small country. Parts of it are desert. You’re not particularly worried about defending the desert. You might be worried about defending very specific sites in the middle of the desert, but you’re not worried about defending just the sand. So Israel is basically a point defense situation, and point defenses are much, much easier [to employ successfully] than area defenses.

Actually, for a while now I’ve been quite bullish about point defenses, not least because US point defenses, like THAAD and Patriot, perform pretty well. Israel’s success in defense tells us very little about the feasibility of area defenses. Indeed, the testing of the of US area defenses—here I mean specifically the ground-based midcourse system—those tests have been unrealistic, staged. And the US interceptors, even under those conditions, have a pretty bad track record.

So if you’re a fan of point defenses, which are expensive but I think pretty useful and not destabilizing, Israel’s success should make you pretty bullish on those point defenses. I don’t think it should change your opinion about area missile defenses much at all.