Share this page!

Enter URL or ID to Unroll

You can paste full URL like: https://x.com/threadreaderapp/status/1644127596119195649
or just the ID like: 1644127596119195649

How to get URL link on X (Twitter) App

  1. On the Twitter thread, click on or icon on the bottom
  2. Click again on or Share Via icon
  3. Click on Copy Link to Tweet
  4. Paste it above and click "Unroll Thread"!
  5. More info at Twitter Help
(((James Acton))) Profile picture
(((James Acton)))
@james_acton32
Co-director, @carnegienpp and senior fellow, @CarnegieEndow; nuclear policy, national security & politics; occasional agonizing about @Nationals; RT≠E etc.
5 subscribers
Oct 1 • 6 tweets • 2 min read
Bad move. As U.S. policies make the ME more unstable--including by increasing the likelihood of Iran's acquiring nukes--we pledge to defend Qatar.

If Trump believes the U.S. should defend Qatar, he should negotiate a treaty and present it to the Senate, not issue an EO. (1/n) There's a real debate to be had about U.S. interests in the ME (one that the China firsters within USG are losing).

But, even if you think the U.S. should defend Qatar, it's indefensible--as a matter of basic democratic accountability--to make the commitment via an EO. (2/n)
Jul 28 • 12 tweets • 5 min read
On Friday, I argued in @ForeignPolicy that Trump admin claims it had set Iran back by "years" were true but disingenuous.

The same day, the normally excellent @IgnatiusPost amplified the misleading U.S. and Israeli narratives.

Let's take a look...

(Links below). (1/n) First off, here are links for my @ForeignPolicy and @IgnatiusPost's @washingtonpost piece.

foreignpolicy.com/2025/07/25/ira…

washingtonpost.com/opinions/2025/…

(2/n)
Jul 21 • 13 tweets • 5 min read
🧵How much damage was done to Iran's nuclear program?

An analysis of Friday's reporting of the U.S. government assessment. I'll focus on the @nytimes, which was clearer than the earlier but confused @NBCNews story. (1/n) Image U.S. officials continue to argue that it would take Iran years to rebuild the facilities that were hit. That seems right to me.

But a key question is this: How long would it take Iran to build the bomb? (2/n) Image
Image
Jul 9 • 10 tweets • 4 min read
🧵Introducing optimal deterrence.

Here's my proposed U.S. nuclear strategy for managing escalation, arms racing, and proliferation with two nuclear peers, published through @CFR_org.

Health warning: If you like counterforce, take a DEEP breath before reading on. (1/n) Image The U.S. faces a security environment that's bad and getting worse. There's an emerging bipartisan consensus that, in response, the U.S. needs more nukes. The justification stems from the current policy of "counterforce" targeting. (2/n)

cfr.org/report/optimal…Image
Jun 26 • 11 tweets • 4 min read
How effective were strikes against Iran?

U.S. and Israeli officials stress how long it would take Iran to rebuild its previous nuclear program.

But there's a MUCH more important issue: How long it would take Iran to get the bomb? For that, a small program would suffice. (1/n) As in yesterday’s Axios article, these officials stress the damage done to Iran’s major declared facilities.

But the main issue isn't whether the U.S. & Israel severely damaged what they attacked; I'm sure they did. It's what they did NOT attack. (2/n)

axios.com/2025/06/25/ira…Image
Jun 24 • 7 tweets • 2 min read
How far has military action set back Iran's nuclear program?

Netanyahu says 2-3 years.

JD Vance says "substantially."

In @politico, I argue that Iran could build the bomb in a year. And it now has much stronger reasons to do so. (1/n) Image Iran still has its own nuclear triad: highly enriched uranium, centrifuge components, and expertise.

That's right.

A war waged nominally for nonproliferation has left Iran with several bombs worth of un-safeguarded highly enriched uranium. (2/n)

politico.com/news/magazine/…
Jun 17 • 5 tweets • 2 min read
The claim that Iran had 60 days to make a deal is disingenuous since the United States was hardly ready, and did not conduct, an intensive, detailed negotiation. (1/n) Image For example, there were times when Iran was willing to talk but the United States wasn't (presumably because it needed more time to prepare).

reuters.com/world/middle-e…Image
Jun 16 • 14 tweets • 4 min read
🧵The stated goal of Israel’s military campaign is to prevent Iran from acquiring nuclear weapons. With operations ongoing, it's too early to reach a conclusion.

But I can highlight outstanding questions that will determine whether Israel will achieve its nuclear goal. (1/n) Let's focus on three areas:
👉Fordow
👉Enriched uranium stockpile
👉Centrifuge components

BLUF: As of now, key facilities, material, and equipment are--or are likely--intact.

CAVEAT: The situation is fluid and uncertain. (2/n)
Sep 25, 2024 • 13 tweets • 4 min read
🧵Initial thoughts on Russia's new nuclear doctrine, based on a machine translation of Putin's comments. (I have not seen the doctrine yet; my assessment may change when I do).

BLUF: It's not tied as directly to the Ukraine war as others think. (1/n)

kremlin.ru/events/preside… Here's the key passage (thanks, @ChatGPTapp).

(2/n) Image
Aug 5, 2024 • 19 tweets • 8 min read
This thread is a transparent attempt by Colby to foreclose scrutiny of his behavior around the insurrection on Jan 6, 2021. I want to explain what he did and why it shouldn't be ignored.

(Like all my tweets, these are my personal views.)



(1/n) My criticism of Colby is nothing to do with his conservative views. We were friends—good friends—for years. We regularly had lunch or dinner. I went to Rio for his wedding. He came to mine. Throughout that time, we disagreed a lot about policy—domestic and foreign—but...

(2/n)
May 24, 2024 • 21 tweets • 6 min read
🧵Entanglement in action.

Ukraine has reportedly attacked the Russian early-warning radar at Armavir. This strike was a bad idea. It provided only limited military benefit to Ukraine and exacerbates nuclear risks.

It's time for the U.S. to act. (1/n)

The Armavir radar is part of Russia's "strategic" early warning system. It is designed to detect an incoming nuclear strike and enable Russia to launch its nuclear forces before they are destroyed. Image
Nov 8, 2023 • 6 tweets • 3 min read
At @CSIS @csisponi yesterday, Frank Miller engaged with my recent essay on targeting. In a collegial spirit, I'd like to respond to him here--he has mispresented me--and accept his offer to a public debate.

Link:

Here's the transcript. (1/n) csis.org/events/project…


Image
Image
Image First, I do not claim or imply in my recent essays that changing U.S. targeting policy would induce Beijing or Moscow to make changes to theirs.

I do not believe that our making changes would induce them to do so. (2/n)

warontherocks.com/2023/11/two-my…
Jun 24, 2023 • 15 tweets • 4 min read
<THREAD>There's been evidence-free speculation that Prigozhin may try to seize a nuclear weapon or two. I think this fear is misplaced. That said, there is a nuclear risk associated with this insurrection--one that isn't yet being discussed. (1/n) CAVEAT: The situation on the ground in Russia is clearly very fluid and unpredictable, and I'm certainly not going to try to predict what's going to happen. Here I aim to assess what we should worry about NOT how worried we should be. (2/n)
Mar 28, 2023 • 23 tweets • 10 min read
<THREAD>China now appears to have a comprehensive early-warning system against a U.S. first strike, allowing it to operationalize launch-under-attack options.

Stay tuned a new discovery: a new-ish Chinese radar that rotates (no, not Korla).

(1/n) As Chinese sources and the Pentagon have reported, China has been developing an early-warning (EW) system for a decade-ish so it could launch its nuclear forces before being destroyed by a US first strike (launch under attack). (2/n)

media.defense.gov/2022/Nov/29/20… Image
Mar 13, 2023 • 16 tweets • 4 min read
<THREAD>As @POTUS, @RishiSunak, and @AlboMP announce AUKUS submarine plan, here’s my assessment of the technical and proliferation risks.

BLUF: They’ve made serious efforts to mitigate those risks, but those that remain are real and significant.



(1/n) Here’s the plan (in brief):
1. 🇬🇧 & 🇺🇸 deploy SSNs* in🇦🇺(from 2027)
2. 🇦🇺deploys Virginia-class SSNs purchased from 🇺🇸 (from ~2032)
3. 🇦🇺deploys AUKUS SSNs, designed and produced with UK (starting in early 2040s)

*SSN=nuclear-powered attack sub.

(2/n)

whitehouse.gov/briefing-room/…
Mar 9, 2023 • 4 tweets • 2 min read
If these reports are correct, this program is a goat rodeo in the making. 🇦🇺, which has never operated an SSN before, now plans to operate two different classes. Plus modifying 🇬🇧-built Astute SSNs and introducing 🇺🇸 technology will add significantly to the technical risk. Prior to today, I gave a ~50% chance of 🇦🇺 fielding SSNs by 2040, on the basis of reporting that it would go for an unmodified design. I think the odds of deploying Ozstute are now ~30% (though maybe still 50% of fielding Virginia).
Feb 23, 2023 • 14 tweets • 7 min read
Last week, Australian FM @SenatorWong refused to say explicitly that B-52s visiting Australia do not carry nukes, causing a minor furor.

The thing is that, best I can tell, official, stated U.S. policy is that bombers are not loaded with nukes. (1/n)

breakingdefense.com/2023/02/austra… Some history. Back in the 1960s, the U.S. did keep aircraft--B-52s actually--loaded with nuclear weapons on airborne alert. This posture had obvious disadvantages as became apparent when... you know... aircraft crashed. (2/n)
Feb 21, 2023 • 12 tweets • 3 min read
<THREAD>Russia has suspended its participation in New START--does it matter? (Spoiler alert: Yes!)

Andrey has a great thread below on the details of the Russian announcement; I'm going to assess its significance. (1/n) The immediate effect of Russia's suspension will be the end of notifications. As of 2/1/23, the US and Russia had exchanged over 25,000 notifications (not a typo). These notifications continued even in the absence of inspections and... (2/n)

state.gov/new-start/
Feb 19, 2023 • 11 tweets • 3 min read
<Threads>Some thoughts and suggestions--for both governments and journalists--on news that the IAEA has detected 84% enriched uranium in Iran.

This is a serious and worrying escalation, but we don't yet know how serious and worrying. (1/n)

bloomberg.com/news/articles/… It's serious because it's entirely possible to manufacture a nuclear weapon out of 84% enriched material. Although 90% is often called "weapon grade," it's not a special threshold; rather, you just need a bit more 84% enriched material to compensate for lower enrichment. (2/n)
Feb 16, 2023 • 11 tweets • 4 min read
<THREAD>Some personal recollections of Catherine Kelleher--surely the only person ever to have met both the satirist Tom Lehrer and the long-time dictator of Turkmenistan, Turkmenbasi.

Catherine has just passed away. She was a legend. (1/n) You'll be able to read about her incredible career in formal obituaries. Here, I'll just mention that she was a trailblazing woman in national security, serving as a deputy assistant secretary of defense and the SECDEF's representative to Europe. (2/n)

cissm.umd.edu/our-community/…
Feb 2, 2023 • 6 tweets • 3 min read
<THREAD>My new oped "The Case for Caution in Crimea" is out in @WarOnTheRocks.

It makes two points about enabling Ukraine to threaten Crimea:
1. It will NOT advance real negotiations.
2. It might lead to a nuclear war. (1/n)

warontherocks.com/2023/02/the-ca… The @nytimes reported that the Biden administration believes Ukraine's threatening Crimea could advance a negotiated settlement to the war.

In theory, this argument is sound--giving Putin a lot to lose could make him negotiate seriously.

BUT... (2/n)

nytimes.com/2023/01/18/us/…