Transcribed automatically: it certainly contains errors. It’s pretty good, just not perfect. We provide it here because, for accessibility, it’s better than nothing.
Jeffrey Lewis 0:06
You are listening to the Arms Control Wonk podcast, a podcast on arms control, disarmament, and nonproliferation. I'm Dr. Jeffrey Lewis, a distinguished scholar of global security at Middlebury College and a distinguished research fellow at the Foreign Policy Research Institute. Getting better, Aaron. How are you doing?
Aaron Stein 0:25
How are you, Aaron Stein, president of the Foreign Policy Research Institute?
Jeffrey 0:28
It's good to see you again. Two in a row. My boss. My boss. Can I talk back to you?
Aaron 0:33
Not if you know what's good for you.
Jeffrey 0:37
All right. I'm going to get lippy. Well, you're going to have to sign the HR handbook.
Aaron 0:42
You haven't done that yet.
Jeffrey 0:43
I'm going to have some thoughts about the way you run this place.
Aaron 0:46
Well, it's good to see you again, my friend. So what's your segue there? I'm trying to, I don't know, maybe I'll throw you a big hole in the ground, big enough to control your explosion. Does that work? Oh, I don't know.
Jeffrey 1:01
Okay, listeners, we are going to talk about the allegations that China has conducted a series of, let's call them low yield nuclear explosions, including one on June 22nd, 2020.
Aaron 1:22
Well, I think we should set the context here. Donald Trump, when was it? Two months ago? A month ago? Fairly recently, basically came out and previewed this news and said that the US would do everything else, would resume some element of nuclear testing on parody of the others, prompting speculation. And I think a lot of us had in the general vicinity of what we thought he was talking about, but this adds a new level of specificity.
Jeffrey 1:50
Yes, when Trump said that, I pointed out that this was a difficult and awkward phrase for him because while the US had openly in the annual State Department compliance report accused Russia of conducting low yield explosions, which we can talk about whether they were or weren't maybe on the side of this conversation. The language about China was more circumspect.
The language about China was simply that a lack of transparency gave rise to concerns. And so a lot of us said, You know, based on the public information, you can't really define what an equal basis is. And Tom DiNanno, the Undersecretary of State for International Security and Arms Control, a Middlebury grad said, hold my beer, dude, and said, this is what President Trump meant and said the following at a big speech in Geneva.
Today, I can reveal that the US government is aware that China has conducted nuclear explosive tests, including preparing for tests with designated yields in the hundreds of tons. The PLA sought to conceal testing by obfuscating the nuclear explosions because it recognized these tests violate test ban commitments. China has used decoupling, a method to decrease the effectiveness of seismic monitoring, to hide their activities from the world.
China conducted one such yield-producing nuclear tests, sick, on June 22, 2020.
Aaron 3:32
I mean that's almost for our field, full frontal X-rated revelations.
Jeffrey 3:40
It is very explicit and I will say there were rumors going around that he was going to say something like this. Those rumors came because – and I do want to defend him here. Those rumors emerged that he had had a meeting with some Middlebury College students. And that he had previewed the speech. Yeah, I have it on good authority that he simply said there would be a speech.
But I think people were – in a time-honored Washington tradition, people were using the fact of his meeting with undergraduates to cloak their leaks, which is how the game is played. That is how the game is played. Do not blame the kids. The kids didn't leak anything.
Aaron 4:26
Were there any sort of suspicious bets on one of these polymarket things about this? Because that's the real indicator. That's the real indicator. OK, so let's start at the beginning.
Jeffrey 4:37
Boy, there's so much to talk about.
Aaron 4:39
OK, so let's start at the beginning. We've done this before. But I think for listeners who maybe tuned just in, What do they mean when they say low-yield nuclear tests? Because when you think about nuclear tests, it's the old 1950s footage and boom and the mannequins and the stuff blurring and blah, blah, blah. These are a little different.
Jeffrey 4:58
Right. So all nuclear tests these days are conducted underground. Yep. And there is a treaty that bans them. And it is called the Comprehensive Nuclear Test Ban Treaty. But Despite the name, Comprehensive Nuclear Test Ban Treaty, it doesn't ban tests. You can test weapons. It bans nuclear explosions. It does not define nuclear explosions. On purpose. On purpose.
It's probably a good idea that it doesn't. We can talk about that. But that creates a murky situation where... Countries that don't have nuclear weapons and don't know how nuclear explosions work are just like, we won't do nuclear explosions. And countries that do have nuclear weapons and do know how nuclear explosions work have their own technical definitions.
Now, to further complicate the situation, the US refused to ratify the treaty in 1999, and so China also refused to ratify it, and Russia, which had ratified it, waited for like 20 years, and then a couple of years ago, unratified it. Correct. So we have commitments to abide by the ban on nuclear explosions. But we don't actually have, we have political commitments and we have a legal commitment not to defeat the object and the purpose of the treaty, but we don't actually have like the treaty, we don't have any of the infrastructure that would allow for like an inspection, and we don't even have an agreed definition.
And so what has happened is we have said that they are conducting low yield, I kind of prefer the phrase events, Because we don't really have a definition of an explosion, but there are events that we believe are inconsistent with our definition.
Aaron 7:05
And it's about the amount of yield created from each test.
Jeffrey 7:09
Right, because the treaty doesn't define an explosion. And so that leads to different definitions. And I think the mind bending thought is if an event Like a nuclear reactor goes critical. It most certainly does. But we don't think of that as an explosion. No. There are some instances where you use explosives to compress nuclear material and you get some nuclear yield in the sense like there are neutrons produced, but not enough for it to go critical.
Aaron 7:49
Correct.
Jeffrey 7:49
So it doesn't become a chain reaction.
Aaron 7:51
And those tests do happen.
Jeffrey 7:53
Yeah, they are called subcritical tests.
Aaron 7:54
Subcritical tests, yes.
Jeffrey 7:56
Now, we in our own mind, and I think in consultation with the Russians and the Chinese and the French and the British, I think we kind of all have a sense like there's some stuff that's okay, and then there's some stuff that's not okay, and then the nub of this is there's maybe some stuff in the middle.
Aaron 8:16
Yeah, it's the gray area at the outer edges. How much –
Jeffrey 8:21
Yield do you produce without trying to produce a large yield? Right, and so you know, the way that the weapon years wanted to do this back in the day was they were like, you know, if I can contain the thing in a canister. Oh yeah, I remember this. Yeah, it's not really a nuclear explosion, you know, like a nuclear explosion you know, to quote from Nano Diamonds, yeah, like it sucks the paint off your house and gives your family a permanent orange afro. This is like This is something else.
And so we did not define those events as explosions, even though they produced yield that could be measured. The Russians and I presume the Chinese designers felt the same way, but the politicians were like, well, it would be better just to call it zero yield and nobody really wants to give the Indonesians a briefing on how to do subcritical nuclear testing.
That's the nub of our problem.
Aaron 9:21
No, it makes sense. What's interesting about the announcement is, what was it? Hundreds of tons?
Jeffrey 9:27
Well, OK, so this is very carefully worded.
Aaron 9:30
Yes.
Jeffrey 9:31
Has conducted nuclear explosive tests, including preparing for tests with designated yields in the hundreds of tons.
Aaron 9:44
Yeah.
Jeffrey 9:44
So that implies that they did not detect an event like this seismically. Someone told them. Right. Or they intercepted a communication
Aaron 9:57
And this is where that word decouple is actually important.
Jeffrey 10:03
Right. Because, so let's say you have one piece of intelligence and I, you know, I have no idea. No, we're guessing.
Aaron 10:10
You're in the Middlebury student lecture?
Jeffrey 10:13
Precisely. Well, I just think like, whatever, like you might not like the man's politics, but like the fact that he took time out of his day.
Aaron 10:23
I'm joking. I was joking because people, people are ascribing how they were laundering their own intelligence. I know. Blaming it on your students is ridiculous.
Jeffrey 10:31
Well, like, cause it blames the students and it like, you know, like it attacks him personally, which is, you know, for doing a good thing, but I'm fine attacking him, but I only want to attack him for bad things he does. All right. So. You imagine that they, on the one hand, have some kind of humint or sigint that says, this event is supposed to be 100 tons.
Right. But then you get the event.
Aaron 11:05
Without the seismic signature.
Jeffrey 11:06
And the event is either lower or you never get the event. Right, exactly. Right. And so that raises the question of like, well, how far can how far can you monitor at Lopnor? And then that's where decoupling comes in. And boy, howdy, I teach a whole class on this problem and I use Lopnor as the literal example.
But I think the first thing to start off by saying is there is no event. There is no event on June 22nd, 2020 in the CTBTO catalog that Rob Floyd put out a statement that says we looked, we don't see it. There is no event in any of the catalogs I looked at. And if you're new to this, you know how an earthquake is a 7.6?
Yeah. Same thing, same stations, same everything.
Aaron 11:58
The CTPTO uses seismic monitoring all over the world, fed into a hub in Vienna. And nuclear explosions look different than earthquakes, the little squiggly lines. They just look different. You can tell the difference. And so that's how you monitor whether or not compliance.
Jeffrey 12:16
A different ratio of waves. I call them squiggly lines, my friend. Okay. But this is really important because we can, with very high confidence, using unclassified networks, see events down to an earthquake size of about 3.8, which is really, really small. You're from California. You should know. You know, it's a little, it's a little complicated to turn 3.8 into a real number, but that's like a hundred tons.
Using regional networks, so not just the kind of global monitoring, that number goes down to 2.8 or at lop nor even 2.2. 2.2 is tiny. That would be one ton. So we really should see anything bigger than a ton in the catalog.
Aaron 13:25
In the open source catalog you can look at.
Jeffrey 13:28
Right. And we're not even talking about AFTAC, which has its own classified monitoring network, which I presume would do better, but probably not that much better because seismology is seismology. And why would you work for AFTAC if you could work for, you know, Berkeley or the University of Nevada, Reno?
Aaron 13:44
Well, Reno. I mean, you say Reno, I say Tahoe.
Jeffrey 13:52
I say Tahoe.
Aaron 13:54
Tahoe, there you go. So that goes back to the statement, right? So that goes back to the statement of indications of hundreds of tons. And so the way I interpret that is somebody was talking or somebody told them that they're going to do a test for hundreds of tons for whatever reason on so-and-so date, June 20th, 2020, right?
June 22nd. June 22nd, pardon me, June 22nd, 2020. And presumably there were secondary indicators. My guess is they were watching from space and there was a lot of movement, although you could tell me that.
Jeffrey 14:27
A bus shows up at the test site. I haven't looked yet, but sure, something like that.
Aaron 14:31
Something shows up at the test site because we know where it is. So presumably the test happens or not, and there was no corresponding seismic data.
Jeffrey 14:43
I would think, because if they, and we'll talk about decoupling in one second, but if they set off something that was 100 tons in that mountain, we really ought to have seen it.
Aaron 15:00
Right. Somewhere, either on the information centers you were just talking about, and presumably on even the classified stuff, or somewhere, right?
Jeffrey 15:12
The first North Korean nuclear test was a failure, and it was, I think, a few hundred tons. And, you know, we saw it clear as day.
Aaron 15:22
I remember that day vividly because the debate's playing out on an old blog called Arms Control Wonk about how big it actually was.
Jeffrey 15:27
Like, we would have seen it. I mean, and I know the U.S. would have seen it because inexplicably some genius declassified the monitoring threshold at Lop Nor in the mid-1990s.
Aaron 15:42
Oh, it must've been like with the Epstein files. So some names don't get redacted.
Jeffrey 15:46
Like, yeah. And it, you know, it, it, it, it, it's for decoupled yields, which now we can talk about, but like it would suggest that if it was a hundred tons, we should see it.
Aaron 15:55
So decoupling, it's actually not that complicated, but they make it sound really complicated, but go ahead.
Jeffrey 16:02
Yeah. Well, Albert ladder who hated the test band, but was a really great physicist. Just, just how it shakes out sometimes. Figured out that if you exploded a nuclear weapon inside a cavity created by another nuclear weapon, it would muffle. It would prevent you from fully coupling the energy of the explosion to the surrounding geology, and it would make it look smaller.
He guessed it would make it look 100 times smaller.
Aaron 16:31
Basically, rather than drill a small little hole and then sort of wedge it in there and then the surrounding earth absorbs the entirety of the corresponding shockwave, I'm sure I'm screwing it up somewhere, you dig a really big hole and you stick it in the, in the middle of the really big hole, a lot of the energy dissipates before it hits the walls. Well, some people are like, hooray, no arms control with the Soviets. Screw those guys.
Jeffrey 16:53
But we actually did do a test. We did a series of tests, but there was a real nuclear one in Mississippi of all places. There's literally a picture of a Confederate flag fluttering from the freaking bomb, which is just not one of my favorite visuals. And they figured out that in a salt dome. I remember this. Yeah.
You could decouple it actually by a factor of 70. In rock, all the data suggests it's more like 20 to 40.
Aaron 17:23
That makes sense.
Jeffrey 17:26
So basically a 100-ton test is going to look 20 or 40 times smaller. So a 100-ton test is going to look like – You know, a few tons.
Aaron 17:43
A few tons. Don't ask us to do that math that quickly. Wait, you know what?
Jeffrey 17:48
I should do this. Wait, wait. Let me divide 100 tons divided by 40. Yeah, it's a few tons. It's a few tons. It should be fine. Several tons, I guess. Better word.
Aaron 18:02
And so this was known, right? I want to be clear that this was known. I think, again, because sometimes political reality is you sweep some things under the rug. It was just deemed impractical, I think, for full-scale nuclear, the size of the cavity one needed to make.
Jeffrey 18:16
Yeah. Basically, you can really only do this for really small tests in the first place because the cavity gets bigger and bigger and bigger, and eventually the cavity collapses. Right. And- You pretty much have to use a cavity created by a prior nuclear test just because otherwise we'd see you dig the giant cavity.
Aaron 18:40
Right. There'd be corresponding earth.
Jeffrey 18:42
Right. So like if you watch the, you know, I mean, and the Chinese have dug some new cavities, but you can sort of estimate the size and so on. So, and even then, if you see a small event near the cavity, you know, it, you didn't really fool anybody. And it's also, right, so it's not as good in rock.
It also, the effect is not exactly uniform over all frequencies. So I don't want to be a seismologist, but just, it's not, it's not the killer solution you think, and there's almost no data on it. In fact, this is how little data there is on it. When I went to NNSA, when I went to the Nevada National Security site, they showed us the tunnels they were digging in P-Tunnel, a former nuclear test tunnel.
Because they were going to do decoupling experiments. They were going to set off conventional explosions, one that was fully coupled and at least one that was fully decoupled as a verification exercise.
Aaron 19:44
Now- Do you think there was a link? I'm actually, I didn't know.
Jeffrey 19:49
I think A, like the Chinese certainly would see these events.
Aaron 19:54
And probably wonder. Assuming that what we were told is accurate, and we can come to that later, do you think that there was perhaps some urgency to try and figure out to verify the absence of data?
Jeffrey 20:06
No, I think this was good governance. I think DOE had consistently done good CTBT work, and DOE under the Biden administration increased that work. Okay. Now, I will say, like, Like, I think the transparency they did was great. Like, the reason we know about it is they said we're doing this thing. If you're the Russians or the Chinese, it probably looks like we're evasively testing.
Aaron 20:36
Well, yeah, we've talked about this.
Jeffrey 20:38
And if you're the Chinese or the Russians and you are doing the same thing, it definitely looks like you're evasively testing because you're not even telling us. What you're doing. But all that's aside, I just really don't think we have great data on decoupling. So I think we would have expected to see an event. And the fact that we didn't, even if it was fully decoupled, like a factor of 40 in rock means this thing must have been pretty small.
Aaron 21:09
Right.
Jeffrey 21:12
Then that raises the question like, well, why would you do – what's the purpose of a mouse fart in the desert? The answer is it's good for things like safety, security, stockpile stewardship. You're not going to be developing new crazy killer weapons with this thing.
Aaron 21:31
The very real fact, we talked about this previously, the Chinese just have far less test data than we do.
Jeffrey 21:37
Yeah, I mean, yeah. So I mean, I do think I do think, well, yes, it's the Chinese only conducted 45 tests, whereas we conducted more than 1000. Most of their tests were above ground, because they didn't have fiber optic cable and other things that would allow them to, I would say, properly instrument them. So they got very little data.
Out of most of those tests. So it does strike me that they have a really strong incentive now that they have proper supercomputers to populate their computer codes with experimental data. So if you told me they were doing low-yield testing, I'm not going to like, they would never. Of course they might. They have a real incentive.
Aaron 22:18
To do that because they are at a huge disadvantage and they want to close that disadvantage, but this is what then makes me nervous about the policy response, right? No, that, that's where we're getting to, because we just spent how long on this thing? Hold on, explaining this took us 22 minutes to actually explain this, right? And yes, we have a little bit of witty banter back and forth, so it takes 17 minutes to explain this. I think when people will read this headline, it will be, Russia, China conducts nuclear test in tunnel, US doesn't detect.
And the idea that populates in people's mind is a boom. You know what I mean? And how dare you?
Jeffrey 22:55
Well, by the way, this is one of my like right wing opinions is the old freaking designers were right when they refused to call these things explosions. Oh, no, no, you're completely right. Yeah, no, absolutely. Like those crusty old guys, when they're like, God damn it, that's not an explosion. We know what an explosion is, and that's an experiment.
I think they're right, and I wish we had listened.
Aaron 23:19
They are right, because Because again, it evokes this thing in your mind. It's these old Cold War films and what it is. Again, we're just – I'm inferring this from the statement and I'm not trying to absolve the Chinese. It's something else entirely than what one would conjure up in your mind as to a nuclear test.
Jeffrey 23:37
Right. I mean I think if we – so we've long – I and some colleagues have long argued that the US and Russia and China should just – Along with the British and the French, just agree on what an explosion means. And I don't even care what agreement we reach. If that agreement permits what some people would characterize as low yield testing, that's fine.
I would prefer it didn't because I would prefer to maximally constrain the Russians and the Chinese. I don't think that matters so much. What I am very worried about is we're now going to respond with doing our own low yield tests, own low yield explosions, and it's going to spiral. Then the Russians and the Chinese are going to be like, hey, you said zero means zero.
Now you're out of the deal. So now we can just go ahead and do full yield tests. And the Chinese are going to be like, oh my God, we have a... We have like 300 tests we want to do. Right. No, no, totally. Yeah. They're going to be like, hey, do you remember that one you did in 1977?
Anyway, we're going to do the same one. And it was weird. We were looking at the data we had.
Aaron 24:49
Yeah. No. Right. Because then. Well, I guess again, because it sort of picks up on the theme from the last podcast, if you were listening to it, is the actual constraint on the US is we're just not quite ready to rapidly go to these types of tests. Well, I guess these ones we may be able to, but to rapidly go to these types of tests, we're just not quite ready.
So perhaps we could drag it a little bit. We could do it.
Jeffrey 25:09
I mean, we could six months, a year. But what the Trump people talk about is doing an uninstrumented test, which is just to show that We need business.
Aaron 25:19
It's the hillbilly – well, I guess it's the hillbilly what turns into Typhon. Oh, right, right, right. It's bolting a MK-41 launcher onto a truck and driving it out onto an island and then just shooting something just to say you can't.
Jeffrey 25:35
Right, and then the problem is you – Create the conditions for them to respond more quickly and more ably.
Aaron 25:42
And then to appropriate money on our side, most importantly, and then the money starts going to DOD budget because you demonstrated capability with the hillbilly test.
Jeffrey 25:52
Yeah. And I, you know, so I, I do think, I will also say if we had ratified the CTBT, we could have asked for a special inspection. CTBT allows using national technical means. We could 100% have asked for a special inspection.
Aaron 26:12
Have you ever participated in a CTPTO exercise or at least simulation like Wargame where they actually exercise it?
Jeffrey 26:18
No, I really want to. I know they do do them, man.
Aaron 26:20
I did it once. I did it once a long time ago. I'm not violating confidence. It was quite clear that the fictitious country conducted a low-yield nuclear test and then the other country accused them. The arbiter of this was U.S. intelligence, and it wasn't Americans in the room. I think it was one of like five Americans of like 30, and every single one of the participants was like, we don't trust the American intelligence.
No special inspection. And then the CTVTO exercises were like, no, we can't publish that.
Jeffrey 26:51
I mean, you know, like the US, we did stack the procedure.
Aaron 26:56
Well, this was much closer to 2003. I want to say this is around 2006. There's a hangover.
Jeffrey 27:03
The – I'm sorry. The council – well, we – I mean, this was actually an issue, right? Because there were issues about like – Do you have a so-called red light inspection, which is a method where you're basically allowed to see BTO will do the inspection unless people vote to stop. Or you have a green light method in which you have to get approval from the council in order to go and do it.
But the compromise ends up basically stacking the population of the – I forget what the CTBTO's executive council is called. I think it's the executive council. It's basically stacked with US allies. For now, yeah.
Aaron 27:48
Yeah.
Jeffrey 27:48
Well, I mean, right. Yes. We did not anticipate a world in which everyone hated us. And like maybe we should not have brought that world value. It's just like, well, well, well, if it isn't the consequences of our own actions – So I do think though that like it almost doesn't matter because if China sought to block the inspection, you would have the answer to your question.
Right. Of course. Right. If you are like, hey, there is an anomalous seismic event. We believe it is consistent with a nuclear explosion. We have other reasons to believe it may have been a nuclear event. Here's maybe a satellite photograph showing a bus at the site on that day. This is pretty ambiguous. We want to do an inspection.
If the Chinese were innocent, they would, to a first approximation, they would be like, oh, hey, we don't want the treaty to collapse. This is in all of our interest. Sure, we'll let inspectors come. If the Chinese are like, how dare you accuse us? No way. You don't get to look at our tunnel.
Aaron 28:58
Well, yeah. We have accused people of nuclear tests in the past.
Jeffrey 29:04
Yes. We demarched the Russians over what turned out to be an earthquake in 1997. It was humiliating.
Aaron 29:11
But I guess in this case, there's the absence of the earthquake.
Jeffrey 29:14
That's what's interesting about this is it's the – Well, there's the absence of the earthquake in the catalogs I see. But maybe there's – maybe Aftak picked something up. I don't know.
Aaron 29:24
But go back to the original statement. That is lawyer language in the original statement. It's very lawyered. That is lawyer language in the original statement. It would suggest to me, and again, that one of the indicators to them of decoupling is that they saw nothing. They had probably very clear indications that something was going to happen and then they saw nothing.
Jeffrey 29:51
You know, I hope it is not that. Me too. Because that is the curveball fallacy. Of course. You know, that's where you have human intelligence, which is important and we should have human, but you have to be able to verify and check it. And if you can't verify it and if you can't check it, Then you are in real danger.
And if you get in the habit of saying, I was told a thing, but then I don't have any ability to verify it, my monitoring must be much worse than I thought. Like that, you are in self-licking ice cream territory, self-licking ice cream cone territory, right? You are building a non-falsifiable worldview in which nothing convinces you. And that's just analytically, that's a bad place to be.
Aaron 30:40
No, totally, right? I mean, that's the negative outcome. The other one is that the initial intelligence is so exquisite that they're convinced that the test actually did happen. You know what I mean?
Jeffrey 30:53
So there is the other side. But I don't know.
Aaron 30:57
My point is the nuance in the language.
Jeffrey 31:00
Go ahead. Yes. No, this is exactly the problem we face with the Syrian reactor, where it comes right after Iraq. And so we are inclined to not believe them one bit.
Aaron 31:11
That was the second best case of early OSINT. The first one was the Iranian soccer field. Al-Kubar was really good, too.
Jeffrey 31:18
But then, I mean, I remember saying, there's no way you could tell from a satellite image if this thing was a reactor. It was a box. Right. You'd have to have exquisite human intelligence. It later turns out that like, yes.
Aaron 31:33
Yes, they did.
Jeffrey 31:34
They did. I was like, I remember thinking, well, you would have to have pictures from the reactor construction site to be sure. They did. It's like, yes, about that. I mean, the thing is that's pretty rare. That's true. I'm pretty happy being wrong about that historically. Because it was such an extraordinary piece of intelligence gathering. So unless we are talking about the same level, unless it literally turns out that one of these guys in the Chinese military who's been purged really was on the CIA payroll, and every two weeks he was sending a full debrief on PLA activities that included decoupled nuclear testing in the desert, Which I, like, I, I'm not excluding it.
That's just, that's extremely rare for that to happen. And, and if it, if, if that turns out to be the case, then, you know, then that will explain it. But it just, that, that seems really, really, really unlikely because you would expect to see something in the tens of tons to say nothing of the hundreds of tons.
Aaron 32:54
It's not good, whatever it is, because to agree to go to this granular level of detail probably does reveal some source and method. And so they made the decision to do this Yeah, I don't know, man, because they've been sitting on this since 2020, it seems. I mean, then Trump tweeted it out. I mean, the thing is, did they expect the president, after he got a briefing, to do something and say something?
Again, we're speculating. We're getting too far down into the rabbit hole. Yeah, you're right. We don't know. We just don't know what happened here, and then I don't even know how we'll respond. I mean... Yeah, well, I mean, I think we're going to do...
Jeffrey 33:33
We're going to loosen our definition of experiments. Because again, we don't have a shared definition with the Chinese. I'm going to put an asterisk with that. We might actually have had a shared definition at some point, but that seems to have been lost or broken down. So we really only have our definition, right? Our view is you are violating our definition.
Aaron 33:53
Yeah.
Jeffrey 33:53
Change your definition. We could change the definition. Yeah. I mean, this is my like, you know, Trump could get a Nobel Prize if... They loosened the definition and the five just all agreed on the new loose definition and then brought the treaty into force? I think that could actually happen too. He could do it in a second, right?
He could just literally send Steve Witkoff. Witkoff will be like, how about release of energy from an explosive canister? We all know what that means, right? And everybody will be like, yeah, yeah, yeah. We all know what that means. That's going to make an Epstein joke. Yeah. Good God. Don't. And they'd be done. They'd be done, right?
They'd have a deal, and they'd be like, yep, still zero yield. We've just defined yield as the release of energy from a canister, and then bang. Bang. Bang. We measured in kilotons, but we round down.
Aaron 34:52
It's actually the definition of obliterate.
Jeffrey 34:55
Yes.
Aaron 34:56
That's a callback to an old podcast. Ding. We finally did that. You did it. You did it.
Jeffrey 35:02
I don't know. I don't think any of this matters except how we collectively, and it's not just us, it's the Chinese and the Russians, how we collectively decide to approach this. Do we want to be in a world in which we are all pretty much unconstrained and off doing big yield tests? Do we want to be in a world in which we are doing Experiments that really flirt with that line, or do we want to be in a strict world?
We can pick. I think we're going to pick the dumbest one because that's what we do, but we could just pick.
Aaron 35:51
I think we hit it. I mean, I don't know. Maybe the lasting image will be the Confederate flag sort of fluttering down the Mississippi salt mine.
Jeffrey 36:01
Some sort of metaphor. It was a salt deposit. They didn't excavate it. They used a nuclear explosion to create the cavity.
Aaron 36:07
But, you know, maybe that's some sort of metaphor either for the past or the future. I don't know. So I think we're going to leave it there. Patreon.com slash ACW Podcasts for the show. $3 if you're a student, $5 if you are not. Join the coolest Slack channel where I can promise you they know exactly how to decouple.
Jeffrey, good to be with you.
Jeffrey 36:28
It was a pleasure. Thank you.
