‘Hey, how are we going to make this work today?’
September 17, 2022 7:28 PM   Subscribe

Inside the $100 Billion Mission to Modernize America’s Aging Nuclear Missiles. "Walking into Moffett’s capsule at Alpha-01 is like walking into the past. Banks of turquoise electronics racks, industrial cables, and analog controls have been down here since the U.S. military installed the equipment decades ago. Look closely at the machines and you’ll find names of manufacturers like Radio Corp. of America, defunct since 1987, and Hughes Aircraft Co., defunct since 1997. Some systems have been updated over the years, but these advances are unrecognizable to anyone who lived through the personal-computer revolution, let alone the internet age. The entire ICBM fleet runs on less computational power than what’s now found inside the smartphone in your pocket. When something breaks, the Air Force maintenance crews pull parts from warehouse shelves, pay a contractor to make them to specifications, or even occasionally scavenge them from military museums."
posted by storybored (62 comments total) 27 users marked this as a favorite
 
If there's anything that it would be good for humanity to forget how to do...
posted by clawsoon at 7:34 PM on September 17 [7 favorites]


Dear person reading this thread,

If someone gave you $100 billion to modernize America's aging infrastructures, what would you do with it? Please show your work.

Thanks!
posted by aniola at 8:11 PM on September 17 [5 favorites]


If someone gave you $100 billion to modernize America's aging infrastructures, what would you do with it? Please show your work.

Step 1: Ask vague yet leading questions in order to provoke the evergreen leftist circling-firing-squad to buy me some time to think
Step 2: Bury all the 50+-year-old weapons of death plagued by failing rubber seals/fucked-up rocket fuel/whatever-else-is-wrong in a giant hole in the ground
Step 3: Convert 10% of the country's highway miles into bike paths, a la Rails to Trails. Use savings on road maintenance to finance light rail systems in small-to-medium cities.
Step 4: Blockchain?
Step 5: Profit!
posted by Mayor West at 8:20 PM on September 17 [21 favorites]


The entire ICBM fleet runs on less computational power than what’s now found inside the smartphone in your pocket.

I mean, I get the point trying to be made here, but I’m not convinced this is necessarily a bad thing.
posted by Thorzdad at 8:30 PM on September 17 [28 favorites]


I mean, I get the point trying to be made here, but I’m not convinced this is necessarily a bad thing.

I'm not convinced this is a good thing. What's your reasoning?
posted by bendy at 8:53 PM on September 17


Something that is good about these systems is that, fortunately, you can fabricate most (though not all, such as 8 inch floppies) of the parts if you want to. Nothing is trapped because of software licensing or patents. If there had been a successful rebuild of the ICBM systems in the 90s, the whole thing would be trapped running weird versions of windows 3.1.
posted by rockindata at 9:07 PM on September 17 [31 favorites]


With respect to other options, the science agency I work for could run for the better part of a century on 100 billion. The amount of data we could collect with that kind of funding, and the way we could use that data to help Americans and the rest of the world deal with impacts of climate change…the mind boggles. It would be amazing.
posted by rockindata at 9:11 PM on September 17 [19 favorites]


It's not accidentally going to pocket-dial China or Russia or something.
posted by aniola at 9:12 PM on September 17 [8 favorites]


You can’t have this kind of war,” Eisenhower said at a national security meeting a couple of years later. “There just aren’t enough bulldozers to scrape the bodies off the streets.

-Eric Schlosser.

posted by clavdivs at 9:26 PM on September 17 [25 favorites]


The Voyager spacecraft performed far far beyond their design lifetimes and mission scope with less computational power between them than my kid’s school calculator. I am skeptical of whether that would have happened if their creators had modern computing resources and operating paradigms to act lazily within.

I’m with Thorzdad. I don’t want anything even remotely capable of the modern day advanced automation which all the cool kids [mistakenly] call “AI”, anywhere near a stickpile of thermonuclear planet-enders. With great computing power comes great responsibility, and it remains to be seen if humanity is up to the task. We can’t even successfully manage “social media” apps safely without them being appropriated by those who want to strangle democracy.
posted by armoir from antproof case at 9:35 PM on September 17 [22 favorites]


How about a nice game of chess?
posted by not_on_display at 10:15 PM on September 17 [21 favorites]


Of *course* these clowns want replace an antiquated system with *another* system soon to be antiquated. it is absolutely unnecessary but hey, that's never stopped the war machine contractors before.

Those nonstop subs and jets which fly all day every day are enough to deal with anything that comes our way. Disassemble the current underground nukes, recycle what you can, turn the sites into really cool bungee jumping attractions.

If it were to happen it's kiss your ass goodbye. Not just Washington and some submarine bases, every city with any industry is gone, I'm guessing here but I think it's a safe bet -- every city otn the east coast. Plus: Every city on the west coast. Seems to me it would be a lot easier to upgrade weapons launched from subs and/or jets.

Trump woke a lot of people on that pretty afternoon when he came walking out of his golf clubhouse and pretty much out of nowhere threatened to blow N. Korea into bits of rags and atoms. He sure woke me up. The president of the US can make the end of the world happen just because he has gas. (correction: *Not* the entire world; the world will be just fine. But most life forms as we know them -- adios, amigo.)

I don't guess Biden is reading this. Just in case he is "Hey, Pres -- Don't be a dope. Don't replace old junk with new junk which will be old junk FAST. Build more subs, more planes, each sub and plane loaded up with MIRVs. Problem solved. Don't be a dumbo."
Sincerely
dtb
posted by dancestoblue at 10:47 PM on September 17 [2 favorites]


I'm not convinced this is a good thing. What's your reasoning?

Something something Battlestar Galactica
posted by kaibutsu at 10:51 PM on September 17 [11 favorites]


Hard agree on simple, old tech being preferred: the ideal is the least complexity that gets the job done. Fewer potential sources of fuckup, easier to replace with purely domestic-sourced components if shit breaks or ages out. Last thing I want in our nukes is 118 APIs for a similar number of half-vetted libraries. Modern software engineering mostly works but it’s a fucking miracle we don’t have major infrastructure dependency failures on a daily basis.

The true ideal is decommissioning all of them, which we should totally do the day one of our major competing nuclear powers does democracy less-shitty than we do. Despite how fucked we are, that is not yet the case for China or Russia so even on just purely abstract ethical grounds I support us disarming last. Shoutouts to China for taking the no-first-strike policy serious enough that they don’t keep their land-based ICBMs mated to their warheads, though. I know there are reasons (population) that’s almost uniquely an option for them, but still it’s pretty cool they put their money where their mouth is on something that important to their survival.
posted by Ryvar at 10:51 PM on September 17 [16 favorites]


Metafilter: 118 APIs for a similar number of half-vetted libraries
posted by armoir from antproof case at 11:10 PM on September 17 [1 favorite]


Despite how fucked we are, that is not yet the case for China or Russia so even on just purely abstract ethical grounds I support us disarming last

If ethics as a field of human inquiry is to have any purpose at all, it should be exposing and refuting the genocidal logic behind the idea of incinerating a few hundred million people for the crime of living under an allegedly shittier political system.
posted by derrinyet at 11:57 PM on September 17 [13 favorites]


I don't guess Biden is reading this. Just in case he is "Hey, Pres -- Don't be a dope. Don't replace old junk with new junk which will be old junk FAST. Build more subs, more planes, each sub and plane loaded up with MIRVs. Problem solved. Don't be a dumbo."

The impression I got was that upgrading all that stuff (the other two "legs of the triad") would be $900 billion of the $1 trillion they're asking for, while this is just $100 billion of the $1 trillion.

Just $100 billion to destroy most of the life on our planet, what's not to love?
posted by clawsoon at 12:10 AM on September 18 [4 favorites]


If China for whatever reason decides to nuke the US, the ethical outcome is absolutely not a retaliatory nuking of China. To put it in perspective, if you are a person of European descent you are alive because indigenous inhabitants of the Americas did not have the ability (or presumably willingness) to press a button to wipe out an equivalent 95% of Europeans with deadly biological agents in the sixteenth century.
posted by derrinyet at 12:32 AM on September 18 [3 favorites]


There is no situation where it is ethical to launch nukes in anger, regardless of who you are. If you're nuked to fuck, blowing the rest of the world up too isn't going to help you any at all. At that point it's already too late for you, why take everyone else with you?
posted by Dysk at 1:54 AM on September 18 [3 favorites]


The tip-toeing round Russia, even as it invades its neighbours, is surely the best argument for nuclear weapons ever. But they don't have to be very modern: surely better spend the money on cool weapons to help our little allies who are not under the nuclear umbrella (Taiwan, Ukraine) to be too costly to invade? If Taiwan could tie up the Chinese military for years that would be a good result (for the USA and its allies: the best result would be China not invading, of course...)
posted by one more day at 1:55 AM on September 18 [3 favorites]


Around forgetting, we forgot how Fogbank works and completely reinvent it as part of warhead maintenance. lol

Around humans semi-loosing technology, we no longer make machines that do 200nm chips, which contributes to the electronics shortages.

Around nuclear war..

Owen Toon and policy wonks have always exaggerated nuclear winter.

Yes, we should avert nuclear war between the U.S. and Russia, who both have 1000s, but humanity survives relatively okay if the U.S. and China nuke one another, mostly because the U.S. must save most of their arsenal for Russia. Assuming functional countermeasures on both sides, we're talking fewer detonations than happened during atmospheric nuclear testing, and far less acreage burned than in our climate change driven wildfires now anyways, aka not enough for nuclear winter.

There exist a risk of nuclear summer aka wrecking the ozone layer for 15 years, but Anderson says climate change shall create a wet stratosphere that destroys the ozone layer, presumably for way more than 15 years.

In other words, if it came down to a choice then nuclear war is preferable to climate change. In particular, one could imagine China vs U.S. fights over oil drive both into nuking only-ish the refineries feeding one another and their allies. I'm afraid this does not prevent burning crude oil in ships, and worse coal & some gas continue burning, but it'd ends cars & planes, slows down coal mining, etc.
posted by jeffburdges at 2:25 AM on September 18 [6 favorites]


If it really is this hard to maintain the US arsenal, how well maintained do we think the russian arsenal is?
posted by DreamerFi at 2:46 AM on September 18 [26 favorites]


Anderson's Harvard page has another more current explanation of ozone destruction caused by climate change, which suggests climate driven "nuclear summer" begins developing now. I'd never heard the starting now part before.. It likely takes a while but water vapor must continue increasing even after net zero.. sounds bad..
posted by jeffburdges at 3:46 AM on September 18 [2 favorites]


There is no situation where it is ethical to launch nukes in anger, regardless of who you are. If you're nuked to fuck, blowing the rest of the world up too isn't going to help you any at all. At that point it's already too late for you, why take everyone else with you?

This is entirely true. However, there's a big problem with this idea. Let's call this the 'no strike' option, where you will not fire nuclear weapons if attacked (obviously you'd normally just dismantle them, but maybe you're keeping them to fend off extinction level asteroids or something). Where first strike is you might shoot first, and second strike is when you hit back with nukes after having been hit by a first strike.

Assume everyone but one country has no nukes, or a no strike policy, that country has a HUGE incentive to use them in any conflict, because it's such a force multiplier. As the USA demonstrated on Japan. Imagine the current world, but all western nuclear powers had a no-strike policy - how long do we think Putin could resist nuking Kyiv?

So once *anybody* has nukes, everybody else who doesn't want to risk getting hit by a first strike can either be too small or far away to be worth attacking, or have their own 2nd strike capability, and say that they would use it if attacked (or a trustworthy alliance with someone who'd use them on your behalf). Which can also deter conventional attacks, or course. Which is also why e.g. North Korea pursued them. Regardless of whether you actually would use them - and it's entirely possible some or many nuclear powers do in effect have a no strike policy in secret - you have to be credible that you would use them at least for defence to prevent somone risking a first strike against you.

Of course, you can still give up on nuclear weapons and become a non nuclear armed state - unilateral disarmament. Ukraine did this in the early 90's, when they gave up their 1/3 of the soviet nukes they had after independence - in exchange for security guarantees from Russia and the West. Which of course turned out to be worthless in the end. I cannot see any major nuclear power doing so in the forseeable future.

The only realistic way out of this terrifying standoff is for all nuclear powers to have a long-standing public and credible no-first-strikes policy. At which point, some states can go further to no strike (or just disarm) because they no longer need to deter a first strike, and we might eventually end up in a world that is no longer MAD.
posted by Absolutely No You-Know-What at 4:42 AM on September 18 [14 favorites]


For years, analysts said it was actually the bomber based 'leg' of the nuclear triad (which consists of bombers, missiles, and submarines) that was actually the obsolete one.

But the bombers at least can serve a dual role, and given that we're probably going to have heavy bomber aircraft for the conventional mission, it seems a bit odd to not use them, particularly since a manned bomber is (potentially) recallable and re-targetable in a way that an ICBM isn't.

So the need for the ICBM fleet looks a bit suspect these days.

The only conceivable argument I can come up with is survivability: the silo-based missiles are very hard targets, and would require a massive first strike to take off the table. If you replace them in that role entirely with submarines (as the UK and some other states have done, and is occasionally proposed), you have to consider the low-probability but very high-consequence scenario that an adversary makes some sort of significant 'leapfrog' advance in naval technology and comes up with a way of destroying the entire SLBM fleet at once (e.g. very high-speed nuclear torpedos), leaving the US without a viable retaliatory-strike capability.

That's bad, if you don't like the idea of nuclear war, because having a viable second-strike option is considered stabilizing on nuclear war decision-making. If you have a second-strike option, and you see what you think is a massive first strike coming your way, you can sit tight and hope it's a computer glitch, secure in the knowledge that if it really is The Big One, that you'll be able to reach up from the grave and take them with you. But if you don't have that capability, then you get into the War Games scenario, where you might feel you have to "use it or lose it", or—more realistically—maintain a dangerous state of hair-trigger readiness, with armed bombers in the air or on the flightline at all times, etc., with increased risk of accidents.

I'm not sure that there exists in the unclassified world enough information to make the "are ICBMs really worthwhile Y/N" decision, although I think if I was President tomorrow, I'd want to take a real hard look at it, and in particular whether it would make more sense to put all that money towards submarine-force modernization to prevent an adversary from gaining a first-strike advantage there. That such an option never seems to be on the table reeks of inter-service rivalries rather than real strategic calculus, and that's a real stupid reason to spend money when you're counting by the billion.

But hey, if we're going to have ICBMs, as is apparently the case, I'd like those suckers to have the absolute minimum viable level of technology necessary to get the job done. Build a missile that's designed to sit there in its silo for decades at a time, do its job if necessary, but otherwise not cost a bunch of money and not use any dumbass proprietary parts or software. Ideally, 50 years from now nobody will be getting out their government credit card while shopping on eBay (a BezosMusk Family Company™) for Intel i7 processors or 10GbE Ethernet cards just because they were hot shit in the 2020s. The C2 networks for such things probably need to be continually upgraded, but assuming there's always going to be a couple of people sitting in a bunker who have to turn a key and press a button (as there should be), the weapons themselves ought to be air-gapped, entirely isolated systems. If they can be operated from paper punchcards, run them from punchcards.

if you are a person of European descent you are alive because indigenous inhabitants of the Americas did not have the ability (or presumably willingness) to press a button to wipe out an equivalent 95% of Europeans

I have no idea where you get the "presumed willingness" in that statement, and it seems ridiculous on its face. There's no reason to assume that the great pre-Columbian civilizations of the Americas would have just shrugged and gone peacefully into the long night, if they'd had the ability to stave off the Great Dying, any more than any other civilization in a similar position would. The idea that Europeans were/are particularly ruthless, or especially unique at all, is just an inversion of the same old "noble savage" narrative.

However, there is something to be learned from the pre-Columbian American civilizations' collapse that bears on strategic planning: the Aztecs, Inca, Muisca, Quilmes, Mississippians, and all the other dominant cultures of the Americas c. 1491 probably thought they had their shit together in terms of national defense. What threats they understood and could foresee, they defended against vigorously, and to that end they had militaries and tactics, defensive fortifications, food reserves, command/control systems for raising and fielding armies, strategic partnerships and alliances, etc. etc. It did not, in the end, greatly help except perhaps at the very margins. Smallpox burned through the most well-fortified cities regardless. Call it a 'Black Swan event', or an Outside Context Problem, or whatever, but the point is that you can only defend against threats that you're capable of envisioning (and think are realistic enough to warrant the defensive expense).

It seems probable to me, just looking at what has brought down other dominant civilizations, that what will end us won't be the things we have carefully planned and wargamed and spent endless amounts of resources defending against, but something that we don't foresee and thus aren't able to adequately react to or cope with.
posted by Kadin2048 at 4:55 AM on September 18 [30 favorites]


The idea that Europeans were/are particularly ruthless, or especially unique at all, is just an inversion of the same old "noble savage" narrative.

Societies have waxed and waned in their levels of ruthlessness and violence and genocidal intentions. I think you can safely lump the military and leadership classes of Europe in the 1450-1950 half-millennium in with the Neo-Assyrians or the Aztecs or whatever other genocidal assholes of history you'd like to mention. They're not unique in their desire to enslave and enserf and eliminate, but they're one of the peaks.
posted by clawsoon at 5:17 AM on September 18 [4 favorites]


There's something to be said for keeping it as low-tech as it can be. And definitely keeping it as free from IP-related barriers to modification/repair as possible. There's some value in avoiding the issue of submarines running Windows XP.

Kadin2048: what will end us won't be the things we have carefully planned and wargamed and spent endless amounts of resources defending against, but something that we don't foresee

I don't know about the whole civilization, but I strongly suspect we're going to have some cascading collapse of IT/computer/communications infrastructure due to things progressing towards a monoculture of things built on multiple layers of abstraction, mostly made up of different linux boxes. Another contributing factor is that we're ending up with mission-critical software unwittingly ending up sitting on top of fragile software components.

Not as big-badda-boom dramatic as nukes, but still likely to kill a lot of people.
posted by rmd1023 at 5:59 AM on September 18 [1 favorite]


I liked the photo of the board roughly mapping out the clusters of Minutemen and their location near a river, helpful for OSINT to label stuff on satellite imagery.
posted by k3ninho at 6:03 AM on September 18 [2 favorites]


As a technologist, it would be really interesting to work on a system like this if you could pry it away from Lockheed Martin and Boeing and instead do it in house, with a goal of building a extremely robust, maintainable system that would never need a complete rebuild.

With essentially infinite funding, you could spend a billion dollars and build a new fab to make simple, extremely durable silicon. I’m talking about, say, early 80s type chips with transistors around a micron, where the patents have all expired. This could actually be useful all over the place, since there are lots of things that need a little bit of logic but don’t need what a modern processor can give you.

From there, you could build out a whole stack. Call it The Boring Stack. Keep the interfaces at the command line, the people working on missiles don’t need a GUI. Use commodity hardware where it is easily swappable. And on and on. Building such a toolset could actually be useful in all kinds of places, such as industrial control tooling that is intended to last decades.

Unfortunately, instead it’ll be a Lockheed Martin or whatever boondoggle, drop billions of dollars, and end up with something LESS reliable than what is there today. But it would be great if we were going to spend the money, we could also get something useful out the other end.
posted by rockindata at 6:24 AM on September 18 [9 favorites]


I strongly suspect we're going to have some cascading collapse of IT/computer/communications infrastructure due to things progressing towards a monoculture of things built on multiple layers of abstraction, mostly made up of different linux boxes.

See _A Deepness in the Sky_ for a future history where the computing develops that way but doesn't have a cascading collapse (until it does). People tens of thousands of years in the future with starships that run on unix systems so ancient they have software archaeologists who try to figure out old tricks and APIs for competitive advantage.
posted by GCU Sweet and Full of Grace at 6:34 AM on September 18 [4 favorites]


As president, your first interdimensional meeting with the SCP foundation they explain why these silos are needed, and why you don't want to know what's in them.
posted by adept256 at 7:13 AM on September 18 [3 favorites]


… helpful for OSINT to label stuff on satellite imagery.

The map's over 170 km across, and the prominent river is the North Platte. With a bit of fiddling about, I got that photo to georeference to within about 500 meters or so of satellite imagery. Yes, you can pick out some obvious locations from USGS public imagery, but some might be little more than scratches on a hillside
posted by scruss at 7:34 AM on September 18


As a technologist, it would be really interesting to work on a system like this if you could pry it away from Lockheed Martin...

I do BI/data analysis and the other day the admin-person at my dentist expressed their interest in the field. They thought it would be 'interesting'. I told them that I had to spend most of my time thinking about the legal and ethical implications of what we were doing, rather than the coding.

I suppose the 'refit the nukes' engineers are chosen because they don't think too much.
posted by pompomtom at 7:37 AM on September 18


Oh absolutely, there are massive ethical issues with working on technology to support what is a world ending weapon. I am perfectly happy working in my chronically underfunded science agency instead.
At the same time, we are not going to be disarming. What I am terrified of is some shitty touch screen interface built on layers of abstraction leading to someone accidentally hitting launch when they thought they were doing an exercise, or someone replaceing lockouts that are currently physically operated with keys doing it in software instead. It is so very important that these systems work well and are maintainable and understandable by not just today’s humans, but those who are here 20, 40, and 60 years from now.
posted by rockindata at 7:46 AM on September 18 [2 favorites]


I'm not convinced this is a good thing. What's your reasoning?

Simplicity.
posted by Thorzdad at 7:51 AM on September 18 [1 favorite]


If you haven’t listened to the TAL episode about the time in Arkansas when a nuclear missile exploded, you’re in for a treat! It has a little something for everyone. Episode 634: Human Error in Volatile Situations.
posted by amanda at 8:00 AM on September 18 [1 favorite]


Something something Battlestar Galactica

As it happens, many of the computer panels in the 1979 Battlestar Galactica are from the AN/FSQ-7, the first generation of the kind of equipment this story is talking about.
posted by offog at 8:02 AM on September 18 [1 favorite]


I think that maybe 1% of people who write software would be comfortable with placing modern computation on ICBMs. I got a analogue timer to keep my plants alive on vacation. These things just don't need that much computation; ballistic missile calculations are not that complicated. The only worthwhile thing that you could consider adding would be more secure cryptographic verification of arming commands and target commands, and the military places a high value on the ability to do these things in a crisis with minimal time and working tools. There's little point in even adding GPS, because satellite infrastructure can't be on relied to exist and would be a huge external hacking / spoofing vulnerability (except for a first strike!) .

silo-based missiles are very hard targets

This became irrelevant in the 70's / 80's. The accuracy of missiles and reentry systems of that era is good, and small improvements in accuracy negate large improvements in bunker hardening when you're talking about a nuclear strike. The advantage of bunkers is that they make it easier to control and maintain the weapons and a little easier for the US to hide exactly where they are. By spreading out the real missile sites in a 500 sq mile area that is tightly controlled, opponents are forced to dedicate more warheads if they want to guarantee getting most of them in a first strike. This has probably gotten less valuable as satellite imagery and signals intelligence have advanced (but still forces a 1-warhead to 1-missile tradeoff). As a strategy it requires keeping secrets for a long time and perfectly. The USSR / RF instead heavily uses mobile launchers because they have a lot of thinly populated territory to shuffle launchers around in, and a relatively good ability to keep out spies. In contrast, if the US drove launchers around large sections of utah, wyoming, arizona we would have little ability to prevent listening devices from being planted or civilians from disclosing when they see them moving.

bombers

The B2 was a terrifyingly destabilizing weapon. It was designed as a first-strike nuclear bomber that could simultaneously decapitate / destroy USSR nuclear forces without the warning offered by satellite-based launch detection or ground radar detection of missiles. That it was kept secret so well and was too expensive to manufacture in significant numbers probably saved the world.
posted by a robot made out of meat at 8:13 AM on September 18 [2 favorites]


>> silo-based missiles are very hard targets

> This became irrelevant in the 70's / 80's.


Well, I don't know that I would say it's entirely irrelevant.

First, there are nuclear-armed nations in the world who we could potentially label "adversaries" other than the USSR/Russia, and their missiles may not be as accurate. I don't think India or Pakistan have missiles that can hit the US (they're more concerned about each other), but if they did, they might be in the hundreds-of-meters CEP rather than tens. Same with NK, although I don't know how serious a strategic threat NK really is. (They probably get to fire, what, one missile before South Korea gets turned into an island in response?)

Second, the hardening of missile silos isn't to defend against a direct hit by a nuclear weapon: it's to protect them against a nearby strike. In other words, they're made hard enough to keep the enemy from wiping out an entire field of missiles with one warhead. Having to dispatch and individually target one warhead per silo is the point. Doing so, particularly if you have an arms treaty that caps the total number of missiles, "soaks up" a lot of enemy warheads incinerating North Dakota rather than other counterforce or countervalue targets. At least, such is the theory.

The last time the US went down the ICBM-replacement path, we ended up with the Peacekeeper, which was withdrawn in the early 00s for reasons that I've never quite understood but seem to be maintenance-related. (But we got a nice satellite-launch system out of it.) But it's fair to say that the US has already tried a 'forklift upgrade' of its ICBM force once—and failed.

The alternative to Peacekeeper, which for some reason wasn't ever pursued, was the MGM-134 "Midgetman", which frankly seems like a much better design than a silo-based missile altogether. It's also what most other nations seem to be pursuing these days, rather than silos. (Except for the Russians, who have the same boner for silo-based missiles that the US does.)

Basically you build a whole bunch of relatively inexpensive, single-warhead (or single-warhead-plus-decoy) missiles on road-mobile launchers (TELs). Most of the time they live indoors, with their crews sitting around watching Netflix in the same building. If things get ugly, the crews hop in and drive them out a few miles to one of a dozen or so prepared launch sites, with the actual site chosen at random, moving around periodically. This gives you the same advantage of silos (each launch site "soaks up" one enemy warhead, since they don't know which one will be used) but at much lower cost. The missiles are also much easier to maintain and swap over time, since they're not in silos. Not that the DoD would do it, but you can buy the TELs basically off-the-shelf from any number of manufacturers. And the launch system is basically identical to the way a SLBM is launched ("cold start" with a gas generator popping the missile out of a tube-like container, rather than a "hot start" with the rocket engines igniting in a silo, like the old-school heavy ICBMs). In a rational defense program, the same missiles, plus or minus a boost stage or an additional warhead, would probably be used both on road-mobile and submarine launch platforms.

The B2 was a terrifyingly destabilizing weapon.

Yes—what's more, it's actually a terrifyingly destabilizing weapon in a long history of terrifyingly destabilizing weapons the US has fielded.

Like, scaring the shit out of the Soviets was basically our thing, over and over. I'm not sure if it was really something the US did intentionally, but we certainly did it.

Soviet missiles with big warheads and relatively low accuracy are inherently countervalue, really only useful as I'll-see-you-in-hell city-killer check-out-my-nuclear-dick retaliation weapons. The US's emphasis on accurate weapons with low yields is much scarier, from a planning perspective. Highly accurate, low-yield, MIRVed SLBMs are a really ideal first strike vehicle. As are cruise missiles, both land-based like the GLCM and air-launched like the ACM and ALCM. Particularly in the case of a stealthy cruise missile like the ACM, which can be launched in large numbers from a stealthy platform like the B-2. In theory, with the right combination of cruise missiles and SLBMs, plus our advantage in submarine warfare (SOSUS, Seawolf, etc.) to take out Soviet missile boats, the US could have credibly conducted a first strike / counterforce decapitation attack against the Soviets throughout most of the 1980s, if we'd wanted to.

One of the weirdest things about looking at the Cold War in retrospect is the degree to which each side thought the other side desperately wanted to attack them. The Soviets were (at least in some cases) really pants-shittingly concerned about a US first strike. The US was similarly pants-shittingly afraid of a massed mechanized invasion of Western Europe. Neither one seems to have been much desired by the other.
posted by Kadin2048 at 9:05 AM on September 18 [10 favorites]


At that point it's already too late for you, why take everyone else with you?

Purity of essence, Mandrake.
posted by They sucked his brains out! at 9:19 AM on September 18 [10 favorites]


If you're nuked to fuck, blowing the rest of the world up too isn't going to help you any at all.

to adherents of MAD, making the other guy think you'll incinerate them no matter the outcome is a critical bit of the strategy.

perennial rec for Command and Control (book).
posted by j_curiouser at 9:43 AM on September 18 [5 favorites]


I think that maybe 1% of people who write software would be comfortable with placing modern computation on ICBMs

But how are we going to get them IoT-enabled if we don’t?
posted by atoxyl at 10:48 AM on September 18 [1 favorite]


I want to be able to use my phone to track my ICBMs.
posted by atoxyl at 10:50 AM on September 18 [4 favorites]


"If we combine BEM conventions with some neat flexbox tricks we can actually re-order all the MIRV target tracks dynamically, without even having to listen to their JS events."
posted by protorp at 11:01 AM on September 18 [4 favorites]


The alternative to Peacekeeper, which for some reason wasn't ever pursued, was the MGM-134 "Midgetman"

By the time it had a successful test flight, the cold war was nearly over. There was no longer anyone who could launch a plausible successful first strike against the US. Very reasonable (cost wise) to ride out the useful life of the existing systems.

First, there are nuclear-armed nations in the world who we could potentially label "adversaries" other than the USSR/Russia, and their missiles may not be as accurate.

None of them had (or have) the number of weapons or geopolitical motivation to launch a first strike. Plus, if they wanted to, upgrading to 80's level of accuracy over time would be possible. I don't know that, but it seems pretty likely. It was clear that over reasonable timescale an adversary interested in having first strike capabilities should be assumed to adapt to overcome hardened silos.

Second, the hardening of missile silos isn't to defend against a direct hit by a nuclear weapon: it's to protect them against a nearby strike.

Sure, as I said it forces a 1-1 tradeoff (but worse if you rely on high-MIRV count silo missiles, as it is a 1-warhead to 1-silo trade, part of why current posture has 1 warhead per missile) with incremental gains in required CEP for pretty large outlays in silo construction. As you agree, it's not a huge winner compared to alternatives. I haven't paid attention to exactly what alternatives to the LGM-35 were considered, but the AF seemed to think that small launchers didn't meet their survivability needs. This probably depends heavily on the warning-time assumptions; with 10 minutes of warning a road-based launcher can only get a modest distance in a random direction and in a shelter. With 20 minutes it's much more favorable. This assumes that you aren't launch-on-warning.
posted by a robot made out of meat at 11:01 AM on September 18 [1 favorite]


I don't know about the whole civilization, but I strongly suspect we're going to have some cascading collapse of IT/computer/communications infrastructure due to things progressing towards a monoculture of things built on multiple layers of abstraction, mostly made up of different linux boxes

I’m more concerned about supply chain attacks like Solar Winds. Something like that getting into a nuke system … (shivers)
posted by panama joe at 11:02 AM on September 18


Bluetooth: On

Turn Bluetooth Off
-----------------------
Devices:
AF1 speakers ▸
Brandon's Laptop ▸
Scramble Fighters ▸
ICBMs ▸
HAARP ▸
Dominion Systems ▸
Joe's Secret iPhone ▸
---------------------------------
Open Bluetooth Preferences...
posted by NoThisIsPatrick at 11:14 AM on September 18 [6 favorites]


There is no situation where it is ethical to launch nukes in anger, regardless of who you are. If you're nuked to fuck, blowing the rest of the world up too isn't going to help you any at all. At that point it's already too late for you, why take everyone else with you?

No, but given nuclear armed adversaries it is also unethical to let them believe that you won't use yours in anger. The most ethical position, in a world where nuclear weapons exist, is to lie and claim you'll fuck anyone up who attacks you with nukes but secretly have decided that you'll just let it go.

If there isn't a reason not to use nukes, somebody is gonna use them.

Funny thing is that we weren't that far away from a world in which an international body had control of the nukes as a deterrent force against someone secretly developing and using them rather than leaving it to the whims of individual nations. It was seriously considered by some very powerful people just after WWII. Congress said to fuck right off with that.

I don't know that it would have actually worked as envisioned, but it's an interesting alternate history to contemplate.
posted by wierdo at 12:09 PM on September 18 [6 favorites]




I don't really understand the problems described above for the "no strike" premise. We can commit to never using a nuclear weapon and still cause massive damage to another country, in particular their leadership. I sort of thought the point of even Biden and Obama still using drones to target individuals is basically to keep escalating demonstrations of the precision and intelligence of the US military. (While demonstrating the extent of US racism in that the whole country isn't appalled that we just killed actual human beings).

Russia drops a nuclear weapon, and we target all of the Russian leadership with drones, target key transportation and supply points with conventional weapons, etc. The balance doesn't need to be how many people you can kill in a single blow.

But, then again, I perhaps have a higher faith in humanity than is justified.
posted by lab.beetle at 3:28 PM on September 18


goal of building a extremely robust, maintainable system that would never need a complete rebuild
The similarity between such an ICBM silo and The Clock of the Long Now is uncanny.
posted by joeyh at 4:17 PM on September 18 [1 favorite]


The only realistic way out of this terrifying standoff is for all nuclear powers to have a long-standing public and credible no-first-strikes policy. At which point, some states can go further to no strike (or just disarm) because they no longer need to deter a first strike

Well the problem is that war is war. There might be a public "no strike" policy, they may even believe their own policy. But when someone believes they may be in an existential crisis, they do what they think they have to do.

I mean, I might have a theoretical policy that "no crime someone could commit against me deserves the death penalty." But if I've got a handgun in the glove compartment, I might waver on that when the time comes.
posted by ctmf at 4:45 PM on September 18


in particular whether it would make more sense to put all that money towards submarine-force modernization

It's funny reading the article to me, since the Navy does not seem to be having many of these obsolescence issues that the Air Force is. Columbia's IFI is right around the corner, there's modern fire control systems in place. I think there's a modernization plan underway for the actual missiles, but all the hardware around it is pretty new or will be soon.

But I agree, the SLBM leg of the triad is carrying a lot of weight. There are a lot of true believers in the deterrent that we'd be having an off day if we targeted your house and hit your neighbor instead. The big issue right now is "onshoring" electronics productions; we're desperately short of producers that can make rad-hard components.
posted by backseatpilot at 5:33 PM on September 18 [1 favorite]


We can commit to never using a nuclear weapon and still cause massive damage to another country, in particular their leadership.

It's a lot harder, if not impossible, to credibly claim that you'll be in any position to retaliate after a sufficiently large nuclear first strike if you don't have hardened/hidden delivery systems that can be launched during/after that attack. That's much easier to accomplish, and thus easier to convincingly claim as a deterrent, with big freaking nukes mounted on missiles than it is with something like a drone. Drones rely on a bunch of infrastructure that may not be working or entirely gone if the nukes start flying.

To pick the most obvious ones, even if the satellites themselves aren't attacked and a sufficient level of ground-based infrastructure survives to maintain operations, both GPS and satellite communications are likely to be rendered useless for at least many hours by the electromagnetic interference caused by nuclear detonations. You can't control a drone if your comms are down. It won't know where it is without GPS. You won't know where the adversary's leaders are for your decapitation strike without being able to communicate with your spy satellites.

Plus, decapitation strikes are a bad idea. Once you do that there's nobody left to call off the dogs. Nuclear war planners, especially those who insist that it's possible to win such a war may be more than a little oblivious to the reality of what they're definition of "winning" actually means (oh, yay, there's one person left to plant a flag on top of the ash heap before they too drop dead!), but they are by no means stupid. There's over 70 years of intense thought behind deterrence theory.

Everyone agrees that the situation is terrible and there are less existentially fraught ways of maintaining the balance of power between the large nations who are capable of causing global misery with or without nuclear weapons. The problem is getting from here to there without upsetting the apple cart. The planet sized apple cart that could very well crush most or all of us if it is somehow overturned.
posted by wierdo at 6:49 PM on September 18 [3 favorites]




I want to be able to use my phone to track my ICBMs.

I think I’d prefer it if Foone figured out how to play Doom on the warheads but that might create other interesting issues.
posted by ensign_ricky at 7:46 PM on September 18 [1 favorite]


It's funny reading the article to me, since the Navy does not seem to be having many of these obsolescence issues that the Air Force is.

Trident 2 was close to a clean-sheet design from the late 80s, Minuteman 3 is a modification of a 1950s design. Likewise, I assume that Minuteman launch complexes are less easily modified than submarines since you can't drive one into drydock.

I'd also suspect that for institutional reasons the navy is better about buying long-term stocks of replacement stuff.
posted by GCU Sweet and Full of Grace at 3:00 AM on September 19


To be clear, the point of nuclear retaliation (second strike) is ostensibly *not* to spit one’s last breath “for hate’s sake.” Instead, the idea is to inflict so much damage upon an enemy that they cannot exploit a successful first strike. “Oh, you’ve taken out most of our military? Nicely done - but you no longer have much of your *own* military, or the infrastructure it relies upon, so you’ve not really improved your own national security at all. Probably best not to bother, eh?”
posted by Mr. Excellent at 3:37 AM on September 19 [2 favorites]


Ugh, I hate all of this so much. Humans need to evolve. What a colossal waste of money/resources/thought/and potentially life. (Yeah, yeah, yeah, this is the world we live in, blah blah blah.)
posted by nikoniko at 1:40 PM on September 19 [1 favorite]


The obvious solution is lots of signs that says NO FIGHTING IN THE WAR ROOM
posted by Jacen at 1:30 AM on September 20 [2 favorites]


⛏️⛰️✋🤚
posted by clavdivs at 2:46 PM on September 20


~ "Nuclear weapons can wipe out life on Earth, if used properly." - David Byrne

~ I grew up in place where Minuteman silos had been built around the time that I was born but then abandoned after just a few years. There were six or so scattered outside of town, all on private farmland. In high school and college we would visit them in the night; there were 2-3 that were accessible. Extremely unsafe with rusted platforms and stairs. But for an 18 year old male, extremely cool.
posted by neuron at 12:39 PM on September 21


« Older Moderation Is Censorship, Sayeth The Fifth Circuit   |   An Extended Erudite Discussion of Campus Comedy in... Newer »


You are not currently logged in. Log in or create a new account to post comments.