> 1) The reported starts (5 in total) happened from a single base in the US which didn't make sense from the military point of view: the US would have put themselves at a grave disadvantage by sending just a few missiles

There was an interview with Petrov (in the 90s or 00s) when he spoke about this point, and the interviewer said that in 1983, the US plans for a preemptive strike started with a "beheading strike", in which a few missiles would be sent to hit the Kremlin and a handful strategic military bases. Then the US would wait if the Soviets were still twitching, in which case the second, more massive round would've been fired. Petrov admitted that, if he had known that, he would've decided differently on that fateful day.

(Don't have a direct source link to back this up, sorry. My source is one of the earlier episodes of https://alternativlos.org where they talked about the Cold War.)

One would think proper first strikes would be compromised of SLBMs.

Then again, it's not a proper Cold War strategic game theory play unless you're banking on your enemy to hesitate due to the unusual nature of the attack. SLBMs scream first strike, while a few birds out of many launching from farm country in the central United States in sequential fashion, not so much.

That said, I'm not sure what would have consituted "twitching" after Soviet C&C is wiped out, and if I was a planner at the time I don't think I'd want to find out. It would likely have been prudent to immediately follow up the decapitation strike with complete annihilation of Soviet nuclear retaliatory capability. Every deployed Soviet boomer accounted for and shadowded, with firing solutions computed. Every missile silo targeted. Presumably strategic air bases would be targeted in the first volley, though there would remain the possibility of strategic bomber patrols already airborne.

First strikes are really messy and entail a ridiculous amount of risk. Considering the stakes, it's no wonder neither side ever tried.

s/compromised/comprised/

s/comprised/composed/

Interesting, I never knew that grammatical holy war existed. Learn something new every day.

The original mistake was just an autocorrect error. Some day I'll learn not to write posts from mobile devices.

Also, I noticed you created an account just for this. Assuming it isn't a throwaway, welcome to HN. That's a cool username.

Whilst "comprised" is a bit clunky, it's also not wrong in the slightest.

https://www.merriam-webster.com/dictionary/comprise

"The factory was to be a vast installation, comprising fifty buildings" is the same kind of use.

FWIW, I assumed autocorrect and wasn't trying to start any wars. :)

No offense taken, I even upvoted both corrections.

This plan strikes me as idiotic TBH. The Kremlin is situated smack dab in the center of the most populous city in Russia, Moscow. You can’t take out just the Kremlin without taking out millions of people in the process. And even if you could, the USSR would have no choice but to respond. It’s like someone nuked the White House. Rockets would be flying within minutes.

Nowadays there’s an automated system in place that ensures a response even if the entire high command is taken out: https://en.m.wikipedia.org/wiki/Dead_Hand_(nuclear_war)

Well... to that page's credit somewhere halfway down it does mention that the existence of a dead hand is only a hypothesis. I suppose it's just one more onto the pile of things from Dr. Strangelove that people interpret as absolute fact.

Of course it’s a “hypothesis”. No one wants to drink polonium tea after confirming it.

The whole point of a doomsday device is lost if you keep it a secret! But, a little bit of polonium in your tea is nothing compared to the international conspiracy to sap and impurify all of our precious bodily fluids.

Not really. It’s like Israel’s nuclear arsenal. Not confirming its existence allows Israel to stay out of the various treaties, but is there any doubt that they have it?

> started with a "beheading strike"

Wow, so given Petrov's response, this plan might have actually worked. Those clever generals. (Insert emoticon of me being horrified right now..)

It's difficult to imagine Soviet SSBN commanders realizing their country had been nuked and thinking "If only my superior officer was around to give me the order to strike back!"

There isn't really a workable first strike plan, and that's probably a good thing.

>Don't have a direct source link to back this up, sorry.

So why you're spreading FUD, if you have nothing to prove your point?

Yesterday I read an article about this man, and I would say that the year'83 was not actually without (possible) escalations. On the 1st of September, a Korean airplane with 200+ civilians was shot because an "ordinary" officer simply sent a report that a "jumbo jet" was flying over soviet area - and it was "probably a US spying airplane" (that was the idea). Now, if this guy had mentioned "it's a civil airplane, a boeing", I am quite sure things would have been different.

The reason Petrov is to be considered a hero of old times is not just because he was able to spot a mistake, but to report something different from what he saw. He took the liberty to decide for an entire nation, and as he also admitted, the 12-13 minutes during which these potential nukes were supposed to reach their final destination were "really long". If he had been so certain, he'd have had no issues - he based his decision on his instinct.

If he had simply followed the rules (pretty common in a military hierarchy), we'd probably not be here today. This is what makes this man exceptional - the fact that in the end he even got scolded for not obeying the rules, and that he died "almost" in misery.

I wonder if there are other, still classified instances where the weapon has almost been launched. If lt. colonel Petrov was aware of the limitations of the early warning system then there must have been other faults as well - I bet there have been problems with the early warning system on both sides...

I recently read the non-fiction book Command And Control by Eric Schlosser. It's structured around an incident when a Titan II missile blew up in its silo in Arkansas, throwing the warhead out of the silo and into the air.

Anyway, it uses that incident to go back in history and explain the development of nuclear weapons and how the command and control systems we have today came about. A lot of it was driven to prevent all sorts of near-miss accidents that have happened in the past. We've been really, really lucky.

Highly recommend the book, especially for the HN crowd. It's good prose, but gets deep enough into the tech and system details that you come away with a solid understanding of how it all works.

This was made into a documentary and is currently available on Netflix. If you can spare 90 minutes check it out, it's a hell of a good watch.

https://www.netflix.com/title/80107656

The Okinawa missiles of October was a story that looked like a close call (to me at least!). It was on HN couple of years ago

https://news.ycombinator.com/item?id=10452983

There was an incident in 1979 where a US military center thought Russia was firing.

Ever hear of the bomb the US accidentally dropped on the mainland? When they recovered it, they discovered the bomb had armed itself and that all the safety mechanisms failed except one.

What the hell?! How many times are we going to be lucky before something goes very wrong. I hope they are a lot more careful now, and I mean all sides.

That's why redundant systems are good. It worked as designed.

Ignoring the warning for 1) and 3) would have been a mistake in a Dr Strangelove scenario.