It is Petrov Day again, partially thanks to Stanislaw Petrov.

"Today is September 26th, Petrov Day, celebrated to honor the deed of Stanislav Yevgrafovich Petrov on September 26th, 1983.  Wherever you are, whatever you're doing, take a minute to not destroy the world."

http://lesswrong.com/lw/jq/926_is_petrov_day/


New to LessWrong?

New Comment
25 comments, sorted by Click to highlight new comments since: Today at 1:38 AM

Note that Petrov suffered very few consequences after the initial backlash. His US counterpart Harold Hering was discharged from the Air Force, drove a truck for a while to make ends meet and watched his personal life crumble, for asking the question

How can I know that an order I receive to launch my missiles came from a sane president?"

I don't know if his stand forced any changes in the missile launch protocols, but I admire his courage to take his oath literally and seriously and his refusal to back down under pressure from his superiors more than Petrov's 5 min of agonizing over a decision to disobey a faulty computer algorithm he himself helped design.

Thanks for the link to Harold Hering's article. I just read it, as well as the one on Stanislav Petrov.

Vasili Arkhipov's Wikipedia article is also worth reading. Although most Russian nuclear submarines required only the captain's order to launch, he was the only one of the three officers on his sub to vote against launching their nukes during the Cuban Missile Crisis. Due to his position as Flotilla commander, he was able to win the argument with the submarine's captain, who wanted to launch.

I wonder how many other near misses we know nothing about.

Thanks for the Hering link.

I, too, got into a dispute with the USAF but did considerably better and there was a lot less at stake. True to form for whistleblowers, a Lt. Col. who tried to help me on a related issue got sent to Taiwan. The punishment is worse for higher-ups because they should have known better by then.

Decades later, in my response to questioning how to sue a government agency for negligence, a lawyer told me "No one can sue the King." Questioning a King's sanity, or competence, may be worse than suing.

a lawyer told me "No one can sue the King."

Yes, this is called sovereign immunity.

As far as I know, you can sue a government agency, with some restrictions, but you can rarely sue the people who made the decisions in question while working for said agency. It's worse if the agency is a military branch and the normal civilian safeguards do not apply.

I am glad to now know about Harold Hering, and I think I'm going to add his story to the reading we're having tonight in NYC.

That said - there are a great many people I admire more than Petrov. Petrov is significant not because he's an amazing person, but because he literally was the person who saved the world, for good or for ill.

Note that Petrov suffered very few consequences after the initial backlash. His US counterpart Harold Hering was discharged from the Air Force,

Hering is not Petrov's counterpart.

Petrov, basically, said "This technical system is generating an erroneous result, I'm not going to accept it". Hering said "I will obey an order from a superior officer only if I decide it makes sense".

Hering said "I will obey an order from a superior officer only if I decide it makes sense".

Which, of course, is the oath he swore and what we hold our enemies responsible for.

Which, of course, is the oath he swore

No, I don't think it's the oath he swore.

I honestly can't tell what you're saying here. What do you think is the oath he swore?

Either:

A) An officer's solemn duty to determine whether or not their orders are supporting and defending the Constitution, possibly against domestic enemies, can be rounded off to "I will obey an order from a superior officer only if I decide it makes sense"

or

B) It can't.

In the first case, my use of your language is appropriate. In the second case, your earlier comment is the source of the error, and I was simply not critical enough.

But maybe you think that an officer swears something like the following:

I will obey the orders of the President of the United States and the orders of the officers appointed over me

Which: No! They don't! That's part of the Oath of Enlistment but not the Oath of Office, which the Wikipedia page points out is missing that element. And Hering, as a Major), was an officer.

I think that the B) case is pretty obvious. In particular, because "makes sense" is a MUCH wider criterion than "supporting and defending the Constitution".

In the specific case of Hering, he was was worried about the President being (temporarily?) insane. Given that he was in no position to make a mental health diagnosis, his position essentially boiled down to saying that his perception of the situation overrides his (presumably) direct orders -- and that's not because the President suddenly became a domestic enemy and a threat to the Constitution.

In particular, because "makes sense" is a MUCH wider criterion than "supporting and defending the Constitution".

Then.... why did you say it?

Because I don't think that Hering's preemptive grab of authority ("I will use or not use the nukes depending of what I think is right regardless of what my orders say") falls under "supporting and defending the Constitution". I think it falls under "makes sense to me".

I don't see where your "preemptive grab of authority" characterization is coming from. Officers are not obligated to follow illegal orders; the established doctrine is that they will be held liable if they do so--that they were simply following orders is no defense. That is, they are only supposed to follow orders that "make sense" to them (in a legal and moral, not strategic, sense).

Yes, it's correct that Petrov's resistance was technical and Hering's resistance was moral, and we / the government may have different opinions on how to react to technical or moral resistance. But my point is that Hering is reacting to an inconsistency in the government's approach (simultaneously binding its officers to defy and not to defy), not adding a novel inconsistency on his own.

If you've got a group around and some candles, consider going through the Petrov Day ritual. We'll be doing it in Boston for the third year, and in past years it was quite excellent.

I haven't seen a convincing closure of the debate about whether Petrov was actually the only serious obstacle to WWIII. Obviously he did the right thing, I'm not trying to take that away from him, but I'd be very interested in hearing a more thorough analysis of what would have happened if he had reported up the chain of command that the computer said the US had launched ONE (EDIT: LATER 5) missile(s) but that since one missile launch was a pretty stupid thing to do, it was probably an error.

I haven't seen a convincing closure of the debate about whether Petrov was actually the only serious obstacle to WWIII.

In general, it's good to reward people from stopping a mistake even if they are not the last possible place for it to be caught. Defense in depth!

Yeltsin decided not to retaliate in 1995, during another false alarm. It is not clear Andropov would have acted the same way, given the very different levels of tension in 1995 and 1983.

But I would like to know whether he averted a 1% chance of disaster or a 50% chance or an 80% chance.

I mean would the leadership launch in response to being told that ONE missile had been fired? Were they complete retards?

Were they complete retards?

Looking at the relevant US experience, very likely.

Still, having an insecure PAL code is not the same as pressing what is effectively a 20-minute delayed suicide button based upon the most obvious of false alarms.

Note that the Americans did have the insecure code and the Russians did not press the button X-D

Sure, for what he did he deserves billions of dollars on prize money. The expected utility loss conditional on him making the decision to report the alert must be trillions of dollars.

US had launched ONE missile

It's was five missles.

Does that make a difference? As far as I am aware there is no strategic use for 5 nukes as an opening move...