Hacker News new | past | comments | ask | show | jobs | submit login
Unsung Hero of the Nuclear Age (slate.com)
79 points by _pius on Feb 8, 2014 | hide | past | favorite | 36 comments



I'm a bit confused by one thing in this story: it never mentions the two man rule:

http://en.wikipedia.org/wiki/Two-man_rule

The President cannot order the use of nuclear weapons on his own; he can only issue the order jointly with the Secretary of Defense. The article mentions that Nixon's SecDef asked people to "check with him" before carrying out orders from Nixon, which may be a sort of garbled reference to the two man rule, but if so it's very garbled.

Whether this rule actually answers Maj. Hering's question is a separate issue. But I find it disappointing (though unfortunately not surprising--journalists often get things like this wrong) that the article repeatedly talks as though a single person can issue the order, when that's not the case.


I'm not sure where that fits in either, but the point I took away from the story is; Don't think too hard about Nixon. It was clear or at least not as unclear, that he was not in his best mental condition. One of the problems is, what about the current head of US state? How would you know if he/she isn't in his best condition?

Regarding the two-man rule, my idea of the "Two Person Concept" is to prevent that a single operator would launch an attack by himself or in "error" (whatever that constitutes, I do not know). I interpret Major Hering's question as a separate issue from the "Two Person Concept".


what about the current head of US state? How would you know if he/she isn't in his best condition?

But that is, in fact, one of the reasons for the two man rule: so that just one person being insane isn't enough. Two people would have to be insane: the President and the Secretary of Defense. That at least provides some reduction of risk.

I interpret Major Hering's question as a separate issue from the "Two Person Concept".

At levels below the National Command Authority (NCA, i.e., the President and the SecDef), I agree; the two man rule is just to ensure that, at each level through which the launch order passes, two people have to agree that it's a "valid" launch order, where "valid" means it really does come from NCA and it really was issued intentionally by NCA. The two man rule at all those subordinate levels doesn't tell the two people to ask whether NCA was sane when the order was issued; that's a separate issue.

But at the NCA level, as I noted above, the two man rule does add at least some measure of safeguarding against the President himself being insane, because the SecDef has to agree to issue the order in the first place. That is, he's not just agreeing that yes, the President really does want to launch nuclear weapons; he's agreeing that he, himself, as a separate decision maker, wants to launch nuclear weapons. So if the SecDef is sane, and there isn't a sane reason to launch nuclear weapons, he won't agree to the launch order, even if an insane President wants him to.

It's also worth noting that at the next level down (the Chairman of the Joint Chiefs of Staff, who has to receive the launch order from the President and SecDef), they are, as I understand it, supposed to exercise some judgment about whether the order is valid based on circumstances. For example, suppose an insane President fires his (sane) SecDef when the latter refuses to assent to a launch order. There is an ordered list of people who are "next in line" if someone up the chain is killed or is unavailable; so the President calls the next person on the list and asks them to assent to the launch order. However, even if that person agrees (let's suppose they're insane too), when the two of them transmit the order to the Chairman of the JCS, his first question should be, where's the SecDef? I fired him, says the President. Assuming the Chairman is sane, that in itself would be enough to invalidate the launch order; the President can't just fire the SecDef--he's confirmed by the Senate, and there's a formal process that has to be gone through. (The same goes for all the people on the list, btw--they all have to be confirmed by the Senate, and they can't be summarily fired by the President without due process.)

Of course these sorts of scenarios are the stuff of which thrillers are made (for example, Tom Clancy's The Sum of All Fears); but the point is that officers are supposed to exercise judgment. Which does mean that Major Hering was entirely justified in asking the question he asked.


the President can't just fire the SecDef--he's confirmed by the Senate, and there's a formal process that has to be gone through. (The same goes for all the people on the list, btw--they all have to be confirmed by the Senate, and they can't be summarily fired by the President without due process.)

The rest of your comment is fine, but actually the president can fire any cabinet officer or (almost) any non-judicial appointment at any time. The Supreme Court has agreed that the president has this authority.

Granted, the appointment must be confirmed by the Senate in the first place, but the Senate does not hold a veto on the president removing any of these appointees after confirmation. The impeachment and trial of President Andrew Johnson was based on Congress' attempt to keep him from firing appointees, and that law was later ruled unconstitutional.


actually the president can fire any cabinet officer or (almost) any non-judicial appointment at any time. The Supreme Court has agreed that the president has this authority.

Hm, yes, I see on reading about the Myers v. United States decision that, as usual, Supreme Court jurisprudence makes less sense than it ought to. :-)

I still think, though, that a JCS Chairman in the position I described would question (and would be right to question) a President who conveniently fired a SecDef who did not concur with a nuclear launch order.


Why did the military require human beings to turn the keys if they expected people (and weeded out people unwilling) to turn the key blindly? A wire would have been much more efficient...


Humans are probably more difficult to hack (I hope) and more resistant to noise (I hope).

A weak ago, there was a discussion about the “Letters of last resort” https://news.ycombinator.com/item?id=7165048 (56 points, 6 days ago, 50 comments). They are secret instructions for nuclear submarine commanders in case of something killed the whole UK government.

From Wikipedia:

> According to Peter Hennessy's book The Secret State: Whitehall and the Cold War, 1945 to 1970, the process by which a Trident submarine commander would determine if the British government continues to function includes, amongst other checks, establishing whether BBC Radio 4 continues broadcasting.

This kind of safeguards is easier to implement in humans than in automatic systems. How do you distinguish from the submarine the lack of communication caused by a nuclear attack from a problem caused by a massive electricity outage and a big fire in the BBC building? Maybe you can listen to some Canada and Australia signals to hear if there is a problem.


It is not hard to hack a human, it happens every day - we know it by the term radicalization these days. There are people who abandoned comfortable Western lives to fight jihads and become suicide bombers. Another term for it is brainwashing. In the Cold War people became defectors. You can hack a human by speaking to them or by showing them some text to read, techniques thousands of years old.


It's not possible to know, for sure, if the orders are lawful or sane (since we are talking about MAD, chances are they aren't sane, although they may be legal). That's why humans were designed into the system - they have to judge and face the consequences of their judgement.

It's evident that the original purpose of placing "moral processors" in the loop was lost after the fact.


Does anyone know what TempleOSV2 is talking about? I can't reply to his comment in this thread because it's marked as dead, but it sounds very interesting.


This is the best general explanation I've seen: http://www.techrepublic.com/blog/software-engineer/templeos-...


This account is used by the author of "The Temple Operanting System"[0], who's schizophrenic.

The first and the last two sections of his post are similar to his usual writings (the fourth one comes out of a Markov chain). The second and third ones are unusual, but it's probably gibberish too.

[0] http://www.templeos.org/ IMO both a work of art and a technical achievement. See my sibling comment for a review.


The second part is a quote from: http://www.nist.gov/itl/csd/ct/nist_beacon.cfm

The third part is very similar to the random example in: https://beacon.nist.gov/home

It’s apparently unrelated to the current article.


Not gibberish then. Thanks for pointing this out.



Nuclear weapons are extremely dangerous, not just for the people being targeted but for all mankind. If I had my way they'd never be used. And I want to respect this author's opinion -- after all, they wrote a book about it, and they've been kind enough to write an article.

Having said that, this article reminds me of somebody trying hard to work a problem, giving up, then saying the problem is stupid. That is, trying to work out moral agency in a world where one person can kill hundreds of millions of others, the author (and those like him), give up and pronounce the question wrong -- that is, we should not live in such a world.

Well, that's fine and dandy, but that's the world we live in. So I think we can take "no person should be able to order a nuclear counterstrike" off the table right away. The author doesn't have much to say except running out of gas from waving their arms around so much. This is unfortunate, because we live in a world where more and more individuals have the capability to kill huge numbers of people. Trying to limit the discussion to nuclear weapons does the entire topic a severe injustice. (Though it might sell some more books. Hard to say.) And then effectively giving up and saying that the world we live in shouldn't be like it is does the entire topic disservice.


It's a challenging issue, all right. The important part that I don't see many people discuss is the nature of international relations, most of which is based on how other national leaders and decision-makers will perceive your actions.

Consider: North Korea is run by madmen who have at least some primitive nuclear capability. They regularly make wild accusations and threats against the US. Say they do manage to mount a nuclear device on a long-range rocket. What's to hold them back against launching it at a major US city? Whenever people discuss North Korean (or Iranian or...) nuclear capability, the usual line is that we don't need to worry about it that much, since it would obviously be crazy for them to use them against the US, or any other nuclear power. What is it that makes it crazy, when they've already done so many terrible things to their own people?

It's crazy because, according to MAD, any such attack, or even a specific threat to make such an attack, would result in a full-scale launch against their country. Millions of casualties, the total destruction of their culture and way of life. Everyone in the world, most especially leaders in North Korea, and China, fully believes that the US will carry out this threat if attacked with nuclear weapons.

Now, let's say Iran manages to detonate a primitive nuclear device in a coastal US city. The textbook MAD reply is a total destruction of every Iranian city. You can make the case that this is crazy on it's face - there is no imminent threat to stop, and those millions of people who would die didn't do anything to deserve it. Say that what Hering and the article author seem to want happens - that no weapons are launched, and a more measured, conventional reply is used. What do you think the North Korean leaders will think then? Or China and Russia? That's the more important question to ask.

After that, might North Korea think that they can use a nuclear attack to try and extract some sort of diplomatic concession from us? They're a harder nut to crack with conventional weapons, and they have more firm backing from China. If they get the idea that our MAD policy is toothless, they might try something that could lead to a much greater war, even possibly a much bigger nuclear war.

We've been living in a world for a long time now where the Kim Jong-uns of the world have very good reason to be terrified of using nuclear weapons against the US. Are you willing to see what happens if that is no longer true?

There are terrible people in this world who are prepared to do terrible things to everything we hold dear. To keep the world safe and stable, those people must believe that we will do even more terrible things to them if the situation calls for it. Keeping that belief in place may sometimes require us to actually do some terrible things ourselves,


You keep using that term MAD. I do not think it means what you think it means.

You seem to imply that it's simply a policy of retaliating against any first strike with massive and lethal force. In fact, MAD, as a theory of nuclear deterrence, requires creating a situation in which a first strike from either side results in the total destruction of both sides, that is, mutually assured destruction. To be workable, this requires two things not present in the situations you describe (North Korea and Iran, or any other rogue state):

1. a rational enemy, i.e. an enemy for whom death is a bad thing. As much as American propaganda tried to paint the Soviets as godless monsters, they respected them as rational. Fundamentalists and North Koreans, not so much.

2. strategic parity - a nuke or two held together by duct tape, while certainly capable of causing much destruction, hardly gives anyone the capability of assuring the destruction of the US.


Yeah, the precise definition of MAD and the other policy options that result in the strategy I described are probably wrong. So what? The point is that it is policy as far as I know that any nuclear attack or specific threat to make one is to be responded to by a full-scale nuclear counterattack, regardless of how big or small the attacking country's arsenal is or what we think of their leadership. This is place for the purpose of deterring any attack, including by small countries. If it happens and we fail to follow through, then the deterrent is weakened, and the thinking of other nations about the viability of a first strike may change.


Why would you say Iran fails the first condition, i.e. that they aren't "a rational enemy"? They seem pretty rational to me. I can't be sure about North Korea, but for all the crazy news about them, they probably would behave like a rational enemy as well.


If Kim Jong-Un was faced with personal defeat (threat of losing power), I have no doubt that he would sacrifice his citizens to go out in a blaze of glory worthy of the Great Successor. In a sense, he's sacrificing his citizens now to support his lavish lifestyle, only he's doing it one starving farmer at a time rather than the whole country in one flash.

I'll defer arguing the rationality of the ruling power of Iran to someone with more knowledge. Bear in mind, my comment wasn't a slur against the people of Iran, but the government and military, i.e. those in the chain of command of potential nuclear weapons.


"Starving one farmer at a time" is evil but rational (and, it can be argued, is also done in some West-friendly countries). However, I very seriously doubt Kim Jong-Un would be able to "go out in a blaze of glory" in an apocalyptic scenario. Most likely, North Korea as a whole would act rationally and he would be taken out by a military junta before he destroys everything. No matter how evil he seems, he must have a rational entourage. Even Hitler had one. Stuff simply has to get done in NK; even if the only plan is "blackmail other countries to give us money", do you not see that's a perfectly rational plan?

As for Iran, I understood you were talking about their government. Nothing I've heard of them so far strikes me as irrational -- just hostile to some Western powers.

We must stop confusing "hostile" with "irrational", that's all I'm saying.


You can make the case that this is crazy on it's face - there is no imminent threat to stop, and those millions of people who would die didn't do anything to deserve it.

Some people might argue that it's the Iranian citizens' responsibility to stop their government from doing things that will get them all killed, such as nuking New York.


This is ridiculous. The US isn't going to respond to a nuclear detonation by a hostile power on its soil with a "measured" conventional response.

It will be overwhelming, it's just unlikely to be nuclear for various reasons.

But at the end of the day cities of the aggressor would be levelled, their military obliterated, their leaders be dead, and the country under martial rule by a coalition which would have extraordinary support to occupy and disarm.


Game theoretically, we're closer to US vs. early, pre-H-bomb USSR (1950?) right now, with respect to North Korea.

The "rational" choice is an immediate pre-emptive strike, if it's a single round with only two players. The difference is this time we've got Russia and China who are somewhat independent, and a moderate capability in Pakistan (as well as asymmetric warfare capability in other allies of Iran).

I don't think I'd pre-empt Iran (they're a lot more rational than they seem), but there are definitely situations where I'd use conventional/SOF to lock down Pakistani weapons in an emergency (but not actually attack the country); and definitely situations where I'd use limited nuclear strikes (sub-surface penetrators, etc.) to take out North Korea pre-launch.


I wouldn't say so. The US waved nuclear threats around about of sorts of things in 1950. Nobody does that now. Iran is a totally different situation. They are a developing regional power.

The tone of our own propaganda always makes the pecking order pretty clear. North Korea is always dehumanized, and gets obliterated, if only because nuclear attack is the only way to prevent a complete massacre is Seoul via artillery/etc.

Iran is a totally different situation. They are a developing regional power. We never say things we can't take back about Iran, and fight with Russia out of habit. China isn't a threat -- nobody can afford to lose the trading relationship the China has with the world, including China.


I agree -- China and the US are ultimately rivals in the sense that the US and UK post-1812 have been rivals.

The saddest part about the bad relationship with Iran is it largely stems from one stupid protectionist-for-British-industry decision made in the immediate post-WW2 period, and then a shitty US ally picked due to the Cold War. Otherwise, the US and Iran would probably have been allies to roughly the extent the US and Israel have been allies.


This type of game theory mostly only deals with the principal actors and their direct reactions, though. What's often more important is what everybody else in the world thinks of the actions. A nuclear exchange is not the end of the world, and what that world looks like afterwards will be shaped by what people think of your actions. That's mostly what I was addressing. In the event of a nuclear attack by a 'rouge state', i.e. one that doesn't have enough weapons and delivery systems for a major attack, the world will be looking very closely indeed at what our response is.

A response that is not sufficiently swift and terrifying could embolden our enemies/rivals. It could also make countries allied to us and trusting the protection of our nuclear umbrella doubt the protection of that umbrella. They might seek to ally with our rival countries instead, or build their own nuclear deterrent. Consider Japan, for example. China is their rival, and a nuclear power. Japan has no nuclear weapons of their own, as they trust us to back them if China threatens them with nuclear weapons. If they have reason to doubt that our backing is not sufficient to deter China, they might seek to build their own nuclear arsenal instead.

Of course, a response that is seen as excessively brutal and/or unconsidered could potentially cause diplomatic issues as well. Countries unfriendly to us might become more worried about what we might do, and might seek a better arsenal of their own, or closer alliances with nuclear powers opposed to us. Countries friendly to us or more allied to us might become worried about what we might to do them, or about being associated with our actions, and might become more distant.

In reality, there will probably be some of both reactions, no matter what we actually did.

I'm inclined to be rather skeptical of nuclear pre-emption because of the second. Not quite 100% against, but the situation would have to be very clearly defined as it being a necessity. In the case of North Korea, in the world as it is today, I don't see it being necessary because of Chinese influence. China is clearly okay with having North Korea be a thorn in our side, but they very much don't want to let North Korea do anything that might provoke us into either a conventional or nuclear attack against them. They don't want American troops or American nuclear weapons detonations right on their border. They have a lot of influence with North Korea, being pretty much the only thing between them and starving and freezing in the dark.

I also have a nice essay around here somewhere describing how your stereotypical crazy dictator tends to get very sane very quickly when faced with the prospect of nuclear war. China, India, and Pakistan have toned down the crazy considerably since becoming nuclear powers.

Note that the last 2 paragraphs treat the world as it is today - where any actor launching a nuclear attack could very reasonably fear a swift and devastating response. This is exactly what a nuclear attack by another country followed by a response perceived as weak could change. It might, for example, turn nuclear pre-emption into a necessity instead of an unfavorable option. If we couldn't count on Chinese and North Korean fear to deter an attack, we might have to actually launch a pre-emptive attack to stop theirs.


>Now, let's say Iran manages to detonate a primitive nuclear device in a coastal US city. The textbook MAD reply is a total destruction of every Iranian city. You can make the case that this is crazy on it's face

No way. It is completely rational to kill millions of innocents because a handful of madmen also killed millions of innocents.


"When the President does it, that means it is not crazy."


Even when he is drinking and bragging about his ability to kill 70 million in 25 minutes. It's alarming alright.


World obliterated by a drunken president. I'm not sure if it's alarming or hilarious.


Another unsung... Stanislav Petrov: The man who may have saved the world http://www.bbc.co.uk/news/world-europe-24280831


We've been "singing" about this guy several times around here and there on the Internet and he even made it on BBC. Not to say we shouldn't keep doing it, but just that the term "unsung" may not apply so well any more. :)


in my opinion, no order to deploy nuclear weapons can ever possible come from a sane commander. any such order should be refused.


Well nuclear missiles can be used on military targets, or targeted on infrastructure or to strategically poison areas with radiation. I think what you object to is MAD.

As terrible as it is, it does likely prevent wars from happening in the first place. Even if the world is stable now, it may not always be so and it wasn't in the past.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: