On August 6th, in 1945, the world saw the first use of atomic weapons against human targets.  On this day 63 years ago, humanity lost its nuclear virginity.  Until the end of time we will be a species that has used fission bombs in anger.

Time has passed, and we still haven't blown up our world, despite a close call or two.  Which makes it difficult to criticize the decision - would things still have turned out all right, if anyone had chosen differently, anywhere along the way?

Maybe we needed to see the ruins, of the city and the people.

Maybe we didn't.

There's an ongoing debate - and no, it is not a settled issue - over whether the Japanese would have surrendered without the Bomb.  But I would not have dropped the Bomb even to save the lives of American soldiers, because I would have wanted to preserve that world where atomic weapons had never been used - to not cross that line.  I don't know about history to this point; but the world would be safer now, I think, today, if no one had ever used atomic weapons in war, and the idea was not considered suitable for polite discussion.

I'm not saying it was wrong.  I don't know for certain that it was wrong.  I wouldn't have thought that humanity could make it this far without using atomic weapons again.  All I can say is that if it had been me, I wouldn't have done it.

63 comments, sorted by
magical algorithm
Highlighting new comments since Today at 4:40 PM
Select new highlight date
Moderation Guidelines: Reign of Terror - I delete anything I judge to be annoying or counterproductiveexpand_more

I would have exploded the first bomb over the ocean, and only then used it against cities if Japan still hadn't surrendered. No matter how many arguments I read about this, I still can't understand the downsides of that route, besides the cost of a 'wasted bomb.'

But what's just as tragic as the bomb having been used in anger, is that it wasn't finished 2-3 years earlier -- in which case it could have saved tens of millions of lives.

It seems weird to me that you would expect that if people had never used nuclear bombs before they would be more reluctant to use them. I would expect exactly the opposite (not that I am confident about this, but it's what I would expect). Consider that the US used nuclear bombs immediately after it got them.

Eliezer, there are at least six comments on that directly. They also make the point that you have not addressed: they believe that the world may be safer because of using a pair of relatively small nuclear weapons. The main arguments for that side have been that we really did need to see the ruins and that MAD was necessary once multiple parties had nuclear weapons (nothing says MAD like someone showing they really will do it). I may take as a separate point that we would not have worried so much about proliferation, particularly by smaller-than-state entities, if we had not seen the effects. Against that, you have a norm of "impolite" nuclear weapons and a concern about hypocrisy?

As you say, the p-value could be tiny and still meaningful, and your follow-up comment seems more certain of the sign on it. There are also several comments on that point: people seem very sure in hindsight what decisions should have been made decades ago. This is odd whether the argument is that this was the only possible way or that it was completely unnecessary because we have had no other nuclear weapon use since then.

To put it differently, was a "first time" inevitable? Arguable. What would have happened if that first time came when more and/or more powerful weapons were available? Very bad things. How certain are you (general you, anyone) that this was not a 200,000 death innoculation against the 1 billion death war? I am not 99.98% certain that the bombings made future bombings either less or more likely.

Which I suppose is a separate argument against: if you are not 99.98% certain that your decision will prevent 1,000,000,000 deaths, you probably should not kill 200,000 people based on that decision. Which I suppose is a separate argument for: if we are less than 0.02% certain that there are long-term effects in any direction, short-term concerns dominate. (I'm not fond of the latter, phrased that way, but this comment is long enough.)

I'm amazed how many comments are arguing that World War II contained much worse catastrophes than Hiroshima. Of course it did. Hiroshima isn't even in the top 10, if you consider only the direct effects of a small fission weapon. I've usually seen the total death toll from WWII estimated around 50 million lives, and Hiroshima and Nagasaki combined, on the order of 200,000 lives.

I have no reason to think it hurts more to die in a nuclear firestorm than a chemical firestorm, but chemical firestorms don't scale up the way nuclear firestorms do.

The issue is whether establishing a precedent for using nuclear weapons, increased the probability of their future usage, potentially costing billions of lives.

I am amazed at how many commenters entirely ignore this issue, which was explicitly the whole focus of my original post.

This says something about the problems humanity faces in recognizing and handling its global catastrophic risks. Apparently many people just don't have a mental bin for global risks to humanity, only counting up the casualties to their own tribe and country. Either that or they're just short-term thinkers. They can ask how much use of the Bomb helped America, or even how much use of the Bomb hurt (or helped) Japan, but not how use of the Bomb affected the probability of nuclear war for the next 63 years and beyond.

I can tell you that I've spoken to nuclear proliferation specialists and they say that perceived hypocrisy by America is a major problem in convincing other nations not to develop nuclear weapons. Now we don't know what those nations would be saying, if no one had ever used nuclear weapons. Maybe they would just find a different excuse. But at least to first order, it looks like Hiroshima had knock-on effects on nuclear proliferation and nuclear diplomacy.

If we put the cost of a nuclear war at 1 billion lives, then it only takes a 0.0002 positive probability effect on nuclear war to outweigh the direct loss of life at Hiroshima, and only a 0.05 positive probability effect on nuclear war to outweigh WWII, integrated over branches.

I think it is worth mentioning that two A-bombs were dropped, and according the history I have read, the Japanese leadership were not deterred by the first, nor by the firebombings that razed dozens of other cities to the ground.

"What if the alternative was for the U.S. to firebomb and blockade Japan [...]"

That was probably another possibility, but certainly not the only alternative to nuking cities.

How about nuking somewhere very visible but not so populated with the message: "We have more where that came from. Surrender or the next one won't be in a daisy field." ?

All I can say is that it is a good thing you will never rise into any position of power regarding the security of the country or the world. Some people are made for it and some are not. No blame, but you are clearly not.

What if the alternative was for the U.S. to firebomb and blockade Japan for another year and then with Russian help invade. Over twenty times as many Japanese would have died, a million Japanese women could well have been raped by Russian troops, and one-have of Japan would probably have had to live in a communist dictatorship for at least 40 years. Would you really prefer this alternative to what actually happened?

Or even worse, what if the alternative was for the U.S. to use biological and chemical weapons on Japan and kill most of its population?

In 1937 in Nanking China, the Japanese put to death over 300,000 civilians. Mostly with bayonets and rifles. That's more than twice the number that died at Hiroshima and Nagasaki combined.

Something to think about...

Scott, the cost of a 'wasted bomb' in this case was not having another bomb to drop. They only had the two they used.

Or put another way, the cost was the war being prolonged until they did have another bomb produced (order of months).

I don't know about history to this point; but the world would be safer now, I think, today, if no one had ever used atomic weapons in war, and the idea was not considered suitable for polite discussion.
So firebombing would be a perfectly acceptable tactic, then?

I am reminded of one of the medieval popes who banned the use of crossbows against Christians because it was too 'horrific' a weapon to use against 'civilized' people.

I don't recall a single nuke being used in Vietnam, and none have been used in Iraq, but those wars are still messing people up pretty well. It's not clear to me how killing lots of people all at once in a sudden flash is better than killing them slowly and agonizingly with napalm, or incendiary bombs, or machine guns, or crushed beneath tank treads, or rusty bayonets.

Mayhap there's a form of cognitive bias lurking here to be overcome?

What I'm particularly disturbed by is the suggestion that actions are 'tainting', that the fact that some people used nukes before somehow alters mankind from its previous state, leaving a lasting mark or stain, and especially the idea that this change is transmissible. I don't see how a child born in 2008 is somehow fundamentally different from one born in 1908 in regards to use of nuclear weapons, nor do I see any obvious way in which a society in which the WWII use of nukes had been forgotten would differ from an otherwise-identical society before the use of those weapons.

This appeals to all sorts of contamination instincts, which is why it's a common and popular implicit metaphor in various religions, but it's a very peculiar stance for a rationalist to take.

In a world full of irrational people with availability biases, it is likely that seeing the effects of two weak nuclear bombs against Japan helped prevent the use of an arsenal of hydrogen bombs against Korea, Vietnam or the Soviet Union.

I was trying to understand hindsight bias when a friend first directed me to overcomingbias.com. I still have the link he sent in my favorites and it informs my view of the original post above and many of those that follow. Here it is for those interested in such things:


Apparently many people just don't have a mental bin for global risks to humanity, only counting up the casualties to their own tribe and country. Either that or they're just short-term thinkers.

Eliezer, I certainly worry about global risks to humanity, but I also worry about the "paradoxes" of utilitarian ethics. E.g., would you advocate killing an innocent person if long-term considerations convinced you it would have an 0.00001% chance of saving the human race? I'm pretty sure most people wouldn't, and if asked to give a reason, might say that they don't trust anyone to estimate such small probabilities correctly.

Seconding Scott Aaronson (especially on the tragedy of the US getting nukes too late) and Michael G R, though with the suggestion that the "demo" bomb could have been Trinity.

Eliezer: I think that the hypocrisy of the US is mostly in our maintaining a large arsenal but telling others that they can't have any, not in having used nukes. I don't see a case for .1% probability increase for nuclear war without using the bomb which is stronger than the case for the inverse, but I do see a million dead Japanese. Also, in terms of catastrophic risks, a couple more years of total war to encourage ultimately disastrous cultural changes in the US.

What I really want to know about WWII is why Hitler wasn't assassinated before it was too late. Surely if the situation with Archduke Ferdinand teaches us anything it's that Assassination can solve all of one's geo-political problems ;-(

Frelkins, You shifted rather quickly from what I think is the stronger argument against MAD (greater catastrophic risk due to human error and irrationality) to what I think is a weaker argument against MAD (a claim that some states are suicidal). I think you should focus on the stronger argument.

Also, I think the claim that a world without the type of MAD one gets from nukes is a world where all politics is solved through war is, I think, inaccurate. Some politics seems to be solved through war, others don't, both before and after MAD. It may be true that there's never been direct conflict on sovereign territory between two nations that both have nuclear strike capability against each other, but that's a small swath of history.

I'm not arguing against MAD, or against the concept that nuclear proliferation results in a more peaceful world. But I'm not sold on it yet either. It's worth more study, it seems to me.

Frelkins, I think the main perceived flaw in this line of reasoning is that error and irrational decision making are possible, and with viable MAD set up, the results could be catastrophic.

Time has passed, and we still haven't blown up our world, despite a close call or two.

I can't help but wonder about anthropic effects here. It might be the case that nuclear-armed species annihilate themselves with high probability (say 50% per decade), but of course, all surviving observers live on planets where it hasn't happened through sheer chance.

(Though on the other hand, if an all-out nuclear war is survivable for a species like ours, then this line of thought wouldn't work.)

I'm always wary of claims of type "if we do otherwise moral thing X someone else may react by committing horrific crime Y, so we should not do X"; at the least we should be very explicit about where moral blame lies, and at the most we should discount Y because of game-theoretical incentives (example: not paying off kidnappers). Still, it seems like this doesn't much affect Eliezer's point here.

By the way, I'm always nervous about persuading people of impopular political positions when I agree with them on far more important non-political issues; they'll probably lose social standing.

Apologies, I completely butchered that last paragraph. Too many edits, not enough re-reading or sleep. I have no idea if the intent came through, and I may try again after another sleep cycle.

While your point about a world that hadn't used nuclear weapons being safer today is unclear, I think your claim that 'you wouldn't drop the bomb' is driven by hindsight bias. At the time, the far more pressing issue from Truman's perspective was how to end the war with a minimum loss of US life, and the long term consequences of the bomb were far from clear.

I also think that memorials like Hiroshima Day pervert the overall moral perspective on World War 2. Because it was a large, salient act of destruction, it gets remembered. The Burma Railway and the Rape of Nanking (brutality which didn't even serve any strategic purpose) don't have any memorials. It is a gross distortion for Hiroshima to let the Japanese be primarily viewed as the victims of World War 2. EVERYTHING about World War 2 was horrible, but you can't only emphasise one bit of that horror without affecting the overall perception.

As to the question of why you didn't want to just have a show of force, I remember Victor Davis Hanson arguing that in order to prevent conflicts from re-starting later, there is a psychological importance in the enemy realising that they are well and truly beaten. Without this, he argued, it's possible for revisionists to re-stoke the conflict later. Hitler did exactly this when he claimed that the German army in WW1 was on the verge of victory when it was stabbed in the back by politicians at home, instead of actually being days away from total defeat. Say what you will about the bomb, but it certainly let the Japanese know that they were beaten, and Japanese militarism hasn't resurfaced since.

In a repeated game of Prisoners Dilemma, Tit-For-Tat seems to be a dominant strategy. With Hiroshima, the Japanese found out that payback's a bitch. The only injustice is to the extent that the individuals who bore the brunt of the attack weren't personally the ones who instigated it, but this is true of every war in history. I feel for their suffering, but no more or less than any other civilians in World War 2, or anywhere else.

The Second World War, as a whole, was probably the most catastrophic event in humanity's recorded history. The world was pretty much screwed as soon as it started -- indeed, probably as soon as Hitler acquired control of Germany.

For the purpose of saving humanity in the future, it may not be most effective to focus our attention on particular decisions (even apparently large ones) made during the course of the war (indeed, near the end of it); we risk missing the forest for the trees.

I wrote something a little while ago about how Nagasaki was a secondary target, and Kokura was saved by cloudy conditions.


From the perspective of warding off nuclear apocolypse being the overriding concern, perhaps the play would have been not to drop the bombs on Japan so we could save them to drop on the USSR. If we had invaded Russia at that point we could perhaps have stopped nuclear proliferation with just one nation having the bomb. And after that point, if anybody else even looks funny at an atom, we invade their country and put all their scientists to the sword.

Or is that not a nice enough solution?

You're probably right. But I can see a small benefit: we have become wary. There's still the possibility that someone will develop an effective defense system against the bomb. On the other hand, if we had never used the bomb, it would probably be less widely known and there would be the possibility of a sucker punch from cult of mentally disturbed physicists.

Steve, I think you posted your comment in the wrong thread, not this one.

@Robin H.

It seems that if and only if extinction was occurring, would your approach to be valid. If not, the potential causes would be the ones to address, not the ones that have nothing to do with it.

If that is accurate, what probability do you assign to human extinction being in process today?

Off to the ridges for a hike. :)


Funky, you might be right.

Consider Tacitus:
"To ravage, to slaughter, to usurp under false titles, they call empire; and where they make a desert, they call it peace."

How better to make a desert than with nukes?

As a general rule, real WMDs do not help nations achieve the goals they think of as victory. Imagine for example that we had created plentiful nukes two years earlier, and we had then bombed 20 german cities while the germans surrendered to us. We would then have to deal with russia, and our german ally would have 20 fewer cities to assist us than they would otherwise.

WMDs don't give us what we want. They only help us avoid disastrous defeat by threatening a different sort of disaster.

Usually, threatening to use nukes is an admission of defeat. You don't do it when you're winning. Ex: USA in korea. USA in vietnam. israel in 1973. Exception: USA in invasion of iraq. We said if iraq used chemical or bioweapons we'd use nukes. The word is out now that they didn't have them and couldn't have used them, but it's a rare thing -- as if we thought their mustard gas might give them a decisive victory or something.

You might be right that we haven't used any other WMDs because we used nukes once. There's no compelling evidence in either direction. I want to also point out that we have not had many uses for them. WMDs are mostly good for destroying cities full of civilians, with side effects that might last for many generations in the cases of biological, genetic, ecological etc weapons. All have known long-lasting side effects except for the nerve gases which are not particularly effective. How often have we needed to destroy enemy cities? We bombed Hanoi and Haiphong as part of our peace negotiations, but when have we done the like since? We pride ourselves on pinpoint strikes against particular targets.

It's only losers who use WMDs, and only when the surrender they face is worse than the side effects of the WMDs. Why would it be surprising that it hasn't happened again since our ill-advised single use?

Unlike J. Thomas, I do believe Hiroshima and Nagasaki did become “deterrents” of sorts for later use of nukes (for nation states at any rate). Nukes were unprecedented, the first weapon that could potentially spell the end our species. In spite of this, there were hawkish generals on both sides in the 50’s and 60’s who were eager to use nukes (attempting to sidestep MAD scenarios with talk of “first strikes” and “acceptable losses”). The horrible pictures from 1945 went a long way towards discrediting this kind of talk by showing the true impact of using nukes – not just powerful and acceptable tactical tools to use against an entrenched enemy (as reels and reels of atomic weapon tests would suggest) but something nightmarish that carries a devastating human cost.

We’ve managed to avoid use of other horrible weapons (biological & genetic warfare, etc.) because of the atomic example, which forced us in no uncertain terms to deal with the fact that we had at last developed the technological means to end our existence. This type of awareness did not (and could not) exist in 1945. It evolved over time, in the shadow of Hiroshima and Nagasaki.

Politics, yuck. Stop it please.

I don't have much time for Eliezer's point regarding 'once you've used nukes in anger you can never go back'. Sure, the world was stunned, and rightly so, but what did we learn about humanity that we didn't already know? The most powerful, secular, free nation in the world built a weapon of terrifying potential and used it straight away. That's what we are, and now we know it. Silver lining? We'd better hope so.

'Was it right or wrong?' is a wrong question. It was deemed right at the time, and so it happened. Anyone who claims that they have looked over the evidence and have decided that actually, no, it was the wrong choice, is wasting their time. Equally so if you think it was right. What's important is what we take from it, what lessons we learn from it; the part that has yet to be written.

So, how do we make sure that next time, it's perceived to be the wrong decision?

it's not the wheapons that kill people, but the people who use them.

There's a level where that's kind of true.

But consider the chicken. In the usual way of things, when two cocks meet they do some threat displays and likely one of them runs away. If not they fight each other a little and then likely one of them runs away.

If you strap razor blades to their feet and put them into a pen where they can't run away then you have something you can sell tickets to. Except it's illegal in this country. You could say "Razor blades don't kill fighting cocks, other cocks do" but it wouldn't be very true, now would it?

The people who strap on the razor blades convert a dominance ritual into a sort of ritual murder. The chickens just follow their instincts, never thinking out the consequences.

It would be possible to attach the trigger for a nuclear weapon to a cock so that when it killed its opponent the nuke would go off. We wouldn't actually do that, but we could. And then you could say "Nukes don't kill people. Roosters kill people." It would be just as true as the current version.

EY: "I am amazed at how many commenters entirely ignore this issue, which was explicitly the whole focus of my original post."

It's very difficult really to answer this issue. I think that yes, if we hand't come to use nuclear wheapons, if we were living in a nuclear-free world, it would certainly be safer.

However, don't forget.. it's not the wheapons that kill people, but the people who use them. This war is a clear demonstration that people are unable to resolve issues rationally and in a civilized manner all of the time.

So I think your question should really be 'would we be capable of living in a world without such threats'? And my answer to that question is no. Power has too much appleal to people. History doens't teach us enough. Humanity often comes second.

That is unless we change, this world will never be that safe. No matter what wheapons are available. That's why teaching history in schools is so important. It allows us to learn from our mistakes and change.

J Thomas: Surely no more than say, 75% of the country went crazy.

Mark, no one has used biological weapons even though we have developed them. (There may be some unpublicised exceptions, maybe south africa used some against africans etc.) No one has ever used genetic weapons. The idea that every weapon gets used except for MAD is wrong.

You say that we cannot have disarmament. As long as the USA prevents disarmament, you are right. But after the next nuclear war, we will have nuclear disarmament provided the world economy still exists. You say it can't happen on no evidence. I say, wait and see. When it comes, you won't be able to stop it.

Which does not say the world will have universal peace. Making the world safe for conventional warfare may not be exactly pleasant. But it will happen, pleasant or not.

I have the sense that you imagine two sides here, the idealists who imagine wonderful worlds that can never be versus the realists like you who face the gritty truth that life is tough but if we're tough enough we can prevail. But see, your sort of realism came out of the Cold War, and it hasn't adapted to the modern world yet. In the new world aircraft carriers are big fat targets, and within your lifetime we will have to face the weather without the comfort of our nuclear umbrella. It won't be a utopia, it's just the next step.

Actually, though I am an american citizen, I don't bother to vote. I can't stand either major party, and I don't "identify" with our current president, either of his likely replacements, or any other recent president.

My preference for america not losing (that is, not being at a strategic disadvantage relative to other countries) is because (for all its flaws), I think america is (mostly) (compared to our enemies) "the good guys". If I were brought up in some other system, I'd undoubtedly feel different. Obviously I couldn't be on the right side in both cases, so by what logic do I think I'm on the right side of things? Because I was brought up in this system, and I do think we're "the good guys" (at least relatively speaking).

Now I readily accept the possibility that I'm wrong in my worldview (if such a phrase actually means anything at all), and I'm all in favor of questioning my allegiance and considering that either I was mistaken in the past, or perhaps I was right in the past but conditions have changed. I do this regularly, but so far I still come out on the "we're the good guys" side.

Even granting the possibility that the wool is being pulled over my eyes, I still have to come out on the side I think is right. I assure you "the bad guys" spend very little time engaging in all this navel-gazing, so for us to become all high-minded and neutral and indifferent to the outcome virtually guarantees that the bad guys will win.

America-bashing is quite popular, but it's really rather juvenile (compare "amerikka" to a REAL police state) and quite irresponsible. America's enemies are FAR harsher on the environment, women, minorities, the poor, homosexuals, etc than even the wingnut's version of "bushitler" and the rest of the "neocon conspiracy" . Can any sane person seriously NOT care who comes out on top in the long run?

Mark, I think you over-identify with whoever controls the nuclear weapons in the US arsenal. I think their existence is a complex phenomenon, and I'm not sure it can be reduced to "I am an American citizen and voter, therefore I exert partial control and ownership of the weapons in the nuclear aresenal".

Beyond that, I think a major source of bias is people who let the status quo and power/hegemony alignment do a lot of their argumentative legwork for them. I think you're doing that here, but it's a much bigger problem warping our models of reality than this instance.

J Thomas, I state again that a world without nuclear weapons is NOT a realistic option. If we didn't develop them, someone else would. If not now, eventually. And even if you make the unlikly assertion that no one would develop them if they didn't exist now, the objective fact is that people HAVE developed them and they DO exist now. Putting the genie back in the bottle is simply impossible, and wishing it were otherwise won't make it so.

We have enemies. We always have, and if you seriously believe that we won't always have enemies, you're far deeper into santa-claus-land than I am. There's simply no evidence whatsoever for the position that people will somehow permanently put aside all of their differences and live happily ever after until the end of time.

Now don't get me wrong - I wish there weren't nuclear weapons at least as much as you. Probably more, because I don't think our enemies can be talked out of trying to kill us. And sooner or later they'll get ahold of a nuclear weapon and detonate it, probably in a major US city. The difference between us is that I'm willing to face the fact that there are and always will be nuclear weapons, whereas you seem to think that if you hide from this admittedly uncomfortable fact you'll somehow be immune from its logical conclusion. Good luck with that - just be sure to put on some SPF-1,000,000 sunscreen and I'm sure everything will work out just fine.

Given MAD, why would anyone adjust their probabilities of nuclear attack, even after considering the precedent of nuclear bombardment set by the U.S? In the same act of initiating a nuclear launch, one also signs his own death sentence. As long as that statement is true, why should we touch our probabilities?

Scott, I'll take your bait. Yes of course I should consider my probability calculation carefully and skeptically, but if even so I estimated killing one innocent person would reduce extinction by one part in a hundred thousand, and if those were the only consequences worth considering, then yes I would support that.

Mark, you have the right to your untestable opinions. No one can ever show whether we would have used nukes other times if we hadn't that time, or that somebody else would have used nukes if they had them, or that if you were in Truman's place you'd do the same thing he did.

There's no way for anybody to know about any of these things, so you have the perfect right to believe whatever you want just as you do about how many Santa Clauses there are in Heaven and whether the Yankees would have won the series in 1947 if they had Joe DiMaggio, and whether the germans could have won WWII if they'd pushed forward to take Moscow and they got their winter uniforms and if they tried hard to make friends with the ukrainians etc.

This idea that the best way to prevent a nuclear war is to persuade the world that we're crazy enough to kill everybody so they'd better do what we say -- there's something kind of screwy about it.

If we actually want a world where nobody sets off nuclear bombs, we do much better to create a world where nobody builds nuclear bombs. There's something kind of, well, obvious about that reasoning.

We've had less than 62 years when we have avoided nuclear war despite MAD. We have had 5,000 or 15,000 or 1,000,000 years where we avoided nuclear war by not having nukes, depending on how you count. Which looks like a more reliable way to avoid nuclear war?

I think the use of the bombs is probably the only think that has PREVENTED their use ever since... without an object lesson in just how powerful they are, no one would belive it. And no, blowing up a bunch of fish probably wouldn't have had the same impact.

Besides, if the japanese or germans or russians had acquired nuclear weapons first, does anyone seriously doubt that they'd have used them? The question isn't whether or not nuclear weapons would be used - it's who would use them. Me, I'm 110% OK with the outcome. No moral qualms whatsoever, and if I were in truman's shoes (with or without knowledge of the last 63 years) I'd do exactly what he did without a moment's hesitation, or a moment of second-guessing myself afterwards.

That's just me, your mileage may vary. But if it varies very much, I hope I never have to depend on you to defend anything important.

Myself: I can't help but wonder about anthropic effects here. It might be the case that nuclear-armed species annihilate themselves with high probability (say 50% per decade), but of course, all surviving observers live on planets where it hasn't happened through sheer chance.

Just to expand on this (someone please stop me if this sort of speculative post is irritating):

Imagine there are a hundred Earths (maybe because of MWI, or because the universe is infinite, or whatever). Lets say there's a 90% chance of nuclear war before 2008, and such a war would reduce the 2008 population by 90%. In that case, you still end up with 53% of observers in 2008 living on an Earth where nuclear war didn't occur.

This implies that we might be overconfident, and assign too low a probability to nuclear war, just because we've survived as long as we have.

But: The argument seems to implicitly assume that I am a random observer in 2008. I'm not sure this is legitimate. Anthropic reasoning is irritatingly tricky.

Frelkins, let's consider MAD in action.

In 1973 israel lost a war. Egypt wasn't ready to take half of the sinai much less the whole thing, but still it was clear that israel had lost and would have to negotiate. instead, israel threatened to nuke egypt.

The USA detected nuclear material crossing the dardanelles, and "we" initiated DefCon 1 and announced it as DefCon 3. "We" told the russians that unless they backed down and let israel threaten egypt with nukes when there was no countermeasure available to egypt, we would kill everybody in the world and let the russians kill us all too.

Luckily, the russians weren't as crazy about egypt as the USA was crazy about israel, so they backed down. We then shipped to israel everything the israelis needed to win -- our best planes, our best tanks, our best ECM, our best analysis of our best satellite photos, things we'd been keeping secret from the russians, etc. Israel then won the war and we negotiated with egypt for them, giving israel tens of billions of dollars in echange for their agreeing to accept the peace we negotiated for them.

Given this background, how could it be acceptable to have MAD between israel and iran? Far more pleasant for israel to be a nuclear power with only non-nuclear nations as enemies. When you can do anything you want and nuke em if they can't take a fuck.... And the alternative is to have to negotiate? Negotiate with somebody who could destroy you?

If iran gets nukes there will be a big US uproar about it, kind of like "who lost china?". The republicans will say that they were 100% ready to take care of the problem but the democrats wouldn't let them. And every time israel and iran have harsh words the uproar will start up again. The only reason israel is in danger of total destruction is that the democrats made it happen.

We could make an argument similar to the one about hiroshima. We could say, if only there is one nuclear exchange between a couple of small nations -- israel and lebanon would be ideal -- and the whole world gets to observe the result, then we will get disarmament. The world won't stand for anybody holding nukes. If one nation tries to do it their own public won't stand for it. We could clear up the continuing MAD madness, but it takes one graphic example. Without that, too many people will keep beleving that MAD works.

See the problem with MAD? It isn't just that wars can start from accidents or mistakes. More fundamentally, we have strategists who try to carefully judge just how far they can push our enemies before MAD will break down. The better we are at persuading the world that we are willing to kill everybody for trivial reasons, the more we can get away with. And the more we get away with, the more we try to get away with. The only way to find out how far we can push before MAD breaks down, is to push a little bit too far.

However it gets described, this is not a stable strategy.

But it will take a small nuclear war before we can get out of it. Therefore a small nuclear war is necessary, and beneficial, and if we do something to promote one we will be doing the world a great and necessary service.

I tend to feel there is something wrong with this reasoning. And yet, it looks so true....


"error and irrational decision making are possible"

Oh sure, HA: also it is possible that one side is actually suicidal, which some argue is a notable feature Tehran's Shi'ism. Then clearly MAD fails.

On the other hand, without MAD, we have the tragic Hiroshima situation - there is no real compunction to prevent immediate use upon technological achievement. Suppose Robin Hanson is right and people really don't care about morality as much as they say. It's von Clausewitz time, right? "War is a continuation of politics with other means." Politics is many things, but moral isn't one of them, sadly, since I might argue that we are not ruled by philosopher-kings but rather classic demagogues.

I can't guarantee that Nash always works in every situation, but in an imperfect world, I'd prefer to go with what has been proven to work before in roughly similar conditions. Once the weapons exist, aren't we in an ugly world with no good choices?

I think careful analysis may show that there is no nuke-potential state that is in fact suicidal. The mullahs are keen to retain power - they are politicians too. Thus despite the dangers, I am unsure we actually have any other serious alternatives but to return to MAD. But hey I have an open mind.


"Time has passed, and we still haven't blown up our world, despite a close call or two."

Not at all a non sequitur. It is exactly what Nash would predict, isn't it? The entire crazy logic of the Cold War game-theory - as satirized by Kubrick in Dr. Strangelove - was that arms reduction and war prevention weren't even possible rationally until MAD. Once the weapons are invented, we need the balance of terror - only when MAD is basically proven can we reveal the exact numbers and agree to reduce.

Since MAD was achieved with the USSR, nuclear war then became suicide, basically impossible. Rational behavior was forced upon us. Even the "close calls" are needed; if the Soviets weren't convinced that the USA was willing to use, they had no reason to respond by building up, since basically spending all your GDP on second strike capabilities doesn't appear at first to be a good idea. But without clear second strike, we don't have MAD.

As repulsive as it seems, we were lucky to have real grown-ups in charge at the time who were capable of doing this crazy logic - we have after all prevented what could have been true horror, a real nuclear war.

MAD worked for our Cold War and it will also work for other problem areas, like India and Pakistan. I might argue that rather than wring our hands about Iran, its possible capabilities, the Shahab-6 missile and what do about it, we may in fact be better off encouraging Israel to reach MAD with Tehran ASAP. . .

It's absurd to argue that "we" did the right thing because the results happened to turn out well. You can make that argument about anything. For example, if Hitler had not started WWII when he did, there would inevitably have been a world war after both sides had nuclear weapons and it would have been far, far worse. Hitler might have done it for the wrong reasons but we owe him our lives for doing it.

All it takes is to look at what happened, and make up a worse alternative, and then say that what happened was better than the alternative. You can do that about anything. Unless you can repeat the experiment and get odds, it's bogus.

There's the argument that without a horrible example somebody would inevitably have used nukes because they couldn't wrap their minds around it. This is bogus. We have the example of biological warfare, which we have avoided so far without a horrible example. Ditto genetic warfare. However, a variant of it can be used to excuse Truman. Truman had been kept out of the loop, he didn't know about the nukes until very late. Then he had to make a choice quickly. If national leaders couldn't understand it without the graphic example, that would apply to Truman even more. Forgive him, he didn't know what he was doing.

There is the argument that the japanese would not have surrendered without being nuked. This is the first bogus argument all over again, a completely bogus claim about what would have happened otherwise. But this one is special. We had no method set up to accept a surrender from japan. If all along, since Pearl Harbor, we had been negotiating with them about what it would take to end the war, we might at some point have negotiated a peace. But we didn't want to. We wanted an unconditional surrender and nothing else. So there was nothing to negotiate about and no method arranged to do it, and it took the japanese days after our nuke to reach an agreement about how to surrender and we bombed them again while they did it.

So rephrase that. "We nuked japan because we would not accept a conditional surrender." We could have ended our war with japan at any time after Pearl Harbor if we had been willing to negotiate terms. Right after Pearl Harbor the japanese might have insisted on some sort of favorable terms. After they took Manila they would probably not have wanted to give it back. By Midway they might have given back a lot, but of course we wanted them to give back all of china etc. Maybe we could not have agreed on a peace. But we were totally unwilling to find out. It isn't fair to say that we nuked japan because they refused to surrender. It is fair to say that we nuked japan because we were unwilling to negotiate a surrender. We'll never know whether we could have arranged a surrender without nuking japan because we made absolutely no effort to do so. We were unwilling.

There are arguments that we should not negotiate peace and fight at the same time. The basic reason is that peace negotiations weaken our resolve. Apply that this time. We were unwilling to negotiate a surrender with japan because that would weaken our resolve to nuke them.

VD Hanson says that to keep a war from starting again you really need to smash the enemy, show them that they're completely beaten. Japan was already beaten that way, without nukes. The world would be a better place if Hanson got into an argument where he had to admit that his arguments were worthless and perhaps suffer some mild consequence like a YouTube bare-bottom flogging to prove it. The reason he keeps blathering on is that no matter how many times his bogus claims are definitively refuted, he doesn't have to notice.

Bogus justifications aside, people who weren't adults during WWII probably can't understand what it was like. I've read old Life magazines from during the war, and it seems like an alien culture. We haven't had anything like it more recently, except in the weeks after 9/11. For few weeks there, normally-sane people were saying we would have to kill every muslim who didn't renounce islam, or turn the middle east into a glass parking lot, etc. I'm not talking about the insane people who were still saying that in 2002. For awhile there the nation as a whole went crazy. And in WWII it went on for years. We lost close to half a million dead and their families had the grief to deal with. Rationing. A whole lot of men away, in the military. Censorship, and particularly censorship of anything that hindered the war effort. I particularly remember the full-page photo of the steelworkers' union organizer. He tried to start a strike when we were at war! He got a bayonet wound on his butt, and they photographed it and put it in Life magazine along with his recantation. He said he was wrong to hurt the war effort and he apologized.

Say what you like about what you would have done. But you weren't there. People aren't very good at predicting what they would have done if things were different.

So, it's been 63 years and people are real intent on what "we" should have done. Why don't we ever talk about what Genghis should have done, or what Tamerlane should have done, or maybe what Andrew Jackson should have done? We never give a lot of thought to what MacLellan should have done -- if he'd done the right things. maybe the Civil War would have been cut short with far less loss of life. What should we have done in the philippines or the mexican war?

What should Hitler have done? Clearly his approach was suboptimal, but if he had been sane -- if you had been in his place, say -- what might work for getting good results, mercy or at least much reduced injustice for germany?

Why do we only think about what Truman should have done -- one single old dead guy -- and not the other old dead guys?

I'm appalled at the consistent bad quality of the comments in this thread. Eliezer, you should refrain from writing similar posts in the future now that you know many of your readers are incapable of engaging with your arguments when they bear on issues which they are so emotional about. Your talents are wasted with such an unreceptive audience.

"The Second World War, as a whole, was probably the most catastrophic event in humanity's recorded history. The world was pretty much screwed as soon as it started -- indeed, probably as soon as Hitler acquired control of Germany."

WWII was just a continuation of WWI, which was a much less 'noble' war, if such a thing can even be said to exist.

War begets more war.

This looks like it estimates order-of-magnitude 100M deaths from full-scale US-Soviet nuclear war. I guess future arms races might be much worse though.

It's the easiest thing in the world to armchair-quarterback the events of the past. The real challenge is to understand why so many people felt that it was a necessary, though I'm sure as abhorrent to them as it is to you, thing to do.

The truth is that you can't say what you would have done, because you weren't there. You can guess, and of course your guess will have all the smug self-righteous moral overtones of someone who has never held life and death in his hands and had to decide.

Dropping the atomic bombs made America look like a bunch of irrational cowboys. This was a good thing. The Cuban missile crisis (probably) did not end in nuclear war because the Russians were convinced the Americans were irrational enough to use nukes and had shown this by using them before. Playing chicken after convincing the other guy your irrational may not be a great strategy but it does seem to have worked.

There is a good documentary on the thought processes of those that bombed Japan and were involved in the cuban missile crisis called "the fog of war" excerpt here http://www.youtube.com/watch?v=er2xCn3_QcQ

stefan Thanks for the interesting comment

What I've never understood is why it's so much worse to be vaporized/die of radiation burns than it is to be incinerated/die of fire burns. Or why it's "wrong" to nuke people cheering on soldiers who used pregnant women as bayonet targets but "just an unfortunate part of war" to blast the same people into little bits. I'm being coy, of course. There is no difference. The fact is that the "sin" of dropping one bomb versus another was concocted, and has since been propagated, by the anti-American Left and Right.

I'm with James Miller and Caledonian on this one, and I want it taken further. Caledonian, I think the cognitive bias is good old repugnancy bias. How I'd like it taken further: I think what we want to avoid is not (1) horrific outcomes due to war from a specific type of technology, nor (2) horrific outcomes due to war generally, but (3) horrific outcomes generally. As such, beyond using nuclear weapons (which I'm not convinced prevents any of the three, though it may), how about greatly increasing the variety of human medical experimentation we engage in, including medical experimentation without consent, and breeding and cloning people, and making genetic knockout people and disease models of people to the extent that there will be a net decrease in horrific outcomes (death, suffering, etc.)? Sort of Dr. Ishii meets Jonas Salk.

Who cares, I want to hear about building AIs.

I agree with jsalvati. I suspect that seeing the damage that nuclear weapons can create is a significant part of the reason that we have such a large aversion to using them.

@jsalvati "Consider that the US used nuclear bombs immediately after it got them."

Well, not because the inventors wanted it. To be fair, Einstein wrote Roosevelt only out of fear the Nazis would build it first. And of course at the test site, Oppenheimer, Rabi, and Bainbridge all immediately expressed their moral dismay; Oppenheimer most eloquently with the famous quote from the Gita. Indeed, after the test results were seen, several of the Project team wrote a petition begging the bomb never be used.

But from a Cold War game-theory perspective, wasn't MAD considered necessary for safety? With that logic in mind, the US would use the bomb, right, with no counterbalance in place for the Nash equilibrium?

Caledonian, agreed. Whatever we say are the inevitable results of that slaughter, whether it's that we prevented a later nuclear war or we poisoned the chance for peace, they're all bogus.

We don't know what would have happenned instead if only things were different. We can only guess by making netaphors from other situations.

Here's a metaphor--

Pre-nuke: You have a neigbor who annoys you. He plays his stereo too loud. He throws garbage over the fence into your yard. He doesn't mow his grass, you get bugs and another neighbor has trouble selling his house.

You annoy him too.

Every now and then the two of you have a big argument that perhaps goes to the level of a fistfight. If one of you is enough better, he can intimidate the other. Maybe you can make him leave his stereo low and mow his grass, for fear you'll beat him up again. Maybe he can make you stop cooking with garlic and pick up his trash and stop mowing your lawn. Maybe other neighbors will get involved and who wins depends on how the allies fall out. Maybe the allies aren't stable enough to get any resolution.

One-nuke: You buy a gun. You're not sure he'll be properly intimidated by the gun so after you knock him down you shoot off on one of his toes to prove you mean business. Also this helps to intimidate one of your other neighbors who's really good with his fists. Your neighbors argue some about whether you should have done that, but they aren't ready to do anything about it.

Two-nukes: You have a gun. Your ugly neighbor has a gun. You practice your quick-draw and your targetting because next time there's a fight one of you is going to die. The first one to get his gun out and aimed properly wins. So it makes sense to draw quick at the first sign the other guy might draw. It makes sense to draw and shoot at the first sign the argument is getting intense. It makes sense to draw and shoot at the first provocation. It makes sense to draw and shoot the next time you see him.

Many-nukes: You have a gun. Your ugly neighbor has a gun. You have a bunch of catapults that will throw tanks of gasoline onto your neighbor's house and yard. Also thermite. Anybody there will be killed. Your neighbor has the same thing aimed at you. It's a standoff, neither of you can kill the other without at least losing his house. Your neighbors don't like what the situation is doing to their property values, but they are neither willing nor able to kill you over it. You work hard to persuade your neighbor that you're crazy enough to die killing him, so you can intimidate him even though you can't really win. Meanwhile you work at more catapults designed to knock his oil drums out of the air before they reach your house. It doesn't seem like it would work, but if you can persuade him that you believe they'd work then you can intimidate him.

And you've arranged your catapults so that you can in theory hit any house in the neighborhood. It's all very expensive but you know it's necessary. You try to keep any of your other neighbors from building catapults themselves and mostly they don't want to. But if they do try, and they're a bit unfriendly, you threaten to fire-bomb them to make sure they don't, or maybe you shoot them or just beat them up. Whatever it takes, since you sure don't want another important enemy.

Many of your neighbors think you are crazy. But as you point out, you can't expect to have peace. There's always going to be an enemy, and you have to do whatever it takes to win.

If this metaphor fits, it makes the whole thing look insane. You have to accept that any of your neighbors can shoot you while you're opening your front door, or burn your house down around you during the night. It just isn't practical to prevent that, and yet it hardly ever happens. But maybe this isn't a good metaphor. Maybe nations aren't like people, and neighboring nations aren't like people's neighbors. Maybe the best chance for national survival is to threaten everybody else with nukes.

"What line?" would I think have been bomb-virgin USA's attitude. "It's just a bigger kaboom".

It's hard to count the living. I think it's almost certain the cold war would have gone nuclear, somewhere, perhaps only a proxy war at first - but I think it would have become a tit-for-tat habit. I think the damage would have been incalculably worse. Not just the usual stuff like climate and radiation, but also the whole experience of a city as a community rather than as ground zero waiting to happen. It would be a loss of subjective security comparable to the collapse of protected trade routes at the fall of Rome - and that was what wiped out their unitary culture. The Americans in the real world lucked out, so to speak - the war ended so soon, they got to feel guilt.

On the other hand... if only the pacific war hadn't started! It was such a stupid, avoidable war. Innocent people (and beautiful cities) should not be blasted to ash, even for useful lessons. Circumstances should never have reached that pass!

EY: 'Until the end of time we will be a species that has used fission bombs in anger.'

The decision was a calculated political decision made to let the Soviet Union know that we were(are) crazy enough to use nuclear weaponry even when unnecessary. Truman knew the Japanese were running out of time:

"Truman himself eloquently noted in his diary that Stalin would "be in the Jap War on August 15th. Fini (sic) Japs when that comes about."

EY: 'Time has passed, and we still haven't blown up our world, despite a close call or two. Which makes it difficult to criticize the decision - '.

Non sequitur.

Whether or not we blow ourselves up, either with a nuclear device or an AI, has no relationship to that first use of a nuclear weapon. The facts at the time of those 'close calls or two' carry no similarity to the facts of the first use, staring with ALL sides had access to the big nukes during every subsequent 'close call'; that at the time of the 'close calls' that it was common knowledge the effects of nuclear war would have even without the example of the first use; that it was commonly accepted by all that one launch would result in a retaliation.

The decision is easy to criticize. It was used not for any purpose of ending the war, but rather just to send a 'message' from one 'super' power to another 'super' power. It is easy to criticize the decision even if it 'saved' some projected number of one nation's population at the cost of some of another nation's population. It is always in the hands of the 'forecasters' of those about to kill others in the name of some cause, war, religion, etc. to be able to proclaim saving lives or acceptable losses. It's standard propaganda of the winner. (and of the loser, until they lose.)

The first implementation AI is analogous to the the first strike nuclear situation. Fortunately, from what I gather, there will only be one implementation because I'd hate to see an AI Cold War... :)

Off for my hike into the wilderness. Remember: exercise, eat less, sleep well and be good to others. And have some fun.


Maybe I'm falling victim to ex post rationalization, it strikes me that it was pretty much inevitable that the US would use the atomic bomb once it had it.

The thing that saved us from a nuclear war beyond the two bombs dropped in Japan is that it took the US until the early 1950s to get its nuclear weapons production line working smoothly enough to assure sustained high production levels of weapons, and by then the USSR had nuclear weapons as well. The US had originally aimed at being able to produce three weapons a month after August 1945, gearing up to much more in 1946, but the end of the war and problems with uranium and plutonium production meant effectively a two year hiatus in large scale nuclear weapons production, and slow halting production for a few more years, with sustained high level production only after 1950.

So we never got to a situation where the US had both enough atomic bombs to cover its initial target list for the USSR (about 250 targets) with assurance of follow on weapons, and the USSR did not have atomic weapons as well.

So it seems that the real question is how dropping the bombs in Japan affected how management and resources for the US atomic bomb production line were handled. If the US had gotten 1000 weapons assembled by 1950 it might have handled the USSR and China differently. As it was, the US didn't get up to those numbers of weapons two years later.

Actually what is striking is that when confronted with this issue, most people feel superior enough to it to bluntly state what is the right thing to do and the wrong thing to do in this situation, forgetting that wrong and right are actually only relative to the people in question, and there are millions people in question.

That's sort of like setting a standard, and expecting people to fit in this standard, and neglecting those that don't. Who knows that the Hiroshima bomb didn't actually kill the next ten scientists and Nobel Prize winners who would have revolutionized the world, and made it a totally better place to live in? Who knows the real reason why the bombs were dropped anyways? Isn't it bias, that compells you to think that people who did it would have done it for the same reasons you would have done it?

And what was the war about anyways? Isn't it just the result of people putting too much power into a handful of individuals, stupidly believing that their choices are cleverer then their own, forgetting that they are just human, and just biased as them?

Isn't that why democracy is supposedly better?

But everyone prefers to believe that their opinion is the righest one about anything. This is just TOM bias though.

Ah yes, America murders civilians day. Not German civilians, they look too 'American'.