Doomsday debated

Eliot Gelwan in his blog Follow Me Here posted a piece, entitled “They're just lying, I'm sorry to say,” which links to Impearls' depleted uranium article (permalink), calling it “A pretty damning indictment of concerns from the left over depleted uranium weapons, if you can believe the science cited here, well-documented and footnoted.”

In that same posting, however, Gelwan goes on to link to and criticize Impearls' subsequent piece on Doomsday (permalink), writing:

McNeil […] scoffs at the idea that a nuclear holocaust would mean doomsday for the human race, with a much more selective appraisal of the evidence [than in the depleted-uranium article], clearly in the service of his ideological biases.

I initially posted a reply as a comment on Eliot's blog, to wit:

Speaking as the person targeted by these comments, no doubt I do have ideological biases; however, I do not believe the idea that nuclear arsenals as they existed during the cold war and today were/are insufficient to trigger human extinction is very controversial.  Notice that renowned physicist Freeman Dyson in his reply had no criticism to my comments in this regard.

Moreover, I have no doubt, as I clearly expressed in that piece, that human extinction could be caused by means such as “gigaton mines” or similar, relatively cheap device — a Doomsday Machine — deliberately designed to accomplish that goal.  It could do this perhaps by, among other things, shattering the earth's crust and releasing huge volcanic eruptions, surpassing even the geologic “Deccan trap“ occurring around the time of the demise of the dinosaurs.

Soon after, Gelwan replied again, in his comments, which I'll quote in full:

I have concerns about your confidence.  The human track record for anticipating the full range of cascading, immensely complex sequelae from our acts of violence against the ecosphere is not very good.  Inherently, one can easily underestimate but it is difficult to overestimate, since we continually demonstrate we know so little about the consequences of our actions.

And, for every esteemed Freeman Dyson (to whom, I might add, the question does not appear to have been put directly; you infer his assent from the fact that he had no criticism) there have been perhaps fifty scientists who have actually worked on nuclear weapons development and related fields who disagree.  This is what, for example, the Union of Concerned Scientists were concerned about.  This is what the Bulletin of the Atomic Scientists issues bulletins about.  Should I anticipate your dismissing all their concerns because they are leftists?

One way to do risk assessment is to weigh possible outcomes by, roughly, the product of likelihood × severity.  Likelihood alone is not a sufficient basis for planning to meet risks.  Even if I agreed with your low likelihood estimation (and I do not), in and of itself it is not a sufficient basis to reassure thoughtful people about the doomsday risks.

I don't want to get into a flame war with you; this is a little like arguing about religion.  We'll have to agree to disagree on most things.  If the contemptible Bush administration has its way in reversing the progress the world has made toward abolishing the risks of nuclear annihilation, you're right, I probably won't see you On the Beach, but I probably will in Nuclear Winter.  Thoughtful people should be questioning questionable assumptions and pat reassurances on this issue, especially when they arise from a partisan ideological agenda, as if their life depended on it.  It does.

Eliot says he has “concerns about your confidence.”  My “confidence,” as he calls it, my “scoff[ing] at the idea” as he characterized it in his earlier posting, consisted of my saying I considered it “unlikely” present or cold war-level nuclear arsenals would instigate human extinction — I then went on to discuss potential ways the arms race might have proceeded which could, had things developed differently, have triggered Doomsday.  This is hardly scoffing at the very idea or danger of Doomsday, as he implies, quite the contrary.  Moreover, my piece describes the extreme effects even a non human-extinctifying (to coin a term) cold war-era nuclear exchange would certainly have had for the participant nations, going on to quote Freeman Dyson that “nuclear war means death” and “it is a truth which we must never forget.”  Getting from there to the conclusion that I'm minimizing the risks from nuclear war is a monumental non sequitur, in my opinion, and Gelwan's comments display a sizable “ideological bias” (to pick up a deprecating term he tosses around) of his own — not to speak of that exhibited in the last paragraph!

Gelwan proceeds with this breathtaking prose:  “The human track record for anticipating […] is not very good.  Inherently, one can easily underestimate but it is difficult to overestimate […].”  On the contrary, it is environmentalist disaster predictions which have failed miserably (fortunately for us all) over the last several decades in living up to their expectations.  While I agree in principle with the first part of Eliot's statement, the converse is also true: environmentalists' records in anticipating, e.g., the manifold, interlocking strengths of the ecosystem, or the technological dynamic in the search for mineral resources (finding new sources even as known reserves are depleted), etc., etc., are also quite poor.  Given the disposition among environmental activists to keep the rubes ignorant (as Gelwan baldly says, “in and of itself it [truth] is not a sufficient basis to reassure thoughtful people about the doomsday risks”), as well as environmentalists' inclination to listen only to each other's (given the previous point, what ends up being) spin, in my view guarantees overestimating the risks.

Eliot wonders what Freeman Dyson's views are with regard to the extinctifying potential of present or cold war nuclear arsenals.  In his book Weapons and Hope (1984), Dyson wrote:  “I am unable to imagine any chain of events by which our existing nuclear weapons could destroy mankind and leave no remnant population of surviviors.”  Gelwan imagines that fifty times as many physical scientists would agree with him as Dyson in this regard.  In my view that's a ridiculous assertion; I believe the actual ratio for physicists would be well in the reverse direction, but that's neither here nor there — being a nuclear scientist doesn't give one special expertise or aptitude for predicting most environmental “sequelae,” as Gelwan terms them, to nuclear war.  As for Gelwan's attempted put-down of Dyson, implying he never “actually worked on nuclear weapons development and related fields,” on the contrary Freeman Dyson was one of the principal actors in the Orion spaceship project.  Eliot might recall what was to propel Orion: mini-atomic bombs.

Getting back to Doomsday, the real danger, I think — and Dyson agrees, or rather I agree with him in this regard (see Weapons and Hope, pp. 22-23) — is not that a nuclear exchange say between Pakistan and India, or even between Russia (or the old Soviet Union) and the United States, would trigger subsequent human extinction.  The danger, Dyson writes, is that we would survive an initial nuclear exchange; but decide, more or less, hey that's not so bad, we can take this… then go on to fight more nuclear wars, exchange after exchange, waged upon an ever increasingly poisoned and dessicated earth — until ultimately human extinction would follow.

That's why it is so important that even a limited nuclear exchange (such as possible between India and Pakistan, clearly short of Doomsday) must not be undertaken in the first place.  Trying, as Gelwan seeks to and “atomic scientists” at Bulletin of the Atomic Scientists do, to scare thoughtful people by suggesting that Doomsday is plausible in such a scenario will simply shred further the already much-tattered reputation of environmentalist “doomsday-mongers.”

As for how reasonable it is to suppose even a cold war-level nuclear arsenal would be capable of instigating human extinction, let's do a little sanity check.  The total size of the nuclear arsenal as it existed during the cold war may be estimated at about a gigaton; that's close enough, a factor of a couple either way wouldn't make any difference to the discussion.  The explosive power of a ton of TNT is equivalent to 4.184 × 10⁹ Joules of energy.  A gigaton of TNT therefore contains 4 × 10¹⁸ Joules.  According to this 1997 article in the journal Nature (requires subscription or pay-per-view), the impact which created the Chicxulub crater on the northern tip of the Yucatan peninsula of Mexico 65 million years ago (exterminating the dinosaurs along with three quarters of all species alive, land and sea, killing all creatures larger than about 10 kg or 22 lbs. body weight, which included all the dinosaurs) had an energy of 5 × 10²³ Joules.  Thus the Chicxulub impact — triggering the only known “nuclear winter” during the last more than hundred million years on earth — had an energy over 100,000 times greater than the cold war nuclear arsenal.  How likely is it then a “mere” gigaton (comparable to a magnitude 8 earthquake, which occurs every few years) could accomplish it?

No, I suggest the real danger lies as laid out above: that nuclear war might “catch on” as both nationally survivable (at least when taken in moderation, like fugu) and politically advantageous.

Doomsday postponed

The Common Sense Pundit (of Melbourne, Florida) e-mailed this comment in reply to Impearls' earlier piece on Doomsday:

I read Herman Kahn's book while in high school.  Amazing that a small town library in rural Iowa would have such a book.  If I remember correctly, he had a preface or introduction explaining that it was important for nuclear armed nations to understand the implications of nuclear weapons.  Looking back, his analysis pointed to the eventual outcome — a MAD pact and an arms race.  He made a good argument that if two nations are relatively matched in weaponry, then he who fires first loses.

I am not familiar with Dyson's Weapons and Hope.  But I think we should send a copy of Herman Kahn's On Thermonuclear War to all nations developing nuclear weapons.

I also wonder that had it not been for the two bombs dropped on Japan, would the US and USSR have effectively and unwittingly created a Doomsday Machine.  What then would have happened during the Cuban Missile Crisis?

Dyson suggests there wasn't time enough by the 60's — fortunately — for any nation to have assembled the megatonnage necessary to sterilize the earth.  Unfortunately, on the other hand, I'd say that Common Sense Pundit's point remains for the somewhat longer term.  Dyson's comments as to the relative cheapness and technological feasibility of even extreme weapons such as gigaton mines or a Doomsday device are eye opening.  It leaves one troubled that yes, the (original) cold war is over, but obviously the danger has not evaporated for the long haul.  (Let's get those Big Trees growing on comets!)

Common Sense Pundit goes on to write:

Also, given how many strategic errors Saddam has made, you see why the world must do everything possible to keep nukes out of the hands of absolute dictators surrounded by yes men.  [2003.08.31 18:47 UT.]

Absolutely!  If Saddam were to have gotten operational nukes and done something stupid with them (use your post-September 11 imagination…), America very possibly would have been forced — by public opinion if nothing else — to nuke him “good” in return.  Even so inherently gentle and civilized a being as Eugene Volokh, in a thought-provoking piece in The Volokh Conspiracy last year entitled ”THREATENING MASS NUCLEAR KILLING,” stated the case thusly:

[T]he proposal [to nuke Iraq if Saddam threatens us with nuclear blackmail] — like nuclear deterrence generally — would require us to kill hundreds of thousands of Iraqi civilians.  It would also require us to threaten to kill millions more, by bombing Iraqi cities.  These would be millions of Iraqi civilians slaughtered, not just as collateral damage caused by an attack on a military target, but precisely in order to kill them, with the intention of getting the Iraqis to surrender.

This is not a terribly nice thing to do.  Would we have to do it?  Yes.  Would we be justified in doing it?  I think so.  But it's the sort of justification that might justify a starving man in a lifeboat killing an innocent fellow passenger in order to eat him.

We would, I think, consider it morally excusable, but surely it's not morally wonderful.  As America keeps pointing out, there's a big difference between (1) killing civilians as an unfortunate side effect of an attack on a military target — an attack that is made as precise as possible, in order to kill as few civilians as possible — and (2) killing civilians in order to inflict damage on the other side, and to thus beat it into submission.  And the logic of nuclear deterrence (as opposed to the logic of a preemptive attack today on Iraq) is precisely #2.

Now if nuclear deterrence fully succeeds, then the moral difficulty is avoided:  The threat of killing civilians by the millions prevents the need to actually kill civilians by the millions.  But the problem is that the more humanitarian one appears, the more one appears to be concerned about the opinions of “the international community,” the more one appears to care about the views of “the Arab street,” the more one subscribes to the view that in this interconnected world we ought not try to make too many mortal enemies, the less credible the deterrent will be.

After all, to make good on the deterrent, we have to be willing to kill hundreds of thousands, or millions, of innocents.  We have to be willing to make mortal enemies of those who will understandably be upset by those deaths.

Read the whole thing.  At the time I replied to Eugene's piece with a personal e-mail (as Impearls had not yet been launched), which I'll restate here:

I too have been concerned that the deterrence value of our nuclear arsenal (which after all is merely a matter of confidence among the peoples of the world that we'd use that weaponry) has been eroding, and that if we actually needed to use those arms, after a nuclear attack on America say, we'd become mass murderers.  As you suggest, it might be understandable perhaps that we would use those arms to attack the cities of the enemy after a nuclear attack on us, but we would still be attackers of innocents in the mass — and, as you say, civilians in this case would be the primary targets, not unintended consequences.

To avoid such a moral disaster for us, I too believe preventive war before Saddam achieves nuclear status carries far less risk for the American people.  It's both physically safer for us, and morally safer for our reputation and consciences through the rest of time.

My view is that even if America were attacked with nuclear forces, we ought not to massacre the enemy's people en masse, but rather (through judicious use of nukes and much other weaponry and soldiery) invade, occupy, and root out the regime that attacked us.  Let the certainty that that will happen be the new deterrence against an attack on us with a weapon of mass destruction in the first place.

Thus the invasion and occupation — the liberation (without scare quotes) — of Iraq, in my view, was an entirely justifiable act whose overriding justification lay in forestalling the likelihood of having to take much more “vigorous” (and morally stained) action in the future.  Thank God (and the United States of America) that such a fate, a tragedy for the entire world, was avoided in the case of Iraq.  Let the deterrent value of this innovation in international relations make dictators pause before embarking on such reckless adventurism as Saddam Hussein's in the future!