Commenting on “doomsday” scenarios such as illustrated in the 1950s film On the Beach, in a discussion on a science fiction mailing list, several people argued that human extinction instigated by thermonuclear warfare, caused directly by either a "nuclear winter" (a result of kicked-up dust blocking sunlight for extensive periods) or ionizing radiation from fallout, is just not in the cards.  The thread went this way:

My point (and I'm not altogether certain I made it well) was that we don't possess the kind of power that's required to do much more than affect things locally or in the very short term — even with our entire nuclear arsenal.
<snip>
Apparently, someone has calculated the amount of radiation needed to kill off the human population and compared that to the amount of nuclear weaponry we have.  This supports (but does not prove) my supposition.

I disagree.  It does not support your supposition, because it is unrelated.

Actually, it's not unrelated at all.  [His] original point was that there isn't enough “horsepower” as he puts it, using ANY weapons effect mechanism, to sterilize the earth.  He may have said this in a thread about nuclear winter, but he did not limit his comments to that mechanism.
 

There are at least three modalities of human extinction that should be considered here (more, if say extraterrestrial sources such as asteroidal impacts are to be included) — only a couple of variants of one of which have been discussed so far:

1. a.  Extinction due to a “nuclear winter,” instigated by dust thrown up by a thermonuclear exchange based on present nuclear arsenals, or those which existed during the cold war.

1. b.  Extinction due to radiation spread around the world by fallout, due to a thermonuclear exchange based on present nuclear arsenals, or those which existed during the cold war.

2.  Extinction due to a thermonuclear exchange based on nuclear arsenals contemplated around the time of On the Beach.

3.  Extinction resulting from the triggering of a “Doomsday Machine” a la Dr. Strangelove.

I agree with commenters that 1a & b above are unlikely to result in anything like human, much less biosphere, extinction.  As an aside, however, the effects on the protagonists themselves of tens of thousands of nuclear weapons going off in essentially all population centers in the land would still be extreme.  I heard once (I've lost the reference, sorry) that those tens of thousands of weapons were directed so broadly that at least one nuclear bomb was targeted on a vacant lot that might have been used as an airfield in an emergency.  It's hard to survive that, unless you're far from any, even the very smallest, populated centers.  That's still not extinction for the human race, however.

Physicist Freeman Dyson discussed the nuclear arsenals contemplated during the late 40s and 50s in his thought-provoking book Weapons and Hope: ¹

The nightmare, which caused Conant and Oppenheimer to oppose the hydrogen bomb so passionately, was an arms race driven by the forces of economics and deliverability toward monstrously large bombs.  They expected that each bomb would require a large investment of scarce and expensive tritium to ignite it, so that no country could afford many such bombs.  Economics would then dictate that each bomb be fueled with a huge quantity of cheap deuterium to justify the initial investment.  The bombs would then become so large that they could barely be carried in airplanes; Oppenheimer remarked in a letter to Conant shortly before the 1949 committee meeting:  “I am not sure the miserable thing will work, nor that it can be gotten to a target except by oxcart.”  The problem of deliverability would then dictate that the bombs be installed in submarines or surface ships and detonated offshore, devastating great areas of country with tidal waves and radioactive fallout.  But then the installation of shore defenses would force the bomb designers to move the point of detonation farther out to sea, and the bombs would have to be made still larger to do the same damage from a greater distance.  The vicious cycle of the arms race would then continue, bombs growing less deliverable as they became bigger, and growing bigger as they became less deliverable.  The only end to such an arms race would be bombs the size of submarines, each having a yield of many thousands of megatons.  The cheapness of deuterium makes the construction of such bombs technically and economically feasible.  They were called “gigaton mines” by the people who worried about them in the 1950s.  A gigaton is the technical term for a thousand megatons.  The construction of gigaton mines would indeed be, in Conant's words, “a threat to the future of the human race which is intolerable.”

This nightmare produced a literary response which has continued to reverberate ever since.  In 1957, only two years before his death, the novelist Nevil Shute Norway published On the Beach, a description of mankind wiped out by radiological warfare.  […]

The myth of On the Beach, like Jonathan Schell's myth, is technically flawed in many ways.  Almost all the details are wrong: radioactive cobalt would not substantially increase the lethality of large hydrogen bombs; fallout would not descend uniformly over large areas but would fall sporadically in space and time; people could protect themselves from the radioactivity by sheltering under a few feet of dirt; and the war is supposed to happen in 1961, too soon for even the most malevolent country to have acquired the megatonnage needed to give a lethal dose of radiation to the entire earth.  Nevertheless, the myth did what Norway intended it to do.  On the fundamental human level, in spite of all the technical inaccuracies, it spoke truth.  It told the world, in language that everyone could understand, that nuclear war means death.  And the world listened.

If the hydrogen bomb had led to an arms race of the kind which Conant and Oppenheimer most feared, with undeliverable bombs growing bigger and bigger until they became gigaton mines, then the scenario of On the Beach might ultimately have come close to reality.  Gigaton mines could, one way or another, make our planet uninhabitable.  This is the truth which Norway's story brought home to mankind.  And it is a truth which we must never forget.

The last possibility (literally) to be considered is that of a “Doomsday Machine.”  Dyson went on to discuss this: ²

Gigaton mines were also in Herman Kahn's mind when he published his book On Thermonuclear War in 1960.  In that book he created a new myth, the Doomsday Machine, which was intended to be a reductio ad absurdum of the idea of deterrence.  The Doomsday Machine is a device designed to deter nuclear war with absolute certainty by making the cost of aggression infinite.  It is the final theoretical step in an arms race which begins with hydrogen bombs and runs in the direction of gigaton mines.  Let Herman Kahn describe it in his own words:

A Doomsday weapons system might be imaginatively, and entirely hypothetically, described as follows:  Assume that for, say, ten billion dollars we could build a device whose only function is to destroy all human life.  The device is protected from enemy action, perhaps by being put thousands of feet underground, and then connected to a computer which is in turn connected, by a reliable communication system, to hundreds of sensory devices all over the United States.  The computer would then be programmed so that if, say, five nuclear bombs exploded over the United States, the device would be triggered and the earth destroyed….  […]

Herman Kahn's On Thermonuclear War may have been a classic, but it was not a best-seller.  His historical erudition and ironic style were not designed to appeal to a wide audience.  The public learned about Doomsday Machines through the art of Stanley Kubrick, whose film Dr. Strangelove or: How I learned to Stop Worrying and Love the Bomb presented Kahn's idea in the language of black comedy.  Here is Soviet Ambassador De Sadeski, speaking to American President Muffley:

“A Doomsday Machine, gentlemen.  That's what I said and that's what I meant.  When it is detonated it will produce enough lethal radioactive fallout so within twelve months the surface of the earth will be as dead as the moon….  There were those of us who fought against it, but in the end we could not keep up in the Peace Race, the Space Race and the Arms Race.  Our deterrent began to lack credibility.  Our people grumbled for more nylons and lipsticks.  Our Doomsday project cost us just a fraction of what we had been spending in just a single year.  But the deciding factor was when we learned your country was working along similar lines, and we were afraid of a Doomsday Gap.”

From what I've seen, in Dyson's book and elsewhere, a Doomsday Machine is technically not all that difficult, but would as “Ambassador De Sadeski” said, probably cost much less than the Pentagon's current strategic budget repeated year after year, decade after decade.
 

Postscript.  Since I find myself in the astonishing position of being one of Freeman Dyson's publishers (and for my favorite work of his!), I shared the above exchange with him.  Freeman graciously sent back a response, to wit: ³

I am delighted that somebody still reads Weapons and Hope, although history has made a lot of it obsolete.  The main thing that has changed since the book was written, and is relevant to your discussion, is the retirement of all the multimegaton weapons.  Both in the USA and in Russia, the military people understand that half a megaton is big enough for any conceivable military purpose.  So there are no longer any weapons bigger than that in the stockpiles.  The irony of this situation is that the stockpiles today are almost exactly the same as they would have been if the hydrogen bomb had never been invented.  It is easy to make a pure fission bomb with a yield of half a megaton.  So the whole fight of Oppenheimer and his friends against the hydrogen bomb was in the end unimportant.  If they had won it would have made very little difference.
 

P.P.S.  After getting the above kind note from Freeman, I forwarded the following back to him, which I believe with conviction:

With regard to your book Weapons and Hope.  History may have left it behind — but the book, in turn, helped to create that history.  It's fashionable nowadays to credit Ronald Reagan (or Mikhail Gorbachev) for bringing about the situation whereby the cold war came, mostly peacefully, to an end.  In retrospect I've come to believe that I misjudged and underestimated Reagan (though I'm not sure how you care for being placed in the same sentence with him), but in my view your book (and you) are deserving of considerable credit.  I recall an editorial in Science strongly emphasizing its importance.  I believe the book had an impact, and thank you most effusively for it.

The vital, flickering flame that Dyson's Weapons and Hope encouraged and sustained during those blood-curdling days of the cold war (as indeed was its intent) was hope — hope that there could be a non-cataclysmic end to the terrifying nuclear standoff, which frankly makes anything nowadays seem trivial.  (Take the India-Pakistan faceoff, for example.  A handful of Hiroshima-style bombs threaten in that theater, versus in the cold war tens of thousands of precisely aimed, super-powered hydrogen bombs.  There's virtually no comparison.)

Freeman Dyson the savior of civilization?  Certainly, credit for the safe ending of the cold war must be spread widely.  I'm convinced, however, that Dyson's effect was significant and enduring, and even without actual human extinction having been at risk, many people might reflect that, to an appeciable extent, they owe a debt to Freeman Dyson for their lives.

Am I being too melodramatic here?  Perhaps.  One thing is for certain, however: now we have the opportunity to spread out to the stars, and grow the Big Trees that Dyson envisioned — on comets! ⁴
 
 

References

¹ Freeman Dyson, Weapons and Hope, Harper & Row, New York, 1984, ISBN 0-06-039031-X (U.S. and Canada), LOC U21.2.D94 1984; pp. 32-34.

² Ibid., pp. 34-36.

³ Freeman Dyson, personal correspondence, 2003.08.14.

⁴ Freeman Dyson, The World, the Flesh, and the Devil, Impearls, 2002.11.12 (originally presented 1972.05.16).
 

UPDATE:  2003.09.03 22:16 UT.  See follow-up article Doomsday postponed (permalink).

UPDATE:  2003.09.04 18:30 UT:  Reader M. Simon comments that this “neglected to mention the effect of improved ICBM accuracy on bomb size.”  He's perfectly correct, that was an important factor in the decline in required megatonnage (or should we now say kilotonnage) for nuclear weapons.

UPDATE:  2003.09.10 16:52 UT.  See follow-up article Doomsday debated (permalink).