During the Cold War, there were some cases where nuclear exchange nearly resulted due to faulty technology, which faults and nuclear war were only narrowly averted. 1 case IIRC was in 1979, when a US tracking station in Alaska apparently mistakenly read a flock of birds as a Soviet incoming nuclear strike. How would the world have looked had the superpowers launched nuclear strikes on each other based on such mistaken intel ? Could 'accidental' nuclear war still have stopped short of MAD had such faults been later discovered and rectified by the power in question ?
'Accidental' nuclear war
- Thread starter Melvin Loh
- Start date
Gad, that's a pretty nightmarish scenario. I guess it depends on when it happens and what goes wrong. In the 50s, you can call back thebombers (and even if one or two don't believe you AND the other side can't taske them out even with your help, you lose a few cities). By the 60s you have missiles to reckon with. There's no 'Undo' button on a missile launcher. And by the 70s, you've got overkill capacities in the thousands.
The other question is, what goes wrong. If, frex, the US assume a massive Soviet strike as they did in '79 (or the Soviets trace a NATO surprise attack in the making because their foreign ministry didn't inform them of scheduled maneuvers, as happened IIRC 1982) then the doctrine of MAD calls for immediate retaliatory strike. Use them or lose them. At whixch point it almost doesn't matter if the other side gets warned or not, launches or not. The difference between New York and Moscow will, at best, be a few weeks more to live.
However, if a mistake results in a limited launch - say, single submarine or a single missile - then you have a bare chance of making it. If the communication systems work. Think about it: the Army Missile Forces need to call the Pentagon, the Pentagon the President, the President the Secretary General of the CPUSSR, he his strategic rocket forces, and all in less time than it takes for an ICBM to cross that Atlantic. It would make a great spy novel, except it's been done.
And the guilty nation will SO pay through their noses for this...
The other question is, what goes wrong. If, frex, the US assume a massive Soviet strike as they did in '79 (or the Soviets trace a NATO surprise attack in the making because their foreign ministry didn't inform them of scheduled maneuvers, as happened IIRC 1982) then the doctrine of MAD calls for immediate retaliatory strike. Use them or lose them. At whixch point it almost doesn't matter if the other side gets warned or not, launches or not. The difference between New York and Moscow will, at best, be a few weeks more to live.
However, if a mistake results in a limited launch - say, single submarine or a single missile - then you have a bare chance of making it. If the communication systems work. Think about it: the Army Missile Forces need to call the Pentagon, the Pentagon the President, the President the Secretary General of the CPUSSR, he his strategic rocket forces, and all in less time than it takes for an ICBM to cross that Atlantic. It would make a great spy novel, except it's been done.
And the guilty nation will SO pay through their noses for this...
The real question is where the failure occurs. If it is in the early warning systems (the flock of birds, or a gas plume from an oil refinery, etc.), then there is a real chance of a mass launch. If that happens, we are all toast (albeit glowing toast), as it is unlikely that the damage/response could be contained at an acceptable level.
I am not sure how you can get a 'single' launch (i.e., some sort of failure in the chain of command), were one missile/bomber/etc. is used against the 'bad guys', since these systems all have PALs, and thus require some sort of authorization for use. Subs and missile silos both get orders to execute orders, as it were (Subs listen for messages at preset times, missile silos get the orders from their command centers), while bombers have their well-known 'fail-safe' points. Short of some sort of massive software screw-up (unlikely, given how many layers of checks and double-checks exist, and how procedures have been evolved to deal with this), I think that it is unlikely you would see this sort of failure event.
Now, rather than looking at the later part of our atomic age (say 1975 and later), what about the early period? Single launch failures, runaway bombers, rogue commanders, etc, would have been far, far more likely in the 50s and early 60s. Everything was slower (bombers took hours to get to their targets, missiles tended to be liquid-fueled, and thus took quite some time to launch, etc.), and the comm networks of the time weren't geared to rapid-fire response times, hence giving leaders some time to think about their actions...
I am not sure how you can get a 'single' launch (i.e., some sort of failure in the chain of command), were one missile/bomber/etc. is used against the 'bad guys', since these systems all have PALs, and thus require some sort of authorization for use. Subs and missile silos both get orders to execute orders, as it were (Subs listen for messages at preset times, missile silos get the orders from their command centers), while bombers have their well-known 'fail-safe' points. Short of some sort of massive software screw-up (unlikely, given how many layers of checks and double-checks exist, and how procedures have been evolved to deal with this), I think that it is unlikely you would see this sort of failure event.
Now, rather than looking at the later part of our atomic age (say 1975 and later), what about the early period? Single launch failures, runaway bombers, rogue commanders, etc, would have been far, far more likely in the 50s and early 60s. Everything was slower (bombers took hours to get to their targets, missiles tended to be liquid-fueled, and thus took quite some time to launch, etc.), and the comm networks of the time weren't geared to rapid-fire response times, hence giving leaders some time to think about their actions...
