Interviewer:
TOWARD THE END OF THE
SIXTIES, SAC HAD REALLY PRETTY LARGE NUMBER OF WEAPONS. AND PRETTY WEAPONS AT THAT. AND THERE
WERE A LOT OF CRITICISMS VOICED BY PARTICULARLY THE NAVY, ARMY, AND THE PRESIDENT'S SCIENCE
ADVISERS WHEN THEY GOT A LOOK AT THE SAC WAR PLANS...THAT THIS WAS LUDICROUS OVERKILL. THAT THE
AMOUNT OF WEAPONRY THAT WAS PLANNED TO BE LAID DOWN AND THE MEGATONNAGE WAS JUST FAR GREATER
THAN COULD POSSIBLY BE HAVING ANY MILITARY PURPOSE, WHAT WAS SAC'S ANSWER TO THOSE CRITICISMS?
AND WHY DID YOU NEED SO MUCH?
Martin:
That is probably one
of the most complicated questions that you could raise from the standpoint of the technicalities
of strategic targeting. I think the subject of overkill had several fathers. First though, let
me just say why so many weapons were judged to be needed to knock out a number of targets. And I
might add, a great number of targets. In the first place, it's a, it's an exercise in
mathematics. And without computers, I dare say that it would be almost impossible. But all of
the probabilities have to be taken into account. For example, will the airplane or the missile
be ready to take off if we have to retaliate? You assign a percentage of likelihood to that,
based on maintenance conditions, and other factors that might prevent that weapon from
launching. Then from the point of launch, right on through the entire mission, the probability
of successful completion of that phase of the mission must be assessed, estimated. And I might
say that all of the many operational exercises are analyzed to arrive at the best possible
estimates of those probabilities. Because you rarely accomplish a mission in one hundred percent
perfect completion. Perhaps that's an illustration of Murphy's Law. If something can go wrong,
it'll go wrong. But in any case, the delivery vehicle is just one part of the problem. The
weapon itself, whether again, whether a warhead on a missile, or a bomb on an aircraft, has the
same number of probabilities of successful performance assigned to every phase of it's
operation. Another factor is weather. So when you got all through, you find that you don't have
any real confidence in a one on one, that is, one weapon against one target, probability of
success in accomplishing what you're trying to do in terms of damage to the enemy. Therefore, to
use a very simple example, you might find that it takes two or three weapons, against one
target. Now that's one phase of the targeting problem. The other is somewhat in reverse. You
take a look at the target. How accurate is your information on that target: (a) the location;
(b) the size; (c) the nature, or, in the jargon, how hard is the target? What kind of a blast
effect, or force from a weapon could it stand and survive? These things are all quite
questionable, and they must be based on the best intelligence available from all sources. So
again, the probability of kill to use the targeting language for a moment, the probability of
kill, depends on delivery vehicles, weapon characteristics, and target characteristics. And when
you got all through, looking at the nature and number and types of targets, that should be hit
in order to accomplish the mission, a great many more weapons are needed than you would suspect
from just looking at the problem on the surface. Now, that adds up to a sizable force of
strategic forces. A sizable investment in defense dollars, in tax payers dollars. If the
priority of those strategic forces tends to interfere, or reduce the budget of other forces, you
can expect objections to that expenditure for strategic forces. There are only so many budget
dollars, even though there are often a great many. There still is a limit. So in those parts of
the military structure of the United States that do not participate directly in the strategic
forces, one could expect objections to great numbers of forces, great numbers of weapons. That
is one illustration of I think where the term overkill was born.