ARMS-D-Request@XX.LCS.MIT.EDU (Moderator) (11/19/86)
Arms-Discussion Digest Wednesday, November 19, 1986 1:22AM
Volume 7, Issue 63
Today's Topics:
Launch on warning
LOW
Reykjavik Mystery
Administrivia
24 hour endurance for bombers
counterstrike?
treaty verifiability
Invading the Soviet Union
ground based missile defenses
----------------------------------------------------------------------
Date: Tue, 18 Nov 1986 12:42 EST
From: LIN@XX.LCS.MIT.EDU
Subject: Launch on warning
From: Clifford Johnson <GA.CJJ at forsythe.stanford.edu>
What will be done is that the President (or military conferees) will
be provided with the opportunity to take a retaliatory decision
based on a 5-way categorization of the attack. That's the "as
pursued" launch on warning policy. My (Oxford American) dictionary
states that a "policy" is "any general plan or course of action
engaged in by a goverment" etc.
We agree that the US government will have the opportunity to take make
a retaliatory decision. Thus, according to your definition, we have a
policy of deciding what to do. We don't have a policy of executing an
LOW, but rather a policy of deciding whether or not to execute an LOW.
There's a preemption targeting policy, and a set of procedures,
which, as with launch on warning, provide for the taking of a
preemption decision.
So by your definition, the U.S. has a "policy of X" if it has a set of
procedures that provide for the taking of a decision of whether or not
to do X. Do I understand you?
In a previous message, you said to me that IF the reports of
incoming missiles could be guaranteed to be true, you
*might* be willing to execute a launch on that warning. Do
you still hold to that position?
No. I used to think a strategic (non-sensor) warning could
conceivably give rise to a certainty. Having since studied the
matter, I agree with the standard authorities who uniformly
conclude that a strategic warning is much flakier than a tactical
warning. So, there being no possibility of such certainty, this
is a discussion I can't logically conceive of any more.
I'm not talking about strategic warning -- I agree that strategic
warning is much flakier than tactical warning. I mean
guaranteed-correct tactical warning.
I don't believe that the operation of an LOWC indicates a
threat of imminent hostilities.
"Imminent"= likely to occur at any moment, according to my
dictionary. I construe the emphasis to be on the "at any moment"
quality, without particular regard to the size of the likelihood,
so long as it is not vanishingly small.
By imminent hostilities, do you mean that the operation of LOWC iteslf
indicates imminenent hostilities (i.e., hostilities that the
capability itself would lead to), or imminent hostilities due ot
other influences, such as Soviet actions? If the first, then you are
right by definition. If the second, then I don't see how.
It's the imminent risk of unintentional
launch due to such things as unreliable sensors that makes
operating a LOWC a form of first-use *at law*.
So if you were guaranteed to have perfect sensors, operating LOWC
would be OK? BTW, I don't understand what "at law" means.
You make my point that game theoretic structures cannot be applied
to *bridge the gulf* between conventional and nuclear war games.
So the simple question is, do you start the game or not?
You try not to. But if deterrence fails, and the Soviets begin to
march into Europe, what do you do? What is your recommendation? If
the choice is nuclear use or surrender, what then? I certainly agree
that we should do all we can to avoid coming to that point, but what
happens if you do?
------------------------------
Date: Tuesday, 18 Nov 86 15:36:23 EST
From: sclafani (Michael Sclafani) @ a.psy.cmu.edu
Subject: LOW
It seems to me that an implication of Bayes Theorem in probability
might have some relevance to this discussion. When a test is performed
for an event of low probability, if the test has even an small
probabilty of returning a false positive, then the chance that a positive
result is valid becomes small. Specifically:
P(attack|warning)
P(warning|attack) * P(attack)
= -----------------------------------------------------------------
P(warning|attack) * P(attack) + P(warning|noattack) * P(noattack)
An arbitary example: Assume we will always detect a real attack, or
P(warning|attack)=1. Assume that P(attack) is 1/10000, since it's not
happened yet. Assume that P(warning|noattack) is 1/1000, since many
cases of malfunction exist. Then, the chance that a warning received
indicates a real attack is about 9%.
Assigning values here is a game that anyone can play. Kennedy is attributed
with estimating the chance of war at 1 in 3 during the missile crisis. The
chance of a false warning is easier to get a handle on. I think this might
be useful to keep in mind when dealing with Launch On Warning.
Michael Sclafani
Carnegie-Mellon University
Undergraduate Student
Internet: sclafani@a.psy.cmu.edu
------------------------------
Subject: Reykjavik Mystery
Date: Tue, 18 Nov 86 15:10:49 -0800
From: crummer@aerospace.ARPA
>>>Gorbachev: Let's eliminate all missiles, and you don't deploy SDI
>>> within 10 years. Reagan: Let's eliminate all missiles, and
>>>I have to deploy SDI
>>> within 10 years.
>>> =====> NO DEAL <=====
>>>
>>>The mystery that I don't understand is: If all missiles have been
>>>eliminated, why would somebody want to deploy SDI?
Perhaps R. Reagan, as apropos his own communication, doesn't take
Gorbachev seriously; just takes his words as he would a used car commercial.
>>>Without balistic missiles, it would be foolish for Americans to deploy
>>>SDI. Therefore Reagan seems to have made a mistake, trying to
>>>maintain SDI even after it (SDI) has been made obsolete and aimless by
>>>missile ban.
This sounds like breakthrough stuff: "A verifiable treaty that
makes Star Wars impotent and obsolete."
>>>However, Gorbachev has also made a mistake by preventing Reagan from
>>>deploying SDI. He should have said, "Go ahead; make my day." If the
>>>deployment were at all feasible. The Americans would have looked very
>>>dumb to the world sending up weapons to shoot at retired missiles.
Unfortunately for our self-esteem Gorbachev's fear of SDI is quite
sensible. According to Maj. Dr. Pete Worden of the SDIO: "... some
of the very same directed energy technologies that we are looking at,
lasers and particle beams, if you will, could be turned against your
defensive components.", and also satellites. Satellites are in
predictable orbits; sitting ducks.
>>>Isn't this conclusion easy to reach? Or, alternatively, maybe there
>>>is a hidden purpose of SDI which would explain their behavior at
>>>Reykjavik? Could somebody illuminate me on the reasons why Reagan,
>>>Gorbachev, and 65% of the American public supporting Reagan's stance
>>>at the pre-summit, believe that the right thing was done at Reykjavik?
My guess is that people think of him as a responsible, truthful,
serious person who wouldn't lie to the public. The evidence seems to
show, to me at least, that he uses words to sway public opinion, not
to communicate the truth.
>>>More concretely:
>>>
>>>1. Why did Reagan insist on SDI when there would be no missiles?
He wants superiority to, not parity with the Soviet Union.
>>>2. Why did Gorbachev insist on the non-deployment of SDI when there
>>> would be no missiles?
He doesn't want his enemy to control space. He wants to launch
rockets without having to clear the launches with Washington.
>>>3. Why did 65% of American public believe Reagan did the right thing?
The American people like Reagan, they WANT to believe him.
>>>
>>> -Calton-
>>> Fearfully ignorant
(I don't think so; not any more so than thi kid who
didn't understand why the emperor was naked.)
--Charlie
------------------------------
Date: Tue, 18 Nov 1986 21:26 EST
From: LIN@XX.LCS.MIT.EDU
Subject: Administrivia
Ignore duplicate issue of #61. The second issue is an unedited version.
------------------------------
Date: Tue, 18 Nov 1986 21:36 EST
From: LIN@XX.LCS.MIT.EDU
Subject: 24 hour endurance for bombers
From: Lin
> Are you saying that if we wished to spend the money, we could not
> maintain an airborne alert? If you concede this, then it seems my
> suggestion is technically feasible; expensive, perhaps, but feasible.
From: hplabs!pyramid!utzoo!henry at ucbvax.Berkeley.EDU
... Airborne alert is feasible, but
doing it on any significant scale is very hard on planes and crews.
Doing it as a matter of policy would probably mean major expansions in
the bomber and tanker forces, to keep the fatigue-life problems under
control.
We agree. It costs something to buy a 24 hour response time. You
have conceded my point.
------------------------------
Date: Tue, 18 Nov 1986 21:40 EST
From: LIN@XX.LCS.MIT.EDU
Subject: counterstrike?
From: osmigo1 at ngp.utexas.edu (Ron Morgan)
In the above event (a large nuclear attack on the US),
specifically WHAT is the rationale for a
counterstrike?
GIVEN an attack on the US, there may not be any rationale. But you
don't base your defense strategy on the stated premise that you would
do nothing if they attacked. Rather, you practice and prepare for
doing something that they would not like you to do them IF they did
anything to you.
------------------------------
Date: Tue, 18 Nov 1986 22:18 EST
From: LIN@XX.LCS.MIT.EDU
Subject: treaty verifiability
From: rutgers!meccts!meccsd!mvs at seismo.CSS.GOV
Brezhnev promised its production
rate [Backfire] would not exceed 30 a year. There is some evidence that
production is higher, and that at least two squadrons are positioned
close to the artic circle where they can have extensive coverage over
US targets.
The latter is legal, and the former is not militarily significant,
even if true.
There are still questions about the actual capabilities of the SS-20
missile. There is some speculation that its range is actually greater
than the claimed 3000 miles.
There is also speculation that the Soviets try to use telepathy to
find submarines.
If the Soviet's
aren't hiding anything on the SS-20 then the question is why they
conducted tests under the cover of night and encrypted the telemetry
data. Without on-site verification we simply do not know these sorts
of answers.
Even with OSI, we could not tell. We also encrypt telemetry. Testing
at night doesn't hide much of anything.
There is some question as to whether the Soviet's have been
hiding research into a strategic ABM system under the guise of the
allowable air-defense program.
No evidence on this score is conclusive. The only possibility is the
SA-X-12, which even the DoD has said in testimony does not violate the
treaty.
... the ABM treaty says that the Soviet's are not permitted to
test their air-defense missiles against incoming ballistic missiles
and must not deploy radar that could track incoming warheads.
You don't know what the treaty says. It specifically says STRATEGIC
ballistic missiles. Radar is not limited in the way you describe it
either.
During the late 70's evidence started coming in that the Soviet's were
working on a particle beam weapon ABM system based on low yield
nuclear weapons. This provided the inital impetus to what has become
known as SDI.
No one knows the first. There were reports of particle beam weapons
in 1977, but no one knows that it was for ABM. The impetus to SDI was
hardly that.
Supposedly during the inital SALT I discussions, the US told the
Soviet's about the size of our arsenal and asked for them to do the
same. But the Soviet's couldn't tell us the size of their arsenal,
because they claimed they didn't know.
That's not surprising. In fact, the story is that the U.S. began to
tell the Soviets about the Soviet arsenal, and some Soviets asked the
U.S. to not reveal such information, since not all Soviets were
cleared to have such information.
The CEP for a Trident I missile is about 450 meters.
Trident II will be on line in 1989, and will have a 475 KT warhead
with 400 foot accuracy. That IS accurate enough for a hard target
kill counterforce capability.
... the probability of arrival for
submarine missiles is equal to about 80% of the reliability of the
overall system. This makes them (and the eventual successors like the
Trident II) even less usable for a first-strike. You can be certain
that the Soviet's are also aware of this.
Such is true for the Soviet Union as well, thereby making a Soviet
first strike less plausible too.
------------------------------
Date: Tue, 18 Nov 1986 22:25 EST
From: LIN@XX.LCS.MIT.EDU
Subject: Invading the Soviet Union
From: Eugene Miya N. <eugene at ames-nas>
...Depending on damage level of the Soviet Union, do we have invasion
plans (conventional with perhaps some tactical nuc capability?)
Actually, the new naval maritime strategy postulates Marine amphibious
assaults in two places on the SU during the latter phases of the war.
------------------------------
Date: Wed, 19 Nov 1986 00:52 EST
From: LIN@XX.LCS.MIT.EDU
Subject: ground based missile defenses
From: cfccs at HAWAII-EMH
... under the ABM treaty, both sides are allowed certain anti-missile
defenses. The Soviet Union currently has a defense system of this type
around Moscow (I don't know if other cities are protected).
No other Soviet sites are protected.
No one has mentioned this in their
talks on firing missiles at the S.U.
100 interceptors are allowed. Those are irrelevant in a world of 8000
strategic ballistic warheads.
------------------------------
End of Arms-Discussion Digest
*****************************