Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/17/84; site mhuxt.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxt!js2j
From: js2j@mhuxt.UUCP (sonntag)
Newsgroups: net.politics
Subject: Re: poll  (nuclear disarmament verifiability)
Message-ID: <692@mhuxt.UUCP>
Date: Mon, 18-Mar-85 14:22:01 EST
Article-I.D.: mhuxt.692
Posted: Mon Mar 18 14:22:01 1985
Date-Received: Tue, 19-Mar-85 05:47:25 EST
References: <5202@ucbvax.ARPA> <386@abnji.UUCP> <5544@ucbvax.ARPA> <634@tty3b.UUCP>
Organization: AT&T Bell Laboratories, Murray Hill
Lines: 42

> I disagree that we are "beyond verification".   If your requirement is 100%
> reliable verification, then we've always been beyond verification and arms
>control is impossible.  I think most people who want 100% verifiablity actually
> want 0% arms control. 
> Mike Kelly

    I'd agree with a slightly changed statement:  I think most people who want
100% verifiability actually want 0% arms.  In other words, Mike, I'd rather
not have 90% verifiability and 10% of enough nuclear  weapons to destroy the
earth N times in the hands of governments who are 90% sure that the other
guy is disarmed.
    As an example, suppose you were in a Mexican standoff with someone who
you trusted not at all.  Suppose a cardboard barrier prevented you from 
seeing whether they still had a gun trained on you or not.  A precarious
situation, no?  Would you be happy to put your gun down without some way
of being sure that they were also putting theirs down?  Disarming without
100% verifiability entails a risk that one or both of the governments involved
will seize the chance to blow the other away.  What is the cost of a risk
like this?  How about:  (value of most of our lives) * (1 - verifiability) *
(1 - government trustworthiness).  Notice that government trustworthiness
decreases with the percieved strength of the other country.  (in other words,
one country is unlikely to try something if it thinks that the other country
is strong enough to retaliate.)
    You seem to be of the opinion that 100% verifiability is impossible to
even approximate.  Why?  If each country were allowed to send, say 2000
inspectors to the other country, and if the inspectors were changed regularly,
and were given free access to whatever they wanted to inspect, wouldn't that,
in conjunction with already existing spy satellites, give close to 100%
verifiability?  (before you say: 'ha.  He's stupid to think that they'd
ever allow *that*.  If we want disarmament, we'll have to take a risk and
not insist on 100% verifiability.', consider the mexican standoff again.
If you said: 'Let's stop this foolishness and put our guns down.  But first,
remove your screen, so that I can see whether or not you are really putting
yours down.'  and they said: 'No way.  You'll just have to trust me.' 
would *you* trust them?  Would you trust them, if it was not just your life
you were risking, but the lives of us all?
    Here's to the day when the sword of Damocles no longer hangs above our
heads.  May we find a way to remove it without dropping it.
-- 
Jeff Sonntag
ihnp4!mhuxt!js2j
    "I am what I am, and that's all that I am."-Popeye the sailor man.