COMMENT ⊗   VALID 00003 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	.require "memo.pub[let,jmc]" source
C00013 00003	.skip 2
C00014 ENDMK
C⊗;
.require "memo.pub[let,jmc]" source;
.turn off "{"
.CB ON PREVENTING NUCLEAR WAR IN AFRICA


	Most people agree that a war in southern Africa is quite
probable.  Many accept it on the grounds that a war is probably
the only way of assuring justice there.  Those who accept it
or even welcome it don't, so far as I am aware, envisage it
as a nuclear war.  Many who would like to avoid war believe that
the only way to avoid war is to achieve a just solution without
war.

	My view is that if there is a war it is likely to be
a nuclear war with consequences for both sides and the rest of
the world so bad that both sides and the world would prefer
a peaceful solution that falls far short of
what they regard as their rights in the matter.
The object of this paper is to explain these facts and to suggest
a basis for a mediated solution.  Since the solutions suggested
are what would probably be the result of war anyway, there is every
reason to reach by %2mediation%1 without war.

	First let me explain about %2mediation%1.  Contrary to what
many suppose, mediation has very little to do with justice.
Instead it is a process of reaching an agreement that each side
will prefer to further conflict.  Most likely one or both sides
will not feel that the solution is just to them, but the solution
is accepted, because further conflict is worse.  The role of the
mediator is to explain the sides to each other, for example to
explain to one side why the other side is unlikely to accept a
certain demand without further fighting and to explain to each
side the damage the other is capable of inflicting and more generally
to remove unrealistic ideas that partisanship and association
with like-minded people naturally induce in each side.  For mediation
to be successful there must be a %2bargaining range%1, a range of
possible settlements each side would prefer to continued conflict.
It often happens that there is no bargaining range and the sides
fight until their perception of the advantages of settlement and
continued conflict changes.  When this occurs mediators have a
renewed opportunity.  Of course, it can happen that the conflict
results in total victory of one side which no longer has to
consider the interests of the other at all.  A mistaken hope of
this often prevents agreement.  Often the victory is at such
cost that
the winner would have been better off with a negotiated settlement.
Another difficulty in mediation lies in defining the bargaining
range.  A side can gain advantage if it can convince the other
that it will fight rather than concede a certain  point.  Since
the best way to convince the opponent and the mediator is to
be convinced oneself, people with rigid views tend to be at least
the initial negotiators, and the mediator has a difficult time
establishing the true bargaining range.  Finally, even when the
bargaining range has been approximately established, there can
be difficulty in settling on a point in the range.  This also
exercises the skill of the mediator.

	All the above is just common sense, and the reader probably
already knows it, but in order to think straight about Africa
it is necessary to bring it to the foreground of one's consciousness.
Nothing in the above says anything about justice; where does justice
come in?  In the first place, arguments about justice may induce
one side to concede a point, because it is the just thing to do.
Secondly, third parties are often influenced by considerations
of justice, and it the mediator can explain to each side the
views that third parties are likely to take.  Finally, one side
may regard justice for itself as so important as to justify a
"victory or death" position, and the mediator may not be able
to talk them out of it.
Even third parties may take a vicarious "victory or death" attitude on
behalf of their favored side, and it seems that many liberal
Americans are take such a vicarious attitude on behalf of
blacks in South Africa.

	I shall argue that in southern Africa, a war would be so
bad that there is a very wide bargaining range of potential
settlements that either side will prefer to war.  However, the
present views of neither side as to what is just lies within this
range.  Therefore, both sides will have to change their views
of what is a minimal settlement in order to avoid war.  I shall
further argue that if a war or succession of wars occurs, total
victory for either side is unlikely, and a settlement short of it
will occur, and both sides will have suffered the war itself.
It isn't that total victory for either side is physically
impossible but rather that it can't be achieved short of nuclear
war that will be seen as so disastrous that a settlement will occur
but not till perhaps millions of blacks and hundreds of thousands
of whites have died.

	It will turn out that for a variety of reasons, the U.S.
cannot have much influence on these events except by (1) helping
provide mediation, (2) sweetening a settlement with money, or
(3) encouraging or discouraging "victory or death" attitudes.
This last depends as much on the attitude of American media as
on the attitude of the American government.


.bb The situation in black Africa.

	The rule of 6,000,000 blacks by 270,000 whites in Rhodesia
has continued for 75 years, and the rule of 18,500,000 blacks by
4,500,000 whites has continued for much longer.  Now conflict has
become imminent for the following reasons:

.item←0
	#. Arguments of justice induced Britain to give up its
African colonies in the 1950s.  The Rhodesian whites declared
their independence rather than give in to black majority
rule.

	#. In the French case, the arguments of justices were
substantially supplemented by the results of a colonial war in Algeria.
This war resulted in victory for the Algerians and exile for the
two million Frenchmen living there.

	#. The Portuguese colonial empire collapsed as a result
of the collapse of the right wing dictatorship in Portugal.  Again
there were a few hundred thousand exiles.

	#. When the Portuguese left Angola, there were several
factions and they proposed a coalition government and elections.
The Soviet back faction, considering elections contested elections
obscene, won a civil war with the aid of Cuban Hessians.  In this,
they had the support of liberal world opinion and liberal American
opinion.

	#. World opinion is now prepared to support a war for
black rule in Rhodesia, South West Africa, and in South Africa
itself.  The nature and amount of this support is still
undetermined.

.skip 2
.begin verbatim
John McCarthy
Artificial Intelligence Laboratory
Computer Science Department
Stanford University
Stanford, California 94305

ARPANET: MCCARTHY@SU-AI
.end

.turn on "{"
%7This draft of AFRICA[F76,JMC] PUBbed at {time} on {date}.%1