COMMENT ⊗   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	.device XGP
C00029 ENDMK
C⊗;
.device XGP;
.font 1 "non44"
.font 2 "non44i"
.height←50;
.page frame height high 115 wide;
.title area heading lines 1 to 3;
.area text lines 4 to height;
.oddleftborder←evenleftborder←1500;
.turn on "%";
.place text
.spacing 50 mills;
.Begin center
PARTICIPATION OF A WOULD-BE EMIGRANT IN A SCIENTIFIC CONFERENCE 
IN THE SOVIET UNION
.end
.skip 3
.page←1;
.every heading(,%1{page!});

	At the Fourth International Joint Conference on Artificial
Intelligence (IJCAI), held in Tbilisi, USSR from 3 to 8 September 1975,
Alexander Y.  Lerner, a cyberneticist who had declared his intention to
emigrate from the USSR to Israel four years previously, participated
without incident in a panel discussion entitled "Artificial Intelligence,
Cybernetics and Mathematics".  Since this was the first time that a
scientist refused permission to emigrate was permitted to take part in a
scientific meeting in the Soviet Union, there has been curiosity about how
this came about, and whether a more humane Soviet policy more in keeping
with the customary openness of scientific meetings can now be expected.

	%2Artificial intelligence%1 is the study of intellectual processes
with emphasis on how computers can be made to carry them out.  The IJCAIs
(conferences) have been held every two years since 1969 in Washington
D.C., London, Stanford California, and Tbilisi USSR.  Each meeting has had
several hundred attendees - about half Americans, since most research in
AI is American.  The IJCAI (committee) is an informal organization of
individuals representing only themselves that meets at each conference to
select one or two individuals responsible for organizing the next
conference.

	At the Stanford conference in 1973, a delegation from the Council
of Cybernetics of the USSR invited the Committee to hold the next
conference in the USSR at a location to be subsequently determined.  At
first, there was considerable skepticism about whether the IJCAI's
informal habits were compatible with the way things might have to be done
in the USSR, but the Soviet delegation accepted all our proposals
regarding organization including the complete authority of the Program
Chairman, Professor Patrick Winston of M.I.T., over the program and the
overall authority of the General Chairman, Professor Erik Sandewall of
Uppsala University (Sweden).  It was agreed that Professor Winston would
publish an English language advance proceedings and that translations of
submitted Soviet papers would be supplied in time for this.  It was agreed
that delegates from all countries would be allowed to attend, because of a
previous case in which an Israeli invited to a conference was subsequently
denied a visa.  No issue concerning the attendance of Soviet citizens was
raised.  There was some behind the scenes muttering about treatment of
dissidents and problems with Intourist, but after an initial vote that
conferences should alternate between North America and elsewhere, there
were 8 votes for the USSR against 3 for Japan and 1 for West Germany.

	An inquiry in Israel established that Israelis received visas for
international Conferences held in the Soviet Union, although when an
Israeli working in the United States was invited to a Soviet domestic
conference (perhaps in the belief that he was an American), no visa could
be obtained.  (This raises questions for people receiving such
invitations).

	About a year before the Conference, Professor Jack Minker of the
University of Maryland and the Union of Concerned Scientists (They are
concerned about Soviet treatment of dissidents and would-be emigrants),
proposed that the Conference be moved from the Soviet Union because of the
ill treatment of these people.  The organizers declined on the grounds
that the state of these issues had not changed since the decision to hold
the meeting in the USSR.  Professor Minker pursued this goal with a letter
to the SIGART newsletter, and an ad hoc meeting held in connection with
the ACM meeting in San Diego voted 31-0 to propose moving the meeting, but
the organizers didn't change their position.  (SIGART is the Special
Interest Group on Artificial Intelligence of the Association for Computing
Machinery (ACM), and most researchers in artificial intelligence belong to
SIGART, but the IJCAI has no official connection with it).

	Then Professor Minker proposed to organize a panel on Artificial
Intelligence and Cybernetics for the meeting and to invite Dr. Alexander
Lerner of Moscow to take part.  Dr. Lerner had been employed by the
Institute of Problems of Control in Moscow until he tried to emigrate in
1971 and was rather prominent in Soviet cybernetics and held many official
positions.  In 1971 he was fired, remained unemployed, and had his name
removed from a new edition of his book.  He is one of the more prominent
%2refuseniks%1 and has been interviewed by foreign newspapers and visiting
U.S. Senators and Representatives.

	Several people accepted Minker's invitation to be on the panel,
and the Program Chairman accepted the panel for the program on determining
that the panel topic was suitable for the conference and that Minker was
qualified to organize it, i.e. without giving any weight to conjectures
about motivation.  In April 1975 three Soviet scientists who came to
M.I.T. to discuss the program expressed their surprise that someone who
opposed the agreed arrangements for the meeting should chair a panel.  In
deference to their position, Minker was persuaded give up being chairman
of the panel, and it was agreed that Professor John McCarthy of Stanford
University, who had accepted membership in the panel, would be asked to be
co-chairman with a Soviet scientist subsequently to be appointed.  The
sense in which the Soviet scientists agreed to Lerner's participation -
whether they said only that they had no personal objections or whether
they committed the Soviet Council on Cybernetics - is unclear.

	In May the %2New York Times%1 reported that Lerner was told by the
KGB (a Soviet organization combining the functions of the FBI and CIA with
political control functions not carried out in the U.S.) that he would not
be allowed to go to Tbilisi and participate in the conference.  This
caused Minker to renew his campaign to move the conference with a mailing
to the members of SIGART getting about 500 cards supporting his position.
The Organizing Committee members agreed by telephone to send a letter to
the Soviet Organizing Committee stating that we took the issue of Lerner's
right to attend and the Conference's right to hear him very seriously and
that his inclusion in the program was within the prerogatives of the the
Program Chairman.  The letter went on to say that there would be a "public
protest" at the conference if Lerner were not allowed to take part.  The
form of the protest was not stated, partly for tactical reasons and partly
because it would depend on what the participants would want to do at the
time.  In any case, it was planned to obtain a tape of Lerner's speech and
play it at the panel, but additional actions were contemplated.  The
letter was signed by all but one non-Soviet member of the Organizing
Committee; he was not informed of enough of the background in time to make
up his mind.

	A letter was received  from Professor I. M. Gelfand,  who was
appointed  Soviet co-chairman, asking about  the suggested content of
the panel and  suggesting that  good participants be  obtained.   The
reply discussed the  scientific question and emphasized  that part of
the   membership   of  the   panel   was  already   determined,  that
participation was not organized  on national lines and  that McCarthy
was continuing  Minker's invitation to  Lerner, and that  the Program
Chairman had the final authority. 

	No reply was received to either of the above letters.

	When the members of the Organizing Committee arrived in Moscow,
the Soviet Committee had apparently still not decided what to do about
the issue, except that they offered several arguments against the
participation of Lerner in the panel and in the Conference.  The main
argument was that the Soviet %2delegation%1 was made up of the nominees
of various organizations, and no organization had nominated Lerner.  The
reply was that the AI Conference was of individuals not delegates, and
his registration for the Conference had been accepted and he had been
invited to participate in the panel.

	When Lerner  was visited in  his Moscow  apartment, he had  a
tape of his  proposed remarks and an airplane reservation to Tbilisi,
but he thought that the KGB might prevent him from going to Tbilisi. 
The  %2New  York  Times%1  correspondent Christopher  Wren,  who  had
written  the May article about Lerner being  told that he couldn't go
to Tbilisi was  contacted and expressed  journalistic interest.   The
U.S.  Embassy's scientific counselor  was contacted, told  what might
happen, and he expressed a desire to be kept informed. 

	On the morning when the conference was to open a breakfast was
held at which the Georgian host, Dr Chavchanidze, Director of the
Institute of Cybernetics in Tbilisi, proposed a compromise.  Lerner,
who had arrived in Tbilisi without incident, could take part in the
Conference and the panel, but as a guest and not as a member of the
Soviet delegation.  In return for this, both Lerner and the foreign
members of the Organizing Committee were to promise to discourage any
political activity at the Conference.  Both Lerner and the Organizing
Committee were quite satisfied with this.

	Once this was agreed, Dr. Chavchanidze and the other Soviet
organizers acted with great courtesy and consideration.  Dr. Lerner
was greeted courteously by his old friends, his request that his
wife be admitted too was granted without hesitation, and a prominent
Soviet computer scientist, Dr. A.P. Ershov, replaced Gelfand, who was ill,
as co-chairman of the panel, and another prominent Soviet scientist
agreed to participate and took part in a preliminary get-together of
the panel, but had to return to Moscow before the panel took place.
The panel took place normally in a room crowded with both Soviet
and foreign participants.  It was agreed that both sides had lived up
to the letter and spirit of the agreement.

	Two other problems should be mentioned, because they interacted
with the Lerner issue.  The first was that the one Israeli who tried
to come was denied a visa by the Soviet Embassy in Washington less than
a week before the beginning of the meeting.  On the one hand, the
Soviet organizers didn't immediately realize his passport was Israeli
because he worked in the U.S.  On the other hand, his visa application
stated clearly that the was attending an international conference.
After the denial, telephone calls from the General Chairman to Moscow
resulted in an assurance that a cable had been sent to the
Soviet Consulate in San Francisco that would result in a visa, but it didn't.
A further assurance, obtained after the Organizing Committee arrived
in Moscow, that the problem had finally been solved and a visa was
available in Vienna was not checked, because the scientist in question
decided to give up.  This issue had the effect of making the Western
scientists more militant on the Lerner issue and may have made the
Soviets more willing to make concessions, since they had given
assurances two years previously that Israelis would have no problems.

	The second problem concerned the  brothers, Issai and Gregory
Goldstein  of Tbilisi,  who also  had lost their  jobs for  trying to
emigrate and who  also wished to  attend the  conference.  They  were
allowed to attend the panel involving  Lerner where one of them asked
a  question identifying  himself as from  the State  of Israel, which
caused no reaction at the time.  According to their report, they were
picked up the next day by the KGB and told that they had been invited
by the Americans which was improper and compelled on threat of arrest
to promise not to  return to the Conference unless  properly invited.
This caused some  Conference attendees to propose moving a session to
their home, but the Organizing  Committee declined to do this on  the
grounds that our agreement concerned  Lerner and had been adhered to,
but we  stated that the psychological atmosphere would be improved if
something could be done for the Goldsteins.   Dr. Chavchanidze stated
that he had been unable  to contact anyone in authority, but that the
Goldstein's could  come  as far  as  he was  concerned  provided they
"behaved themselves", but he couldn't  guarantee what the authorities
would do.   This was Saturday and the Goldstein's attended the Monday
morning session after which they  were again picked up and told  they
couldn't come no  matter what anyone said.   The Conference concluded
Monday afternoon in an atmosphere of mutual good will.

	Here are some conclusions that we have drawn from this affair,
but we must emphasize that they are all tentative:

	1. Sometimes foreign pressure can get Soviet organizations to
behave more in accordance with international scientific and humanitarian
norms than they will without the pressure.  Success in any instance cannot
be predicted in advance of trying.

	2. It was important that the Organizing Committee itself took a
stand.  The delegation of authority to the General Chairman and the
Program Chairman helped, because they could act on what they felt was the
consensus rather than wait for formal meetings.  There is evidence that
our unified willingness to go as far as a demonstration if necessary,
combined with an equally unified desire to avoid all incident if a limited
objective was achieved, was decisive.

	3. The IJCAI involved a number of circumstances that will not
recur.  Thus it may have been easier for the Soviets to make concessions
to the Organizing Committee that had resisted attempts to move the
Conference from the Soviet Union.  The fact that the Organizing Committee
had to fight to keep the meeting in the Soviet Union is a symptom of
growing Western impatience with the way the Soviets conduct international
scientific relations and with their treatment of their own nonconformists.

	4. No person on the Soviet side expressed personal objection to
Lerner, but none, so far as we know, took independent action on his
behalf.  We think it was easier for them to advocate concessions to a
foreign viewpoint than to express personal agreement with that viewpoint.

	5. It was never clear who made the final Soviet decision or on
what level the decision was made, though we were told that the matter was
discussed in the Presidium of the Academy of Sciences.  According to the
model of maximum avoidance of responsibility, higher authorities might
have told the lower levels to optimize several variables simultaneously -
effectively leaving to the lower level the decision of what to do when it
became clear that simultaneous optimization was impossible.  This model is
supported by the fact that if "loss of face" was a consideration - and
several Soviets said it was - then less face would have been lost by
making the same decisions earlier.

	6. In our opinion, the motivations of individuals and the local
conditions played an important role, but we have not reported our
conjectures about this.

	7.  Many   of  us   are  still   uncertain  how  to   balance
considerations  of non-interference in another  country's affairs and
not mixing politics  with science with  considerations of the  proper
conduct of scientific meetings and with humanitarian considerations. 
Soviet scientists with whom the matter was discussed also differed in
their opinions of whether  our action was good, bad,  or irrelevant. 
This report merely tells what we did and how it turned out. 

	8. Other scientific organizations may resolve such issues
differently, but maintaining any but a purely passive position will
require good communications, keeping the Soviet side informed, and a way
of making prompt decisions, since matters may go better or worse than
anticipated.
.tabs 45; turn on "\";
.begin nofill
\John McCarthy
\Artificial Intelligence Laboratory
\Stanford University
\Stanford, California 94305
.end