LETTER TO BART, EARL, HAL, RUSS ET AL REGARDING DRAFT PROPOSAL OF 28 MARCH

• Document Type: CREST
• Collection: STARGATE
• Document Number (FOIA) /ESDN (CREST): CIA-RDP96-00787R000400070005-8
• Release Decision: RIPPUB
• Original Classification: K
• Document Page Count: 4
• Document Creation Date: November 4, 2016
• Document Release Date: March 29, 2000
• Sequence Number: 5
• Publication Date: April 4, 1973
• Content Type: LETTER

Approved For Release 000/08/07: CIA-RDP96-00787R00049070005-8
1+ April 1973
Dear Bart, Earl, Hal, Russ et al:
First let me thank' you all once again for making my visit so pleasant and
so profitable. I hope that, by the time you receive this, I will have been able
to call to inform you that we agree in principle with the 28 March draft proposal.
Naturally, there have been many discussions about the matter since my return
and the purpose of this letter is to acquaint you with the views that are emerging
here. As I'm sure you also realized later, the precise details of how best to
proceed if our mutual goals are to be achieved (while protecting our mutual inter-
ests) are more complex even than we perceived during our discussions. So try hope
you will excuse the length of this letter (as well my typing errors) as
flesh-out possible procedures--with the understanding that nothing is definitive
unless you find it workable and acceptable. To the extent you do find it both
workable and acceptable, however, you may want to initiate appropriate modifications
of your procedures with our Subject (hereafter S--not to be confused with "S") as
soon as possible.
As I see it, there are three over-riding considerations
- that the tests be of such a nature and conducted in such a fashion that, if
'successful', we achieve the kinds of validation which our people can understand
and accept as a basis for possible further research;
- that, consistent with that objective, the tests be conducted in an atmos-
phere 'as fair to S (as little changedddf m youraearlier procedures) asdpossible;
- and that the tests be conduc in
either the existence or the nature of the outside sponsor.
In the latter% context, while you may wish eventually to inform him of the
new sponsor you mentioned to me, I find that it could be highly embarrassing for
us if you gave him any reason to believe that our tests were being conducted on
behalf of that sponsor--unless we coordinated the entire matter with them before-
hand, which we do not now intend to do._
In any case, if all of the above is agreed, the following procedures seem
d
s.
best designed to satisfy our nee
To ensure that our material looks no different to S, you might
(1) Materials. ship to us about 30 pieces each of the inner and outer envelopes (preferably manila)
which you will be using with him and the cards on which you draw your material, as
well as the.kinds of paper you would use to guard against transparency. In addition,
please send a couple of examples of your sealed envelopes--so that we can seal ours
in the same manner (this may seem like art-for-art's-sake but S may be highly
sensitive to such matters); also, we'd like a couple of exemplars of how your own,
typical line drawings appear on the cards. Finally, it might be best if you sent
one or two of the magic-markers you use so the color, width will be the same. s
you may have guessed we're actually not interested in the research at all but, what
with retrenchments, are simply eager to build-up our uunr any sttrue blanks ockrooms.
with S as yet it
(2) Use of Blanks. If you have not ye
might be wise to try it during the warm-up period since, if it is unduly upsetting
or threatens the even tenor of your relationship with him, we won't want to use
them in our series. On the other hand, it would be a quite useful control factor
Approved For Release 2000/08/07: CIA-RDP96-00787R000400070005-8
 - 2 -
Approved For Releas%000/08/07: CIA-RDP96-00787R00040 0005-8
if you can get him to accept that additional wrinkle. By the way, you will recall
that we discussed our using some 'specially prepared' blanks but we have decided,
for several reasons, not to do so. In the first place, it could get quite sticky
if he 'succeeds' with such material and you were then forced to give him false
feedback by using the mental-image ploy. In the second, the statistical sample
would be so small as not to materially affect our primary purpose--which is
validation, pure and simple. If there is a follow-on, however, it might be useful
to consider that technique again--with the difference that you prepare the samples
with simple (and commercially available) systems and that you level with him about
it (which we couldn't do using our systems). In any case, if the warm-up period
shows that use of blanks is acceptable, any blanks we employ will be just that.
(3) The Question of Feedback. This is the most delicate aspect of all since
it is the one area in which, for optimum effect here, you must at least try to
modify earlier procedures--and, probably, the earlier in the new series that you
are able to introduce the changes the more natural it will seem to S. Obviously,
the manner in which you do so and the rationale you use is entirely up to you but
I would like to outline the optimum from our point of view and, perhaps, suggest a
few possible approaches. Ideally, we'd like delayed feedback of the nature which
would permit you to tell him at the end of a series of, say, ten tests (whether it
takes him one or several days to complete it) that he gottien so many 'right', so
many 'wrong' and so many 'near-misses'--without giving him any more specifics or
access to the stimuli material until we have checked them out here and mailed them
back to you. Since that might mean an elapsed time of 10 days, we recognize that
this might not be feasible. On the other hand, if you can impress upon him that
you are applying much more stringent controls now so as to increase the scientific
validity (as, I believe, he has recently said he is interested in), he might buy the
idea that other and totally disassociated SRI offices (perhaps in other cities) are
being used to create/control/validate the stimuli material--and that he and you will
be able to compare the results together only after all of the sealed-envelope work
is completed, thus making it impossible for scoffers to kick holes in the procedures
afterwards. This would also explain the double-blind routine and the fact that
you don't know the answers yourselves. You could obtain the right/wrong/near-miss
data by calling a number we will provide anytime during our working day (1200- to / 4 Q0
2OW hours your time) on the agreed testing dates. c'6?c)
If, however, you find during the warm-up period that S simply cannot function
under that system we would not want you to persist to the point where you would
clearly jeopardize the atmosphere and your working relationship. In which case we
could fall back to the system where, after his drawings and our material %% are in
our representative's hands, you would be told telephonically what each drawing was
and--if necessary--draw a mock-up to show him. By the way, we do not know whither
in the past the material has always been opened in his presence. If it was, it
would be desirable to modify that (using the 'other SRI lab' rationale) as soon
as possible. Otherwise, we'd be forced to 'fix' our envelopes so that the very
edge could be cut off with scissors (after our man has S's drawings) and then be
returned to us for validation that they hadn't been tampered with otherwise. But
this procedure would give our scoffers so much latitude that it should be avoided
at all reasonable cost. I'.F it cannot be avoided, the scissor-opening routine had
better become an early part of your SOP.
Naturally, all of these questions should be settled during your warm-up period--
before we start preparing our material. In this context, it might be wise to plan
another get-together (preferably here) after you have a feel for the way thing's are
going.
(4) The Double-Blind Aspect. I cannot recall whether any of the earlier. testing
was of the double-blind variety, where no one present w t nts. If not,
of course g~p~r 6iliFo t eleaswo 8iji7 t1 I -F~ q ,~ ~'O gh the die-in-
 -3-
the-box kpipewiffAftff sec jW46W-7e: gA tRpp9&g01@i ,q0@gf q@ 05A any case,
your warm-up should tell us what we can hope for in this respect. Our feeling is
that, if he bombs completely during warm-up, we should switch the testing focus re
our material immediately to the ESP mode; on the other hand, if he does even reas-
onably well in the warm-up double-blinds, we should proceed on that basis for our
first batch and then consider going to ESP for the second. Should we go to the ESP
mode, feedback will no longer be a problem but it will be replaced by even thornier
'validation' problems. I suggest that, in that case, our sealed (but un-'fixed')
envelopes be sent to our representative who will carry them to you on the testing
date. Before each test one of the investigators will join our man wherever he is
watching the live video monitor, they will open the envelope and reseal the card
in another one after the investigator has memorized it. He will then carry it back
to the testing room and it will be critical that he be on-camera from the moment he
enters and remain on-camera until S has drawn his response and put it on-camera in
front of the investigator. Naturally, it will be even more important here than in
the double-blind mode that there be no exchanges with S of any sort during the
testing sequences--so as to avoid later suggestions that there might have been
subliminal cue-giving. We would, of course, want to have copies of all testing-
situation video tapes sent back with our material--whether the testing is double-
blind or ESP. By the way, could you refresh my memory as to the time it normally
takes S to complete one such test--and how many you expect to be able to cover in
one day's testing?
(5) Miscellaneous Procedures. We are thinking of some slight modifications as
regards the use of our material. For instance, if you feel that he can handle ten
envelopes in one day, it might be best if we run our material in two (rather than
four) batches--and not mixing-up your and our material at all. This should prove
less confusing at both ends and, should the going be slow on a given occasion, you
could simply give the material back to our man and continue the next day. We are
also thinking of using nothing but simple line drawings since that is what he seems
to be most comfortable with--although we might use other material (digits/letters)
in three or four of our second batch of ten. When our material is used we'd like
to see what he can do without first being told the category and then, if he baulks
or has difficulty, you can tell him that they are line drawings (or, alternatively
as regards the second batch, 'mixed'). Our outer envelopes will each bear a small
(roughly quarter-inch) number (e.g., 1 through 10 in ball-point pen) on the upper
right-hand corner of the facing (non-flap) surface--and I suggest that you so number
each of your envelopes for each test series, designating the series by A/B/C setter
prefixes; you and our man can then add the appropriate letter prefix to our envelopes
before the tests begin. We should probably allow at least two weeks between our
test series, setting the date for the second at a mutually convenient time--and in
such manner as will seem natural and innocuous to S. As regards wrapping the cards
in paper within the envelopes, it would be wise for you to experiment with variations
during the warm-up. Ideally, we'd like to wrap them in black (or some other dark)
paper but it would be absurd for us to do so if, for whatever reason, it frustrates
or inhibits performance; perhaps wrapping in white paper would be sufficient (espes-
ially if you are using fairly thick gauge manila envelopes--in which case we might,
if necessary, dispense with wrappings altogether.) Finally, I believe we should
leave the question of long-distance-telephone work until we have this immediate
business well in hand. But if you wish to try something on that score for reasons
of your own at an earlier date please let me or =know. SG11
Well, I warned you about the length--and I don't flatter myself that, even now,
everythi,ig of consequence has been covered. Please let me have your views when it
is convenient. Even at the smartest pace, I don't see how we can hope to run our
Approved For Release 2000/08/07 : CIA-RDP96-00787R000400070005-8
 Approved For Release OO/08/07: CIA-RpE96-00787R00040Q~7~D005-8
first test much before mid-May.
By the way, I haven't yet received those video tapes--and I hasten to add
that I'm not bugging you for them, only alerting you in case they've gone astray
in the mails.
When you write you can use either the regular mailing channels to my office
(I believe that someone in SRI has that information or to office--or if
you wish, to my home address :
Best regards,
SGFOIA3
Approved For Release 2000/08/07 : CIA-RDP96-00787R000400070005-8