RPKP Update No.6
July 17th, 1996
This was originally sent out to members of the RPKP mailing list. If you
would like to receive future updates, please
Klaus Scharff, a German enthusiast with an extensive knowledge of unorthodox
theories of time, has contributed a fourteen-page bibliography of serious
academic articles related to backwards-time phenomena. This has now been
posted at the website, and is gradually being added to. The articles
are drawn from journals of physics, philosophy, psychology, cybernetics,
parapsychology, etc. and include such intruiging titles as
"Acausal phenomena in physics and biology", "Tachyons, backwards causation
and freedom", and "Should physicists say that the past really happened?".
This is a very helpful step forward in the initiation of serious inter-disciplinary
discussions of acausal and retrocausal systems and phenomena.
Topher Cooper has contributed a lucid explanation of the significance of
Dick Bierman's (recently added)
RPK database. This is quite accessible
to non-statisticians. Although it has been
revised a couple of times since first appearing, it seems that Cooper's final word on the matter
is that the odds are roughly one in 630 thousand million that the distribution
of Z-scores which appear in the
database arose by chance! Due to the
nature of meta-analytical methods, any of the nine major experimenters
included could be removed without reducing this to a significantly less
Finally, York Dobyns of
PEARlab has summarised his recent Journal of Scientific
Exploration article on selection vs. influence. This is part of the ongoing
debate over the validity of May, et.al.'s decision augmentation theory (DAT).
Dobyns argues against DAT by introducing the method of "rank frequencies".
We hope to have a more detailed description of his argument before too long.
Jack Sarfatti has been in communication with us over his interpretation of
DAT and Dick Bierman's recent suggestion that the existence of "influence"
(contrary to DAT) could be demonstrated using repeated viewing of the random
data in the experimental protocol. According to Bierman, if DAT is correct, and the apparent
"retrocausal influence" is an entirely precognitive mechanism, then repeatedly displaying
the data to a sequence subjects should not affect the results.
Consequently, some correlation between repeated viewing and improved results
would suggest that DAT is somehow incorrect (or incomplete). Sarfatti has
been questioning the defintion of the term "biased sample" which May, et.al.
use, and has argued that if DAT is correct, repeated viewing could improve
results (in accord with his "back-action" theory of consciousness). The
confusion over DAT has lead us to believe that it might be better to work
towards a demonstration of "retropsychokinesis" without worrying about
whether it is due to "selection" or an "influence". As May has pointed out, whatever
the mechanism is, it involves an (acausal) "flow of information from future
to past". We can't help thinking that the selection/influence distinction
might have more to do with our own misunderstanding of time and consciousness
than anything which is actually "happening". Therefore, we believe, a reliable and publically verifiable
demonstration of the effect, without the inclusion of DAT-tests, might be sufficient to force a much needed debate in the
academic community about the nature of time, consciousness and objective reality.
We've started to get some feedback about the RPK experiment programs which
are now available via anonymous FTP. As it takes quite a few sessions (in
the appropriate setting and psychological state) to discern any effect, most
of this has been to do with personal preferences for certain programs (there
are six available). If you have not yet looked at these programs you
please do - they are available as DOS executables, easily downloaded and
We will begin remote monitoring of formal RPK experiments once these programs
have been converted into Java applets (this is still our major obstacle).
A moment of false hope occured a few days ago when someone brought to our
attention the "c2j" C++ to Java conversion program, available free via WWW.
Unfortunately, we soon discovered, this can do little with our C++ source code, due to the amount
of graphical interfacing involved. One C++ programmer currently learing Java
estimated 400 hours work per program, which is discouraging to say the least.
On the other hand, John Walker, creator of the HotBits random number server,
managed to put together a rough Java applet based on the program SWING (before
he'd seen the actual source code) in one evening. He has offered to
lend a hand in the near future, so there is hope.
Topher Cooper has suggested the use of PGP encryption in the experiment
protocol. This would allow us to post "predictions" of well-publicised
random events which are to occur in the future (a single number from the
coming week's UK National Lottery, for example). The "predictions" would
in fact be randomly generated before being encrypted and posted on several
widely-read USENET newsgroups (with a plaintext explanation). After the
random event in question had been agreed upon (e.g. the lottery draw), we could set up
RPK trials with talented subjects attempting to introduce subtle biases
into the original random data. If properly coordinated, this would result
in the correct number "having been" encrypted in the past. This could
bypass the need to introduce high-profile "independent observers", allowing
virtually anyone with Internet access to take involve themself in this role.
There are certain technical considerations to do with the preservation of
indeterminacy in encryption scenarios, but we are fairly close to finding
a satisfactory solution to these.
As ever, questions, comments and suggestions are very welcome.
The RetroPsychoKinesis Project (http://www.fourmilab.ch/rpkp/)