Skip to first unread message

Alix

unread,
Jul 16, 2012, 5:43:26 PM7/16/12
to gcp...@googlegroups.com
One of the most controversial topics in our world today is that of the degree to which our existence is deterministic. The modern objective sciences attempt to find a Grand Unified Theory which can completely predict the happenings of our Universe. This thought can cause great disturbance among many, because of its implications for the existence of Free Will, Choice... or simply the element of Randomness. 

But in the face of Statistical Field Theory– which combines probability theory, quantum mechanics, and field theory– combined with Lattice Field Theory– which views the fields over a spacetime which has been discretized over a lattice– we can imagine a universe in which at each point in space-time there exists a probability of a given point (or collection of points) as having the quality of being subject to deterministic laws, and contrarily a probability of each point being able to, in essence, "choose", allowing for randomness. 

It is my hypothesis that the more aware (or concentrated or conscious) a point (or particle, string, or being) is of the possible states which it can assume, the greater the probability is for that being to be able to make choices about which state it is to assume; points which are less conscious will then follow a more deterministic trajectory (or existence). Of course (using point particles as a pedagogical example), each particle has its own properties (such as electric charge and mass), which will influence the behavior of the other particles around it (by way of the exchange of force particles); a particle that has become aware enough to chose its (lets say, seemingly random) trajectory will in turn cause a domino effect of deterministic trajectories of the other particles around it. 

In this sense, both randomness and determinism can exist. Any thoughts, or experiences?

Ed Acker

unread,
Jul 16, 2012, 7:36:07 PM7/16/12
to gcp...@googlegroups.com, gcp...@googlegroups.com
Hi Alice,

I am exited to learn more about your theory. My personal view is that both determinism and randomness are products of a story-telling machine trying to describe a multidimensional parallel universe with an extremely limited serial language.

Sent from my iPhone
--
You received this message because you are subscribed to the Google Groups "gcpwork" group.
To post to this group, send email to gcp...@googlegroups.com.
To unsubscribe from this group, send email to gcpwork+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msg/gcpwork/-/1J78f4yx6GoJ.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

Rocco

unread,
Jul 16, 2012, 8:39:11 PM7/16/12
to gcp...@googlegroups.com

Down at the subsurface level of existence, what we see is not necessarily a reflection of how things operate.  And what has to be taken into account at all levels is "effect of the observer'".  What seems to get lost in trying to communicate this little quirk of existence is that it is responsible for the entire perceived package.   Without the observer there is no 3rd dimension. Is the dream state any less real than the waking state since both require an observer?

Of course you can insist that there is a linear real world just chuck full of random events that at any moment could tag you as the next victim. However, that would require ignoring what we know regarding subsurface physics. Once you start looking at the Universe as a gathering of photons you begin to understand that chaos is just not in the cards.  (Do photons communicate, decide on a course of action and we call that free will?) In fact it seems almost impossible to find anything that is a random event. Even the once touted random event of radio active decay seems to lack scientific agreement about it being random: it appears to have a pattern.

Roger Nelson

unread,
Jul 17, 2012, 8:22:26 AM7/17/12
to gcp...@googlegroups.com
The article on randomness is interesting, but it is from 1973 I think. I
am not aware of an update, but if there is further work in the area it
could be important. In the GCP we use electron tunneling as the
fundamental source of randomness, and as far as I am aware, the bit
sequence produced by research grade RNGs of this kind show *no*
structure. On the other hand, a phenomenological "consciousness field"
based on some form of interconnection or coherence somehow introduces
subtle but statistically detectable structure. Could there be a link to
the patterning induced by asking the "right" question in examining
radioactive decay?

On 7/16/2012 8:39 PM, Rocco wrote:
> Down at the subsurface level of existence, what we see is not
> necessarily a reflection of how things operate. And what has to be
> taken into account at all levels is "effect of the observer'
> <http://en.wikipedia.org/wiki/Observer_effect_%28physics%29>". What
> seems to get lost in trying to communicate this little quirk of
> existence is that it is responsible for the entire perceived package
> <http://www.vision.net.au/~apaterson/science/observer_effect.htm>.
> Without the observer there is no 3rd dimension. Is the dream state any
> less real than the waking state since both require an observer?
>
> Of course you can insist that there is a linear real world just chuck
> full of random events that at any moment could tag you as the next
> victim. However, that would require ignoring what we know regarding
> subsurface physics. Once you start looking at the Universe as a
> gathering of photons you begin to understand that chaos is just not in
> the cards. (Do photons communicate, decide on a course of action and we
> call that free will?) In fact it seems almost impossible to find
> anything that is a random event. Even the once touted random event of
> radio active decay
> <http://pubs.acs.org/doi/abs/10.1021/j100644a019> seems to lack
> scientific agreement about it being random: it appears to have a pattern.
>
>
> On Monday, July 16, 2012 3:43:26 PM UTC-6, Alix wrote:
>
> One of the most controversial topics in our world today is that of
> the degree to which our existence is deterministic. The modern
> objective sciences attempt to find a Grand Unified Theory which can
> completely predict the happenings of our Universe. This thought can
> cause great disturbance among many, because of its implications for
> the existence of Free Will, Choice... or simply the element of
> Randomness.
>
> But in the face of Statistical Field Theory� which combines
> probability theory, quantum mechanics, and field theory� combined
> with Lattice Field Theory� which views the fields over a spacetime
> which has been discretized over a lattice� we can imagine a universe
> in which at each point in space-time there exists a probability of a
> given point (or collection of points) as having the quality of being
> subject to deterministic laws, and contrarily a probability of each
> point being able to, in essence, "choose", allowing for randomness.
>
> It is my hypothesis that the more aware (or concentrated or
> conscious) a point (or particle, string, or being) is of the
> possible states which it can assume, the greater the probability is
> for that being to be able to make choices about which state it is to
> assume; points which are less conscious will then follow a more
> deterministic trajectory (or existence). Of course (using point
> particles as a pedagogical example), each particle has its own
> properties (such as electric charge and mass), which will influence
> the behavior of the other particles around it (by way of the
> exchange of force particles); a particle that has become aware
> enough to chose its (lets say, seemingly random) trajectory will in
> turn cause a domino effect of deterministic trajectories of the
> other particles around it.
>
> In this sense, both randomness and determinism can exist. Any
> thoughts, or experiences?
>
> --
> You received this message because you are subscribed to the Google
> Groups "gcpwork" group.
> To post to this group, send email to gcp...@googlegroups.com.
> To unsubscribe from this group, send email to
> gcpwork+u...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msg/gcpwork/-/db04bkN-Iz0J.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>

--
Roger D. Nelson
Director, Global Consciousness Project
rdne...@princeton.edu http://noosphere.princeton.edu


Rocco

unread,
Jul 17, 2012, 10:05:07 AM7/17/12
to gcp...@googlegroups.com

Without an agreement of what will be called ‘random’ it would be difficult to communicate an event that WE BELIEVE is ‘determined’. An example of this is in pharmacological studies. Either a pill worked or did not work (at a predetermined level of significance). Where reality becomes difficult to define is when ‘a sugar pill’ also worked (placebos). We can spin around forever trying to find a one size fits all form of randomness…or we can simply agree to accept a measure as random even if we cannot prove conclusively that it is random. And of course that is what we do when we agree on level of significance. Trying to prove a uniform random event in all of existence is not much different than trying to prove the existence of God.



Roger

unread,
Jul 17, 2012, 6:32:47 PM7/17/12
to gcp...@googlegroups.com
My pragmatism pushes me to experimental designs that work for their intended purpose even if the tools we must employ are not perfect. So for me, randomness is defined in the context of the research. I look for the best sources available with needed characteristics. Then I calibrate to a fare-thee-well, with 200 million bit samples and a sophisticated suite of statistical tests. When the device passes muster (shows normal, expected distribution of parameters) it goes to the field. The experiment runs as designed, and each device produces a continuous sequence of data which is examined to characterize its really long term behavior. Finally, the composite of the long series of formal hypothesis tests (which select particular moments in time) can be compared with the distribution of simulated series generated by re-sampling the 98% of data not used in the formal tests. So, finally, although we can speak of deviations from perfect theoretical behavior, we don't depend on that. Instead we look at a comparison of experimental data with well-conceived control data (re-sampling).

I am not sure this is actually relevant to the discussion of "randomness" but I feel strongly that operational definitions and empirical comparisons can win the day when we are trying to learn something.

Now, for the sake of experiment, I will try to attach a figure showing the kind of comparison I am talking about. If that doesn't work, it can be seen at http://teilhard.global-mind.org/control.distribution.html  Sure enough it didn't work apparently because too large ...
Reply all
Reply to author
Forward
0 new messages