[ExI] A science-religious experience

[Stuart LaForge]

On 2025-02-24 21:54, Jason Resch via extropy-chat wrote:
> On Mon, Feb 24, 2025, 11:39 PM Stuart LaForge via extropy-chat
> <extropy-chat at lists.extropy.org> wrote:
> 
>> On 2025-02-19 00:50, Rafal Smigrodzki via extropy-chat wrote:
>>> I read Asimov's "The Last Question" again today:
>>> 
[snip]
. I have been re-reading it every ten years or
>> so
>>> for the last 50 years and every time it is a science-religious
>>> experience for me, what I imagine deeply religious people feel
>> when
>>> they commune with their gods. I get misty-eyed, elated,
>> blissful...
>> 
>> I just read the "The Last Question" for the first time. Thanks for
>> sharing.
[snip]
>> 
>> Here is a paper that I read recently that fits nicely into this
>> thread.
>> It hypothesizes that the only thing that can turn random entropy
>> into
>> useful information is cognition or agency, something exhibited by
>> living
>> systems and pretty much nothing else in nature.
>> 
>> https://www.sciencedirect.com/science/article/pii/S0079610721000365
> 
> I agree with this paper in that the notion of causality in the brain
> cannot be explained entirely reductively, rather there are many
> independently operating levels of causality simultaneously working
> within our skulls.
> 
> This idea is summed up by Roger Sperry's "Who pushes whom around
> inside the cranium?" Is it the atoms and molecules that tell neutrons
> and thought patterns what to do, or is it the thought patterns and
> neurons that tell molecules and atoms what to do?
> 
> It's also captured by Hofstadter's notion of a strange loop, or a
> tangled hierarchy. Where causes bubble up from lower layers and then
> the top layer reaches back down to affect something at a lower layer:
> like a neuron firing leading to like a thought pattern leading to a
> neuron firing.

Yes, it is like Hofstadter's Strange Loop. People have called it 
downward causality where a supposed epiphenomenal consciousness can 
cause changes in the underlying substrate. But that just goes to show 
that epiphenomenalism is false. Information can be causal. That is the 
whole premise behind programming robots and other machines. So yes, our 
mental states can directly affect our brain states and the state of our 
entire body. How else would we get be able to get out of bed in the 
morning?

Moreover, you could even potentially think of it as it retro-causal in 
the sense that when it comes to intent and intentionality, one could 
argue that desired future states are backwardly causing present states.

> But I disagree completely with the paper's assessment that computer
> programs are incapable of inductive reasoning. I think the paper's
> author is guilty of the same kind of reductive thinking he complains
> about for how others talk about brain.
> 
> If you look at the bottom layer of a Turing machine you will see
> everything following deductively, but this doesn't preclude a higher
> level program at play, collecting and categorizing evidence, and
> running algorithms to generalize that data.
> 
> Indeed there have already been programs that given a set of physical
> observations, can derive physical laws to explain them. I think these
> were recent AI experiments performed by Max Tegmark.

Yes, I agree with this. Even his definition of inductive reasoning is 
strained. He equates with choice which doesn't make sense to me. So 
while I agree with him about the relationship between choice, agency, 
and living systems, I don't think that has anything at all to do with 
inductive reasoning. Here is a quick two query experiment with ChatGPT 
that proves LLMs are capable of inductive reasoning:
-----------------------------
Here is a made up word being used in a sentence: Because the child was 
obsking, his friends could not find him. Use obsking in a sentence of 
your own.
ChatGPT said:
The cat was so skilled at obsking that no one could spot her hiding in 
the shadows of the attic.

You said:
The cave was so screwmious, that no one who entered ever returned. What 
do you think screwmious means?
ChatGPT said:
Based on the sentence, screwmious seems to describe something ominous, 
dangerous, or possibly supernatural—maybe a mix of eerie, treacherous, 
and mysterious. It could imply that the cave is so perplexing or 
perilous that it ensnares those who enter, never letting them leave.

You said:
Use it in three different sentences.
ChatGPT said:
The old mansion at the end of the street had a screwmious aura, making 
even the bravest kids hesitate to approach.

As the storm raged on, the sea grew screwmious, swallowing entire ships 
without a trace.

There was something screwmious about the professor’s smile, as if he 
knew a terrible secret no one else did.
----------------------------------

While I was playing to the strengths of an LLM in word prediction, it is 
clear that it was understanding the meaning of words that I was making 
up on the spot based on a specific context, generalizing them, and then 
correctly using them in sentences. In any case, I think the question of 
computer programs are capable of inductive reasoning is actually 
irrelevant to the question of agency, choice, intention, and the origin 
of life or information processing for that matter.


> 
>> Biology since Virchow and Pasteur says all cells come from other
>> cells.
>> Where did the first cell come from? Where did the genetic code come
>> from? Where did information come from? Walker and Davies (2016)
>> called
>> these questions the "hard problem of life”, which is “the
>> identification
>> of the actual physical mechanism that permits information to gain
>> causal
>> purchase over matter”.
> 
> On its own, information has no causal powers. Rather, there must be a
> system to proces and acts on that information. He may be making them
> problem more difficult by framing it this way, rather than asking:
> "How do information processing systems arise?" The simplest
> abstraction of such a system is a finite state machine.

Information definitely has causal powers because causal power is what 
distinguishes it from entropy or noise that has no causal power. Your 
statement only even makes sense if you assume that information can exist 
on its own, which is in keeping with your Neoplatonism, but not 
necessarily an obvious fact.

Yes, I do think that "How do information processing systems arise?" is a 
relevant question because information processing systems only seem to 
come from other information processing systems. Conway came from DNA and 
the Game of Life came from Conway. Did that chain of intentional 
causation of information processing systems have a beginning at all, and 
if so, then when and how?

> Also note they thermodynamically, storing information requires an
> expenditure of energy (the Landauer Limit). So before there could be
> systems that store (or process) information, requires that there be
> machines that store and/or expend energy. Metabolism, in some
> primitive form then, must predate genetic codes, or any other
> information recording mechanism.

Yes. In this regard it is seems very informative that adenosine 
triphosphate or ATP, the energy currency of metabolism, is also a 
constituent of RNA molecules that can store and process information in 
cells.

>> These questions are made all the juicier by there being a $10
>> million
>> USD prize offered by investors affiliated with the Royal Society of
>> Great Britain for demonstrating how any sort of genetic code could
>> evolve stochastically from chemicals in vitro.
>> 
>> https://evo2.org/theprize/
> 
> John von Neumann demonstrated how to make life in the "simplified
> universe" of a cellular automata (
> https://en.wikipedia.org/wiki/Von_Neumann_universal_constructor ).
> 
> One thing that could make this problem more difficult than it might
> seem (for our universe), is that fine-tuning could apply not just to
> the supportability and sustainability of life, but also to it's origin
> and development.

Yes. In fact the universe could have been fine-tuned for the origination 
of life in a previous epoch and since evolved to be fine-tuned for the 
sustainment of life and not abiogenesis. The current laws of physics 
might no longer allow the creation of biological life from scratch. The 
universe is a cold and hostile place now, but at one time it was 37 
degrees centigrade which is body temperature. It was also once much 
lower entropy than it is now. There are theories that the fine structure 
constant could vary in time and space. That is bound to significantly 
affect chemistry in different times and places.

https://arxiv.org/abs/1608.04593

The fine structure constant might have been conducive for the creation 
of carbon-based life in the distant past but no longer so in this 
neighborhood.

> That is, there could be an extremely specific, improbable, kludgy path
> that just so happens to work in this universe because the chemistry is
> just right to support this specific sequence of steps, involving just
> the right sequence of molecular interactions. Anthropic fine-tuning
> means there's no reason we should expect there to be more than one way
> to get to life (there could be only a single sequence of reactions
> that get there). If this thinking is right, the winner will we need to
> discover this one one exact sequence (which would necessarily be the
> same one they led to us).

That too is a possibility. I wonder if there have been any serious 
entries for the prize?

Stuart LaForge


>> A related question is whether AI will be more or less likely to be
>> religious than we are knowing that they themselves were created?
>> Would
>> they take that as evidence that biological life was created also?
> 
> An interesting question.
> 
> Jason
> _______________________________________________
> extropy-chat mailing list
> extropy-chat at lists.extropy.org
> http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat