[Fis] An Unbeatable Tradition?

[Stuart Kauffman]

Hi Marcus and All. Marcus thank you. Consider: Deduction, Induction, and Abduction - John Stuart Mill and human rationality. Great!! 

Now consider this: If I am using an engine block as a paper weight, I cannot deduce that I can also use the engine block’s sharp rigid corners to crack open coconuts. More generally we cannot deduce the different uses of any single thing alone or surely with indefinitely many other things. 

So I am being  a good scientist, even a Bayesian.  I have now used 1,429 engine blocks as paper weights and used a Bayesian approach to improve the engine blocks to be better paper weights. I have even used AI to construct a Markov Blanket! …Cool.  My mind evolved to predict the world for survival purposes. Thank whatever gods.   I have make an induction with a universal quantifier: “All Engine blocks within some constraints can be used as paper weights!” Then, given this covering law, I find an engine block and deduce: Because all engine blocks given these constraints can be used as paper weights, and this engine block fits the constraints, therefore I can use this engine block as a paper weight.  All true. Thank Darwin and some neurobiologists, philosophers, and cognitive theorists that my mind evolved to predict, Bayesian and all!

But with all that, I have no “logical or algorithmic” way whatsoever to get to, “Hot damn! I can use this engine block and its rigid corners to crack open coconuts.! “  Uhoh, evolving life is about open thermodynamic systems that reproduce so must eat, thus must find, stumble upon, grok, intuit ever - novel affordances. Were we only Bayesians we would be dead.  Evolution is not limited to Bayesian. The evolving biosphere invents novel morphologies all the time.Evolution invents novel behaviors all the time. 

This is also correct for us humans with MIND. We invent all the time. Mind is not limited to algorithmic. General AI seems ruled out. The evolving biosphere is a propagating non deducible construction, not a computation. Why do we continue to think that everything that becomes complex is some computation, even one that must be run to see what it produces? No. 

Andrea and I argue that we are trapped in our categories, hence in most of our deductive mathematics. There is no logical procedure to discover, stumble upon, find out the invention. Einstein did not deduce SR or GR. 

The evolving world is not trapped in its algorithms. Andrea and I published, The World Is Not a Theorem”. 

Wittgenstein realized all this nearly a century ago  From Russell Whitehead and his Tractatus to his Investigations and non reducible language games.

Marcus, I do think we may find statistical laws of the process. For example, I sent all of us the Stu Andre recent paper, “Is the emergence of life….” In that we mention the TAP process. That simple mathematical recurrence relation predicts the power law descent distribution of 3,000,000 patents. So TAP is capturing something about human non-deductive invention. These is a statistics of ‘jury rigging”, somehow related to constraints and what they next enable.  What is next enabled is the unprestatable non - deducible Adjacent Possible. Here we do not know the sample space so have no probability measure. We cannot even define random. Yet TAP seems to work for descent distributions.  ODD, how can both of these claims be true, if they are?

Best, 

Stu

> On Jan 24, 2024, at 2:29 AM, Marcus Abundis <55mrcs at gmail.com> wrote:
> 
> Pedro and Stuart,
> 
> Thank you for the usefully provocative posts.
> 
> Pedro's 22 Jan post points to THE central matter in a Third Transition and Phase Transitions, no?
> < Not aspiring to any universality about the conceptions of this independent informational exploration of life--at least shouldn't it be attempted?? . . . So, an unexpected[?!] convergence might be found finally >
> — If I understand Pedro correctly: creation of new adjacent possibilities surely follows `some type of logic'! Why is THAT not discussed here? But I am also unsure why this should be `unexpected’ (Pedro?). Still, is it really NOT POSSIBLE to formally deduce `base elements of creativity'? Does one simply claim it is a `mystery’ (a failure of scientific imagination), to throw our hands up and walk away?
> — I am unsure if this is what Stuart (and others?) suggest. If, as Stuart says, it is `not deductive reasoning’, it surely is not inductive reasoning (nor analogic reasoning). So EXACTLY what type of reasoning is needed?! Intuitively I think `abductive reasoning’, but then that does not seem to fully `fill the bill’. Alternatively, perhaps a blended vista is needed. Further to Pedro’s note about Plamen, Joe, and Eric, I too have offered my own Theory of Meaning view on the matter, much earlier.
> 
> ERIC < . . . at a given point in time we have real possibilities and we have pseudo possibilities . . . I did develop a theory of changing sets to make possible the formalization of changing possibilities. >
> — I am unsure how one effectively differs between real and pseudo possibilities, as mostly everything is possible until `the hand is dealt’ (a wave function collapse?). Still, a theory of changing sets sounds interesting (link?:< my theory of information and ability of agents >)
> Also, < in writing by the time you get to the end of the sentence you already have a better understanding of what you are thinking. >
> — In writing it down again, this time on a computer . . . you should have EVEN BETTER clarity of your intent and language, no? This would be useful. Still, my own QM view is skeptical, I am uninterested in things prior to a wave function collapse as all is too much like `noise’.
> 
> JOE < information is characterized by its quality as well as by scalar quantities. I have not seen any algorithmic theory that captures quality, but would of course be glad to know of one. >
> — How does Signal Entropy’s variable X^n logarithmic base NOT present a scalar (simple-to-complex) role?
> 
> STUART — when someone mentions Aristotle’s four causes I am too often left with more questions than answers. Foremost, it requires a Prime Mover — which we are to call `What?’ exactly, thermodynamics?! Also I do not see how it gets us to simple-to-complex material reality, it does not even rise to the level of dialectic logic born of Pre-Socratic views. Lastly, it seems awfully anthropocentric, given what we know of cause-and-effect in this modern era. So I am unsure what to do with Aristotle’s four causes in current exchanges.
> 
> Marcus
>  

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listas.unizar.es/pipermail/fis/attachments/20240124/603f84e3/attachment-0003.html>