[FRIAM] loopiness (again)

[glen ☣]

On 02/07/2017 11:44 AM, Marcus Daniels wrote:
> It seems like you are just saying that the phenotype is not knowable, and that there is no inherent meaning until many individuals act and there are consequences.   No problem with that.  The phenotype is coupled to the environment of the community members, so it needs to be carried along as an instantaneous parameter or dynamically evolving state object.

Well, I did conflate the two issues.  Sorry.  The simpler issue is the co-evolution of the individuals and their co-construction of the environment.  The more complex issue is that individuals are self-referential.  The implications of that self-reference might be trivial.  But lots of people have gone to great lengths to address whether or not it _is_ trivial.  (E.g. Rosen, Wolpert, Gödel, etc.)  Even our own Nick has tried to distinguish pseudo-circular from circular reasoning.  So, whether individuals _can_ be flattened is interesting.  And it relates to this conversation is the extent to which that depth allows or facilitates some sort of semantic slippage... ambiguous grounding of symbols so that things like fake news can prematurely register with the individual.  Or so that reprehensible speech like we see from Trump can be chalked off as a joke or "locker room talk".  Or so that an individual can identify with a group solely through reddit or youtube comments.

> Even so, it seems to me that the individuals, via their control program (which also can be updated), can update the bits that relate the action to the outcome.  It may or may not be the case that the individuals develop the same representation of that correlated event, but it seems unlikely that memory would take on a non-compressible representation involving a hugely different number of bits.   I speculate that the encoding could be normalized across individuals to some common subset and that one would find some individuals encoding more or less detail and that those encodings could be reliably mapped to words like blue or soccer ball or delight.

Right.  That speculation is reasonable.  But what I'm trying to distinguish is whether or not such a commonality/normalization can occur _without_ actions, solely with thoughts, memes, and non-face-to-face words.  I think not.  I think the actions in the richer medium of meat space is required for that normalization.

> Given the post-fact world, we'd expect to find predicates with free terms to be grounded randomly against fictive chaff terms, so long as there weren't immediate pain cause by doing so.   

Yes, but not just immediate pain, immediate anything (head-nodding, facial expressions, interruptions, etc.).  I can recall back in school, walking across campus with a friend of mine who, being an atheist, was very well versed in Judaism.  I made some mistake in the way I pronounced Hava Nagila and he flat out laughed at me.  That immediate correction toward consensus reality became canonical to me at the time.

-- 
☣ glen