Currently enjoying some Pai Mutan, the Court Jester of Teas, while listening to the oddly overrated Ani DiFranco's 「Freakshow」, which dances the line between irritating and cool. Earphones allow me to upgrade to Dar, one of my favourite songs.
Idea of a general system for the "how" of learning in speculation: interesting. Why do we have these types of conversation? Two types:
- Counterfactual - What if X-at-variance-with-reality were true?
- Reality-additional - What if X-we-dont-know were true?
Hypotheticals: approach reality at right angles to attempt to find consistency that holds across conceptual boundaries - hunt out soft and hard edges of a judgement(*). Sometimes this is not explicit - exposure of vague ideas of a concept, æsthetic exposure to 「themes of life」 in novels (Ayn Rand? L. Ron Hubbard?, N's 「Also Sprach」?).
Some dislike the very process of trying to figure out what one believes in the abstract, out of their fear of offending someone, or deciding to oppose something somebody they like is involved with - social thinkers, they abandon decisions based on ideas, and instead profess anything that will make the people around them the most confortable, waffling when between people. We reject and condemn wholeheartedly this practice - those who commit to values and thoughts, even those who commit enough to say "I do not care about that", or who say "I don't know enough to have an opinion" are better than dishonest intellectual cowards who fake existence in the world of ideas. Note as well that taking a position does not prohibit comprimise, nor does it mean one holds all one's values as precious as each other or that one has figured out how they fit together. Philosophy is naturally like Origami - a given position can result from the folding of many values together, and it could specify any wrinkle desired in practice, but the metaphilosophical aesthetics people have, along with the values so expressed, act together to judge one's system and allow arguments between such systems.
Initial temptation: think of transactional layers in a database as roughly similar to challenge of coherently operating in variant (either flavour as described above) Weltanschauungen. Issue of course is that there is never an intent to commit changes in the same way - one doesn't (or at least shouldn't) inject the test as true into one's worldview ("suppose we we were in Europe, ...." should not lead us to change our notion of where we are, except in the case of a very atypical use of this style of argument where we try to match the content to reality to suggest the strong or unique plausbility of the test), instead we either use these en-masse to "feel out" the possibility-space-size of a conviction, or we poke at what we think are value-complex hypotheticals to weigh the relative strength (or presence!) of the values and styles of reasoning involved in producing a judgement.
Sometimes surprising: that we are able to correctly(ish) able to extract meaning using this style of reasoning. It is no wonder that we don't do it well when we are not very careful, it is surprising that we are able to do it at all... although it's not hard to see how useful it would be. Another flavour of this style of thought: contingency planning, presumably useful in the EEA. In contingency plans, do we leave the plans to "hang in the air" as a type of Reality-additional, or is modifying one's instincts to an appropriate response for reasonably-likely outcomes and then dropping the explicit hypothetical idea enough? To what extent is this how we think about morality or construct the foundations for guilt? (reference: the potential gap, when one is actively shifting one's value system, between the values "stored" in one's foundation for guilt and the new ones one is adopting - a christian convert from islam might feel guilt for a failure to perform ablution in some circumstances until they have adapted those foundations - foundations are *supposed* to lag behind one's values because guilt acts to guard the "big picture us of our values" from the us-of-the-moment)
Another potential tie: experiences with video games occasionally inducing violent behaviour and situation-inappropriate responses - a kind of abstraction violation.
Feeling: May also has something to do with supersampling in statistics, how can we maximally learn from our experiences given a world-model and statistical intuitions?
It's stuff like this that makes me doubt most cognitive models I've seen come close to high human reasoning - they may capture some of the basics of lower thought, but I've never seen them come close to topics like this. (Reader can supply the obligatory-and-worthwhile slam of the "Semantic Web" reductionists/formalists here).
Note: I have always found the British reformed spelling, "judgement", prettier than the pre-reformed/American spelling "judgment".
Also: Just realised that a cherished bit of silliness I've had for a few years is based on a misremembering - I sometime back glued together a moment in Super Mario Brothers (stage 1-2 - the first underground stage) where on can jump up a set of platforms and bump a block so a starman (invincibility item) pops out and bounces forward across that area until the player can make it to an area where a gap in the ceiling lets it fall - I glued that with David Bowie's 「Starman」 (where it's waiting in the sky, it'd like to come and greet us but ...). Bit of a bummer - I just realised that it was in fact a 1-up mushroom on the ceiling, not a starman. A quick google search for super mario 1-2 map confirms it - the starman is earlier in the stage. Oh well. Passing off a secondhand quirky memory here seems a suitable end to that (weird) misconnection.
I kind of wish there more people around here who liked talking about these kinds of things...