it's clear to me that 'thing-ness' is a layer that can turn on and off, and can be wildly wrong, and I've observed the label on a thing change in realtime often :)
below are my answers:
I see red, then very shortly after the word resolves (the idea of green gets activated in various sense modalities in thought)When you look at the word GREEN’, what is the actual experience?
visual experience is the colour red, yeah. thoughts stirred up linguistically have colour experiences in them but they're not actual visual experiencesIs the red color "experienced" or is color green "experienced" as the label suggests?
oh my no. most labels are infinitely sub-divisible and also able to be subsumed into larger parts indefinitely (my arm, my whole body, a crowded room, a riot might all apply to 'me' somehow at any point)Do the labels have a one-to-one correspondence with "reality?"
yes there's the green idea complex being activated when no green is presentOr do the labels suggest something else other than what is here now (red color)?
the latterIs green-ness an inherent attribute of the "experience" of the red color; or is green just a word label on the experience of the red color?
NoIf the label ‘GREEN’ is replaced with the label ‘GOOD’ or ‘BAD’, is the redness affected in any way as the labels suggests?
No effect whatsoever :)Does redness become "good" or "bad," or do the labels have no effect whatsoever on ‘reality’?
I've been thinking intermittently about how reading works with the interaction of visual perception, a lot of fairly abstract shape, word, structure, etc recognition processes and then thoughts are produced out of each of those processes independently (like I see a word and know what it means, but I also know what it means in this sentence and what the sentence means so far, and so on).
This all happens with basically no lag or perceptible cost. I've been trying to watch the details but reading is an awesome way to distract yourself with a bombardment of potentially sticky ideas and spun-off scenarios.

