Ovinomancer
No flips for you!
I do not know what you mean by "valency" here. It's a word that has meaning to me, both chemically and linguistically, but I'm not at all clear on how you think it's representative of a concept here. You seem to be using it in the sense of positive correlation or attraction?I want to acknowledge an adjustment to my position. In saying the cues represent the fiction, I am not saying they are the fiction. Rather I say they can (should) have a strong valency such that manipulating one, manipulates the other. [You could say "crucially informs" or "drives discoveries about" here, in place of "manipulates".]
That appears to be a central point of contention. I say that arrangements of symbols and rules for manipulating them has consequences for our fiction: is productive!
And where it is not productive I sincerely question the point of having them. @pemerton might say they help us remember, and I say they help us remember and (via thinking by analogy) systematically produce a progressive fiction.
As for your last, your summation of @pemerton's arguments here is badly skewed, as that's not at all close to the entirety of what he's been saying but rather a single bit out of context. Cues help focus fiction, yes, but that doesn't mean that cues are the fiction or that manipulation of the fiction requires manipulation of a cue or vice versa. For example, I might have a cue in the form of a miniature of my character. This miniature can be closely representative of how my character appears or it could be a penny. The actual fiction of how my character appears isn't tightly related to the token I'm using, nor does a manipulation of the token mean a manipulation of my character's fiction. For example, my character can lose an arm in the fiction and yet the figure retains both arms. Or, the figure might become damaged and lose an arm, yet my character doesn't lose an arm in the fiction. Cues are there to help share complex constructs that are hard to accurately share and hold in your head, but they're suggestive to the fiction at best, and there's no required one to one manipulation from one to the other. Yes, a given ruleset might have more use for cues than another, but the scale that this rest upon doesn't require that cues be directly representative of the fiction or vice versa. Cue exists to help share imaginations, they are not part and parcel of them.
In 5e, I can have a combat run with a grid and miniatures, and I can require that all miniatures be representative to some degree or another, but in doing so, in creating a detailed and modeled combat space with minis that were hand painted and sculpted to match the fictional descriptions of the participants, I'm still not to the point where the cues completely define the fiction. They represent general locations (within a 5' space) and don't mimic the thrusts of spears, swipes of claws, roars and screams of battle, etc. They cue me to imagine certain details, yes, but I still must imagine them, and I must imagine more than the cues provide. And this is at the high level of detail, because I could also run a 5e combat with none of these cues, or one where it's a hand drawn sketch and we use pocket change to get a general idea of what's where.