Postby janKipo » Fri Aug 19, 2016 5:22 pm
I don't see what fractals have to d with it, except, I suppose, it is the same all the way up and all the way down. But the point about evaluations is just that they are not winners and losers; everybody might get an A (or an F). So what you say about "evaluate" -- aside from the stuff about proto-semantics (whatever that is) has some point: it is about scoring against a standard. "Standard" is very likely 'nasin' and scores tend to become nanpa eventually, so 'nanpa nasin', "number according to a standard" seems in the right area, at least. And, of course, we need what the standard is for, "quality" , 'pona', so 'nanpa pi nasin pona;.
But groups don't have to have leaders or heads or tails or even definite parts; they're just things somehow set off from ohers. To be sure, most of the important ones do have some or all these added details, but that makes them more than groups.
Inversion changes sense clearly with directional concepts, like 'ilo', but less obviously for other cases. 'soweli moku' can be the same as 'moku soweli' (though also very different, of course) and I expect the same applies to 'kasi' (more so, since we don't talk as much about food for plants). The fact is that modification is so vague that almost anything is possible. 'moku tomo' is usually home made food or house specialties, for example (never fuel or electricity, though I can see why it might be).
I don't get the last bit, but it looks nice.