> OK, I think I am beginning to get your point of view a little more, and
maybe you are getting more of mine. I still don't think you've quite
defined what you mean by FL2 - a parse tree is not a definition, like Logan
said - but I see that part of your working definition is that it assign a
meaning to any syntactically valid word order, and that it be able to
translate potentially any sentence from another language.

Yeah, a parse tree-based language can be unambiguous although we don't understand it completely. I was thinking about the organisation of nouns in FL1, and realised the parse trees give you more sub-sentences than are necessary. It's easiest to manually arrange the words into noun-case-noun-case-noun-case-action, and that way it becomes much easier to speak and understand, e.g.


This reads something like "The girls kicks the cat in the house because the boy sees a house on a hill." The because-clause doesn't have to be embedded. And if you read it inversely, it can become:

kick: topic 
+++ subject: girl
+++ object: cat: in: house
see: reason
+++ subject: boy
+++ object: house: on: hill

> What would FL3 mean?

FL3 adds one more dimension to an FL2 grammar. It could be another lexical category such as verbs, conjunctions, adjectives or determiners, or it could be the option to omit a constituent; such as a conjunction, a subject marker or an object marker no longer being obligatory. There's only one recursive minimalistic FL1 and perhaps a couple of minimalistic FL2 languages, but probably quite a few FL3 languages. And each option you didn't use for your FL3 will be available for your FL4 etc., so the number of options is going to expand exponentially. There will probably be a mathematical equation for it.

> Even if someone analyzed it that way, I would say it's because they're
comparing with languages they're familiar with and seeing something that's
not there.

On the other hand I did construct it by using what I learned from natural languages. FL2 is an a posteriori language. For instance in English many prepositions can function as conjunctions, e.g. "So I will know before you" (before is a preposition) vs. "So I will know before you see" (before is a conjunction); many examples can be found in a general dictionary. So if it feels like FL2 sentences don't state anything, the same could be argued about the English ones above.

> I would instead translate the sentence differently. [House [glance boy]
hill] or, if you want it to be clearer the house is atop the hill, [house
[top hill] [glance boy] hill].

That's pretty brilliant! So the boy is the subject because it's *his* action. The generated parse tree looks weirder than before, but the written form seems to work just fine. I suppose the word order in the long sentence is now OVS.

>> >> FL2.4 (I'm going to start naming these for easy comparison)
>> S -> aXb: (X there is a, namely b)
>> S -> SS:  S and S
>> X -> S:     S, within which,
>> X -> (empty): (empty)
>> >> Then we'd get:
>> - ((There is nom, namely boy) and ((there is on, namely hill), within
>> which, (there is acc, namely house)), within which, there is topic,
>> namely see.)
>> On is not hill. The on is on the hill; the hill is actually under, which
>> makes it the exact opposite of on.

> In FL2.4, the word "on" is sort of like "platform" - it's saying that the
hill is the type of thing which can support other things. Or actually, the
type of thing which is currently supporting other things. Maybe it makes
more sense if I say it like:

> The house is in an "on", specifically, the hill.

I still don't get it. Unless we're looking for a place-word that expresses vertical relation, whether on or under. That would work although I don't know any examples from natural languages.