[YG Conlang Archives] > [engelang group] > messages [Date Index] [Thread Index] >


[Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: [engelang] Xorban: Termsets



On Sat, Sep 22, 2012 at 8:38 AM, Jorge Llambías <jjllambias@hidden.email> wrote:
 

On Sat, Sep 22, 2012 at 4:30 AM, Mike S. <maikxlx@gmail.com> wrote:
>
> The pseudo-production rules are:
>
> termset-structure := termset-coordinator termset termset
> termset-coordinator := "ne'e" | "ne'a" | "ne'o"
> termset := termset-delimiter | termset-structure | term termset
> termset-delimiter := "ne"
> term := modifier
>
> Mind you, all TSS modifiers are formally unary operators. Those
> productions are ONLY for illustrating the idea. Nevertheless, the TSS is
> the only production here that belongs in a sentence. I have other ideas,
> but I'm going to stop and wait for feedback. I hope this clarifies my idea.

It's an interesting idea, but the pseudo-production rules still leave
some productions from the production rules unexplained. For example,
how do we understand "la mlta ne'e xkra" or "la mlta ne xkra"?

To spell out what you've already puzzled out, the first observation is that, whatever else, "xkra" _can't_ be in a TS, because only a modifer (that's not a TS modifier) can be in a TS.  So "xkra" has to be the trailing formula (not shown in the pseudo-rules).  The second observation is that each TS coordinator has to get two "ne". The third is that each "ne" has to get a "ne'V".


The first one probably should result in "la mlta je xkra xkra"="la
mlta (ni) xkra", which makes "ne'e" and "ne'a" equivalent to "ni" in
that situation, and "ne'o" equivalent to "nu": la mlta ne'o xkra"="la
mlta jo xkra xkra"="la mlta nu xkra".

Yes. "la mlta je xkra xkra" < "la mlta ne'e [ne] [ne] xkra", etc.

 
I suppose "la mlta ne xkra" would be just "la mlta xkra"?

That's what we'd want, because "ne" means "put trailing formula here".  What our semi-formal TSS rules so far say is,

"la mlta ne xkra" > "[ne'V] la mlta ne xkra" > "[ne'V] la mlta ne [ne] xkra"

That would mean "jV la mlta xkra [la ?a] xkra".  If "jV" were "je" and the implicit binding rule were agreed on, it's possible that we'd be done.  Since it's not agreed on, we can say that a sentence with _solitary_ "ne" is allowed to group with "ni" placed in front of it instead of "je ... ne" treatment.  Or that solitary "ne" just means 0.


And "la mlta ne la grka ne xkra" is equivalent to the two sentences
"la mlta xkra la grka xkra"?

Yes, that's why I first tried to put "ne'V"="ne'e" in front of the sentence in the solitary "ne" example. 

What's nice about these rules so far is that we won't need "ne'e" in sentences with no leading serial termset;  a formula with N "ne" would be interpreted as starting with N-1 "ne'e".  Even when people make "mistakes" omitting "ne'e" where it should be, e.g.

la ma djna [ne'e] le nnle li crdi ne le nxle li ckti ne dndakike

=> je la ma djna le nnle li crdi dndakike [la ma djna] le nxle li ckti dndakike

... things will often work out.  Yet again, the implicit binding rule proves its usefulness.

--
co ma'a mke