View source for allow forethought sumti tail connection



Nobody has thought of any reason why this should be ungrammatical, as it currently is. -- mi'e [[User:And Rosta|And Rosta]]

''Care to give us any examples? -- Adam''

''le ge nanmu gi lojbo''

= ''le nanmu je lojbo'' (corrected from: ''le nanmu gi'e lojbo'')

= ''le du poi ke'a ge nanmu gi lojbo'' -- [[User:And Rosta|And Rosta]]

That should be written ''le gu'e nanmu gi lojbo''.  See [http://www.lojban.org/files/reference-grammar/chap14.html#s12].

* My (perhaps incorrect) understanding of GUhA is that it conjoins selbri/brivla rather than sumti tails. If ''le gu'e pendo be la djak gi mamta be la djil'' is well-formed, then GUhA does forethought sumti tail connection. It remains the case that there is no good reason why GA shouldn't do the same.

If your construction were grammatical, I would tend to interpret ''mi viska lo ge nanmu gi lojbo'' as "I saw a man and a Lojbanist".

* That would be a common mistake, yes, due to glico influence. Cf. Woldy p350.

''le nanmu gi'e lojbo'' is clearly ambiguous. Consider "mi viska le nanmu gi'e lojbo". If you are referring to only one individual, then tanru logical connection is best.

* How is that ambiguous? It means "I see the man, and am Lojbanic."
** According to the proposal, it might mean ''mi viska le ge nanmu gi lojbo''.

*** This, if nothing else, should make it clear that this proposal is a Bad Idea. --mi'e [[.djorden.|.djorden.]]
*** By saying that, you create the impression of not having read the proposal, even though it is at the top of the page, and your response accordingly seems even more unthinkingly kneejerk than one might otherwise suspect it to be. The page is called "'''forethought''' sumti tail connection"; the proposal concerns GA not GIhE. (The paraphrasing example wrongly used ''gi'e'' instead of ''je''.)

**** Introducing a syntax of forethought connection which doesn't translate into afterthought is Bad.
***** Afterthought connection is a high-cost accommodation to users' needs. As I said on another page, "it's a privilege, not a right". It is reasonable to wish for forethought counterparts of every afterthought, but not reasonable to demand afterthought equivalents of every forethought.

****If all the 'proposal' deals with is forethought version of tanru connective "je", we already have that, as someone mentioned. --mi'e [[.djorden.|.djorden.]]
*****See above: "My (perhaps incorrect) understanding of GUhA is that it conjoins selbri/brivla rather than sumti tails. If ''le gu'e pendo be la djak gi mamta be la djil'' is well-formed, then GUhA does forethought sumti tail connection. It remains the case that there is no good reason why GA shouldn't do the same."

****** Umm; "je" joins inside selbri/brivla also....  Your example sentence is well formed, but it has to do with the selbri rules, not sumti tails.  The link args with be happens inside tanru-unit, i'm not sure if you consider them relevant or not....  Looking at the grammar, it seems sumti-tails are *never* connectable, which makes sense since you do that in the selbri rule that it calls, so I'm guessing you might have meant "bridi tails":  But je has nothing to do with that either.  It is safe to say your 'proposal' needs some clarification: I'd guess it was made without fully understanding the existing rules of connection inside of selbri and/or connection of bridi-tails. --mi'e [[.djorden.|.djorden.]]
****** That's right. Convert ''da poi ke'a ge mamta la djak gi pendo la djil'' to a semantically-equivalent form with ''lo''. If I understand you, GUhE does this. I'm not sure what you mean by ''link arguments with be'' -- ''mamta be la djan (bei) pendo'' is just a plain tanru, isn't it, so not relevant. --[[User:And Rosta|And Rosta]]

******* gu'e does *not* do this; but neither does 'je'.  Your proposal isn't even internally consistent, so I'll stop commenting on it for now (unless/until you fix it).  --mi'e [[.djorden.|.djorden.]]  (ps: I agree link arguments in a tanru unit are irrelevant).
******* At least I was right in thinking that GUhE doesn't do this. I don't know what you want me to fix. The proposal asks for a way to convert ''da poi ke'a ge mamta la djak gi pendo la djil'' to a semantically-equivalent form with ''lo''. It suggests that there is no obstacle to making licit ''lo ge mamta la djak gi pendo la djil''. Is your complaint that this isn't formulated in terms of specific changes to the formal grammar?