Oh yes, I see. Still I'm not sure it would be worth it to change the
grammar to make this work. For example

"broda be mi je brode" would change from "(broda be mi) je (brode)" to
"broda be (mi je broda)"

(in experimental grmmar) and that would be weird.

Comment #14: Re: Interaction with Modals and Connectives

gleki (Sun Dec 28 06:50:01 2014)

durka42 wrote: > Oh yes, I see. Still I'm not sure it would be worth it to change the > grammar to make this work. For example > > "broda be mi je brode" would change from "(broda be mi) je (brode)" to > "broda be (mi je broda)" > > (in experimental grmmar) and that would be weird. It won't change since there is no zi'o generator there. su'ai is one.

Comment #15: Re: Interaction with Modals and Connectives Alex Burka (Mon Dec 29 04:45:59 2014) > It won't change since there is no zi'o generator there. su'ai is one. Hmm so su'ai modifies the behavior of the following connective so that it no longer connects things but just attaches to the sumti before it. But if su'ai has that power it can't be in SE, since the grammar will treat ``su'ai jo the same way as ``se ju. Comment #16: Re: Interaction with Modals and Connectives Alex Burka (Mon Dec 29 04:47:51 2014) durka42 wrote: > > It won't change since there is no zi'o generator there. su'ai is > one. > > Hmm so su'ai modifies the behavior of the following connective so that > it no longer connects things but just attaches to the sumti before it. But > if su'ai has that power it can't be in SE, since the grammar will treat > "su'ai jo" the same way as "se ju". There's no winning when working with undocumented wiki syntax, is there. See above quote for corrected version. Comment #17: Re: Interaction with Modals and Connectives gleki (Mon Dec 29 06:53:43 2014) durka42 wrote: > durka42 wrote: > > > It won't change since there is no zi'o generator there. su'ai is > > one. > > > > Hmm so su'ai modifies the behavior of the following connective so that > > it no longer connects things but just attaches to the sumti before it. > But > > if su'ai has that power it can't be in SE, since the grammar will > treat > > "su'ai jo" the same way as "se ju". > > There's no winning when working with undocumented wiki syntax, is there. > See above quote for corrected version. su'ai deletes x2 in selbri and moves its value into a unified x1 where former x1 and x2 are connected using ce or jo'u depending on your dialect. It works exactly the same with connectives: x2 is deleted. SE changes selbri and that's what is needed here: a new selbri with a new place structure appears. If it can't be in SE then okay. Neither it put syntactically a zi'o into x2 since zi'o can be reached with fe but in "su'ai broda" former x2 can never be reached again. So su'ai is rather similar to "zil-" but not zi'o. Comment #18: Re: Interaction with Modals and Connectives gleki (Mon Feb 16 16:42:35 2015) gleki wrote: > durka42 wrote: > > > Hmm so su'ai modifies the behavior of the following connective so > that > > > it no longer connects things but just attaches to the sumti before it. > > But > > > if su'ai has that power it can't be in SE, since the grammar will > > treat > > > "su'ai jo" the same way as "se ju". > > > > SE changes selbri and that's what is needed here: a new selbri with a new > place structure appears. > > If it can't be in SE then okay. Neither it put syntactically a zi'o into > x2 since zi'o can be reached with fe but in "su'ai broda" former x2 can > never be reached again. > > So su'ai is rather similar to "zil-" but not zi'o. xorxes thinks that it's okay to be in SE since e.g. zo'e and zi'o are both in SE. Not even a subselmaho since zi'o and zo'e are both in koha7. Thus somewhere in subsubselmaho. And of course in su'ai jo the word su'ai modifies jo.