Re: [lojban] "gi" in place of "zi'e"

[tijlan <jbotijlan@gmail.com>]

On 14 August 2011 12:45, Pierre Abbat <phma@phma.optus.nu> wrote:
> On Sunday 14 August 2011 06:40:11 tijlan wrote:
>> Multiple relative clauses (NOI) / phrases (GOI) can be joined up,
>> officially by "zi'e":
>>
>>   da poi [ broda ] zi'e noi [ brode ]
>>   da poi [ broda ] zi'e pe [ de ]
>>
>> I wonder if "gi" could substitute for that joiner:
>>
>>   da poi [ broda ] gi noi [ brode ]
>>   da poi [ broda ] gi pe [ de ]
>
> Semantically, this doesn't make sense. "zi'e" is a logical and of the zihek
> class; there is no logical or, nxor, one-side-irrelevant, or question in the
> class.

So, "zi'enai" would mean "jenai" and not "non-zi'e" (whatever that could mean)?


2011/8/14 Jorge Llambías <jjllambias@gmail.com>:
>> On the other hand, "zi'e" was coined before we started using PEG. It's likely
>> that the word was deemed necessary because of the limitations of LALR1. PEG
>> has no problem seeing that "noi" or "pe" follows, so one of the other classes
>> of conjunctions should be usable in PEG.
>
> Yes, "relative-clauses <- relative-clause (jek relative-clause)*"
> makes more sense than "relative-clauses <- relative-clause (gik
> relative-clause)*" since we could also have "gek relative-clause gik
> relative-clause".

The other ziheks were eliminated because ... they were deemed
unnecessary, I suppose.

All "GOI sumti" except perhaps "goi sumti" can be converted to "NOI
selbri sumti", and of course multiple selbri can be coordinated by jek
etc. within one clause started by the same NOI. Then it may be said
that a joiner for multiple clauses is most grammatically wanted when
we want different NOI for each clause. That is when we want to have
the clauses differently as
  veridical non-restrictive (noi)
or
  veridical restrictive (poi)
or
  non-veridical restrictive (voi).

In other words, however, if we could indicate those differences
through other (existing or non-existing) discrete cmavo, there would
be no absolute need for different NOI and then for "zi'e" either, I
guess. (Such semantically discrete and perhaps primitive cmavo for
veridicality and restrictiveness would also make "le/lei/le'e"
redundant and even help disambiguate our expressions for the purpose
of which we traditionally use those gadri, e.g. "le cribe" as "a
particular population of lo cribe" [restrictive] and "what I
subjectively consider to be a bear" [non-veridical]".)

mu'o

-- 
You received this message because you are subscribed to the Google Groups "lojban" group.
To post to this group, send email to lojban@googlegroups.com.
To unsubscribe from this group, send email to lojban+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.