2010/10/10 Jorge Llambías
<jjllambias@gmail.com>
>> A: .i ie pei xamgu
>> B: nai .i na xamgu
>
> Okay, two questions.
>
> 1) What does nai by itself even mean? and
It's the answer to the "pei" question.
Ah. I see. I would've answered {.ienai}.
> 2) Has anyone, ever, used nai in that way in conversation, as opposed to as
> an example of "something that causes this to break"?
I have no idea. What I'm asking is if you are aware that you are
proposing a change to the currently official language, or whether you
think you are describing the official language. If the latter, you are
wrong. If the former, I'm not convinced by the arguments you have
given.
Then as you so succinctly put it, I'm wrong.
>> > {.i lo broda cu brode lu .i broda lo brodi li'u .i li'o} is a sinlge
>> > jufra
>> > followed by one or more omitted jufra, as indicated by {li'o}.
>>
>> Syntactically, "li'o" is not a jufra, it just attaches to ".i", but
>> I'm not sure how this has to do with anything anyway.
>
> Hence my use of the word "indicated", which is a synonym for "symbolized".
Well, syntactically there is nothing omitted. I'm not trying to be
difficult here, but we seem to be mixing up semantic and syntactic
issues. We are discussing a syntactic issue, right? We are talking
about the syntactic construct "text", not about what we may or may not
choose to loosely call a "text" in a more general sense. We are
discussing the kind of thing we feed to the parser in the hopes that
it is accepted as a valid "text".
I'm sorry, I thought {li'o} meant "omitted text", as in "crap I don't feel like taking the time to make up for such a simple example, so I'll just put '...' there instead."?
>> > {la.alis.} is a single text. It is composed of a large multitude of
>> > jufra.
>>
>> Yes course, with a single speaker/author, Lewis Carrol.
>>
>> (Strictly speaking, it won't completely parse with the current
>> grammar, but the breaking points are very few. In principle it could
>> have been a single text, yes. It is not a conversation where texts are
>> exchanged between two or more peoiple.)
>
> Neither of those points seem to have any relevance to this discussion. I was
> merely providing an example of what I consider the difference between a
> jufra = sentence and a text to be.
OK. Yes, of course a text can consist of several sentences. That has
never been in dispute. So we agree about that.
> Unlike you, I consider an entire conversation to be a single "text", in the
> same way that I consider a thread on these groups to be a single text.
The second part is irrelevant to the Lojban syntactic construct called
"text". The first part does not work in general. Some valid
conversations cannot be reduced to a single "text" construct.
> Also
> apparently unlike you, I don't think that {mi}, {do}, etc. must remain the
> same throughout a single text, but can - and do - change referents with each
> new sentence.
Normally they (especially "mi") only change referents with each new
speaker, not each new sentence. But right, if you parse a whole
conversation as one text, such text will not have a fixed speaker.
That was my point, wasn't it? You lose that property for texts, of
having a fixed speaker.
You and I disagree that texts ever had that property to begin with. I will agree with you that a single sentence will have those values fixed, but it entirely possible for them to change within the same text:
A: "doi B xu do djica lonu do jo'u mi klama lo zarcu .i doi C xu do djica lonu mi te vecnu lo cukta"
> A: "(.i) [bridi] .i [bridi]" two jufra, one text.
>
> A: "(.i) [bridi]"
> B: "(.i) [bridi]" two jufra, one text.
>
> A: "(.i) ma klama"
> B: "(.i) lo zarci (go'i)" two jufra, one text.
We are obviously talking about different things, since ".i" cannot be
omitted by B in the last two cases for them to be a single text, so
what's the parenthesis for? The parser will not insert the ".i" for
you the way it will insert an elidable terminator,
mu'o mi'e xorxes