Is "i" the end or the beginning of a statement? what about "niho"?

46 views
Skip to first unread message

vpbr...@gmail.com

unread,
Jul 13, 2017, 1:43:45 AM7/13/17
to lojban
The question of whether "i" terminates a statement or begins one may not seem to make a lot of difference,
but for me it determines whether I place the "i" at the end or beginning of a written line and determines when I pause in speech between statements.
From the syntax rules, there seems to be no distinction, because "i" mainly serves as a statement separator,
and at the beginning of an utterance as well as at the end the "i" is permitted but optional.

The critical question for interpreting the "i" may be:  what happens when there is a long pause between statements, or perhaps a speaker turn?
At what point is the prior statement complete, ready to wrap and ship, fully committed?
It is clear that no statement is complete until you have seen the "i" (or "niho"...),
and if it hasn't emerged yet, we are still waiting to see how the statement may yet turn out.
Most bridi do not have every sumti place filled, but even if all the places are filled and then a delay stretches on and on,
you cannot tell whether there is still a "vau zoho" to come, or a "fau lo nu lo xarju ba vofli", or a "giha mi bebna".
Until you hear the "i", you are just left hanging.

That is why "i" terminates statements, and why I place it at the end of written lines, and before a spoken pause if I pause.

Syntax like "i je bo" makes it seem like "i" may be starting this statement, but I think not.
The "i" is terminating the previous statement, even if that statement is empty.
"i i je bo fagri" and "i je bo fagri" are equally grammatical utterances,
while "je bo fagri" is ungrammatical because "je" does not connect with anything on its LHS.

This made me wonder whether "niho" and "nohi" should be treated the same as "i", and now I think not.
First, even though people try to do it, you cannot say "... niho ba bo ...".
The "jek/joik/stag BO" connects two statements together, and this does not work over a paragraph break.
Separate paragraphs do not connect this way.

Second, when you are finishing one statement, you often do not know whether the next statement is going to be in a new paragraph.
You haven't decided what you will say next yet at that point.
Only after the delay for thought in between statements might you newly realize: now I'm gonna turn to something different.
And since it is different, it won't be connected to the previous statement logically or with a tag.

That is why I now begin new paragraphs with the niho or nohi at the start of the written line, and after a possible spoken pause.
The "i" that optionally terminates the prior statement may or may not be included just before the "niho".

E.g.

mi pu gleki lo nu penmi do i
je bo mi terpa lo nu rinsa do i

niho mi ba finti lo se cukta poi srana lo nu terpa i

mihe bremenli



Jonathan Jones

unread,
Jul 13, 2017, 1:03:30 PM7/13/17
to loj...@googlegroups.com
It's the beginning. vau is the end of a statement and can normally be left out.

Same for ni'o.

--
You received this message because you are subscribed to the Google Groups "lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban+unsubscribe@googlegroups.com.
To post to this group, send email to loj...@googlegroups.com.
Visit this group at https://groups.google.com/group/lojban.
For more options, visit https://groups.google.com/d/optout.



--
mu'o mi'e .aionys.

.i.e'ucai ko cmima lo pilno be denpa bu .i doi.luk. mi patfu do zo'o
(Come to the Dot Side! Luke, I am your father. :D )

Ilmen

unread,
Jul 13, 2017, 5:06:42 PM7/13/17
to loj...@googlegroups.com
Yeah, I've been annoyed by this issue of ambiguity of end of
transmission. Adding ".i" at the end of your final utterance technically
doesn't solve the problem, as the listener will probably expect you're
going to say another sentence… Even {mu'o} is unsatisfying: when one
hears mu'o and then a pause, they can't be 100% sure the speaker isn't
going to add "do" for {mu'o do} or even {mi'e ~~~} after that. {fa'o}
has been designed for marking an unconditional end of transmission, but
I find that rather too rigid. I'd have preferred a particle saying
"Right now I've nothing more to add, yet I might change my mind
afterwards and say something else. But so far please assume I've
finished and feel free to speak up if you desire so"; maybe with the
same grammar as ".i".

As of now in spoken Lojban, people seem to most often rely on the length
of silence after an utterance for judging whether the time is
appropriate to speak up to their turn. Yet Lojban was intended not to
rely and prosody and sound length for carrying meaning, so using silence
for signifying an end of transmission isn't great. On the other hand if
you use {fa'o} it's prescribed anything you'd say afterward should be
dismissed, which is a little too extreme I think.

—Ilmen.

suke...@gmail.com

unread,
Jul 13, 2017, 7:38:18 PM7/13/17
to lojban
coi

Hm, maybe beginner questions (yes, I am), but:
  • What about the hesitation ".y." that we emit, in practice, when we did not finish our sentence but are still looking for what to say next (http://lojban.github.io/cll/19/14/)?
  • Isn't it intended that ".i" is at the beginning? That way, a (relatively long) silence is always interpreted as the end of your sentence (as you rightly pointed out). Doesn't this avoid the speaker from "stealing / monopolizing" the communication channel by never closing his/her sentence?
Well, my two cents...

co'o

vpbr...@gmail.com

unread,
Jul 14, 2017, 1:34:28 AM7/14/17
to lojban
A vau is not the end, it can be followed by free elements that modify the preceding statement,
and if there was a gihek, then vau can also be followed by tail-terms and even another gihe clause.
Only I/NIhO is the real end.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban+un...@googlegroups.com.

To post to this group, send email to loj...@googlegroups.com.
Visit this group at https://groups.google.com/group/lojban.
For more options, visit https://groups.google.com/d/optout.

Michael Turniansky

unread,
Jul 14, 2017, 8:26:48 AM7/14/17
to loj...@googlegroups.com
.i is neither the beginning nor end.  It is between sentences.  It is officially a "linker".  If I hear you say "i" I will wait for you to say your next utterance.  OTOH,  If you are really that worried about turn taking, and do not want to really on silences that every single human language (spoken or signed), using "mu'o do'u" will do it.

                  --gejyspa


To unsubscribe from this group and stop receiving emails from it, send an email to lojban+unsubscribe@googlegroups.com.

gleki.is...@gmail.com

unread,
Jul 14, 2017, 10:47:40 AM7/14/17
to lojban
{vau mu'o do} and {vau vau mu'o do} must solve the problem. {i} (indeed it's formally neither a start of a paragraph nor necessarily its end) must also alleviate the problem. If they don't that means Lojban simply lacks enough means to handle conversations with precise tools to reference uttterances that appear asynchronously.
{fa'o} as its definition says is not for humans 

.i le jbobau zo'u do cusku zo fa'o ca le nu do pu'o morsi

Vincent Broman

unread,
Jul 14, 2017, 4:33:03 PM7/14/17
to loj...@googlegroups.com
I'm not so much worried about turn-taking, which can be done with
vocatives or suprasegmentals, but rather about chunking.

If I write in lines, and the reader processes by lines instead of word
by word, then what should go on a single line so that the reader gets
complete thoughts? If his parser is not back to the top-level at the end
of the line, then he's left hanging, unfinished, waiting for those zohe
and pending terminators.

If someone speaks with pauses between thoughts, and someone listens to
the end of the thought before fully digesting it and formulating a
response (wow, what a concept!), then it is friendly and helpful for the
speaker to put his pauses at points where the listener knows the
statement is complete, wrapped, and shipped, so he can then parse,
understand, and react.
The place where everyone knows the statement is complete is after the
separator word, not before it.

mihe bremenli


On 07/14/2017 07:47 AM, gleki.is...@gmail.com wrote:
> {vau mu'o do} and {vau vau mu'o do} must solve the problem. {i} (indeed
> it's formally neither a start of a paragraph nor necessarily its end)
> must also alleviate the problem. If they don't that means Lojban simply
> lacks enough means to handle conversations with precise tools to
> reference uttterances that appear asynchronously.
> {fa'o} as its definition says is not for humans
>
> .i le jbobau zo'u do cusku zo fa'o ca le nu do pu'o morsi
>
> Em sexta-feira, 14 de julho de 2017 15:26:48 UTC+3, gejyspa escreveu:
>
> .i is neither the beginning nor end. It is between sentences. It
> is officially a "linker". If I hear you say "i" I will wait for you
> to say your next utterance. OTOH, If you are really that worried
> about turn taking, and do not want to really on silences that every
> single human language (spoken or signed), using "mu'o do'u" will do it.
>
> --gejyspa
>
>
> On Fri, Jul 14, 2017 at 1:34 AM, <vpbr...@gmail.com <javascript:>>
> <https://groups.google.com/group/lojban>.
> For more options, visit
> https://groups.google.com/d/optout
> <https://groups.google.com/d/optout>.
>
>
>
>
> --
> mu'o mi'e .aionys.
>
> .i.e'ucai ko cmima lo pilno be denpa bu .i doi.luk. mi patfu
> do zo'o
> (Come to the Dot Side! Luke, I am your father. :D )
>
> --
> You received this message because you are subscribed to the
> Google Groups "lojban" group.
> To unsubscribe from this group and stop receiving emails from
> it, send an email to lojban+un...@googlegroups.com <javascript:>.
> To post to this group, send email to loj...@googlegroups.com
> <javascript:>.
> <https://groups.google.com/group/lojban>.
> For more options, visit https://groups.google.com/d/optout
> <https://groups.google.com/d/optout>.
>
>
> --
> You received this message because you are subscribed to a topic in the
> Google Groups "lojban" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/lojban/sUWThDLZUxs/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to
> lojban+un...@googlegroups.com
> <mailto:lojban+un...@googlegroups.com>.
> To post to this group, send email to loj...@googlegroups.com
> <mailto:loj...@googlegroups.com>.
Reply all
Reply to author
Forward
0 new messages