Re: ago (LONG and la'a incorrect :))

[Goran Topic <topic@math.hr>]

> coi rodo

.ui coi doi xorxes.

> mi xruti i mi facki le du'u ji'imuno selmri cu denpa le nu mi tcidu

.i .ie .i lojbo liste selmri tolcafne ca'o le slatei

> i pamai mi dunda lei mi re fepni sera'a le du'u tai makau jarno
> le kambra be le'e selku'e

.i ri'ada'i lonu lonu mi tcedjica lenu la lojban. klunu'i cu fanza do ku
ko cusku fi mi

.i lu da dunda lei da re fepni sera'a de li'u ba'e je'a tcegliklu

> Djer:
> > On the wordiness question, dare I mention yet another experimental
> > cmavo?  Would a word for "ago" really rent the fabric of lojban?
>
> And:
> > Yes, it would. "Ago" means "before now", as, I think, does "ba" when
> > it has no overt complement. So "ago" = intransitive ba.
> >
> > Sorry. For "ba", read "pu".
>
> I don't think a new word is needed, but "pu" doesn't really help.
> "Ago" means "before now" but it always comes after a measure of
> time, so it means more than "puku", which cannot be complemented with
> a time length.
>
> "X ago" means "a length of time X before now". I think that the most
> natural way to do this in Lojban is with "za", which is precisely that:
> a length of time.
>
> Consider:
>
>         mi pu za ze'a tcidu
>         I <past> <a while ago> <for a while> read.
>
> To make the interval of reading more specific, we say:
>
>         mi pu za tcidu ze'a le cacra
>         I <past> <a while ago> read <for an hour>.
>
> Similarly, I think that the way to make the offset more specific is:
>
>         mi pu ze'a tcidu za le cacra
>         I <past> <for a while> read <an hour ago>.
>
> We can also add a starting point different from "now":
>
>         mi tcidu pu le nu do klama kei za le cacra ze'a le cacra
>         I read for an hour, one hour before you came.
>
> So, in my opinion, there is already a word for "ago" in Lojban
> (actually three words: zi, za and zu) but instead of coming after
> the time length like in English, it comes before it. (Also, it can
> be used for the future equivalent of "ago", which in English doesn't
> have a single word: "from now".)
>
> I think that the tense paper says that "za <sumti>" means "a while
> before/after <sumti>", but I don't think that is very useful, and
> a way of clearly specifying the length of the offset is necessary.
>
> The exact same thing happens with spatial tenses.

Yes. But I don't think the present is not useful. Especially the
spatial equivalent, VI series. Don't tell me you claim never to need
something like {ko'a kelci va le ckule}?

There is much less meddling and change and trouble if we would just
introduce new sumti tcita, instead of changing the existing ones. I
mean, something like new cmavo, say, {xe'i} (after temci - te'i is not
allocated, if I didn't miss anything). I don't remember {xe'i}
experimental cmavo being in use, too.

mi tcidu pu lenu do klama kei xe'i lo cacra kei ze'a lo cacra
I read <past> you come, <interval> [1] hour, <medium duration> [1] hour.
I read one hour before you came, for one hour.

Spatial would be a bit more wordy, because it would (at least with me)
need {fe'e}, since it would be good to keep spatial and temporal tenses
apart... Much less ambiguity possibilities.

lo verba cu kelci zu'a xe'i fe'e lo mitre be li ci
Child plays <left> <interval> <spatial> 3 metres.
Child plays 3 metres to my left.

Or you could say that it encompasses both spatial and temporal
distance (in which case {fe'e} is not needed), and measures that
which precedes it (saves two syllables in spatial expressions):

carvi pu xe'i lo djedi be li ci be'a la mai,emis. xe'i le minli be li panono
Rain, <past> <interval> 3-day <north> Miami <interval> 100-mile
Three days ago, it was raining 100 miles north from Miami.

Then you could also say, for example,

mi pu te vecnu ti xe'i lo cacra ku joi lo minli be li re
I <past> buy this <interval> hour and 2-mile
I bought this an hour ago and two miles away.

Enough raving. Tell me what you think. Or to shut up. :)
ta'o is there a gismu for distance, like equivalent for {temci}, or is
it necessarily {nilda'o} or {da bi'i de mitre di}-like expressions?

> Iain:
> > How _do_ we say "The window is three metres to the left of the door"?
>
> I would say:
>
>         le canko cu zvati zu'a le vorme va lei ci mitre
>
> (If you don't like {lei ci mitre}, then {le mitre be li ci} or
> {li ci poi mitre ke'a} are equally usable.)

See above.

> Djer:
> > I predict that the problem will not go away. I think it is unrealistic
> > to have awkward and prolix expressions for frequently used concepts. As
> > you know, the reason some expressions are short is their high
> > frequency.
>
> I agree. And I haven't seen any way of expressing "ago" other than using
> "temci" as the main selbri, but that seems a bit too drastic. It's not
> just a matter of conciseness, there should be a clear way of specifying
> offset lengths.

Yes.

> If you order the lojbo squad to march towards the right, they might do
> so without first facing right, no? So first you have to order them to
> face right. My attempt is {le pritu [be do] ko crane}, but the doubt is,
> when is {le pritu} evaluated? Does this mean "make it such that you are
> facing towards what is now your left", or "make it such that you will
> be facing what will be then your left"?

Umm. Make that 'right'.

Ugh. Tough. By the formulation of all explanations of {ko}, it appears
that the latter is a bit more convincing, if not useful, IF we accept
the PRESENT TENSE for {pritu}.

But, maybe the question is wrong - not when is it evaluated, but what
tense context do we assume for {le pritu be do}. I still believe
that each bridi has all the tenses, although ellipsised (what IS the
current consensus on this, anyway? Is there one?), so what is stopping
us to think of the sentence as

le pritu [be do bei pu zi dei] ko crane
left of-you <past> <short interval> this-utterance of-you-imperative be-front
Face that which was your left immediately before this utterance happened.

(assuming I got the dei-as-event thing right :))?

> Jorge

co'o mi'e. goran. poi ko na .e'oga'inai catra vau zo'o

--
Learn languages! The more langs you know, the more incomprehensible you can get
e'udoCILreleiBANgu.izo'ozo'onairoBANguteDJUnobedocubanRI'a.ailekadonaka'eSELjmi