Far away

19 views
Skip to first unread message

Luke Flatley

unread,
Aug 15, 2011, 6:12:04 PM8/15/11
to lojban
How do I say things like "2,000 light years away (from me)", "80,000
leagues beneath the sea" and "Thousands of miles from you"?
Basically, sentences of the format "(number) (units) (preposition)
(location)".

Pierre Abbat

unread,
Aug 16, 2011, 9:10:46 AM8/16/11
to loj...@googlegroups.com

vu lo gusna'a be li reki'o to'o mi. Or should that be "ze'o"?

Pierre

--
li fi'u vu'u fi'u fi'u du li pa

Ian Johnson

unread,
Aug 16, 2011, 10:16:13 AM8/16/11
to loj...@googlegroups.com
That doesn't work, because {vu} specifies the origin. We're (I assume Luke above is lukys on IRC) still trying to decipher this: http://dag.github.com/cll/10/25/

mu'o mi'e latros


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


.arpis.

unread,
Aug 16, 2011, 10:25:04 AM8/16/11
to loj...@googlegroups.com
I seem to recall being told that termsets (being confusing and almost never useful) were deprecated.

I also seem to recall that this (http://www.lojban.org/tiki/BPFK+Section%3A+Distance) is more relevant.

I further seem to recall that {vu} specifying the origin is no longer the case.

ju'ocu'i cai
--
mu'o mi'e .arpis.

.arpis.

unread,
Aug 16, 2011, 10:35:45 AM8/16/11
to loj...@googlegroups.com

Stela Selckiku

unread,
Aug 17, 2011, 12:49:06 AM8/17/11
to loj...@googlegroups.com
On Tue, Aug 16, 2011 at 10:25 AM, .arpis. <rpglover...@gmail.com> wrote:
>
> I further seem to recall that {vu} specifying the origin
> is no longer the case.

Sadly it seems to have recently been the case half the time. :( Which
is to say that we have had an unresolved dispute here. I think I'd
learned both ideas and they'd just harmlessly crossed in my mind,
since my natlangy brain is happy to disambiguate with the fact that
one idea always tags an event or place and the other always tags a
duration or distance. But that's not actually lobykai! So we'll have
to choose.

Today I thought of this idiom: {pu zi ca}. That construction can tag
an event and definitely gives the same meaning as the origin tagging
idea of how {pu zi} works-- right? And that way you can write {pu zi
lo snidu ca lo fasnu noi clani sumti pluja ciksi nitcu}, instead of
{pu lo fasnu noi clani sumti pluja ciksi nitcu ku'o zi lo snidu},
where you're forced to put the distance at the end, which often comes
out silly.

From now on I intend to use the new distance tagging understanding of
ZI and VA, and I've rewritten my song {la cmalu rokci} so it says {vi
bu'u lo rirxe korbi} instead of {vi lo rirxe korbi}.

.i je .a'o la .lojban. zenba lo ka za'e plixaubau

mi'e la stela selckiku

mu'o

Michael Turniansky

unread,
Aug 23, 2011, 9:53:30 AM8/23/11
to loj...@googlegroups.com
On Wed, Aug 17, 2011 at 12:49 AM, Stela Selckiku <selc...@gmail.com> wrote:
On Tue, Aug 16, 2011 at 10:25 AM, .arpis. <rpglover...@gmail.com> wrote:
>
> I further seem to recall that {vu} specifying the origin
> is no longer the case.

Sadly it seems to have recently been the case half the time. :(  Which
is to say that we have had an unresolved dispute here.  I think I'd
learned both ideas and they'd just harmlessly crossed in my mind,
since my natlangy brain is happy to disambiguate with the fact that
one idea always tags an event or place and the other always tags a
duration or distance.  But that's not actually lobykai!  So we'll have
to choose.
  I will go further.  Unless and until a new CLL is written, I will continue to use va/vi/vu for origin, and termsets for distance. So, nanny-nanny-boo-bo!
             --gejyspa, believer in the wirtten "word from on high"
 

Stela Selckiku

unread,
Aug 23, 2011, 12:41:30 PM8/23/11
to loj...@googlegroups.com
On Tue, Aug 23, 2011 at 9:53 AM, Michael Turniansky
<mturn...@gmail.com> wrote:
>
>   I will go further.  Unless and until a new CLL is written, I will continue
> to use va/vi/vu for origin, and termsets for distance. So,
> nanny-nanny-boo-bo!
>              --gejyspa, believer in the wirtten "word from on high"

Avoiding the area until it's resolved is fine, but using two different
definitions isn't! :( I've been using this for marking origins: {PU
ZI ca (origin)} and {FAhA VA bu'u (origin)}. If I understand
correctly-- someone tell me if I'm wrong-- that will properly mark an
origin under *either* meaning of ZI/VA, won't it? It only comes up a
few times a day, so I don't feel like an extra {ca} or {bu'u} is too
much of a price to pay to avoid making this area of the grammar a
newbie-confusing warzone.

mu'omi'e la stela selckiku

Luke Bergen

unread,
Aug 23, 2011, 12:51:52 PM8/23/11
to loj...@googlegroups.com
Sorry, I do things best with examples.

What are the "two meanings" in terms of example sentences and their english translations (or pseudo-english translations involving phrases like "relative to the main selbri" etc...)

I originally thought that {vi} and friends were just like {ca} such that {mi tcidu vi lo tricu} would be understood as "I read [short-distance-from] a tree" and if I had used {vu} instead it would have been "I read [long-distance-from] a tree".  A while back though, I was corrected by someone or other that I want {bu'u} and friends for that sort of thing and that my original sentence actually means something else (what it means I don't remember any more).  .ija'ebo I don't use VI/ZI/etc.. any more as tags.  How SHOULD I be using them (given either school of thought that selkik was talking about)

Stela Selckiku

unread,
Aug 23, 2011, 2:39:30 PM8/23/11
to loj...@googlegroups.com
On Tue, Aug 23, 2011 at 12:51 PM, Luke Bergen <lukea...@gmail.com> wrote:
>
> I originally thought that {vi} and friends were just like {ca} such that {mi
> tcidu vi lo tricu} would be understood as "I read [short-distance-from] a
> tree" and if I had used {vu} instead it would have been "I read
> [long-distance-from] a tree".

That's the old standard definition.

> A while back though, I was corrected by someone or other that I want {bu'u}
> and friends for that sort of thing and that my original sentence actually means
> something else (what it means I don't remember any more).

What it means under the new definition is "I read a short distance
from something, and the length of that distance is a tree." Utter
nonsense, as trees aren't lengths.

This new definition was invented because there's no other convenient
way to specify distances, which is a FAQ and useful thing to be able
to do. No one uses termsets, so no one taught nintadni to use
termsets for it. Someone sometime got the idea-- either accidentally
or intentionally-- to repurpose ZI and VA. Because it's so
astoundingly useful, it's spread to be perhaps the most common way
those tags are used, even though no one bothered to try to make it
official or, apparently, to convince la gejyspa.

> .ija'ebo I don't use VI/ZI/etc.. any more as tags.  How SHOULD I be using
> them (given either school of thought that selkik was talking about)

{mi kelci vi lo tricu}

Old School: I play near a tree.
New School: I play near something, and the distance I'm from it is
tree. (nonsense)

{mi zutse zu'a lo tricu vi lo mitre}

Old School: I sit to the left of a tree, a short distance from a
meter. (nonsense)
New School: I sit one meter to the left of a tree.

{mi zutse zu'a vi lo mitre bu'u lo tricu}

Old School: I sit a short distance to the left of a meter at a tree. (nonsense)
New School: I sit to the left one meter from at a tree. (pretty much
the same meaning as the last one, but allows you to rearrange the
distance and origin)

{zu'a vi bu'u lo tricu mi zutse gi'e kelci}

Old School and New School: Slightly to the left of a tree, I sit and play.

{mi zutse gi'e kelci vau zu'a nu'i lo tricu la'u lo mitre}

Old School: I sit and play one meter to the left of a tree.
New School: Termsets? What the mabla is that?!

.ua nai bu'o cu'i dai

Luke Bergen

unread,
Aug 23, 2011, 3:35:17 PM8/23/11
to loj...@googlegroups.com
cool.  Excellent explanation.  So the distinction between {vi} and {vu} then is not so much a reflection of actual distance but more so what the speaker feels about that distance.  e.g. {mi zutse ne'a lo tricu vi lo mitre} tells us that the speaker feels like a meter a short distance from the tree while if I had said {vu lo mitre} we would be led to believe that the speaker feels that one meter is a very large distance from the tree.

Is that a reasonable understanding?

So at this point, does VI differ from TAG?  I perpetually have trouble when talking about TAG-like constructs.

{broda ba lo nu brode} = "broda occurs {ba (in the future of) } the event of brode"
{broda vi lo brode} = "broda occurs {vi (short distance of) } the brode".  What it is a short distance FROM is unspecified.  For specifying origin we'd need to use a bu'u or some such I guess?  Maybe as is the case with an un-specified TAG, the assumed thing is the speakers here and now.  
i.e. {broda baku} = "broda occurs {ba (in the future of) } ...... speaker's present/location"
{broda viku} ?= "broda occurs {vi (short distance of) } .... speaker's present/location"..... wow... that feels odd.  Clearly genai VI and TAG are entirely different gi I don't understand ga VI gi TAG

Jorge Llambías

unread,
Aug 23, 2011, 6:21:46 PM8/23/11
to loj...@googlegroups.com
On Tue, Aug 23, 2011 at 3:39 PM, Stela Selckiku <selc...@gmail.com> wrote:
>
> What it means under the new definition is "I read a short distance
> from something, and the length of that distance is a tree."  Utter
> nonsense, as trees aren't lengths.

But trees can determine a distance/displacement:

mi pu derse'a lo solji va lo ci tricu be'a lo barda rokci
"I buried the gold three trees north of the big rock."

mu'o mi'e xorxes

tijlan

unread,
Aug 24, 2011, 4:40:34 AM8/24/11
to loj...@googlegroups.com
On 23 August 2011 20:35, Luke Bergen <lukea...@gmail.com> wrote:
> cool.  Excellent explanation.  So the distinction between {vi} and {vu} then
> is not so much a reflection of actual distance but more so what the speaker
> feels about that distance.  e.g. {mi zutse ne'a lo tricu vi lo mitre} tells
> us that the speaker feels like a meter a short distance from the tree while
> if I had said {vu lo mitre} we would be led to believe that the speaker
> feels that one meter is a very large distance from the tree.
> Is that a reasonable understanding?
> So at this point, does VI differ from TAG?

It seems to me that a tag is basically whatever bare cmavo or unit of
cmavo that can be replaced by a "fi'o SELBRI". The new school VI would
be something like:

fi'o cmalu te sepli
at small distance

The old school VI has been something like:

fi'o se jibni se sepli / fi'o jibni
near

There seems to be little syntactical change.

(By the way, could "fi'o cmalu te sepli zo'e" mean "fi'o te jibni zo'e"?)

> I perpetually have trouble when talking about TAG-like constructs.
> {broda ba lo nu brode} = "broda occurs {ba (in the future of) } the event of brode"
> {broda vi lo brode} = "broda occurs {vi (short distance of) } the brode".
>  What it is a short distance FROM is unspecified.  For specifying origin
> we'd need to use a bu'u or some such I guess?  Maybe as is the case with an
> un-specified TAG, the assumed thing is the speakers here and now.
> i.e. {broda baku} = "broda occurs {ba (in the future of) } ...... speaker's
> present/location"
> {broda viku} ?= "broda occurs {vi (short distance of) } .... speaker's
> present/location"..... wow... that feels odd.  Clearly genai VI and TAG are
> entirely different gi I don't understand ga VI gi TAG

'The assumed thing' about each tag doesn't have to be of the same sort
apart from being something specific to the speaker (if we really are
to take TAG KU as implying not just {zo'e} but {lo steci be mi}, {tu'a
mi}, etc.). We can have different me-involving assumptions according
to each tag's meaning. With {ba}, it's naturally a temporal point.
With the new {vi}, it would naturally be a spatial distance:

ba ku = fi'o se balvi tu'a mi
subsequent to some me-involving temporal point (my present)

vi ku = fi'o cmalu te sepli tu'a mi
at some me-involving small (near-my-location) distance

(But TAG KU doesn't always imply the speaker's such frame of
reference, does it?)

mu'o

Pierre Abbat

unread,
Aug 24, 2011, 5:58:44 AM8/24/11
to loj...@googlegroups.com
On Tuesday 23 August 2011 14:39:30 Stela Selckiku wrote:
> {mi zutse zu'a vi lo mitre bu'u lo tricu}
>
> Old School: I sit a short distance to the left of a meter at a tree.
> (nonsense) New School: I sit to the left one meter from at a tree. (pretty
> much the same meaning as the last one, but allows you to rearrange the
> distance and origin)

What would {mi zutse vi lo mitre zu'a lo tricu} mean?

Pierre
--
When a barnacle settles down, its brain disintegrates.
Já não percebe nada, já não percebe nada.

Stela Selckiku

unread,
Aug 24, 2011, 6:03:30 PM8/24/11
to loj...@googlegroups.com
On Wed, Aug 24, 2011 at 5:58 AM, Pierre Abbat <ph...@phma.optus.nu> wrote:
>
> What would {mi zutse vi lo mitre zu'a lo tricu} mean?

My understanding is that it means "I sit just (a) meter(s) in some
direction from somewhere to the left of a tree." My reasoning is that
it's possible to have various legs on our imaginary journey: {ga'u vu
vu'a va zu'a vi bu'u} = "far above somewhat to the west of just to the
left of here/there", or with measurements: {ga'u vu lo minli vu'a va
lo mitre be li ci ze zu'a vi lo gutci be li ci bu'u} = "one mile above
thirty seven meters to the west of three inches left of here/there".
So the FAhA to the right of a VA is where it's a distance from, and a
FAhA to the left of a VA is in which direction to go that distance,
thus: {zu'a va} = "a medium distance to the left" and {va zu'a} = "a
medium distance in some direction from somewhere on the left".

Michael Turniansky

unread,
Aug 25, 2011, 8:50:18 AM8/25/11
to loj...@googlegroups.com

Not nonsense in new school. You just have to interpret "tree" as a
distance (it could be equal to the standard height of a tree, for
example, or where the tip of its shadow lies)

> {mi zutse zu'a lo tricu vi lo mitre}
>
> Old School: I sit to the left of a tree, a short distance from a
> meter. (nonsense)
> New School: I sit one meter to the left of a tree.

Old school isn't nonsense. It means a short distance from the
meter-long thing.

>
> {mi zutse zu'a vi lo mitre bu'u lo tricu}
>
> Old School: I sit a short distance to the left of a meter at a tree. (nonsense)
> New School: I sit to the left one meter from at a tree. (pretty much
> the same meaning as the last one, but allows you to rearrange the
> distance and origin)
>

Old School: I am sitting a short distance from a meter-long thing, at a tree.

> {zu'a vi bu'u lo tricu mi zutse gi'e kelci}
>
> Old School and New School: Slightly to the left of a tree, I sit and play.

This one I think I would called the Old School nonsense. Are we AT the
tree (bu'u) or near it (vi)?


>
> {mi zutse gi'e kelci vau zu'a nu'i lo tricu la'u lo mitre}
>
> Old School: I sit and play one meter to the left of a tree.
> New School: Termsets?  What the mabla is that?!
>

How is it acceptable to say, "I can't be bothered to read the CLL,
therefore I should not use that construct"? People like that probably
vote (pick your favorite deprecated political party). "Oooh... I
don't understand 'klama'. Therefore I'll always use 'cadzu'." Ummm..
just -- No


> .ua nai bu'o cu'i dai
>
> mu'omi'e la stela selckiku
>

Michael Turniansky

unread,
Aug 25, 2011, 9:01:40 AM8/25/11
to loj...@googlegroups.com
On Tue, Aug 23, 2011 at 3:35 PM, Luke Bergen <lukea...@gmail.com> wrote:
> cool.  Excellent explanation.  So the distinction between {vi} and {vu} then
> is not so much a reflection of actual distance but more so what the speaker
> feels about that distance.  e.g. {mi zutse ne'a lo tricu vi lo mitre} tells
> us that the speaker feels like a meter a short distance from the tree while
> if I had said {vu lo mitre} we would be led to believe that the speaker
> feels that one meter is a very large distance from the tree.
> Is that a reasonable understanding?

Yes, in the new school understanding. Which is why I think new
school is *bleep*. If you specify a unit of measure, then your choice
of vi/va/vu becomes arbitrary and subjective, and adds not a wit of
objectivity to "lo mitre be li pa", but nonetheless, you HAVE to pick
one of them (unlike la'u, which is completely objective). So now you
have to think "hmmm...is it near? far? How do I feel about it today?
How will my listener feel?". On the other hand, if you really WANT to
communicate a feeling, rather than a specific number, old or new
school just use a selbri tense.

> So at this point, does VI differ from TAG?  I perpetually have trouble when
> talking about TAG-like constructs.
> {broda ba lo nu brode} = "broda occurs {ba (in the future of) } the event of
> brode"
> {broda vi lo brode} = "broda occurs {vi (short distance of) } the brode".
>  What it is a short distance FROM is unspecified.  For specifying origin
> we'd need to use a bu'u or some such I guess?  Maybe as is the case with an
> un-specified TAG, the assumed thing is the speakers here and now.

Again, like everything else, this is in the CLL. Read chapter 10.
Yes, it's the here and now. As for using VI like TAG, I quote froim
the CLL (12.10): "So far, we have seen tenses only just before the
selbri, or (equivalently in meaning) floating about the bridi with
``ku''. There is another major use for tenses in Lojban: as sumti
tcita, or argument tags" i.e. VI and its friends are subsets of tags.
Grammatically, they pretty much work the same way. The major
difference is how they are semantically intrepreted when used with
"i___bo" where BAI act differnet than tenses.

--gejyspa

Luke Bergen

unread,
Aug 25, 2011, 9:23:38 AM8/25/11
to loj...@googlegroups.com
Right, but I'm assuming that everything under the "new system" came after the CLL so I'm trying to understand how I should NOW understand vi/va/vu to mean?

blech, wasn't this shit supposed to be decided on and set in stone like 15 years ago or something?

Also, yes, you're right.  I haven't read through the CLL in a good year or two.  I need to read through it again to freshen up on technical lojban.

Also, did I use {ne'a} appropriately back there?  I was trying to find the most generic FAhA so as not to indicate whether it's "to the east of", "beneath", "to the left of" or any other specific direction, but rather just "somewhere around".  

.arpis.

unread,
Aug 25, 2011, 9:33:27 AM8/25/11
to loj...@googlegroups.com
On Thu, Aug 25, 2011 at 9:01 AM, Michael Turniansky <mturn...@gmail.com> wrote:
On Tue, Aug 23, 2011 at 3:35 PM, Luke Bergen <lukea...@gmail.com> wrote:
> cool.  Excellent explanation.  So the distinction between {vi} and {vu} then
> is not so much a reflection of actual distance but more so what the speaker
> feels about that distance.  e.g. {mi zutse ne'a lo tricu vi lo mitre} tells
> us that the speaker feels like a meter a short distance from the tree while
> if I had said {vu lo mitre} we would be led to believe that the speaker
> feels that one meter is a very large distance from the tree.
> Is that a reasonable understanding?

 Yes, in the new school understanding.  Which is why I think new
school is *bleep*.  If you specify a unit of measure, then your choice
of vi/va/vu becomes arbitrary and subjective, and adds not a wit of
objectivity to "lo mitre be li pa", but nonetheless, you HAVE to pick
one of them (unlike la'u, which is completely objective).  So now you
have to think "hmmm...is it near?  far?  How do I feel about it today?
 How will my listener feel?". On the other hand, if you really WANT to
communicate a feeling, rather than a specific number, old or new
school just use a selbri tense.

Keep in mind that in the "new school" {va} does not specify _medium_ distance, so if you don't feel particularly short or long, you can just use {va}.



--
Reply all
Reply to author
Forward
0 new messages