Which word is best for intent scrutiny? How would you emphasize
intent-ness there?
-Robin
--
http://singinst.org/ : Our last, best hope for a fantastic future.
Lojban (http://www.lojban.org/): The language in which "this parrot
is dead" is "ti poi spitaki cu morsi", but "this sentence is false"
is "na nei". My personal page: http://www.digitalkingdom.org/rlp/
catlu = troci lo nu viska
zgana = troci lo nu ganse
> Which word is best for intent scrutiny? How would you emphasize
> intent-ness there?
"Intent scrutiny" or "intense scrutiny"?
catlu/zgana both require intent.
I'm not sure what "intent scrutiny" means. I think "lanli" might be
better than catlu/zgana for "scrutinize".
"Intense scrutiny" could be "lo nu carmi lo ka lanli".
mu'o mi'e xorxes
I read the gloss for {catlu} to have visual terzgana only.
I'm not sure what "intent scrutiny" means. I think "lanli" might be
better than catlu/zgana for "scrutinize"."Intense scrutiny" could be "lo nu carmi lo ka lanli".
Wait, *what*? How the hell did {troci} get in there? Neither
definition implies "attempts to" in any way at all.
> > Which word is best for intent scrutiny? �How would you emphasize
> > intent-ness there?
>
> "Intent scrutiny" or "intense scrutiny"?
http://en.wiktionary.org/wiki/intent , adj, sense #1
> catlu/zgana both require intent.
>
> I'm not sure what "intent scrutiny" means. I think "lanli" might
> be better than catlu/zgana for "scrutinize".
Ah! Thank you for that.
The goal of catlu is viska, but it's possible to catlu and fail to viska.
*chuckle*
Come on, xorxes; we've been playing this game long enough that you
should know that you making a bare assertion like that isn't good
enough.
Again: I see no evidence at all in the definitions that what you
just said is actually, ya know, *true*. How do you justify it?
It would be really nice if, instead of making me drag the entire
chain of thought in your head out one link at a time, you would
generate the next 3 links or so and save us both some trouble.
I just didn't think I was saying anything so remarkable. The first
definition of "look" I find in
http://dictionary.reference.com/browse/look is:
"to turn one's eyes toward something or in some direction in order to see"
If you think the relationship between "catlu" and "viska" is other
than the relation between "look" and "see", or if you can find a more
precise way to express that relation than with "troci", please do.
In any case, your original question was about how to distinguish
"catlu" from "zgana", and the important part of my reply was that
zgana is to ganse as catlu is to viska.
To me, to look is to direct a sense at something (for us it's sight, but I'd
accept a bat looking with its ears), and one can look at something and fail
to see it (a star in broad daylight). To observe or behold something requires
actual perception, and does not require direction. To examine or inspect
requires paying attention to details. To watch, which is in the definition of
both words, is to spend some time paying attention to something.
As to ganse vs. viska, one ganse a property but viska a thing. Genesis
1:4: ".i la cevni cu ga'ezga le gusni le ka xamgu"
There's a word "vi'azga" in the notes of "zgana". Presumably this is different
from "catlu".
Pierre
--
sei do'anai mi'a djuno puze'e noroi nalselganse srera
As usual. :D
> The first definition of "look" I find in
> http://dictionary.reference.com/browse/look is: "to turn one's
> eyes toward something or in some direction in order to see"
>
> If you think the relationship between "catlu" and "viska" is other
> than the relation between "look" and "see", or if you can find a
> more precise way to express that relation than with "troci",
> please do.
AFA*I*CT, catlu and viska are pure synonyms; that's why I asked.
Ah, but I see how you're getting that out of catlu. Ok, yeah, I
think I'm with you.
While we're looking at those words, what the hell does "note that
English "look" often means a more generic "observe"" mean?, from the
catlu and viska definitions?
On Tue, Nov 08, 2011 at 08:56:41AM -0300, Jorge Llambías wrote:
> On Tue, Nov 8, 2011 at 4:52 AM, Robin Lee Powell
> <rlpo...@digitalkingdom.org> wrote:
I read it as a warning not to commit malgli by assuming that all senses
of "look" apply; catlu and viska are only ever visual.
look - see
listen - hear
touch - feel
sniff - smell
taste - taste
where the second word in each pair describes an experience while the
first one describes an action one takes in order to have that
experience. ("taste" can work both ways, action "I tasted the soup to
see whether it needed more salt", or experience "I tasted something
funny in the soup". "smell" and "feel" too can work both ways.)
Lojban doesn't have gismu for most of those:
catlu - viska
----- - tirna
pencu - -----
----- - sumne
----- - -----
My understanding is that "zgana" and "ganse" were meant to be the
general words for each column, so we could have:
catlu - viska
kerzga - tirna
pencu - pilga'e
zbizga - sumne
tacyzga - tacyga'e
Or in fully regular form:
kalzga - kalga'e
kerzga - kerga'e
pilzga - pilga'e
zbizga - zbiga'e
tacyzga - tacyga'e
There are however two problems with all of this.
One problem is that all the words used to define "zgana" in English in
the gi'uste are used almost exclusively with sight, and as Pierre
pointed out they seem to mean something more like a prolonged
intentional viska than just directing some sense in order to perceive.
I believe this is just a case of bad gloss words, since English has no
convenient word for the intended concept. It is clear that despite the
gloss words, "zgana" is not restricted to sight, because it has a
place for the means of sensing.
The other problem is the property in x2 of ganse. I think gismu that
have a property argument and no argument for the thing with the
property are just wrong. Fortunately there are not many of them. My
policy is to just ignore the gi'uste about this.
Those are the ones derived from the with-what, but there's also the
series derived from the what:
vinzga - vinga'e
snazga - snaga'e
te'uzga - te'urga'e
panzga - panga'e
vu'izga - vu'irga'e
at least one of which has actually been used.
I completely agree with you about zgana, and I'm inclined to agree
with you about ganse; make sure they get notes in the gismu setion
please. How do you fix ganse when you ignore the gi'uste?
-Robin
> --
> 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.
I'm not in much of a wiki-editing mood lately, so I'll leave that to
someone else.
> How do you fix ganse when you ignore the gi'uste?
Just remove "property" and "(ka)":
x1 [observer] senses/detects/notices stimulus x2 by means x3 under
conditions x4.
Wouldn't that break "ga'ezga"? I coined that lujvo for "perceives something to
have a quality" because "ganse" has a property place.
Pierre
--
Don't buy a French car in Holland. It may be a citroen.
I'd use kairga'e for that: x1 ganse lo nu x2 ckaji x3 kei x4 x5
But you could also not bring up properties: "lo cevni cu ganse lo nu
lo gusni cu xamgu".
2011/11/10 Jorge Llambías <jjlla...@gmail.com>:
>
> Or in fully regular form:
>
> kalzga - kalga'e
> kerzga - kerga'e
> pilzga - pilga'e
> zbizga - zbiga'e
> tacyzga - tacyga'e
Those are the ones derived from the with-what, but there's also the
series derived from the what:
vinzga - vinga'e
snazga - snaga'e
te'uzga - te'urga'e
panzga - panga'e
vu'izga - vu'irga'eat least one of which has actually been used.
On Fri, Nov 18, 2011 at 5:15 AM, Robin Lee Powell wrote:
> This is really interesting stuff! Can you make sure it appears on
> some appropriate BPFK page?
>
> I completely agree with you about zgana, and I'm inclined to agree
> with you about ganse; make sure they get notes in the gismu setion
> please.
I'm not in much of a wiki-editing mood lately, so I'll leave that to
somepony else.
The scheme is still not complete. Let's put all sensory aspects into one table.These are the predicate that we might get (I will give them temporary names in curly brackets).{HEAR} - x1 hears object(e.g. musician)/event (e.g. concert) x2, listens to it and defines x2 as x3 (noisy, loud, sweet, lovely, romantic etc.){SEE} - x1 sees object/event x2, looks at x2 and defines it as x3 (green, that it looks ugly etc.){SMELL} - x1 smells object/event x2, observes x2 and defines it as x3 (e.g. flowery,pungent, musk){TASTE} - x1 tastes object/event x2, observes x2 and defines it as x3 (e.g. tasty, sweet,sour){TOUCH} - x1 touches x2, observes (tries) x2 and defines it as having texture x3 (e.g. rought)As you can see we actually need valsi with three places. "observe" is a variation of each predicate that can be best expressed using additional {jundi} or {zgana} as xorxes suggested.This scheme is consistent. The only problem is that it doesn't correspond to the current mess of sensory gismu.The only gismu that corresponds to this scheme is {te panci}.{sumne} is a superfluous gismu.{te tirna} is something that we can also place in every sensory gismu.{vrusi} overlaps the senses of taste and smell.This is something that definitely needs revision.We have five senses and we must clearly express what we feel.Currently Lojban lacks such power (except in the case of {te panci}).
On Friday, August 10, 2012 5:48:53 PM UTC+4, la gleki wrote:The scheme is still not complete. Let's put all sensory aspects into one table.These are the predicate that we might get (I will give them temporary names in curly brackets).{HEAR} - x1 hears object(e.g. musician)/event (e.g. concert) x2, listens to it and defines x2 as x3 (noisy, loud, sweet, lovely, romantic etc.){SEE} - x1 sees object/event x2, looks at x2 and defines it as x3 (green, that it looks ugly etc.){SMELL} - x1 smells object/event x2, observes x2 and defines it as x3 (e.g. flowery,pungent, musk){TASTE} - x1 tastes object/event x2, observes x2 and defines it as x3 (e.g. tasty, sweet,sour){TOUCH} - x1 touches x2, observes (tries) x2 and defines it as having texture x3 (e.g. rought)As you can see we actually need valsi with three places. "observe" is a variation of each predicate that can be best expressed using additional {jundi} or {zgana} as xorxes suggested.This scheme is consistent. The only problem is that it doesn't correspond to the current mess of sensory gismu.The only gismu that corresponds to this scheme is {te panci}.{sumne} is a superfluous gismu.{te tirna} is something that we can also place in every sensory gismu.{vrusi} overlaps the senses of taste and smell.This is something that definitely needs revision.We have five senses and we must clearly express what we feel.Currently Lojban lacks such power (except in the case of {te panci}).Addition.{te jvinu} also follows this pattern (thanks xorxes for noticing!){te skari} also follows this pattern.However, tirna3, viska3 and skari4 can be easily expressed using {va'o}.
mi tirna lo balzgibendemi tirna lo cladumi tirna lo nu balzgibende cu cladu
When we say "I see you" we actually mean "I see you standing/present in front of me/{co'e}/whatever".
--
You received this message because you are subscribed to the Google Groups "lojban" group.
To view this discussion on the web visit https://groups.google.com/d/msg/lojban/-/CM2jEXP7CeUJ.
On Thu, Sep 13, 2012 at 10:17 AM, la gleki <gleki.is...@gmail.com> wrote:When we say "I see you" we actually mean "I see you standing/present in front of me/{co'e}/whatever".I don't understand why you think "I see you" is short for a longer phrase, especially one that is too specific and often simply not true.
To view this discussion on the web visit https://groups.google.com/d/msg/lojban/-/zjoOg-EmvKwJ.
Actually, those two examples don't relate as much as you'd like. Vision occurs when light reflects off an object into the eye, and is afterward processed by the brain (the brain does *a lot* of work to see). "Seeing events" doesn't make much sense, because sight is just a primitive based on capturing light. You see the object regardless of its state.
To view this discussion on the web visit https://groups.google.com/d/msg/lojban/-/4zuUxnJzg_oJ.
Am 22.09.2012 19:02, schrieb Ian Johnson:
> {mi viska lo plise noi farlu}. With some more twiddling you can force
> simultaneity if you care about it.
Or just "mi viska lo nu lo plise cu farlu".
--
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+unsubscribe@googlegroups.com.
To unsubscribe from this group, send email to lojban+un...@googlegroups.com.
--
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.
On Saturday, September 22, 2012 9:04:40 PM UTC+4, selpa'i wrote:Am 22.09.2012 19:02, schrieb Ian Johnson:
> {mi viska lo plise noi farlu}. With some more twiddling you can force
> simultaneity if you care about it.
Or just "mi viska lo nu lo plise cu farlu".Hey, yes and no. That's the point. If lo se viska can be an abstraction then my new gismu work. If not then I'll have to revert back to less backward-compatible 3-argument predicates. Both my suggestions can be found on http://www.lojban.org/tiki/BPFK+Section:+sensory+gismu