baby words, but general relevance: dai-like cmavo

12 views
Skip to first unread message

Robin Lee Powell

unread,
Nov 5, 2011, 7:59:50 AM11/5/11
to lojba...@lojban.org

OK, so there's {dai}. {.ui dai} means "you are happy, and I am also
happy along with you".

What I routinely find myself wanting to say to the babies is things
like {.o'u [you should be feeling this]}. Something like a UI
equivalent of {ko surla}, or {ko gleki}, or whatever.

Obviously, I can just *say* {ko surla}, but I find myself, in the
flow of Lojban, really *wanting* to use the UI cmavo for this
purpose.

Similarily, I also want {.oi [I see that you are feeling this; I'm
not myself, but I observe it in you]}. Which, again, {za'a do
dunku}, but when I'm in the flow speaking Lojban, I want a UI for
it.

I don't know why I have these urges, but I thought I'd throw them
out in case other people feel similarily and maybe think it's worth
playing with a couple of experimental cmavo for it. As usual, I
probably won't be following the thread.

-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/

Luke Bergen

unread,
Nov 5, 2011, 12:43:57 PM11/5/11
to loj...@googlegroups.com

This has been discussed at length in the past resulting in "da'oi" which rlp may find useful

--
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.

Jorge Llambías

unread,
Nov 5, 2011, 1:23:04 PM11/5/11
to loj...@googlegroups.com, lojba...@lojban.org
On Sat, Nov 5, 2011 at 8:59 AM, Robin Lee Powell
<rlpo...@digitalkingdom.org> wrote:
>
> OK, so there's {dai}.  {.ui dai} means "you are happy, and I am also
> happy along with you".

It doesn't actually mean that, but it is an appropriate thing for me
to say when I feel (or imagine) that you are happy, and I am also
happy along with you.

> Similarily, I also want {.oi [I see that you are feeling this; I'm


> not myself, but I observe it in you]}.  Which, again, {za'a do
> dunku}, but when I'm in the flow speaking Lojban, I want a UI for
> it.

It is fine to just say ".oi" there. You are not saying it *to* the
baby, you are saying it *with* her. It works even better if you don't
use your own normal voice but a baby-like voice imitation. Remember
that the referent of "mi" can be more subtle than just your own old
self.

mu'o mi'e xorxes

Robin Lee Powell

unread,
Nov 6, 2011, 1:47:08 AM11/6/11
to loj...@googlegroups.com, lojba...@lojban.org
On Sat, Nov 05, 2011 at 02:23:04PM -0300, Jorge Llamb�as wrote:
> On Sat, Nov 5, 2011 at 8:59 AM, Robin Lee Powell
> <rlpo...@digitalkingdom.org> wrote:
> >
> > OK, so there's {dai}. �{.ui dai} means "you are happy, and I am
> > also happy along with you".
>
> It doesn't actually mean that,

What does it mean then?

> > Similarily, I also want {.oi [I see that you are feeling this;
> > I'm not myself, but I observe it in you]}. �Which, again, {za'a
> > do dunku}, but when I'm in the flow speaking Lojban, I want a UI
> > for it.
>
> It is fine to just say ".oi" there. You are not saying it *to* the
> baby, you are saying it *with* her.

I'm sorry, I don't buy that at all.

> It works even better if you don't use your own normal voice but a
> baby-like voice imitation. Remember that the referent of "mi" can
> be more subtle than just your own old self.

{mi} isn't relevant here, IMO; UI are for directly expressing the
emotions of the actual immediate speaker, regardless of what {mi} is
bound to. Anything else is ... confusing, at the very least. If
someone says {.ui} and I don't know if they are happy or if some
random person around them is happy, the usefullness of UI has
basically been totally destroyed as far as I can see.

Jorge Llambías

unread,
Nov 6, 2011, 9:47:01 AM11/6/11
to loj...@googlegroups.com, lojba...@lojban.org
On Sun, Nov 6, 2011 at 2:47 AM, Robin Lee Powell
<rlpo...@digitalkingdom.org> wrote:

> On Sat, Nov 05, 2011 at 02:23:04PM -0300, Jorge Llambías wrote:
>> On Sat, Nov 5, 2011 at 8:59 AM, Robin Lee Powell
>> <rlpo...@digitalkingdom.org> wrote:
>> >
>> > OK, so there's {dai}.  {.ui dai} means "you are happy, and I am
>> > also happy along with you".
>>
>> It doesn't actually mean that,
>
> What does it mean then?

It's an expression used to express a certain feeling, not to express a
proposition, like your English sentence. Your English sentence
expresses a proposition that describes the situation in which that
feeling arises. But I had already said something like that, so
presumably what you are saying is that "means" covers that?

>> > Similarily, I also want {.oi [I see that you are feeling this;
>> > I'm not myself, but I observe it in you]}.  Which, again, {za'a
>> > do dunku}, but when I'm in the flow speaking Lojban, I want a UI
>> > for it.
>>
>> It is fine to just say ".oi" there. You are not saying it *to* the
>> baby, you are saying it *with* her.
>
> I'm sorry, I don't buy that at all.
>
>> It works even better if you don't use your own normal voice but a
>> baby-like voice imitation. Remember that the referent of "mi" can
>> be more subtle than just your own old self.
>
> {mi} isn't relevant here, IMO;

Forget I mentioned "mi" then. Whenever you say something, you may be
lending your voice to speak in the name of a group.

> UI are for directly expressing the
> emotions of the actual immediate speaker, regardless of what {mi} is
> bound to.  Anything else is ... confusing, at the very least.

And yet that's how language seems to work. How would you say "ouch!"
to your baby under the same conditions, if you were speaking in
English rather than in Lojban?

> If
> someone says {.ui} and I don't know if they are happy or if some
> random person around them is happy, the usefullness of UI has
> basically been totally destroyed as far as I can see.

We don't usually see people expressing the feelings of random people
around them, so we probably don't need to worry about that.

Robin Lee Powell

unread,
Nov 7, 2011, 2:26:26 AM11/7/11
to loj...@googlegroups.com
On Sun, Nov 06, 2011 at 11:47:01AM -0300, Jorge Llamb�as wrote:
> On Sun, Nov 6, 2011 at 2:47 AM, Robin Lee Powell
> <rlpo...@digitalkingdom.org> wrote:
> > On Sat, Nov 05, 2011 at 02:23:04PM -0300, Jorge Llamb�as wrote:
> >> On Sat, Nov 5, 2011 at 8:59 AM, Robin Lee Powell
> >> <rlpo...@digitalkingdom.org> wrote:
> >> >
>
> >> > Similarily, I also want {.oi [I see that you are feeling
> >> > this; I'm not myself, but I observe it in you]}. �Which,
> >> > again, {za'a do dunku}, but when I'm in the flow speaking
> >> > Lojban, I want a UI for it.
> >>
> >> It is fine to just say ".oi" there. You are not saying it *to*
> >> the baby, you are saying it *with* her.
> >
> > I'm sorry, I don't buy that at all.
> >
> >> It works even better if you don't use your own normal voice but
> >> a baby-like voice imitation. Remember that the referent of "mi"
> >> can be more subtle than just your own old self.
> >
> > {mi} isn't relevant here, IMO;
>
> Forget I mentioned "mi" then. Whenever you say something, you may
> be lending your voice to speak in the name of a group.

(1) I don't think that applies to UI

(2) Even if it did, it wouldn't help; if I'm not hurting, my saying
{.oi} is a lie, whether it's as a group or not.

> > UI are for directly expressing the emotions of the actual
> > immediate speaker, regardless of what {mi} is bound to.
> > �Anything else is ... confusing, at the very least.
>
> And yet that's how language seems to work. How would you say
> "ouch!" to your baby under the same conditions, if you were
> speaking in English rather than in Lojban?

I don't think I ever would? I mean, maybe I'm failing to put myself
in the situation properly, but I can't imagine doing that. I'd say
"Poor dear!" or "You got an owie." or ... actually, yeah, I can
imagine saying "Ouch! I bet that hurts!", but the "Ouch!" would be
{.oi dai}; I would be empathetically experiencing the pain
(something I'm very good at). It wouldn't be a simple
acknowledgement of her experience, which is what I was aksing about.

> >�If someone says {.ui} and I don't know if they are happy or if


> >some random person around them is happy, the usefullness of UI
> >has basically been totally destroyed as far as I can see.
>
> We don't usually see people expressing the feelings of random
> people around them, so we probably don't need to worry about that.

IMO, that's what you just proposed.

Adam Lopresto

unread,
Nov 8, 2011, 12:35:31 PM11/8/11
to loj...@googlegroups.com, lojba...@lojban.org
No answer for the {ko} form. But for the other, I have a very non-standard solution that I think might just work: treat {dai} as a scale. (Glosses below subject to all the standard rigamarole of "attitudinals aren't propositions", etc; work with me here.)

{.o'u dai [ja'ai]}: I see you're relaxed, and feel relaxed with you
{.o'u dai nai}: You're relaxed, but I'm not.
{.o'u dai cu'i}: You're relaxed. Nothing at all about how I feel is being expressed here.

I don't think {dai} as a scale has any defined meaning, but this seems consistent with the standard base form, while adding useful meanings for the others.

Jonathan Jones

unread,
Nov 8, 2011, 1:49:47 PM11/8/11
to loj...@googlegroups.com
I kind of like that. That's a pretty good idea.
--
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 )

Luke Bergen

unread,
Nov 8, 2011, 2:54:39 PM11/8/11
to loj...@googlegroups.com
Hmmm.  If {o'u dai nai} is "you feel relaxed but I do not" would {o'u nai dai} be "you feel not relaxed and feel that way too" or "you do not feel relaxed but I do"?  Maybe one would be {o'u nai dai nai} and the other would be {o'u nai dai}?

This all feels very strange.  The construct seems like it could be useful though.

Jonathan Jones

unread,
Nov 8, 2011, 5:53:35 PM11/8/11
to loj...@googlegroups.com
I think, to make things as simple as possible, UI* dai should always mean "You and I UI*", UI* dai nai, "You UI* but I don't", etc. So {o'unai dai} would be "you and I both feel not relaxed".

Robin Lee Powell

unread,
Nov 9, 2011, 7:41:26 AM11/9/11
to loj...@googlegroups.com
*That* is some fascinating shit right there.

Can you add this as a note to the bottom of
http://www.lojban.org/tiki/BPFK+Section%3A+Attitudinal+Modifiers
please?

-Robin

Robin Lee Powell

unread,
Nov 9, 2011, 7:56:50 AM11/9/11
to loj...@googlegroups.com
For any given UI, {UI} is "I feel X", and then the other 3 cases
(both of us, only you, you but not me) are listed below. {.o'u nai}
is a UI for this purpose, so {.o'u nai dai} would mean exactly what
it means now: "we're both tense".

-Robin

Pierre Abbat

unread,
Nov 9, 2011, 8:19:08 AM11/9/11
to loj...@googlegroups.com
On Tuesday 08 November 2011 17:53:35 Jonathan Jones wrote:
> I think, to make things as simple as possible, UI* dai should always mean
> "You and I UI*", UI* dai nai, "You UI* but I don't", etc. So {o'unai dai}
> would be "you and I both feel not relaxed".

I disagree. I think "UI dai nai" says explicitly that my feeling of UI is mine
alone; to say that my feeling of UI is equal to what I feel you (or someone
else) feel, I'd say "UI dai cu'i".

"pei" is a special case. "pei" asks whether you feel an emotion, so "pei dai"
asks whether you feel someone else (possibly I) feels an emotion.

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

Robin Lee Powell

unread,
Nov 9, 2011, 8:23:27 AM11/9/11
to loj...@googlegroups.com
On Wed, Nov 09, 2011 at 09:19:08AM -0400, Pierre Abbat wrote:
> On Tuesday 08 November 2011 17:53:35 Jonathan Jones wrote:
> > I think, to make things as simple as possible, UI* dai should
> > always mean "You and I UI*", UI* dai nai, "You UI* but I don't",
> > etc. So {o'unai dai} would be "you and I both feel not relaxed".
>
> I disagree. I think "UI dai nai" says explicitly that my feeling
> of UI is mine alone; to say that my feeling of UI is equal to what
> I feel you (or someone else) feel, I'd say "UI dai cu'i".

Well, {dai nai} isn't currently defined at all. Since the default
is that you're expressing your own emotions (as I've said elsewhere,
I feel strongly that that's the *only* valid use of bare UI), I'd
rather use {dai nai} for something useful.

Craig Daniel

unread,
Nov 9, 2011, 9:22:16 AM11/9/11
to loj...@googlegroups.com
On Wed, Nov 9, 2011 at 8:23 AM, Robin Lee Powell
<rlpo...@digitalkingdom.org> wrote:
> Well, {dai nai} isn't currently defined at all.  Since the default
> is that you're expressing your own emotions (as I've said elsewhere,
> I feel strongly that that's the *only* valid use of bare UI), I'd
> rather use {dai nai} for something useful.

Both interpretations make sense to me, and both are useful (and work
precisely the same way except for exchanging the meanings of dai and
nai dai nai). For any given UI there are four variants of UI dai
available: UI dai, UI nai dai, UI dai nai, UI nai dai nai. Obviously
UI dai and UI nai dai say the same thing about opposite emotions, and
UI dai nai and UI nai dai nai do as well. There are also four meanings
for them to cover: we both feel UI (UI dai), we both feel anti-UI (UI
nai dai), I'm pretty sure you feel UI but I don't (UI dai nai proposal
#1) = I alone feel not-UI (UI nai dai nai proposal #2), and I alone
feel UI (UI dai nai proposal #2) = I'm pretty sure you feel not-UI,
but I don't (UI nai dai nai proposal #1). Proposal 2 would also mean
the fact that a bare UI is dai-neutral is more relevant.

I see no compelling argument for one over the other, but do think
there's a slight reason to prefer the first interpretation: I think
we're going to say "you probably feel UI about that" a lot more than
"I don't care how you feel about it, but I'm pretty UI" for reasons of
pragmatics (though both will certainly have their uses). Since the UI
are aligned such that people tend to use the positive form more often,
UI dai nai should be the one that says something more common about our
feeling of UIness. That's a vote for proposal 1, where UI dai nai
means "I don't share your UI, but you probably feel UI."

- mi'e .kreig.

Michael Turniansky

unread,
Nov 11, 2011, 12:32:47 PM11/11/11
to loj...@googlegroups.com
  My two cents -- As a father of five, I quite definitely can/have said "ooh" or "ouch" when my kid feels something (and so yes, I think you can do that in lojban also, if you are hurting because they hurt).  And to ME, "dai" does not imply "..and I feel it, too" but simply what you have first said, "you [or whoever is the subject] feels this" (i.e. sympathy, rather than empathy) ("oidai" = "that must have hurt!").  On the other hand, no one mentioned se'i/se'inai, which I think IS the right way to describe the sense that you are feeling something because someone else has experienced something "oi se'inai" ("Ow!  That fall you took hurt me")
 
                      --gejyspa
 

Jorge Llambías

unread,
Nov 11, 2011, 4:57:03 PM11/11/11
to loj...@googlegroups.com
On Fri, Nov 11, 2011 at 2:32 PM, Michael Turniansky
<mturn...@gmail.com> wrote:
>
> On the other hand, no one
> mentioned se'i/se'inai, which I think IS the right way to describe the sense
> that you are feeling something because someone else has experienced
> something "oi se'inai" ("Ow!  That fall you took hurt me")

I've always interpreted "se'i" as indicating the orientation of the
attitude. The only times I remember using it is in ".oi se'i", when I
mean to complain to myself, not about something you or anyone else
did, or something that just happened to me, but about something I did,
a complaint directed at myself.

The example in CLL is ".au se'i" (I want it for myself) and ".au se'i
nai" (I want it for you, or presumably for someone else).

Michael Turniansky

unread,
Nov 13, 2011, 11:42:40 AM11/13/11
to loj...@googlegroups.com


2011/11/11 Jorge Llambías <jjlla...@gmail.com>

  We seem to be in agreement here.  Who  wants in "au se'inai"?  *I* do.  What's the reason for my want?  Someone else.  Who complains in "oi se'inai"?  I do.  What's the reason for my complaint?  Someone else. Hence, "That fall you took hurt me!"  Whereas "oi dai" to me only conveys "I see that YOU hurt".

            --gejysppa

Craig Daniel

unread,
Nov 13, 2011, 12:20:17 PM11/13/11
to loj...@googlegroups.com
On Sun, Nov 13, 2011 at 11:42 AM, Michael Turniansky
<mturn...@gmail.com> wrote:
>   We seem to be in agreement here.  Who  wants in "au se'inai"?  *I* do.
> What's the reason for my want?  Someone else.  Who complains in "oi
> se'inai"?  I do.  What's the reason for my complaint?  Someone else. Hence,
> "That fall you took hurt me!"  Whereas "oi dai" to me only conveys "I see
> that YOU hurt".

No, oidai expresses a feeling on the part of the speaker. That feeling
is one of empathetic pain, and implies that the speaker feels pain on
behalf of the listener (whether or not the listener actually feels any
pain).

This is important, because the UI (other than xu) are strangely
non-declarative. There is a crucial difference between ".ui" and "mi
gleki". You might be lying about how you feel, so "mi gleki" is simply
false; ".ui" has no truth value, ever. It cannot be affirmed, obeyed,
or answered, as it is not semantically declarative, imperative, or
interrogative. Since I can very readily be mistaken about how you
feel, saying ".oi" on your behalf makes no sense - it's expressing
something that I have no way of knowing even exists, without allowing
you to dispute it. An empathetic feeling, on the other hand, is no
less real just because the person being empathized with feels
differently; that's the kind of feeling ".oidai" expresses.

Thus, we have the following options:

.ui = I express happiness
.uidai = I express a feeling of empathy with your probable happiness
connected with a perception that you may be happy (that is, my
happiness is empathetic, but whether you're actually happy or not
doesn't change the validity of my .uidai)
.uidainai* = I express happiness, and I'm not empathizing with you
(that is, I probably don't think you're happy, but whether you are or
not has no direct relevance)
.uinai = I express lack-of-happiness
.uinaidai = I express empathy with your apparent lack-of-hapiness (so
this does imply, pragmatically, that I believe you're not happy, but
again, my feeling of empathy doesn't actually depend on what you are
actually feeling because my UI are not allowed to be wrong)
.uinaidainai* = I express lack of happiness, and am not empathizing
with you in this regard (that is, I'm unhappy despite the fact that
you're probably not)

*there's another proposal for how dainai works that I find a little
less intuitive and perhaps a touch less lobykai but also a little more
useful, in which the meanings of .oidainai and .oinaidainai are
inverted. I picked one for the sake of this explanation.

None of these assert how the listener is feeling, nor can any be used
to express something not about the speaker. To do so would be to make
an assertion which might be wrong, and UI are not the way to do that.
We do, however, have bridi that can express exactly what you're trying
to do with UI here; instead of ".uidai" you would say "do gleki,"
which means approximately the same thing about your perception of the
listener's mental state (and contains no indication of your own,
unless you made it ".ui do gleki" or "mi'o gleki"), exactly as you're
hoping for. Crucially, it also gives the listener the option to
respond with "na go'i" if you're mistaken.

English "Ouch" works the same way, in that I can't say "*No, you don't
actually ouch at all." I can believe you're just faking your pain, and
dispute the *implied* assertion, but there is no direct assertion of
ouchness with which I can disagree.

".ui" is an entirely different kind of utterance from "mi gleki," and
the two types ought not be used to substitute for one another. If I
want to talk about your feelings, as opposed to my feelings which may
or may not relate to you, then I need bridi.

- mi'e .kreig.daniyl.

Michael Turniansky

unread,
Nov 13, 2011, 12:35:21 PM11/13/11
to loj...@googlegroups.com
On Sun, Nov 13, 2011 at 12:20 PM, Craig Daniel <craigb...@gmail.com> wrote:
On Sun, Nov 13, 2011 at 11:42 AM, Michael Turniansky
<mturn...@gmail.com> wrote:
>   We seem to be in agreement here.  Who  wants in "au se'inai"?  *I* do.
> What's the reason for my want?  Someone else.  Who complains in "oi
> se'inai"?  I do.  What's the reason for my complaint?  Someone else. Hence,
> "That fall you took hurt me!"  Whereas "oi dai" to me only conveys "I see
> that YOU hurt".

No, oidai expresses a feeling on the part of the speaker. That feeling
is one of empathetic pain, and implies that the speaker feels pain on
behalf of the listener (whether or not the listener actually feels any
pain).

This is important, because the UI (other than xu) are strangely
non-declarative. There is a crucial difference between ".ui" and "mi
gleki". You might be lying about how you feel, so "mi gleki" is simply
false; ".ui" has no truth value, ever. It cannot be affirmed, obeyed,
or answered, as it is not semantically declarative, imperative, or
interrogative. Since I can very readily be mistaken about how you
feel, saying ".oi" on your behalf makes no sense - it's expressing
something that I have no way of knowing even exists, without allowing
you to dispute it. An empathetic feeling, on the other hand, is no
less real just because the person being empathized with feels
differently; that's the kind of feeling ".oidai" expresses.

  But I never asserted that "oidai" declares or (as you use later), "asserts".  I said it CONVEYS that meaning.
  In fact, now that I have just reread the lojban reference grammar, that's EXACTLY the example they give. (13.10.9) and then goes on to say, "Both ``pei'' and ``dai'' represent exceptions to the normal rule that attitudinals reflect the speaker's attitude."
 
  I didn't THINK I was making this up.
          --gejyspa

Craig Daniel

unread,
Nov 13, 2011, 12:55:59 PM11/13/11
to loj...@googlegroups.com

Hm, so it does.

The ma'oste gives the following: "dai UI5 attitudinal modifier:
marks empathetic use of preceding attitudinal; shows another's
feelings"

If this were a simple conflict between the Book and the ma'oste (as
with the syntax of "vo'a"), I'd say the Book ought to win, but the
book's example here is actually not self-consistent interpreted your
way - it glosses ".oiro'odai" as "[pain!] [physical] [empathy]", which
implies that the speaker is empathizing with the listener's pain
rather than merely referring to it, and proceeds to translate it more
idomatically as "Ouch! That must've hurt!" (which pe'i implies the
speaker feels something too, even though they're not physically
injured) right before it tells you it's not about the speaker's
attitude. It proceeds thence to example 13.10.10, empathizing with a
non-living object, which can *only* be about the speaker's empathetic
emotions and has nothing to do with an actual belief in the ship's
emotions, before giving the sentence you quoted. The fact that
(barring syntactically-dubious experimental COI) there's no way to
specify who you're empathizing with, only to describe an empathetic
feeling, seems to bolster this understanding as well, and of course
example 13.10.10 makes it crystal clear the empathy may not be with
anyone present.

I think it's fairest to say that "dai" does reflect the speaker's
attitude, but is an exception to the normal rule that attitudinals are
*only* about the speaker's feelings. This one is about that, but in a
way that has reference to others' feelings as perceived (however
implausible that perception, as in the ship example) by the speaker.

- mi'e .kreig.

Luke Bergen

unread,
Nov 13, 2011, 1:19:26 PM11/13/11
to loj...@googlegroups.com
I waffle on this one so much.  I kind of wish that "attitudinals" were not defined as "how the speaker feels" but more generically like "equivalent to the grunts and such from other languages".  e.g. it would be nice (as I'm guessing Robin is making reference to) the sound one makes when they see a kid running and suddenly fall and skin their knee or some such.  In english this sound is "ooooo" while making a "grimace" expression.  Expressing this is not a semantic statement with at truth value so for we jbopre, it feels like it should be an attitudinal.

So what should it be?  I don't feel pain.  I feel sympathy because I observe you feeling pain.

Craig Daniel

unread,
Nov 13, 2011, 1:23:36 PM11/13/11
to loj...@googlegroups.com
On Sun, Nov 13, 2011 at 1:19 PM, Luke Bergen <lukea...@gmail.com> wrote:
> I waffle on this one so much.  I kind of wish that "attitudinals" were not
> defined as "how the speaker feels" but more generically like "equivalent to
> the grunts and such from other languages".  e.g. it would be nice (as I'm
> guessing Robin is making reference to) the sound one makes when they see a
> kid running and suddenly fall and skin their knee or some such.  In english
> this sound is "ooooo" while making a "grimace" expression.  Expressing this
> is not a semantic statement with at truth value so for we jbopre, it feels
> like it should be an attitudinal.
> So what should it be?  I don't feel pain.  I feel sympathy because I observe
> you feeling pain.

I basically agree, and I certainly agree that what you're looking for
in that example is ".oidai" (and/or ".uu,") but I think there's a
distinction to be drawn between me expressing sympathy/empathy with
pain (.oidai) and actually directly expressing pain felt by another
being (which expression inherently runs the risk of being wrong, and
needs a truth value).

- mi'e .kreig.

Michael Turniansky

unread,
Nov 13, 2011, 1:29:42 PM11/13/11
to loj...@googlegroups.com
  Nope. It's called anthropomorphizing.  I'm sympathizing with the way I believe that item would feel, if it could feel pain.  For example, were I of a sadistic bent, And I stepped on a snail, crushing it, I could reasonably cry out, "oiro'osaidai .i uisairo'e" corresponding roughly to English "[high squeaky voice] Oh, ow! [gloating voice] MWA-HAHA!"  My mental state and the snail's do not correspond at all.
 
            --gejyspa

Craig Daniel

unread,
Nov 13, 2011, 1:53:52 PM11/13/11
to loj...@googlegroups.com

I disagree. I think ".oiro'osaidai .i .uisairo'e" implies changing
feelings about having stepped on a snail*, beginning by empathizing
with the snail's physical pain (not, technically, the same as feeling
said pain oneself - obviously, your personal feelings on the matter
are not physical - but if you don't empathize, I would think you would
want something involving "dainai" to directly negate the assertion of
empathy, the semantics of which are a matter of discussion) and
shifting toward happiness (that is not suggested as being empathic at
all, as befits the snail's unhappy state).

There are two interpretations of "dainai" proposed in this thread. In
one, it merely denies that your feeling is empathetic (negating the
"dai"), and you want something like ".uisaidainai", which is both fun
to say (although not as awesome as ".aisaidainai") and also a
statement expressing happiness which is felt in a thoroughly
non-empathetic way, meaning you're glad despite the snail's pain. In
the other interpretation, where "dainai" involves a nai negating over
the entire structure, ".oiro'osaidainai" would mean something like
"I'm feeling a distinct lack of sympathy for your serious physical
pain!" - which seems like precisely what you want to say with the
first sentence there; since you're not actually trying to empathize
*at all,* I have trouble seeing how an empathetic attitudinal fits the
meaning you want to convey.

*CLL, right after the part we were just talking about: "Finally, we
often want to report how our attitudes are changing. If our attitude
has not changed, we can just repeat the attitudinal. (Therefore, ".ui
.ui .ui" is not the same as ".uicai", but simply means that we are
continuing to be happy.)" For a mix of feelings, I'd be inclined to
leave out the ".i" in the middle, but what it's a blend or sequence
*of* is not changed.

- mi'e .kreig.

Michael Turniansky

unread,
Nov 13, 2011, 2:33:35 PM11/13/11
to loj...@googlegroups.com
On Sun, Nov 13, 2011 at 1:53 PM, Craig Daniel <craigb...@gmail.com> wrote:
On Sun, Nov 13, 2011 at 1:29 PM, Michael Turniansky
<mturn...@gmail.com> wrote:
>   Nope. It's called anthropomorphizing.  I'm sympathizing with the way I
> believe that item would feel, if it could feel pain.  For example, were I of
> a sadistic bent, And I stepped on a snail, crushing it, I could reasonably
> cry out, "oiro'osaidai .i uisairo'e" corresponding roughly to English "[high
> squeaky voice] Oh, ow! [gloating voice] MWA-HAHA!"  My mental state and the
> snail's do not correspond at all.

I disagree. I think ".oiro'osaidai .i .uisairo'e" implies changing
feelings about having stepped on a snail*, beginning by empathizing
with the snail's physical pain
  
  [snip]
 
*CLL, right after the part we were just talking about: "Finally, we
often want to report how our attitudes are changing. If our attitude
has not changed, we can just repeat the attitudinal. (Therefore, ".ui
.ui .ui" is not the same as ".uicai", but simply means that we are
continuing to be happy.)" For a mix of feelings, I'd be inclined to
leave out the ".i" in the middle, but what it's a blend or sequence
*of* is not changed.

  I won't bother rehashing my opinions on the matter at hand (dai), because we've already covered it, but about this new point, the reason I used "i" was simply intended as a separator, so that the two emotional states wouldn't "bleed" into one another (not sure if they would have, so I was just forestalling it.  Feel free to read my sentence without it) If I wanted to convey changing emotions, well, that's what bu'o is for.
             --gejyspa



Craig Daniel

unread,
Nov 13, 2011, 2:35:44 PM11/13/11
to loj...@googlegroups.com

Yeah, whether it implies one and then the other or both is a little
unclear (I read CLL as indicating the former, but I'm not certain) and
definitely not relevant to the semantics of dai.

Robin Lee Powell

unread,
Nov 14, 2011, 10:42:11 PM11/14/11
to loj...@googlegroups.com
It turns out that the CLL is *ambiguous* as to whether {dai} means
"I feel it with you" or "I note you feeling it".

-_-

From http://dag.github.com/cll/13/10/

You can of course make a bridi claim that so-and-so felt
such-and-such an emotion, but you can also make use of the
attitudinal system by adding the indicator “dai”, which attributes
the preceding attitudinal to someone else — exactly whom, must be
determined from context. You can also use “dai” conversationally
when you empathize, or feel someone else’s emotion as if it were
your own:

So, um, *yeah*.

I think I'ma go with your plan, since I think it's more used to mean
"I feel it with you" than the other.

-Robin

Robin Lee Powell

unread,
Nov 14, 2011, 10:54:35 PM11/14/11
to loj...@googlegroups.com
Also, I've just seen that this thread descended into a clusterfuck
of "the CLL says X, the ma'oste says Y, usage says Z". Yay.

I have therefore linked it at
http://www.lojban.org/tiki/BPFK+Section%3A+Attitudinal+Modifiers ,
for future use in my making of arbitrary decisions on this topic
which I will then impose by fiat. :D

-Robin

Reply all
Reply to author
Forward
0 new messages