Sensations / qualia (colors etc.)

[Spheniscine (la zipcpi)]

One question that comes up time and again is how to separate "red", the color, from an object that's red, {lo xunre}. E.g. "I like (the color) red". "Red is a beautiful color".

{si'o} is often suggested (mi nelci lo si'o xunre / lo si'o xunre cu melbi), but may not be a sufficient solution. Compare with "I see the color red". This is not necessarily dependent on there actually being anything red; I might be hallucinating, or have synaesthesia.

Selpahi has suggested making new brivla altogether for the colors, but I believe that they need to be regularized from {xunre} etc. Just making zi'evla for the colors ({xu'unre} etc.) is, I believe, an unsustainable solution; what of combinatory color words, or {skasmi} or {skarxetmele}?

Additionally, it isn't just colors that have this issue. How about "I feel dryness"? This is similarly independent of there actualy being anything that's {lo sudga}.

I propose that what is actually needed is a new type of object. I have defined it as {ganseti} on JVS: http://jbovlaste.lojban.org/dict/ganseti

x₁ is the sensation/qualia (abstract) associated with objects with property x₂ (ka), via sense x₃ (ka)

lo ganseti be lo ka (ce'u) xunre = "the color red", lo ganseti be lo ka (ce'u) sudga = "the sensation/feeling of dryness". x₃ is the involved sense, e.g. lo ka viska for sight; this can often be left unspecified. Proposed NU: ga'ei, with rafsi -gel- . See ganse, li'i.

Thoughts? Questions? Alternatives?

Additional thought: If {lo ganseti} / {lo ga'ei broda} is a new type of object, where does it go? nelci2 is flexible enough to accept both concrete and abstract objects of all kinds, but is a ga'ei-type suitable for ganse2 or viska2? If it isn't, we'd need new brivla for those. Perhaps a new gismu, {gelse} "x1 (entity) feels sensation x2 (ga'ei) via sense x3 (ka)". With rafsi -ge'e-, so that lujvo can then be built e.g. {visyge'e} = {x1 gelse x2 lo ka viska}

[Gleki Arxokuna]

2015-07-29 14:35 GMT+03:00 Spheniscine (la zipcpi) <sphen...@gmail.com>:

Additionally, it isn't just colors that have this issue. How about "I feel dryness"? This is similarly independent of there actualy being anything that's {lo sudga}.

I'm sure this example can be refuted. why not just {mi lifri lo ka sudga}?

However, there are other concepts similar to colors, e.g.

"I like randomness of forms" which like with "color red" doesn't cover any particular object continuously enduring in space.

[Jorge Llambías]

On Wed, Jul 29, 2015 at 8:35 AM, Spheniscine (la zipcpi) <sphen...@gmail.com> wrote:

One question that comes up time and again is how to separate "red", the color, from an object that's red, {lo xunre}. E.g. "I like (the color) red". "Red is a beautiful color".

Is that very different from "I like elephants", "elephants are beautiful animals"?  

{si'o} is often suggested (mi nelci lo si'o xunre / lo si'o xunre cu melbi), but may not be a sufficient solution. Compare with "I see the color red". This is not necessarily dependent on there actually being anything red; I might be hallucinating, or have synaesthesia.

Is that very different from "I see elephants" when there are no elephants around? 

Selpahi has suggested making new brivla altogether for the colors, but I believe that they need to be regularized from {xunre} etc. Just making zi'evla for the colors ({xu'unre} etc.) is, I believe, an unsustainable solution; what of combinatory color words, or {skasmi} or {skarxetmele}? 

Additionally, it isn't just colors that have this issue. How about "I feel dryness"? This is similarly independent of there actualy being anything that's {lo sudga}.

I propose that what is actually needed is a new type of object. I have defined it as {ganseti} on JVS: http://jbovlaste.lojban.org/dict/ganseti

So "I like elephants" becomes "mi nelci lo ganseti be lo ka xanto"?

 

x₁ is the sensation/qualia (abstract) associated with objects with property x₂ (ka), via sense x₃ (ka)

lo ganseti be lo ka (ce'u) xunre = "the color red", lo ganseti be lo ka (ce'u) sudga = "the sensation/feeling of dryness". x₃ is the involved sense, e.g. lo ka viska for sight; this can often be left unspecified. Proposed NU: ga'ei, with rafsi -gel- . See ganse, li'i.

Thoughts? Questions? Alternatives?

I don't see a problem with a word for "qualia", but I don't expect people to start saying "I like the qualia of being red" or "I like the qualia of being an elephant" instead of "I like red" or "I like elephants" though. It seems more natural that "lo xanto", "lo xunre" can be elephants (in general) and red as well as particular elephants or particular red objects. I'd rather these ontological distinctions be handled with brivla, not grammaticized with cmavo.

Additional thought: If {lo ganseti} / {lo ga'ei broda} is a new type of object, where does it go? nelci2 is flexible enough to accept both concrete and abstract objects of all kinds, but is a ga'ei-type suitable for ganse2 or viska2? If it isn't, we'd need new brivla for those. Perhaps a new gismu, {gelse} "x1 (entity) feels sensation x2 (ga'ei) via sense x3 (ka)". With rafsi -ge'e-, so that lujvo can then be built e.g. {visyge'e} = {x1 gelse x2 lo ka viska}

A brivla like that could be useful, whether it be gismu or something else. But I'm skeptical of argument places for ungrounded properties. What if you want to say "I saw elephants. They were pink with purple spots." Is there a way to refer back to the elephants in a second sentence?

mu'o mi'e xorxes

[Spheniscine (la zipcpi)]

Is that very different from "I see elephants" when there are no elephants around? 

Hm kinda a good point; guess that comes back to xorlo, doesn't it? 

I'm not sure it's the same though. Seeing the color red is somewhat different than seeing a red thing, real or imagined.

[Spheniscine (la zipcpi)]

xorxes (accidentally replied to me personally instead of the mailing list):

The color red is a less complex than a red thing, it has fewer properties (like spatial extension).

[Spheniscine (la zipcpi)]

There is one possible wrinkle of defining colors as qualia. I don't think cameras can experience qualia; however, they are made by humans to detect the same/similar colors as humans do.

By the way I have decided that {lo ganseti} / {lo ga'ei broda} isn't a new type after all, but a specialized form of {li'i}. I have defined {li'imdu} so that {li'i} is its NU-form (c.f. {nundumu}, {du'umdu}, {su'umdu}). Thus my proposal is that {li'i} denotes subjective conscious experiences. Since {li'imdu} can convert both a {nu} and a {ka}, {li'i} may or may not have a {ce'u}.

[Spheniscine (la zipcpi)]

Proposed Lojban definition of ganseti: x₁ x₄ li'i ce'u ganse lo'e me'au x₂ x₃ va'o lo fadni

[Curtis Franks]

Throughout this response, I use "qualia" rather than "quale" at some points because there may be more than one quale associated with that thing, I do not want to assume.

> "Is that very different from "I like elephants", "elephants are beautiful animals"?"

Yes, at least from a technical standpoint in Lojban. In Lojban "lo xunre" is a red thing. So, we can only say "I like red things" and "red things are beautiful things". To actually capture the essence of redness, we need a new mechanism. We could also use it in order to capture the qualia of elephantness, but that is only sort of what you probably like or find beautiful. There is an added layer of abstraction or instant-experience there. Red (the color) is a sensation, independent of what is causing it (the stimulus). The essence/being of an elephant is similar, but not really a sensation that I, at least, directly register normally. However, if you had never seen an elephant and I described everything about them to you, and then you eventually experienced one in real life, any new information that you gained from the experience would presumably be the qualia of elephantness and that is something which you would have a firmer grasp on and might like. I would not say, though, that most people are thinking this way when saying that they like elephants, although maybe they could or should (or should could).

> "Is that very different from "I see elephants" when there are no elephants around?"

Again, a little bit. Seeing red is a direct sensation. There is no object involved and no processing/deeper understanding. Seeing an elephant is seeing an object which is an elephant and interpreting it as such (even if that object does not exist, there is no way of knowing that by instanteous vision alone). Of course, there are some qualia to it, but there is also, in the normal interpretation of that statement, something more. It is not impossible to see the qualia of elephantness, but that is not what you really mean, I think.

> "So "I like elephants" becomes "mi nelci lo ganseti be lo ka xanto"?"

No. That is "I like the qualia of elephantness". Liking the object of an or several elephant(s) is just "mi nelci lo xanto". An elephant is one object which has the qualia of elephantness (and, arguably, might be the unique class with this property). Analogously, "mi nelci lo xunre" is "I like one or several objects which are red". These objects are things which have the quale of redness (possibly and even probably/arguably necessarily among other qualia). To like redness itself, the very idea and essence and immediate sensation of it, one needs the qualia abstraction.

> "I don't see a problem with a word for "qualia", but I don't expect people to start saying "I like the qualia of being red" or "I like the qualia of being an elephant" instead of "I like red" or "I like elephants" though. It seems more natural that "lo xanto", "lo xunre" can be elephants (in general) and red as well as particular elephants or particular red objects. I'd rather these ontological distinctions be handled with brivla, not grammaticized with cmavo."

It is a matter of what people should say, not what they currently do say due to bad habits and the failings of natural languages and their own education and tendencies in using Lojban. If they are alerted to such issues, then they are more likely to improve. As they currently are, and if they ignore these issues, they are incorrect. Lojban is, in my opinion, meant to bring this realization to the surface and then to fix it and to provide the tools for doing so in a reasonable but uncompromisingly rigorous way. Natural usage has nothing to do with it because it is wrong in this case.

I think that a cmavo is extremely beneficial and probably necessary in this case. In general, we may try to avoid staking out ontological positions via their establishment.

Note that this is not just a philosophical position that is being hardcoded into the language's vocabulary, functionality/support, and grammar. There is already a major flaw in the language that must be patched somehow. This is one solution, and a versatile and robust one at that. And, like I said, it is good to bring awareness to the language's learners about this issue in their own conscious understanding of their cognition.

[Gleki Arxokuna]

2016-05-12 11:43 GMT+03:00 Curtis Franks <curtis....@gmail.com>:

Throughout this response, I use "qualia" rather than "quale" at some points because there may be more than one quale associated with that thing, I do not want to assume.

> "Is that very different from "I like elephants", "elephants are beautiful animals"?"

Yes, at least from a technical standpoint in Lojban. In Lojban "lo xunre" is a red thing. So, we can only say "I like red things" and "red things are beautiful things". To actually capture the essence of redness, we need a new mechanism. We could also use it in order to capture the qualia of elephantness, but that is only sort of what you probably like or find beautiful. There is an added layer of abstraction or instant-experience there. Red (the color) is a sensation, independent of what is causing it (the stimulus). The essence/being of an elephant is similar, but not really a sensation that I, at least, directly register normally. However, if you had never seen an elephant and I described everything about them to you, and then you eventually experienced one in real life, any new information that you gained from the experience would presumably be the qualia of elephantness and that is something which you would have a firmer grasp on and might like. I would not say, though, that most people are thinking this way when saying that they like elephants, although maybe they could or should (or should could).



> "Is that very different from "I see elephants" when there are no elephants around?"

Again, a little bit. Seeing red is a direct sensation. There is no object involved and no processing/deeper understanding.

There is still some object that is red. How can you sense red without an object?

A red apple is on a red table together with a red knife and there is the red sun that you can see through the red window. All of them together (table+apple+knife+window+the sun) can be called "one single red thing (or grand-thing if you wish)".

All of the red things you have ever experiences in your life can be called "one single red thing" (so that's why "set of all things" in gua\spi definition although "set" needs more precise definition here, obviously, the same apple seen two times are two experiences and therefore two mini-things that are parts of one grand-thing).

Seeing an elephant is seeing an object which is an elephant and interpreting it as such (even if that object does not exist, there is no way of knowing that by instanteous vision alone). Of course, there are some qualia to it, but there is also, in the normal interpretation of that statement, something more. It is not impossible to see the qualia of elephantness, but that is not what you really mean, I think.

So what is meant here?

I've seen several pictures of elephants in books and now I sense a real elephant. It has the same set of properties as those elephants have (pictures of whom I read earlier.)

I've seen that red apple, that red knife and now I can see a red ball.

Earlier experiences had red being 640 nm and 660 nm wavelength together (a mixture of two properties, both wavelengths are expressed with "red").

This red ball is 640 nm only (+ many other properties like roundness etc.) I've never seen red balls before but this one matches my earlier experiences so I call it "red". This red ball also has this property of 640 nm as the aopple and the knife.

This elephant also has this property of "has trunk + gray + four legs + ears" just as those elephants pictured in those books I read earlier.

 

> "So "I like elephants" becomes "mi nelci lo ganseti be lo ka xanto"?"

No. That is "I like the qualia of elephantness". Liking the object of an or several elephant(s) is just "mi nelci lo xanto". An elephant is one object which has the qualia of elephantness (and, arguably, might be the unique class with this property). Analogously, "mi nelci lo xunre" is "I like one or several objects which are red". These objects are things which have the quale of redness (possibly and even probably/arguably necessarily among other qualia). To like redness itself, the very idea and essence and immediate sensation of it, one needs the qualia abstraction.



> "I don't see a problem with a word for "qualia", but I don't expect people to start saying "I like the qualia of being red" or "I like the qualia of being an elephant" instead of "I like red" or "I like elephants" though. It seems more natural that "lo xanto", "lo xunre" can be elephants (in general) and red as well as particular elephants or particular red objects. I'd rather these ontological distinctions be handled with brivla, not grammaticized with cmavo."

It is a matter of what people should say, not what they currently do say due to bad habits and the failings of natural languages and their own education and tendencies in using Lojban. If they are alerted to such issues, then they are more likely to improve. As they currently are, and if they ignore these issues, they are incorrect. Lojban is, in my opinion, meant to bring this realization to the surface and then to fix it and to provide the tools for doing so in a reasonable but uncompromisingly rigorous way. Natural usage has nothing to do with it because it is wrong in this case.

I think that a cmavo is extremely beneficial and probably necessary in this case. In general, we may try to avoid staking out ontological positions via their establishment.

Note that this is not just a philosophical position that is being hardcoded into the language's vocabulary, functionality/support, and grammar. There is already a major flaw in the language that must be patched somehow. This is one solution, and a versatile and robust one at that. And, like I said, it is good to bring awareness to the language's learners about this issue in their own conscious understanding of their cognition. 

I think the only flaw here is to be still with morphology of English with its adjective/noun distinction.

E.g. Russian language uses colors as verbs so what? Shall we borrow this thing into Lojban?

The only lack of precision I can see in red/elephant distinction is that elephants are more stable in time but this is a vague distinction as people mentioned many times (e.g. [1]).

 

--
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.
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]

  I mean, I'm reading emails from 3.5 years ago, and Lord knows i don't grok modern dialects of lojban, but.  What's wrong with "mi nelci (tu'a) lo ka xunre"? isn't the whole point of ka to be a quale extractor?

           --gejyspa

[Jacob Thomas Errington]

Why even bother with {tu'a}? What's hiding under that implicit abstraction anyway?

One could just say {mi nelci lo ka xunre}. Even though modern dialects
of Lojban use {ka} in a particular way (mostly unrelated to the
presentation of {ka} in the CLL), {mi nelci lo ka xunre} could make
sense.
I would interpret it to mean something like "I like to-be-red" and I
use dashes here to distance this interpretation from what the English
"I like to be red" means. The Lojban would mean something like "I like
it that things are red" or that "I like redness". This is in contrast
with {lo ka xunre cu pluka mi} which would actually mean "I like to be
red".

mi'e la tsani mu'o

> > <https://mw.lojban.org/papri/Tl%C3%B6n,_Uqbar_and_la_gleki's_fishy_apples>

> > ).
> >
> >
> >
> >>
> >> --
> >> 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.
> >> 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.
> >>
> >
> > --
> > 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.
> > 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.
> >
>
> --
> 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.

> To view this discussion on the web visit https://groups.google.com/d/msgid/lojban/CAKOEKkT2d61OWw0orJN9PCE135cqL3ZB_Bno5ZjwfOn7JH7apQ%40mail.gmail.com.

--
Jacob Thomas Errington
W: https://jerrington.me/


Sent via Migadu.com, world's easiest email hosting

[Michael Turniansky]

  I have no personal problem with leaving out the "tu'a" (Which is why I put it in parentheses) but I know that many in the lojban community are very particular when it comes to which types of of arguments are allowed in the various slots (I'm not. for the same reason I only use weakly-typed programming languages), so I was just covering my bases.

          --gejyspa

To view this discussion on the web visit https://groups.google.com/d/msgid/lojban/20200205173834.u7jlfnisrqsbqeix%40renro.

[MorphemeAddict]

What is the sense of "ka" that is used in modern dialects of Lojban? I didn't even realize Lojban has dialects. 

stevo

To view this discussion on the web visit https://groups.google.com/d/msgid/lojban/20200205173834.u7jlfnisrqsbqeix%40renro.

[Jacob Thomas Errington]

Broadly speaking it's used to form infinitives.

mi'e la tsani mu'o

> To view this discussion on the web visit https://groups.google.com/d/msgid/lojban/CAHiXx%3DP3La9UiyhELbMEU4xjYreU3DTsCxnKPAuX%3DrLqGNe3YA%40mail.gmail.com.

[Alex Kozhevnikov]

So I'm very out of date and don't know any Lojban, but what happened
to xorlo and how it made "lo" extremely generic? Seems to me that this
means that "lo xunre" is just as much allowed to be the
essence/qualia/color of red as "lo cribe" is allowed to be a liquified
residue of a bear.

I guess the distinction is that I remembered "lo cribe" as basically
being anything that is in context sufficiently associated with the
concept of bear, and thus to me "lo xunre" is naturally anything that
is associated with the concept of red, which for me includes the
qualia of redness or whatever.

It has been so useful for me to be able to think of and refer to a
category broad enough to include "anything associated with the concept
of ...", and if the goal is to enable clarity of thought and
communication about cognition, then that seems like an essential
thing, because the activation of cognition due to it being strongly
enough associated with the currently active cognition is pretty much
one of the fundamental mechanisms by which minds like ours work.

I had (perhaps wrongly) interpreted "lo" as being intended to be broad
and flexible enough to support that usage. But if it is not, then I
sure hope there is another couple-phoneme one-syllable thing to use
instead of "lo" for that purpose.

To be clear, I fully support there being some way to explicitly narrow
down meaning to qualia specifically, and have no opinion about how to
do it. I also have no problem with there being a way to narrow meaning
down to "objects" that are in some sense "more object-like" than
qualia, and if "lo" is widely understood or intended to be the way to
do that, then cool.

But I am making the appeal that it is extremely useful and valuable to
be able to very efficiently and as a fundamental language feature like
"lo" to refer to something so broadly that the meaning could be
anything from a specific red thing to the qualia of redness to the
concept of redness or the association of redness.