I asked this on IRC the other day, and it seems like there isn't. (Also,
consensus seems to be that the conflation of gender with sexuality in
cinse is a botch.)
{ma ganti} might work for plants and non-person animals, unless they've
been spayed/neutered/gelded.
Yes, but what does x2 of "gencinse" mean?
> By the way, is there any word for twins in lojban?
"tarbykansa" covers twins, triplets, etc.
Pierre
--
Don't buy a French car in Holland. It may be a citroen.
Where did "gencinse" come from?
Most Indo-European and Afro-Asiatic languages have it. I guess "x1 is a word
of gender x3", but what sexual (or gendered) activities does a word partake
in?
--
gau do li'i co'e kei do
> I guess "x1 is a word
> of gender x3", but what sexual (or gendered) activities does a word partake
> in?
That sounds like it should be "tercinse valsi" or so.
Just to clarify: when I asked where it came from, I meant how it entered
into the conversation, not what its etymology was.
--
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'd say "remei tarbykansa".
Pierre
--
I believe in Yellow when I'm in Sweden and in Black when I'm in Wales.
On 11/07/2011 11:35 AM, John E Clifford wrote:
> Which "gender" should have a gismu? grammatical, phenotypic, social, cultural,
> intentional, genetic, ...
>
> But just because gender is a complex concept, I don't think you can ignore it.
> I'm studying sociology, so I use the word gender a lot in different discourses.
> First I was surprised that gender doesn't have a gismu.
> How can you even discuss gender issues without a word for it?
> How do you say "I'm studying gender science" in lojban? Or "He has no right to
> oppress me just because we have different sex/gender"?
> /jongausib
>
>
> 2011/11/7 John E Clifford <kali9...@yahoo.com>
>
> Gloryoski! Why should Lojban be able to sort out things that the experts in the
> field can't yet get straight? Grammatical gender is defined by concordance and
> has, in a few languages, some more than casual relation to some physical
> features of the referents. Other languages have derivational devices (other
> than concordance) to signal (somewhat more regularly) such physical features
> (along with others, e.g., size, age). Still others basically don't notice. As
> for the features involved, the range is enormous. And when you throw in genetic
> data or cultural norms or internal intentions, you pass well beyond what
> languages manage to deal with comfortably (or even uncomfortably).
>>
>>
>>
>>
>>
>>
> ________________________________
> From: Sebastian Fr�jd <so.co...@gmail.com>
>> To: "loj...@googlegroups.com" <loj...@googlegroups.com>
>> Sent: Mon, November 7, 2011 6:11:47 AM
>> Subject: Re: [lojban] gender
>>
>>
>> I like the word tarbykansa. So twins would be reltarbykansa or remei
> tarbykansa?
>>
>> Den m�ndagen den 7:e november 2011 skrev Sebastian
> Fr�jd<so.co...@gmail.com>:
I'd say all except grammatical, using tanru to specify what kind you mean.
Not grammatical because it's a kind of klesi that has nothing to do
with sex; connecting the others to grammatical gender is a very
eurocentric thing. Most IE languages have masculine/feminine or
masculine/feminine/neuter. Dyirbal has masculine, feminine, vegetable,
and neuter, a system not entirely atypical for indigenous Australian
languages; calling "vegetable" a gender in the way that "male" and
"female" are is odd. Furthermore, neuter nouns in all of these
languages do not lack grammatical gender, though being neuter in
social gender means "none of the above." And that's not even getting
into languages like Dutch, which has two genders, neither of them
usefully describable as masculine or feminine. (They're common and
neuter, with the latter so called because it's derived from the
proto-Germanic neuter gender while common originates from a merger of
masculine and feminine. The result of this is that, synchronically,
there are two genders, neither of which has any clear-cut semantic
domain such that they can be described as relating in any way to
anything like social genders at all except for the fact that a
significant majority of things that have a social gender are common
regardless of what social gender that may be.)
- mi'e .kreig.daniyl.
On Mon, 2011-11-07 at 11:59 -0600, vitci'i wrote:
> One gismu for all of them, with a place to specify which one you're
> talking about.
>
> On 11/07/2011 11:35 AM, John E Clifford wrote:
> > Which "gender" should have a gismu? grammatical, phenotypic, social, cultural,
> > intentional, genetic, ...
> > 2011/11/7 John E Clifford <kali9...@yahoo.com>
> >
> > Gloryoski! Why should Lojban be able to sort out things that the experts in the
> > field can't yet get straight? Grammatical gender is defined by concordance and
> > has, in a few languages, some more than casual relation to some physical
> > features of the referents. Other languages have derivational devices (other
> > than concordance) to signal (somewhat more regularly) such physical features
> > (along with others, e.g., size, age). Still others basically don't notice. As
> > for the features involved, the range is enormous. And when you throw in genetic
> > data or cultural norms or internal intentions, you pass well beyond what
> > languages manage to deal with comfortably (or even uncomfortably).
Here, however, I think that using "cinse" is malglico; derive
the words from "klesi" instead.
Giving it a gismu place is highly unlikely to encourage people to
specify it - the "by criterion/standard x3" and the like places see
fairly little use.
- mi'e .kreig.daniyl.
Nevertheless, it makes the omission explicit.
No more so than saying "ta mlatu" makes the choice to not specify what
sort of cat explicit.
- mi'e .kreig.
Suppose we omit the place, and rely on tanru. How would you ask "what
kind of gender?"?
Sent from my iPad
But sociologically assigned gender, biological sex at birth, biological
sex constructed by surgery, personal identification, etc. *are*
entangled -- not identical, but not easily separable either, and if
nothing else then at least moderately-strongly correlated.
--
--
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/-/NbpyKzs8fEkJ.
There are more than two genders.
How about if we just let {cinse} mean gender, and move sexuality to
{gledji}?
{fetyjavnak-} (female, male, or both female & male) would allow more
than two genders. But I would agree that three mightn't still be
enough.
> How about if we just let {cinse} mean gender, and move sexuality to {gledji}?
Yes, we could generalize {cinse} up to a point where it would cover
the spectrum mentioned by J.E.C. above and would provide a more
productive rafsi:
gincinse - genetic gender
jikcinse - social gender
klucinse - cultural gender
etc.
mu'o
It probably wouldn't. If we want to cover biological sex, then we need
to be able to distinguish (e.g.) a hermaphroditic creature such as a
slug from a member of a species that reproduces asexually. If we want to
cover sociological gender, it's even more complicated than that.
>> How about if we just let {cinse} mean gender, and move sexuality to {gledji}?
>
> Yes, we could generalize {cinse} up to a point where it would cover
> the spectrum mentioned by J.E.C. above and would provide a more
> productive rafsi:
>
> gincinse - genetic gender
> jikcinse - social gender
> klucinse - cultural gender
> etc.
I'm actually talking about making {cinse} less general -- it already
means gender; I just want to make it not also mean sexuality.
I also really want to have the explicit x3 and x4 places of {cinse} when
talking about gender, so that I can say things like {cinse fi so'i da}
or {cinse fo ma}.
How do you understand the term "sexuality" in the definition of
{cinse}? What is the sort of ka that you don't want the x2 to exhibit?
> I also really want to have the explicit x3 and x4 places of {cinse} when
> talking about gender, so that I can say things like {cinse fi so'i da}
> or {cinse fo ma}.
What would those places mean? Would they not be expressible through BAIs?
mu'o
I understand it to mean, and don't want it to mean, gledji.
>> I also really want to have the explicit x3 and x4 places of {cinse} when
>> talking about gender, so that I can say things like {cinse fi so'i da}
>> or {cinse fo ma}.
>
> What would those places mean? Would they not be expressible through BAIs?
I'm not sure where your question is coming from; they'd mean the same
thing they currently do. Possible x4s might include gender of
identification (sevzi), gender performance, sex/gender assigned at
birth, anatomical sex, and chromosomal sex. Possible x3s (belonging to
various x4s) might include masculine, feminine, androgynous, butch,
femme, boi, fetsi, nakni, intersex, hermaphroditic, and of
asexually-reproducing species.
What BAIs would you use? {ci'e} might work for x4, but I think you'd
need a cinse modal for the x3.
What would be an example bridi with {cinse} in that sense? And how
would that possibility be a trouble?
>>> I also really want to have the explicit x3 and x4 places of {cinse} when
>>> talking about gender, so that I can say things like {cinse fi so'i da}
>>> or {cinse fo ma}.
>>
>> What would those places mean? Would they not be expressible through BAIs?
>
> I'm not sure where your question is coming from; they'd mean the same
> thing they currently do.
We already can say things like {cinse fi so'i da} or {cinse fo ma},
but you seemed to be suggesting you couldn't and wanted to add
"explicit" places for that, so I guessed you were talking about some
hypothetical new {cinse} with a different place structure.
> Possible x4s might include gender of
> identification (sevzi), gender performance, sex/gender assigned at
> birth, anatomical sex, and chromosomal sex.
According to the current definition, the x4 is the standard by which
the x3 is deduced from the x2. How would "gender performance" etc. be
such a reference frame? Sounds more like the x2, the event in which a
gender is exhibited.
mu'o
jbovlaste gives "x1 is sexy" as a possible meaning, so {cinse} could
actually qualify.
It's a problem because if I want to unambiguously mean "sexuality" then
I can say "gledji" but if I want to unambiguously mean "gender" then I
have no good options.
>>>> I also really want to have the explicit x3 and x4 places of {cinse} when
>>>> talking about gender, so that I can say things like {cinse fi so'i da}
>>>> or {cinse fo ma}.
>>>
>>> What would those places mean? Would they not be expressible through BAIs?
>>
>> I'm not sure where your question is coming from; they'd mean the same
>> thing they currently do.
>
> We already can say things like {cinse fi so'i da} or {cinse fo ma},
> but you seemed to be suggesting you couldn't and wanted to add
> "explicit" places for that, so I guessed you were talking about some
> hypothetical new {cinse} with a different place structure.
>
>
>> Possible x4s might include gender of
>> identification (sevzi), gender performance, sex/gender assigned at
>> birth, anatomical sex, and chromosomal sex.
>
> According to the current definition, the x4 is the standard by which
> the x3 is deduced from the x2. How would "gender performance" etc. be
> such a reference frame? Sounds more like the x2, the event in which a
> gender is exhibited.
I think you're reading cinse4 differently than I am. As far as I can
tell, the entirety of the actual definition is "by standard"; I had read
that to mean something like "in dimension/attribute/respect".
But "gledji" and "gender" can already be separated into the x2 and x3.
For example:
lo ninmu lo nu gledji lo fetsi po'o cu cinse lo ka [lesbian]
The woman, in the state of sexually desiring females only, exhibits
the quality of lesbianism.
This allows the x3 to always mean "gender" (whatever characteristics
used to distinguish between sets of sexual entities in whatever
dimension -- genetic, phenotypic, social, cultural, linguistic, etc.).
So it seems we can use {cinse} to unambiguously mean "gender".
>>> Possible x4s might include gender of
>>> identification (sevzi), gender performance, sex/gender assigned at
>>> birth, anatomical sex, and chromosomal sex.
>>
>> According to the current definition, the x4 is the standard by which
>> the x3 is deduced from the x2. How would "gender performance" etc. be
>> such a reference frame? Sounds more like the x2, the event in which a
>> gender is exhibited.
>
> I think you're reading cinse4 differently than I am. As far as I can
> tell, the entirety of the actual definition is "by standard"; I had read
> that to mean something like "in dimension/attribute/respect".
"by standard" is also a keyword for "ma'i" of "manri". Eliminate
cinse4 and we could still express it through "ma'i":
cinse lo nu prami gi'u gledji vau lo fetsi .e lo nakni kei lo ka
[bisexual] kei ma'i mi
The state of loving females and males, whether or not wanting to
have sex with them, is bisexual, as far as I can tell.
lo nanmu cu cinse fi lo ka [homosexual] kei ma'i lo certu .enai vo'a
The man is gay according to the expert but not himself.
Consider also the difference between pluja2 ("in aspect") and pluja3
("by standard"). A standard in this sense is a rule by which x1 can be
determined to broda, not a dimension in which x1 brodas. In the case
of {cinse}, such a dimension is referred to by the x2.
mu'o
You seem to be working with a very limited concept of gender and
sexuality. Sexuality includes not just sexual orientation, but a variety
of fetishes such as latex and bondage. A lojbanic view of sexual
orientation should start with androsexual/gynosexual rather than
heterosexual/homosexual. "By standard" means a *dimension* of sexuality,
not a *means of inferring* a single dimension. Your entire perspective
on this problem is inextricably rooted in wrongheadedness, and I no
longer care about your opinion.
As far as the official gismu definitions are concerned, "standard"
generally means "frame of reference", or {manri}. Those places that go
by the phrase "by standard" can generally be replaced by a tag marked
by {ma'i}. I gave you concrete examples.
I would appreciate an elaboration on "a dimension of sexuality". How
is that different from cinse2?
> You seem to be working with a very limited concept of gender and
> sexuality. Sexuality includes not just sexual orientation, but a variety
> of fetishes such as latex and bondage.
I did not say sexuality is only about sexual orientation. My point was
that the distinction between sexuality as a desire-driven activity
(you suggested "gledji") and gender as any sexual characterization can
be made within {cinse}, by the x2 and x3. True, there are different
senses for "sexuality": one has to do more with cinse2, the other with
cinse3, in my view. Consider the following examples:
mi lo nu djica lo nu skori se lasna cu cinse lo [masochism]
I'm a sexual masochist in that I have a desire for being rope-bound.
mi lo nu djica lo nu ckabu se taxfu cu cinse lo [latex fetishism]
I'm a sexual rubberguy in that I have a desire for being latex-clothed.
The x3 is whatever sexual quality or characterization you attribute to
the x1 based off the observation of the x2. The event or state
expressed by the x2 can be cognitive, genetic, social, linguistic,
etc. I'm limiting neither "sexuality" nor "gender" to "sexual
orientation". My suggestion is that {cinse} could cover the whole
practical spectrum of "sexual" stuff. Being attracted to feminity is
sexual; having a vagina is sexual; having a masculine suffix is
sexual; having a boner on being rope-bound or latex-clothed, too, is
sexual. And I would see reasonable convenience in {cinse} covering all
that.
"What is x's gender?" and "What is x's sexuality?" seem to belong to
the question of "What characterizes x's sexual function?" "The event
of being a biological female and being sexually attracted to
biological females" is a sexual function characterized by terms like
"lesbian". Likewise, "the event of being sexually excited by bondage
play" is a sexual function characterized by terms like "BDSM". BDSM
isn't about a male/female fancying males/females, but it is a sexual
trait that an entity exhibits in an event. And that's what {cinse}
could be about.
> A lojbanic view of sexual orientation should start with androsexual/gynosexual rather than heterosexual/homosexual.
That's quite irrelevant to my point. My comments have been neither for
nor against any of such schemes.
mu'o