Cont: Transwomen are not women part XII (also merged)

[Emily's Cat]

Not only men like to look at "eye candy", or do lesbians not count?

In my experience, lesbians look and admire... but tend to not reduce females to being valued for their sexiness alone, en masse.

 

[mumblethrax]

Because sex is not ordinal.

First the minor correction: ordinal data is categorical data.

Second, I don't particularly need (universal) sex to be ordinal.

If you are measuring traits, you are generating evidence. When you have enough evidence to infer the sex of the individual, you categorize it.

Literally the only thing you have to do is remove the step where you categorize the data (on the assumption that two gametes implies two reproductive phenotypes, which I don't think holds), and you have non-categorical data (after multivariate analysis). Because you're necessarily describing where the individual is on the continuum of possible phenotypes in order to find out which side it's on. Of course you can still categorize it...but only after you've gathered the data (formally or informally) necessary to make the categorization in the first place.

Third, there's still a language game going on here.

If I say "In species x, this is the male of the species, right?"
"That's right."
"And this is the female, yes?"
"Yep."
"This one seems to have implemented both reproductive strategies--it has both male and female reproductive traits, and is reproductively viable. Can we say it's a hermaphrodite?"
"Sure."

Well, that's three categories. That's a logical commitment with logical entailments. You can't then go back to insisting there are only two sexes, because that's now a different meaning for a different context. It's possible, in the sense of "the sex of an individual organism", to have more than two categories. That's true even if you want to say, "Well, strictly speaking, we have one organism that has implemented a female strategy, one male, and one has implemented both." That's still three categories of individuals. Maybe you don't want to call it "sex"...call it "glory", it doesn't matter. It's the same data. There's now a different set of rules at play. And it doesn't matter which species we're talking about...the commitment holds. The sex of an organism has to mean something different from the universal definition. Or it's just meaningless. Dealer's choice, I guess.

And if that's the case, the assumption that we must categorize the phenotypical continuum into one of two sexes (because that's how many gametes there are) is just false. We can just look at the continuum.

Further, this hardly matters. No matter what we do, the data will speak for itself. Do a cluster analysis, and you'll get two tightly packed clusters of individuals, and a few stragglers (in humans). I don't get why dissenting from the view that sex is necessarily binary, but can instead be seen as a (strongly) bimodal distribution is treated as such a heresy. It's just a necessary consequence of the continuum of phenotypes.

 

[Emily's Cat]

No, it will look like you don't. Because there's only one mode (cherries). Because there is no ordering, there is no possibility of noticing two distinct groups. So bimodal can mean nothing other than...two modes.

I've never seen a true bimodal distribution where the two modes are exactly equal. They are usually of different hieghts. It's a bimodal distribution because there are two peaks.

The look of the visualization is not the reason it's (roughly) bimodal. You can plot Male, Female, Other, and it would still be.

Because it has, roughly, two modes.

Only if you do population balancing by sex in your sampling, to ensure that the number of females is exactly equal to the number of males. Which you wouldn't do unless you already know that they're functionally different populations for this exercise!

If you were to plot your nonordinal categorical information for males and females with a completely unbiased sample... using the same logic you have used for the fruit example, there would not be two modes. The number of males and females isn't exactly equal.

If you did your unbiased and unadjusted population sample in China, you would "conclude" that the "mode" is male. If you did it in the US, your "mode" would be female. In both cases, you could not claim a "bimodal" distribution because one would have a higher count than the other, resulting in only one "mode".

If the entire nature of the visual changes when you change the apparent order of a non-ordinal categorical variable... you can't claim it to be a specific type of distribution. If bimodality is dependent on the arbitrary order you impose, it's not a bimodal distribution in the first place. It's a visual artifact resulting for an arbitrary arrangement.

 

[Emily's Cat]

In the UK prisoners have the same rights to medical treatment as anyone else. If the cosmetic surgery they are seeking is one that is covered by the NHSs then they would have a right to it. Generally the NHSs do not cover elective cosmetic surgery, so you won't get a facelift but you will get cosmetic surgery such as skin grafts for an injury even if it isn't necessary in terms of your health. Since treatment for gender dysphoria is covered by our NHSs a prisoner would be able to access these treatments. I find it hard to think of a reason why a citizen in prison shouldn't receive the appropriate medical care all us citizens are entitled to.

You have to remember that in the USA we are not entitled to free health care. So it can kind of smart that what a prisoner can get for free, you may have to pay well into the 6 figures for.

Additionally, in the US, reconstructive surgery after a disfiguring event is not considered to be cosmetic.

 

[Reformed Offlian]

I'm not giving you my or my mother's medical records.

I suspect that Darat was asking for *statistical* evidence backing up your general claim, not for documentation of your anecdote.

 

[Emily's Cat]

Yes and I am on a steep learning curve how dangerous it has become.

Just popped up while typing

https://twitter.com/ElonMuskAOC/status/1670152671796383745?t=JI3LxwlhoVlGnfOlrgSDzg&s=19

FYI - that's a parody account, not actually Elon Musk.

 

[mumblethrax]

It's the ACLU, FFS. The American Civil Liberties Union is SUING for a multiple rapist not having been provided "gender affirming care" while in prison. And they're suing after the fact.

If we can't discuss the behavior of the goddamned ACLU as an influence on policy... then I don't know what you think qualifies as "worthy of discussion" here.

The ACLU represented actual Nazis. That's their bag--they sue on behalf of the rights, not on behalf of the people.

And Owen was, in fact, denied the right to healthcare, barring some credible defense. There's some irony in the fact that prisoners are among the few people afforded such a right, but that's the price of locking them up and denying them access to outside care. Florida had a ban on trans healthcare in prisons in place which was overturned on 8th amendment grounds.

 

[mumblethrax]

I've never seen a true bimodal distribution where the two modes are exactly equal. They are usually of different hieghts.

"Mode" does not mean the same thing in both instances. The modes in an ordered bimodal distribution are the modes of the (implied) groupings. For nominal data, there are no implied grouping. It can mean only two modes in the entire population.

It's a bimodal distribution because there are two peaks

That's only true with ordered data. You are taking a property that only exists for one kind of data in a bimodal distribution, and then expecting it to show up in another where it cannot, by definition.

Only if you do population balancing by sex in your sampling, to ensure that the number of females is exactly equal to the number of males. Which you wouldn't do unless you already know that they're functionally different populations for this exercise!

You don't need to.

If you were to plot your nonordinal categorical information for males and females with a completely unbiased sample... using the same logic you have used for the fruit example, there would not be two modes. The number of males and females isn't exactly equal.

Which might be why I said roughly.

If you did your unbiased and unadjusted population sample in China, you would "conclude" that the "mode" is male. If you did it in the US, your "mode" would be female. In both cases, you could not claim a "bimodal" distribution because one would have a higher count than the other, resulting in only one "mode".

I'm not talking about sex in China and the US. Population. I'm giving you a toy example of why moving around around the bars doesn't matter in the least. If the order is arbitrary, it's arbitrary

If the entire nature of the visual changes when you change the apparent order of a non-ordinal categorical variable... you can't claim it to be a specific type of distribution.

Yes, you can. The visualization is not what determines the type of distribution you're looking at. The data is. The visualization reveals things about the data...but obviously the visualization cannot reveal anything about the order of an unorderable dataset.

If bimodality is dependent on the arbitrary order you impose, it's not a bimodal distribution in the first place. It's a visual artifact resulting for an arbitrary arrangement.

It's isn't. The order is arbitrary. It's true no matter how you arrange the bars.

 

[Emily's Cat]

First the minor correction: ordinal data is categorical data.

Minor correction back at you - ordinal data is a type of categorical data. Not all categorical data is ordinal. It's only ordinal if there's a natural order to it.
Sex is not ordinal. You can't order it, there's no natural "greater than/less than" aspect that allows you to impose a meaningful order.

So, you're not really correcting me at all.

Second, I don't particularly need (universal) sex to be ordinal.

If you are measuring traits, you are generating evidence. When you have enough evidence to infer the sex of the individual, you categorize it.

Literally the only thing you have to do is remove the step where you categorize the data (on the assumption that two gametes implies two reproductive phenotypes, which I don't think holds), and you have non-categorical data (after multivariate analysis). Because you're necessarily describing where the individual is on the continuum of possible phenotypes in order to find out which side it's on. Of course you can still categorize it...but only after you've gathered the data (formally or informally) necessary to make the categorization in the first place.

Third, there's still a language game going on here.

If I say "In species x, this is the male of the species, right?"
"That's right."
"And this is the female, yes?"
"Yep."
"This one seems to have implemented both reproductive strategies--it has both male and female reproductive traits, and is reproductively viable. Can we say it's a hermaphrodite?"
"Sure."

Well, that's three categories.

Humans aren't snails. Your argument here is irrelevant. If we were talking about a species that has two sexes, but was a hermaphroditic species (like a lot of plants) it would be an entirely different discussion.

But as we are discussing humans... this isn't making a useful point at all.

And if that's the case, the assumption that we must categorize the phenotypical continuum into one of two sexes (because that's how many gametes there are) is just false. We can just look at the continuum.

The continuum between A and Not-A?

What's the continuum between Real and Imaginary numbers?

 

[Emily's Cat]

No matter what we do, the data will speak for itself. Do a cluster analysis, and you'll get two tightly packed clusters of individuals, and a few stragglers (in humans). I don't get why dissenting from the view that sex is necessarily binary, but can instead be seen as a (strongly) bimodal distribution is treated as such a heresy. It's just a necessary consequence of the continuum of phenotypes.

A cluster analysis of WHAT? Be specific. What are you clustering?

I'm going to jump the gun and say you're clustering characteristics. Some of those characteristics are directly caused by sex, some are merely correlated with sex.

But none of them are actually sex itself.

That's why arguing that sex is "bimodal" is a problem. It's because you're attributing the wrong thing. Sex-correlated characteristics can be bimodal when you lump both sexes together. But those are not sex.

 

[mumblethrax]

Minor correction back at you - ordinal data is a type of categorical data.

That's what I just said...I didn't mean equivalent to.

Humans aren't snails. Your argument here is irrelevant. If we were talking about a species that has two sexes, but was a hermaphroditic species (like a lot of plants) it would be an entirely different discussion.

But as we are discussing humans... this isn't making a useful point at all.

It is. I'm pointing out that if you make a logical commitment, you are obligated to ride that train to the last station. You can't say 'two gametes necessarily implies two sexes...except where it's three sexes.'

The continuum between A and Not-A?

The continuum of phenotypes. If there is no such continuum, then it is not possible to take different measurements of individual traits at all.

What's the continuum between Real and Imaginary numbers?

Zero relevance. We are not talking about different kinds of numbers.

 

[Samson]

FYI - that's a parody account, not actually Elon Musk.

Thanks
I had no idea.

 

[mumblethrax]

I'm going to jump the gun and say you're clustering characteristics. Some of those characteristics are directly caused by sex, some are merely correlated with sex.

Why would you be measuring the ones merely influenced by sex? They're largely irrelevant.

But none of them are actually sex itself.

They are, taken together, measurements of the reproductive phenotype. That is the sex of the individual. It is not sex in the universal sense.

That's why arguing that sex is "bimodal" is a problem. It's because you're attributing the wrong thing.

No, it isn't. It's because you're applying the wrong definition. Which is the language game in question.

 

[mumblethrax]

There's no distinction because up until about five minutes ago, the term "gender" was synonymous with sex.

The term gender has never been synonymous with sex, except in the euphemistic sense. Prior to the 1950s, it had no meaning at all to do with sex. It was borrowed from linguistics for the purpose of describing the cultural, psychological, and behavioral attendants of sex, because of the analogical relationship. If you go back to the 1960s you can find grumpy old coots writing letters to the editors about how you can't call it gender because that only means grammatical gender.

Now it's been hijacked and provided a different meaning altogether... and people are pretending like the new meaning is what was always meant.

It has always meant that, from day it was coined in the sociology/sexology sense.

The law, being a social rather than natural phenomenon, cannot do anything but create attendants of sex.

 

[Emily's Cat]

That's what I just said...I didn't mean equivalent to.

You're correcting me that a non-ordinal type of categorical data is categorical?

It is. I'm pointing out that if you make a logical commitment, you are obligated to ride that train to the last station. You can't say 'two gametes necessarily implies two sexes...except where it's three sexes.'

I haven't said that. I won't say that. There aren't three sexes among any of the species being discussed. And even among species that have genuine hermaphrodites... there are still only two sexes. Some species are able to have two sexes in one individual. That's still only two sexes. Hermaphroditic species don't have a third sex.

The continuum of phenotypes. If there is no such continuum, then it is not possible to take different measurements of individual traits at all.

That's... exactly the point. You cannot take measurements of sex at all. There is no continuum of sex.

Zero relevance. We are not talking about different kinds of numbers.

We're not talking about snails either so...

 

[Emily's Cat]

They are, taken together, measurements of the reproductive phenotype. That is the sex of the individual. It is not sex in the universal sense.

Okay. Let's play your game.

Since you've repeatedly referenced this as a multivariate system, please outline the variables that you're measuring. What specifically and exactly are you measuring? What is the unit of measurement that you're applying to each of those variables? Are they discrete or a continuous measures on each?

 

[mumblethrax]

Elective cosmetic surgery isn't healthcare.

Medically indicated surgery is.

(And that's not what 'elective surgery' means.)

 

[Emily's Cat]

The term gender has never been synonymous with sex, except in the euphemistic sense. Prior to the 1950s, it had no meaning at all to do with sex. It was borrowed from linguistics for the purpose of describing the cultural, psychological, and behavioral attendants of sex, because of the analogical relationship. If you go back to the 1960s you can find grumpy old coots writing letters to the editors about how you can't call it gender because that only means grammatical gender.


It has always meant that, from day it was coined in the sociology/sexology sense.

The law, being a social rather than natural phenomenon, cannot do anything but create attendants of sex.

Are you seriously arguing that in the legal sense... the law is actually and for realsies talking about what sort of gender role and presentation a person likes? And has always done so?

Are you seriously arguing that the government's view of "gender" 50 years ago was NOT as a euphemism for sex?

 

[Emily's Cat]

Medically indicated surgery is.

Elective cosmetic surgery isn't medically indicated surgery. I don't know what you think your point is.

 

[mumblethrax]

You're correcting me that a non-ordinal type of categorical data is categorical?

No. I'm pointing out that ordinal data is categorical because you objected to the use of categorical data.

I haven't said that. I won't say that. There aren't three sexes among any of the species being discussed. And even among species that have genuine hermaphrodites... there are still only two sexes. Some species are able to have two sexes in one individual. That's still only two sexes. Hermaphroditic species don't have a third sex.

Like I said...call it glory, it's still a third category and it's still the same data. Refusing to agree that an individual animal can be a hermaphrodite would be absurd, so ya got three.

That's... exactly the point. You cannot take measurements of sex at all. There is no continuum of sex.

You can take measurements of the individual traits that make up the reproductive phenotype. Taken together...that's sex (of the individual).

We're not talking about snails either so...

The. Logical. Commitment. Holds.