[FRIAM] gene complex for homosexuality

glen gepropella at gmail.com
Fri Jan 14 11:34:52 EST 2022


Well, this reminds me of hidden state machines and the law of requisite variety. If we're trying to explain why humans have persnickety preferences, including state-hiding tendencies like focusing on emotion vs brute fact, rather than argue for a flattening of the collective human/biological machine(s), we should look at both the machine(s) and the environment(s) in which they're grown.

Does the state-hiding machine present a more expressive problem solver than would otherwise be achievable without hiding state? And is that extra expressibility necessary (or more convenient/efficient) than with an in principle equivalent flattened (set of) machine(s)?

I'm then reminded of demonstrations (?) that zero determinant game strats, while able to dominate in ideal contexts, don't do so well in evolutionary contexts. So, perhaps the answer to these questions is simply "No" ... that the hidden state doesn't provide any extra problem solving ability and the tendency to (or advocacy of) avoid the flattening is an operable sign of bad faith? Or, in the lingo of the laity, curmudgeons are a justified cost because they more quickly indicate the bad actors. Get off my lawn!

And if the answer is "No", how do we explain the existence of this "cognitive ease", this tendency to rely on stereotypes and historicity-reinforced signs (perhaps now having lost/changed their referents)?

On 1/13/22 16:48, Marcus Daniels wrote:
> Anyway, the reason I noticed this article is that I posit that the steely harm reduction approach that was discussed recently is in my mind a form of stoicism.   Can one put away their emotional responses and make hard choices based on the greater global good?   If one engages in large intimate social networks, I would say two things are likely to happen:  1) executive decisions become harder because there is diffusion of sensitive information, and thus political complications in making them.  Members in the network may not be sharing the whole factual context (preferring the emotionally laden parts) 2) there are still dominance relations (her language), but they are just manifest in different ways.  Namely by being in the center of a social network and slightly censoring the information that gets passed along.
> 
> As it relates to the subject line, there may be some weak tendency one has to share or not share by default depending on hormones/genetics.
> ________________________________
> From: Friam <friam-bounces at redfish.com> on behalf of Marcus Daniels <marcus at snoutfarm.com>
> Sent: Thursday, January 13, 2022 5:12 PM
> To: friam at redfish.com <friam at redfish.com>
> Subject: Re: [FRIAM] gene complex for homosexuality
> 
> < So, I'd argue against you completely. This essay is talking about how to detect and operate in the presence of bad faith. And, to be clear, the bad faith actor doesn't necessarily *know* that they're acting in bad faith. In fact, it's a more canonical case of bad faith if the actor has simply habituated to it. >
> 
> A contrast she draws is between petulant vulnerability and "real" vulnerability.   That it is "scary" and "any less necessary, for men".
> There's another option which is not to use "the language of vulnerability as a cudgel", but also not engage "the human condition of reliance on others."  She is expressing an expectation for high intimacy, and it is implicit that there is something wrong with keeping your distance.   I've seen this false choice portrayed by other so-called feminists.  I don't buy it.
> 
> Marcus
> ________________________________
> From: Friam <friam-bounces at redfish.com> on behalf of glen <gepropella at gmail.com>
> Sent: Thursday, January 13, 2022 4:55 PM
> To: friam at redfish.com <friam at redfish.com>
> Subject: Re: [FRIAM] gene complex for homosexuality
> 
> What's interesting about that essay is its appeal to character or "virtue ethics", I think. I've tried to address this a few times in past threads, especially when concepts like "bad faith" arise. Rittenhouse' crying looked precisely like bad faith to me. I get accused of it a lot because I enjoy playing roles and believe playing roles (like Devil's Advocate) facilitates healthy reasoning. (E.g. EricC's accusation of illiberalism on my part when condemning the anti-masker's punching of the doctor.)
> 
> So, I'd argue against you completely. This essay is talking about how to detect and operate in the presence of bad faith. And, to be clear, the bad faith actor doesn't necessarily *know* that they're acting in bad faith. In fact, it's a more canonical case of bad faith if the actor has simply habituated to it. Rittenhouse's crying on the stand and Kavanaugh's crying in his confirmation hearings both seem to me to be statements about their *character*. That means whatever ways we have/develop to detect bad faith can be made reflective ... kinda like the Reddit forum "Am I the Asshole?" 8^D
> 
> I doubt one's oxytocin-laced skepticism over such acting is completely arbitrary ... or even a preference at all.
> 
> On 1/13/22 14:33, Marcus Daniels wrote:
>> Well, now that I've taken one extreme position, let me take the other extreme position!   This essay reflects, IMO, an arbitrary preference for social affinities of a certain sort, and it is only one sort of valid class of relationships.  Relationships that have benefits, but also costs.   It's not just overbearing on how men should be, but also on how women should be.
>>
>> https://www.nytimes.com/2022/01/13/opinion/toxic-masculinity.html


-- 
glen
Theorem 3. There exists a double master function.



More information about the Friam mailing list