antonfire

> Therapy isn't supposed to scale.

As I see it "therapy" is already a catch-all terms for many very different things. In my experience, sometimes "it's the relationship that heals", other times it's something else.

E.g. as I understand it, cognitive behavioral therapy up there in terms of evidence base. In my experience it's more of a "learn cognitive skills" modality than an "it's the relationship that heals" modality. (As compared with, say, psychodynamic therapy.)

For better or for worse, to me CBT feels like an approach that doesn't go particularly deep, but is in some cases effective anyway. And it's subject to some valid criticism for that: in some cases it just gives the patient more tools to bury issues more deeply; functionally patching symptoms rather than addressing an underlying issue. There's tension around this even within the world of "human" therapy.

One way or another, a lot of current therapeutic practice is an attempt to "get therapy to scale", with associated compromises. Human therapists are "good enough", not "perfect". We find approaches that tend to work, gather evidence that they work, create educational materials and train people up to produce more competent practitioners of those approaches, then throw them at the world. This process is subject to the same enshittification pressures and compromises that any attempts at scaling are. (The world of "influencer" and "life coach" nonsense even more so.)

I expect something akin to "ChatGPT therapy" to ultimately fit somewhere in this landscape. My hope is that it's somewhere between self-help books and human therapy. I do hope it doesn't completely steamroll the aspects of real therapy that are grounded in "it's the [human] relationship that heals". (And I do worry that it will.) I expect LLMs to remain a pretty poor replacement for this for a long time, even in a scenario where they are "better than human" at other cognitive tasks.

But I do think some therapy modalities (not just influencer and life coach nonsense) are a place where LLMs could fit in and make things better with "scale". Whatever it is, it won't be a drop-in replacement, I think if it goes this way we'll (have to) navigate new compromises and develop new therapy modalities for this niche that are relatively easy to "teach" to an LLM, while being effective and safe.

Personally, the main reason I think replacing human therapists with LLMs would be wildly irresponsible isn't "it's the relationship that heals", its an LLM's ability to remain grounded and e.g. "escalate" when appropriate. (Like recognizing signs of a suicidal client and behaving appropriately, e.g. pulling a human into the loop. I trust self-driving cars to drive more safely than humans, and pull over when they can't [after ~$1e11 of investment]. I have less trust for an LLM-driven therapist to "pull over" at the right time.)

To me that's a bigger sense in which "you shouldn't call it therapy" if you hot-swap an LLM in place of a human. In therapy, the person on the other end is a medical practitioner with an ethical code and responsibilities. If anything, I'm relying on them to wear that hat more than I'm relying on them to wear a "capable of human relationship" hat.

As a fun aside on language, I prefer "they", explicitly, e.g. I pick "they" in pronoun dropdowns, I often wear a "they" pronoun pin, and so on. But insofar as people think in terms of these different versions of "they", I prefer the "'unknown/unspecified' version" over the "'prefers they' version"! Turns out that's hard to communicate.

If being annoying by (not-)answering with "mu" to all gender/pronoun questions communicates it better, maybe it's a point in favor of being annoying!

IMO it points at some cracks in the perspective that there are different versions in the first place.

You said "this makes sense", but the rest of your comment is essentially still in disagreement.

> A subtler approach that started with "why shouldn't boys wear pink?" and progressed from there would have already finished the job by now.

That's happening. There's been a lot of progress, and it hasn't finished the job. Not even as far as cis people are concerned.

(I guess it would be coherent to believe that the subtler approach hasn't worked because of less-subtle approaches like a push for transgender health care. I think that's naive.)

It's a bit cliche, and you might be tired of hearing it, but for what it's worth this conversation brings to mind the "disappointed with the white moderate" paragraph in MLK's Letter from Birmingham Jail.

> But if sufficiently changing and making flexible our expectations for what it means to be male and female would have been sufficient to make most transgender people comfortable, why did we choose the much harder sell instead?

I think your sense of "sufficiently changing" isn't aligned with the sense of "sufficiently changing" I would need to tentatively grant that, if you think such a change is an easier cultural sell than "some people are born in the wrong body".

> We could make a lot of progress in trans acceptance very quickly by just reframing the whole thing in these terms. [...] But arguing that males shouldn't need to live up to an artificial and incredibly outdated standard of masculinity? That would be a much, much easier sell.

It's an "easier sell" of a different thing.

I'm not seeing a "we" forming here as far as "trans acceptance" is concerned.

Judging from your comment history, your perspective on this seems to basically be grounded in an objection to (more generously: apprehension about) transgender healthcare practices. On purportedly scientific grounds, while ascribing a "politically-driven" motivation in terms of groupthink to people who support these practices. So I think your motivation to ascribe this to a "cultural thing" is grounded in a desire to decouple it from the healthcare thing, because you think it serves an overall political project better.

I'm with you on loosening gender roles. I'm not with you on reducing trans acceptance to that.

> why are we doubling down on gender binaries and talking about switching genders instead of creating a campaign that would both get at the root of the issue and be easier to swallow for a larger portion of the country?

In my personal life, I am probably about as far from "doubling down on gender binaries" as anyone you are likely to encounter. In my experience, I find many more people who are genuinely working past "doubling down on gender binaries" in transgender spaces than I do outside them.

My not-doubling-down-on-gender-binaries approach to it is not an easy pill to swallow for a large portion of the country. It may not even be an easy pill to swallow for you. (E.g. "which" bathroom do you want me in? How easily do you think the rest of the country will swallow that?)

Kind of a tangent, but I think the cultural/social side is bigger and messier than your questions suggest, and I think boiling this down to "biological vs cultural" misses that.

> If gender is a social construct, is identifying as transgender the result of feeling pressure to conform to a cookie-cutter definition of what someone with male/female parts is meant to be like?

I suspect "identifying as" cisgender is the outcome of this kind of cookie-cutter pressure to much the same degree, if not more. This tends to go unnoticed even in conversations where people are directly engaging with the ideas. (Even though that's part of what "gender is a social construct" is meant to suggest.)

A rhetorical question for cis people: to what degree do you feel your cisgenderness is a result of feeling pressure to conform to a cookie-cutter definition of what someone with male/female parts is meant to be like? I suspect there's more meat to genuinely unpacking this than you might think. Trans people's answers to this might not be that different from cis people's.

"It's a social construct" is an invitation to peek behind (or at least recognize) an abstraction; but it's just a peek into a quite complex story.

"Social construct" doesn't necessarily mean something to rise above, or to dismantle, or to deny. Money is a social construct. Human rights are a social construct. Friendship is a social construct. Sure, one can usefully imagine oneself "above" those things at times, but it's unclear whether aspiring to that is a good idea, and realistically most people won't attain it even if they do. Culturally we must find some relationship to those things anyway, we cannot ignore them.

As I see it, gender is an aspect of a messy evolving cultural system. Yes, the concept of "transgender" is part of that system, though it certainly doesn't fit into that system in quite the same way as "man" and "woman". Trans people tend to challenge or pressure many aspects of this system in some ways that cis people tend not to, but that's not the same thing as denying the system in whole. (Some people do see "gender abolition" as an aspirational ideal; many don't.)

Broadly, I think "is being transgender also just a social construct that can and maybe should be addressed by loosening up our tight expectations for gender roles" vastly underestimates the scope and scale of this cultural system, the degree to which it's tangled up in our lives, and the difficulty of untangling it.

> Some women are tetrachromats with an extra colour sense, while men are more likely to have red/green colour blindness.

If I'm not mistaken, red/green color blindness is more common in men because it's caused my a mutation on the X chromosome (which men tend to have fewer of). I would guess a similar thing about tetrachromacy.

So those are probably unrelated to color-perception changes due to exogenous estrogen.

> Perhaps this is an insensitive question/comment, but do trans women feel like they have the wrong body or the wrong wholesale gender?

It varies.

> In my experience with trans women I know, they still seem to relate primarily to men (they still gravitate towards male dominated interests) whereas many gay men I know seem to relate primarily with women, and gravitate towards women interests.

For whatever it's worth, I think observations like this are as useful a cue to look inward for an explanation as they are to look outwards.

For one thing, part of the whole "gender" thing is the way people's preconceptions lead them to parse information about others (and themselves!), and your sense of trends is probably influenced by that. (E.g. when a (gay) man gravitates towards "women interests" that may just be more salient than when a woman does, so you notice it more.) For another, you might be in a lot of male-dominated spaces (e.g. this one), so the set of trans women you know is probably not that representative. These might not be the whole story, but they certainly have a role to play in whatever reconciliation you're seeking. Gender is difficult to navigate: we're all swimming in it.

For me personally: I'm "nonbinary", whatever that means. As I see it today, for me being trans feels like more of a "wrong wholesale gender" thing than a "wrong body" thing. (But I'm open to the idea that I'm just not in touch with my body.) Part of the "wholesale gender" thing is the realization at some point in my life that "gender" was playing a much bigger role in my life than I had realized, including how I relate to people, what interests I gravitate towards, and so on. Something I find deeply aversive.

But I'm also averse to, like, rearranging my whole life to retroactively "fix the gender story" around it, just to make myself more legible. You might parse me as gravitating towards interests that line up with my assigned gender at birth (AGAB), and maybe even as relating to people primarily of my AGAB, and so on. I'm sure some people go further and functionally take this as an excuse to continue to relate to me through the lens of my "birth gender" or what have have you. I'm sure it's easier. From my perspective, I suspect those people are underestimating how much of a clusterfuck the whole "gender" thing is.

> But if, instead of extending it, you try to contract it, then without crossing into metaphysics, you'll stop at the brain.

Dunno about the other poster, but I can promise you I will not.

I think it's kind of a pointless exercise to try to draw a physical boundary of what's "me" and what isn't, but carrying that exercise a bit further: How much of the spinal cord can I exclude in my "sense of me" before I "cross into metaphysics"? If I drop my eyes from my sense of "me", seems like I could also drop the neurons in the brain that are responsible solely for visual processing of input from them. Or is that a step too far and "into metaphysics"?

Heck, I think the classic nerd brain-in-a-meatsuit position doesn't really stop at the brain here either. E.g. I think it places an accurate simulation of one's brain running on different hardware on about the same level as the meat brain, and "you" wouldn't know the difference. That's just not a thing we can do (yet?). Does that position cross into metaphysics?

In practice, I "contract my sense of self" when it comes to my thoughts too, which (presumably) all happen in the brain. I often find it useful to ask "where did that thought come from?" and give an external account ("ah, I picked it up from X") and let that have some bearing on the next thought. I also have "intrusive thoughts"; the act of labeling a thought an "intrusive thought" is (arguably, partially) an act of contracting one's sense of self to exclude that thought.

I'm pretty sure this conversation had "crossed into metaphysics" by the time discussion about expansion/contraction of one's "sense of you" was happening; not when the contraction reached the brain.

> But the fact remains that a hand or a heart or a lung can be removed or replaced without loss of self. A brain can't.

Here's a related observation that tipped me a bit away from thinking about my body as the meatsack carrying my brain: cutting off a hand doesn't cut off the experience or sensation of having a hand. (E.g. people with missing limbs have phantom pain.)

Lots of emotions are at least partly "embodied". People feel love in the chest. Fear in the stomach.

Not sure about anybody else, but I feel frustration in my elbows. I strongly suspect that if my arms were removed, I would still feel frustration in my elbows. I don't think removing my physical arms would remove all the ways "having arms" is baked into my experience. Even if an external physicalist account accurately places those phenomena "in the brain" and not "in the arms", that's not really all that relevant to an internal account of someone actually having the experience.

As I see it, a human's brain-from-the-inside (mind? soul?) is "body-shaped" in a lot of ways. When I hear warnings against "reducing ourselves to a bunch of 'computers", I read them as prompts to recognize this. As warnings against recognizing and engaging with *only* the parts of one's experience that can fit into an "abstract reasoner" model.

Throwing blame on "the prevalence of psychotherapy" into this conversation, writing it all off to "pseudo-science" and emphasizing "lifestyle" (as in "if you exercised more you'd be better off", I guess) reads like a just-so story from a brain in a meat sack that's missing the substance of the conversation.

Just like "meditation" is many different things, and "exercise" is many different things, and "healthcare" is many things, and "thinking" is many different things, "psychotherapy" is many different things.

One of the first things by current therapist asked me was more or less how embodied I was. I suspect of all the relationships with other people in my life, the one with my psychotherapist is the one that's most likely to be help me build out a fuller relationship to my body in the way the grandparent comment is advocating.

The therapist before that focused on behavioral "lifestyle" changes when feelings came up, in a way that I guess superficially follows the "I am a body" model, but really felt like it only added distance and abstraction to my relationship to myself. I guess that approach was grounded in a scientific evidence base that those kinds of interventions are what moves metrics.

> It appears this world has become manically trigger-happy to label something as -ist or -istic

Also to label something as maniacal, perhaps.

"Women divorce men more often than the other way around; this seems to bear out that men are more dissatisfied in marriage" seems like a stretch to me.

At any rate, it reminds me of the classic physics joke:

> An experimental physicist comes running excitedly to a theorist's office, waving a graph taken off his latest experiment. "Hmmm", says the theorist, "that's exactly where you'd expect to see that peak. Here is the reason." A long logical explanation follows. In the middle of it, the experimentalist says "Wait a minute", studies the chart for a second and says "Oops, this is upside down." He fixes it. "Hmmm," says the theorist, "you'd expect to see a dip in exactly that position. Here is the reason..."

Leaning into society's stereotypes is one way to bend to people's implicit demands to change your writing (and quite possibly thinking).

But sure, if the metric you're optimizing is raw views and smiles and laughs, then probably the way to go is leaning into stereotypes. There's a reason those views and smiles and laughs are called "cheap".

> It's a trope to say that our "best minds are working on ads".

"The last generation was better because they read _books_ and had _discipline_ and didn't waste their time on frivolous garbage" is also a trope.

> This study found that the "marshmallow test"—as a single measure—is no more or less predictive than a basket of other measures of self-control the study tested, or any of those other measures of self-control taken alone.

Are we looking at the same study? I don't see where "no more or less predictive than a basket" comes from, specifically where "no less predictive" comes from.

My reading of the abstract is that, the study found that a measure based on the "marshmallow test" ("preschool delay of gratification", RND in the article body), is not predictive of the outcomes they measured (11 capital formation outcomes).

It also found that a basket of measures of self-control (collected at various ages, RNSRI/RNCCQ in the article body) is predictive of the outcomes, whether you include the preschool measurement or not.

So from skimming the study without even reading the article, it sounds to me like they found that the preschool measure doesn't predict the outcomes they're measuring by itself, and it doesn't contribute predictive power when it's used as part of an index of self-control measured at a variety of ages.