Said Achmiz

Karma: 16,472

Said Achmiz May 14, 2025, 4:59 PM
2 points
0
on: Moral Obligation and Moral Opportunity

Positive reinforcement is more effective than positive punishment because it tells someone exactly what to do instead of just what not to do.

But being told exactly what to do is much more restrictive than just being told what not to do. So if you make the above claim, and also make the claim that your way is more pleasant and less restrictive, then Bob is right—there is definitely some trick being played here.

Said Achmiz May 13, 2025, 6:22 PM
18 points
9
in reply to: Said Achmiz’s comment on: Interest In Conflict Is Instrumentally Convergent
Here is some commentary on my views on this subject, framed with a bit of literary analysis. (This was inspired by a semi-recent rereading of a pair of Alexander Wales stories, namely Shadows of the Limelight and The Metropolitan Man. NOTE: This comment contains spoilers… well, a spoiler… but only a very mild and general one.)

The key point is: Superman is evil.

For anyone who hasn’t read The Metropolitan Man: it’s a Superman fanfic, wherein Superman is revealed to be evil. That is: it’s not an AU fic where Superman, instead of being good like he normally is, is instead evil; rather, it’s just the regular Superman, but the story shows (or so I claim! it’s not made explicit, of course) that regular Superman is evil.

Now, this is hardly a novel insight, but the trick is articulating just what the nature of that evil is; but I’ve finally managed to do so, to my own satisfaction. The problem with Superman (as depicted in The Metropolitan Man—and also, of course, many, many canon Superman stories) is the chain of reasoning that goes:

“With great power comes great responsibility” → “With great responsibility comes great moral leeway”

In other words, you are responsible for the lives of many people, the fate of a nation/planet/whatever, or some other great and important thing; therefore you cannot be expected to scrupulously follow rules that normal people are held to, and are allowed to break the law, to lie, to imprison innocents, or whatever else.

But why do you have that great responsibility? Well, because you have great power, so you were obligated to take on that responsibility.

But then result of this is: “With great power comes great moral leeway”.

Which is just “might makes right”.

So, through this concept of moral obligation, you have reasoned yourself into being exempt from the rules because you’re very powerful; and of course, being powerful, you also can’t have those rules enforced upon you—very convenient!

That’s Superman in a nutshell. (Also many other superheroes, but Superman is the most famous example.)

The correct solution (morally speaking) is, of course, to deny the first step: no, with great power does not come great responsibility—except the responsibility of being careful not to use that power to do harm. Anything other than that, over and above any responsibilities that any person has, is strictly supererogatory.

And of course one does not have to be Superman to encounter this problem. Whenever you gain power, there is a very great temptation to use this power; and it is a dangerous delusion, to think that the only problematic temptation is the temptation to use your power to aggrandize or enrich yourself directly. The temptation to use your power to benefit others is even more perilous.

So what is to be done?

If you find yourself with power, do not begin by thinking “how can I use this to do good”. Think first:
- How can I ensure that I do not use this to do harm?
- How can I share this power, so that I am not the only one who has it? Can I set things up so that everyone has it?
- Does anyone need to have this power? Can I set things up so that nobody has it?
(Note that these are not numbered—I am not saying that you need to think of these things in that order. Think of them all, simultaneously.)

(There are also auxiliary questions, like “If someone whom I didn’t trust at all were to gain this power, what would I wish they’d do with it?”, and “Suppose that I become crazy and/or evil tomorrow; what should I do to prevent bad-future-me from abusing this power?”. Note that if the above-listed primary questions are addressed effectively, these auxiliary questions also disappear almost entirely.)

(The astute reader may notice parallels between the above perspective and certain dichotomies among different types of economic systems, approaches to the construction of complex abstract systems, etc.)

Said Achmiz May 13, 2025, 3:01 AM
3 points
−2
in reply to: jbash’s comment on: In Defense of a Butlerian Jihad

I still can’t believe some of you would sided with the super-happies !

So you’re siding with the guy who killed 15 billion non-consenting people because he personally couldn’t handle the idea of giving up suffering?

I’m not the OP, but: that was very obviously the correct decision, so… yes, definitely, I am siding with that guy.

Said Achmiz May 12, 2025, 11:27 PM
2 points
0
in reply to: Screwtape’s comment on: Interest In Conflict Is Instrumentally Convergent

I’m in favour of the incremental steps towards that I can make. In the abstract, I agree the procedure for evaluating advice should work even if it comes from bitter enemies. I do not think my personal evaluation procedure is currently robust enough to handle that, though tsuyoku naritai, someday maybe it will be.

Yes, but in the absence of this, every other approach is doomed to failure. And the severity of the failure will be inversely proportional to how seriously the people with the power and authority take this problem, and to how much effort they put into addressing it.

I want good advice or structures to hand to them, and expecting them to be able to evaluate my advice to that standard seems unreasonable.

Respectfully, I disagree. I think that this is the only standard that yields workable results. If it cannot be satisfied even approximately, even in large part (if not in whole), then better not to begin.

Said Achmiz May 12, 2025, 11:23 PM
2 points
0
in reply to: Screwtape’s comment on: Interest In Conflict Is Instrumentally Convergent
The idea that situations/problems that involve conflicts, and require resolving conflicts, are more challenging than situations/problems that don’t involve conflicts, is trivial. If someone thinks that conflict handling should be simple then of course that person is an idiot. If this were all that you were saying, then it would hardly be worthy of a post.

However. In the OP you write:

Now let’s say that the bad actor is not an idiot. They have considered what and who might stop them, and what they might do about it.

It is an obvious, straightforward move to accuse whatever system is responsible for catching them of being corrupt and the people running that system to be horrible or incompetent.

But a more realistic scenario would be:

“Now let’s say that the system is already full of bad actors (as it probably is). They have considered what and who might stop them, and what they might do about it. The system will, of course, be corrupt, and the people running that system will be horrible or incompetent. It is an obvious, straightforward move to promote memes that prevent this from being rectified.”

(You might protest that sure, this happens, but you know that you are not a bad actor, right? Even if nobody else can be sure of this, you at least can! And you’re talking to people who also know that they are not bad actors. And I say that even this is false. You don’t know this. The people in your target audience—other organizers, etc.—also don’t know this about themselves. And certainly nobody else should take your (or their) word for it.)

You write:

I recognize the weirdness around me trying to point out the second mistake—as I said in the post, “Okay, but why should you trust me? Professional interest since complaint handling is part of my role, but good question and don’t be satisfied by that.”

But in fact this should read more like this:

“Okay, but why should you trust me? Good question; the answer is that you definitely shouldn’t trust me—especially since complaint handling is part of my role, so I have a professional interest, which makes me exceptionally likely to be a bad actor. Do not trust! Verify! If you can’t verify, assume treachery until proven otherwise!”

I am trying design a setup that would work even if I was a problem even as I was designing the setup.

This is an admirable goal and I applaud it. My comments are aimed precisely at this goal also. You can read them as saying “your setup does not and cannot succeed at this, so long as you take the approach described in the OP”.

I will very cheerfully bet you my hundred bucks against your dollar that every police force does not criminalize everyone who makes any move towards trying to change the system, unless you’re using some non-standard definitions of police force and criminalize.

I… didn’t claim that any police force criminalizes everyone who makes any move towards trying to change the system, so… no bet! (This is especially a nonsensical formulation since police forces do not criminalize anything; governments—in systems like that of the United States, specifically legislatures—criminalize things. But police forces have many, many tools available to them to apply differential treatment on the basis of evaluated threat level. Heck, even actual criminals can be harrassed and pressured in many ways other than arresting, indicting, and convicting them of a crime.)

I’m going to assume that “every” is colloquial, not literal, and that paragraph is rhetoric, but even if the section in quotes was how police forces and authoritarian enforcement organizations were, I don’t think their logic ends by suggesting constant vigilance against themselves.

It’s not literal only to the extent that I anticipate reasonable disagreements about what qualifies as a “police force”, “security service”, or “authoritarian enforcement organization”. Otherwise, I can’t think of any exceptions.

And their logic certainly doesn’t end by suggesting constant vigilance against themselves. Of course it doesn’t! They’re the ingroup; why should they guard against themselves?

(Although I will point out that the pattern is fractal. The cops treat “civilians” in this way, but then cops-within-the-cops organizations like “Internal Affairs” divisions and the like treat regular cops in this way. Similar patterns involving the relationship of ordinary people, spies, and counter-intelligence agencies are also well known. I don’t know offhand of any examples of this pattern going up another level, but I expect that they exist.)

I do think it would help to have a better name for this bucket, since in my contexts a lot of people aren’t getting paid a regular paycheck to do this and don’t have a lot of training or background. The obvious and central examples are- divorce lawyers are, CEA’s community health team comes to mind as well- but a lot of the time it’s a local meetup organizer in some random town that’s had an issue.

To be clear, this is how I assumed you were using the term as well.

I am not sure whether “this blind spot is not an accident” is suggesting that I’m making a good faith effort but a predictable mistake, or that I am deliberately leaving information which I know to be true and relevant out in an effort to make my argument stronger, or some third thing. Would you please clarify this? As you might imagine, if it’s the second I’m going to say that’s not what I’m trying to do.

First one.

EDIT: I really should clarify this one, sorry. It’s not the second one or any third thing—that’s what my response meant. My comment about “this blind spot is not an accident” was not meant as an accusation of deliberate bad faith argument in your post, in other words—nothing like that!

But your first interpretation isn’t quite right either, because it takes an entirely too mistake-theoretic approach. It would be more correct to say something like: the blind spot in question is a predictable consequence of the logic and incentives of the situation, such that it might be that your intentions are pure, or it might be that your intentions are not pure, but either way the result is ultimately the same, and the distinction usually moot in any case (due to self-deception).

(I am happy to assume perfectly good intentions on the part of actual-you, the actual person I am talking to, for the purposes of this and similar discussions. It’s just that we have to keep in mind the possibility of a hypothetical-you who is in the same situation and is writing the same things but who does not have good intentions.)

Said Achmiz May 11, 2025, 11:50 PM
2 points
0
in reply to: CronoDAS’s comment on: Interest In Conflict Is Instrumentally Convergent
So you trust yourself to decide how to select a doctor; you trust your decision procedure, which you have chosen.

If you were accused of a crime and intended to plead not guilty, how would you choose a defense attorney, assuming you weren’t going to use a public defender?

I’d ask trusted friends for recommendations, because I trust myself to know whom to ask, and how to evaluate their advice.

Said Achmiz May 11, 2025, 11:15 PM
2 points
0
in reply to: CronoDAS’s comment on: Interest In Conflict Is Instrumentally Convergent

In certain domains, I absolutely can and will do this, because “someone else” has knowledge and experience that I don’t and could not conveniently acquire.

And how do you choose who the “someone else” is?

Said Achmiz May 11, 2025, 9:41 PM
2 points
−1
in reply to: NunoSempere’s comment on: Interest In Conflict Is Instrumentally Convergent
Uh… sure, that’s true enough, but this logic requires that we first accept the OP’s categorization scheme—which is part of precisely the meme that I am referring to!

OP notes that “These categories absolutely overlap and intersect some of the time”. This is true, but the trouble is that taking this caveat seriously means discarding the logic of the argument.

Consider an alternate set of categories:
1. Professionals, who might also be bad people.
2. The curious, many of whom are also oddballs, and who might also be bad people.
3. The oddballs, who might also be bad people.
4. Normal people, who might also be bad.
Hmm. Doing a Bayesian calculation on this might be tricky. Perhaps we can separate out some of those, like so:
1. Professionals who are good people.
2. Professionals who are bad people.
3. The curious and/or oddballs who are good people.
4. The curious and/or oddballs who are bad people.
5. Normal people who are good.
6. Normal people who are bad.
We learn that someone has opinions about [whatever]. We now discard categories #5 and #6. But unbeknownst to us, the ratio of bad:good in the overall population was actually lower than the ratio of bad:good among normal people. So learning this fact should reduce our subjective probability of the person being bad.

Is this possible? Well, for example, suppose that it’s 1975, and you learn that a certain person has opinions about how psikhushkas should work. In particular, this individual thinks that said institutions should work differently than they do in fact work. What should you conclude about this person?

In the 1960s Soviet psychiatry, particularly Serbsky Institute Director Dr. Andrei Snezhnevsky, introduced the concept of “sluggish schizophrenia”, a special form of the illness that supposedly affects only social behavior, with no effect on other traits: “most frequently, ideas about a struggle for truth and justice are formed by personalities with a paranoid structure”, according to the Serbsky Institute professors.[9][10]

The logic in the OP is easily recognizable as the logic of every police force, every security service, and every authoritarian enforcement organization. It’s the logic that says “if you’re not one of us, then you’re either a clueless normie who will unthinkingly submit to our authority, or else you’re probably a criminal—unless you can, with great effort and in the face of considerable skepticism, prove to us (and yes, the burden of proof is entirely on you) that you’re one of the rare harmless weirdoes (emphasis on the ‘harmless’; if you give any hint that you’re challenging our authority, or make any move toward trying to change the system, then the ‘harmless’ qualifier is immediately stripped from you, and you move right back into the ‘criminal’ category)”.

(That “professionals” are much more likely than anyone else to be bad actors is another fact that drastically undermines the OP’s thesis—and this blind spot is not an accident. It’s just that “professionals” simply means “the ingroup”—“one of us”. As in, “You know the score, pal! If you’re not cop, you’re little people.”)

Said Achmiz May 11, 2025, 4:40 AM
10 points
−3
on: It’s Okay to Feel Bad for a Bit
Excellent post.

My only (very minor) disagreement with the thesis is that, IMO, Stoicism and Buddhism are not quite the same thing or equivalent, and the difference is relevant to your point. I think that you should take Stoicism slightly more seriously than Buddhism (but still not too seriously). (But I think that your “bumper sticker” advice is basically correct.)

This really is a very minor quibble, though. On the whole, this post is simply correct, period.

Said Achmiz May 10, 2025, 11:34 PM
2 points
0
in reply to: NunoSempere’s comment on: Interest In Conflict Is Instrumentally Convergent
I am not sure I follow. Could you say more? What do you mean by saying that I am discarding that probability mass?

Said Achmiz May 10, 2025, 10:18 PM
0 points
−1
in reply to: tcheasdfjkl’s comment on: Eukryt Wrts Blg

I mean “radioactive” in a descriptive political sense.

… yes, obviously. (How else could you have meant it? What did you think I understood you to be saying…?)

I agree that truth claims ought not be radioactive in this sense and it is a bad thing about the political landscape than they are.

This is all very good, and yet you still haven’t checked whether said claims are true. So what is this opinion (that “truth claims ought not be radioactive in this sense”, etc.) worth, exactly?

Yes, trans people can be discriminated against for being mentally ill. What I meant was that if someone says “trans people are often discriminated against and that’s bad” you should not respond with “well, trans people are mentally ill, what do you expect” as though (a) that’s universally true of trans people (b) that means the discrimination is justified.

Yes, once again, this is just what I assumed that you meant…

But that response makes perfect sense! Yes, one should indeed expect that a population with a substantially above-average prevalence of mental illness will experience substantially above-average discrimination. What in the world does it have to do with anything being “justified”…? Nor is there any implication of universality.

(So your parenthetical is unjustified; I understood you just fine, as you see. I simply disagree.)

I answered the question you asked (in a tone of confident assumption I would not be able to produce an answer)! I thought maybe you wanted existence proofs of me actually believing that saying a true thing can be bad rather than using that as a smokescreen for some reason

To which point it surely is relevant that you (I claim) did not, in fact, produce an answer.

(Possibly you disagree. But then that’s the disagreement, right? Whether your answer was, in fact, sufficient to answer the question. You will agree, at least, that there can be disagreement on this point, yes?)

I wrote:

I ask, of course, because I am skeptical of this notion that the truth of the claim might be admitted by both sides, with only its “appropriateness” being, by itself, evidence of racial animus.

And your examples reinforce, rather than undermining, this skepticism.

The type of person who I think you’re gesturing at would never want to admit that a radioactive claim might possibly be true and if anything might end up using smokescreens to try to avoid admitting that.

You don’t see any connection whatsoever between this description and your comments about how some claims are “radioactive” and you haven’t investigated them?

Said Achmiz May 10, 2025, 9:37 PM
1 point
1
in reply to: tcheasdfjkl’s comment on: Eukryt Wrts Blg

Many of the truly radioactive claims I have not really investigated the truth of

And this doesn’t strike you as being at all suspicious? You judge some claims to be “truly radioactive”, and you haven’t even checked whether they are true? (What, do you think that a topic being “truly radioactive” makes it less likely to be relevant to real-world outcomes?) But never mind that, let’s move on.

The first three claims you list are either politically anodyne or else have the valence of the dominant political faction. And one basically never sees anyone condemned and targeted for shunning on the basis of having such beliefs.

As for the fourth claim—of course it is relevant to bring this up in a discussion of discrimination against trans people! How could it possibly not be? (What, does discrimination against mentally ill people suddenly not exist? Or do you think that no trans person has ever been discriminated against for being mentally ill?) In general, serious mental illness has a pervasive effect on everything about a person’s life. To conclude that someone being mentally ill is not relevant at all to discussion of some major aspect of a person’s life should be quite surprising.

And this illustrates the larger point. Discussing “appropriateness” rather than truth is, in almost situations like this, injurious to our truth-seeking efforts. (Deliberately so, of course.)

Said Achmiz May 10, 2025, 9:25 PM
0 points
2
in reply to: tcheasdfjkl’s comment on: Eukryt Wrts Blg
Suppose that someone has views that I think are “odious”, but which have a totally different political slant (either on the opposite side of the standard political spectrum, or just largely orthogonal) than all this stuff with Cremieux.

Should rationalist gatherings shun this person? If not, why not?

We can even make this more personal: suppose that you have views that I think are “odious”. Should rationalist gatherings shun you? If not, why not?

(I mostly don’t know your political views, and I don’t currently have any reason to think that you should be shunned. But you can easily enough imagine the scenario, I expect.)

Presumably you will answer “no” to both questions. But why? You’re giving reasons why you think that Cremieux is “odious”, on the basis of his views and his public comments about his views—just that, not anything else!^[1] Well, surely I could give reasons why someone (perhaps even you!) is “odious”, on the basis of that person’s views and comments thereon.

So why shouldn’t rationalists shun this hypothetical person? Why shouldn’t rationalists shun hypothetical-you?

Is it a matter of majoritarianism? We should shun anyone whom the majority of rationalists consider “odious”? (But if so—what is the denominator? Who gets to vote in this referendum?)

And if not that—then what? (Note that object-level arguments—“but you see, clearly, this guy really is odious!”—will obviously not suffice.)
1. ↩︎
  This is especially hilarious given that there genuinely seem to be good reasons to, if not disinvite the guy, at least to remove him from the featured-speaker list—the plagiarism, and the exceedingly hostile response to the (quite credible) accusation thereof.

Said Achmiz May 10, 2025, 8:58 PM
1 point
0
in reply to: tcheasdfjkl’s comment on: Eukryt Wrts Blg
Perhaps, perhaps. And do you have any examples of such facts (i.e., true claims) the mentioning of which is sometimes blameworthily racist/antisemitic/etc.?

(Presumably “Jews are more greedy than Gentiles” isn’t such a fact. Or is it, do you think? I am not aware of this actually being established by studies, but perhaps I am wrong about this?)

I ask, of course, because I am skeptical of this notion that the truth of the claim might be admitted by both sides, with only its “appropriateness” being, by itself, evidence of racial animus. It’s very convenient to be able to declare that “it’s not what you’re saying, it’s how [or when or why or how often] you’re saying it”. One notices that, somehow, those who make such declarations never quite get around to investigating the “what”. It’s never the right time, the right context, etc. (Meanwhile, with claims that are favored by such people, it always seems to be the right time and place and context.)

So, what’s an example (better yet, three examples) of such a thing? A true fact, which is sometimes inappropriate to bring up, because doing so would be racist, antisemitic, etc.?

Said Achmiz May 10, 2025, 8:40 PM
2 points
0
in reply to: dirk’s comment on: Eukryt Wrts Blg
…?

This reply seems like a non sequitur. How is it at all responsive to what I wrote?

Said Achmiz May 10, 2025, 8:13 PM
4 points
0
in reply to: dirk’s comment on: Eukryt Wrts Blg
Come now, you are being obtuse. What is the reason why eukaryote claims that Cremieux is racist? It’s his empirical claims, according to eukaryote herself.

Said Achmiz May 10, 2025, 8:09 PM
1 point
1
in reply to: tcheasdfjkl’s comment on: Eukryt Wrts Blg
How can it possibly be a perfect conclusion?? In my scenario, you don’t even know if Christopher Kimball has ever heard of my forum! (Sure, I say that he’s been offered a free ticket, but how do you know whether he’s even gotten the email, or whatever?)

Are you suggesting that I might, right now, at this very moment, be a “community representative” of some community that I’ve never heard of, because they put a link to my blog on their event announcement page, and sent me some sort of offer which went straight to my junk mail folder?

Said Achmiz May 10, 2025, 8:00 PM
4 points
−1
in reply to: Viliam’s comment on: Interest In Conflict Is Instrumentally Convergent

My current best guess about what you are trying to say is something like this: “People should give up on the idea of making systems that are resilient against bad actors on both sides. You should just give unlimited power to one side (the moderator, the meetup czar, the police...) and that’s it. Now at least the system is resilient against bad actors on one side.”

Uh… no. Definitely not.

I guess you believe that community moderation can be solved by an algorithm.

… what in the world?

No, I don’t believe anything like this.

Honestly, it would be hard to get further from my views on this subject than what you’ve described…

Said Achmiz May 10, 2025, 7:06 PM
1 point
−1
in reply to: tcheasdfjkl’s comment on: Eukryt Wrts Blg
I don’t get it. How does any of that make someone a “community representative”?

Suppose I start a baking forum for people who like to apply careful analysis to baking, and I decide to run an event for “rational baking” aficionados. On the announcement page, I write that I love Christopher Kimball’s writings, that he embodies the virtues that we are celebrating, and that he has been offered a free ticket to LessFondant. Would you conclude from this that Kimball is a “community representative” of my forum for baking nerds…?

Seems pretty clear to me that this would be a quite ridiculous conclusion to draw.

Said Achmiz May 10, 2025, 6:57 PM
20 points
11
in reply to: dirk’s comment on: Eukryt Wrts Blg
The problem is not simply the accusation of being racist. The problem is the accusation of being racist, in response to an empirical claim, as a substitute for addressing the empirical claim (and with the implication that the accusation constitutes a sufficient reply to the empirical claim).

Suppose that I say “Jews are more greedy than Gentiles; this is established by the following studies”. Now consider the following possible responses:
1. “You’re an antisemite.”
2. “You are [ mistaken / lying / wrong / otherwise making a false claim ], which I know because [ the studies have poor methodology / they don’t replicate / the operationalization makes no sense / publication bias / etc. ].”
3. “You are [ mistaken / lying / wrong / otherwise making a false claim ], but I have no evidence to back up this disagreement, nor can I provide any good reason to reject the evidence you offer; nevertheless, I claim that you are wrong.”
4. “You are [ mistaken / lying / wrong / otherwise making a false claim ] (which I know because [reasons, as above]); also you’re an antisemite (which I conclude from the fact that you have made this false claim, and other similar evidence).”
Reply #2 is clearly “in the spirit of rationalist discourse”.

Reply #3 is epistemically void—we can and should dismiss it as a basically contentless utterance for the purpose of the object-level discussion (but perhaps it is useful for its author to publicly stake out the position in question, and it provides a data point on the question of “what and how do people think about this topic”). But on the meta level it is at least honest, neither making any unjustified claims nor subverting the discussion itself.

Reply #4 is like reply #2, but with the added claim (accusation) of antisemitism. Is that claim “in the spirit of rationalist discourse”? Eh. It would be counter to that spirit to ban the hypothesis, sure, but what is it actually adding? We seem (in this hypothetical scenario) to be doing fine discussing the object-level issue, so what’s the point of bringing accusations of racial animus into it? Then again, perhaps the question is important for other (social) reasons, so, sure, why not allow it. (But, as with any accusation of some serious personal failing, it ought not be made lightly, and must be well-supported—otherwise, it is the accuser who has transgressed.)

But what the hell is reply #1?

It doesn’t address the empirical claim at all. If we’re talking about “the spirit of rationalist discourse”—well, that spirit has been well and truly violated already, in this case. What’s more, the accusation of antisemitism can’t be supported by the evidence of the accused having made the false empirical claim… because the accuser has not even troubled himself to establish that the empirical claim is false! This sort of reply is strictly an attempt to shut down the conversation.

If it’s “the spirit of rationalist discourse” that concerns you, then something like #1 should be the very last sort of reply that you ever find yourself making.