FWIW I find the book by Smith and Morowitz to be staggeringly illuminating.

However, whenever the problem that you’re trying to solve resides in reality, the epistemic process of “trying to create a map of someone else’s map of reality” is inherently misguided and likely to lead you astray. One of the most important aspects of this, in my experience, is that you cannot interact with someone else’s map of the territory in the same way you can interact with—say, run experiments on—territory.

Recently I’ve begun to ask Duncan, who used to teach taekwondo, to teach me a martial arts technique here or there. I did this tonight, and he taught me a certain way to respond when someone tries to punch your face.

I noticed that I was shutting down right away, a result (I’d guess) of complicated experiences with martial arts in my childhood. There were hints of it as I watched the characteristic karate-like way his body moved, but it was undeniable when he asked me to mimic his movements.

As soon as I recognized “something like frustration”, I asked Duncan to pause. I took several seconds to close my eyes, breathe, and notice what was going on in my head.

In my experience, mentorship relationships are incredibly intense in the martial arts, especially in one-on-one interactions. The other person is standing right in front of you just a couple feet away, looking you in the eye, moving toward you, touching you. Your thoughts in response are not hidden in the privacy of your skull. You don’t get to mull them over until you find just the right words, as you might with a chemistry tutor; instead, what you are thinking is written directly on your posture and movement quality, and your teacher is an expert at reading your movements.

As I stood there with my eyes closed, taking calming breaths, I noticed that the cluster of paralyzing and frustrated feelings seemed to exist in a triangle of me, Duncan, and his perceptions of my movements. At first I didn’t quite understand and didn’t know what to do about it. I just kept breathing until I felt a little better, and then asked him to continue.

I paused us many times in the few minutes we spent on this. Over the course of lesson, I seemed to figure out something about how to thaw this particular kind of being frozen.

The way out is to reach for my true intention. “Why,” I asked myself during one of the pauses, “do I want to learn this technique?” I can’t easily fit the answer into words, but at the heart is a feeling of desire for knowledge, power, and flexibility. I want to know how the technique works, I want to physically embody the knowledge, and I want to feel the fluidity and expansion of learning something new. I am curious about the technique.

There’s a nearby attractor, a pattern of thinking and attending toward which my mind automatically sinks when someone stands in front of me and tries to teach me how to block a punch. I find myself trying to do what they want me to do. Trying to move in the ways that will please them—or, ideally, in ways that will impress them. My thoughts are non-verbal and mostly pre-conscious, but if they weren’t, they might go, “Is this what he means? Does he think this is better? Like this?”

This is what it sounds like to try to map someone else’s map.

Reminding myself of my true intention was like a bridge over a deep chasm. Once I had found a glimmer of my curiosity, I could move from there into thoughts that sound quite different. I thought things like, “Would I turn more easily if my left foot stepped out wider? Maybe I will have more room to move my center of gravity forward if I keep my hands closer to my chest. Should I prepare for the step by shifting my weight to the other foot?”

This is what it sounds like to map the territory.

I sometimes had thoughts that involved Duncan—”What is he doing differently as he sweeps his arms up?”—but they were not about Duncan. They were about the technique, and how I could master it.

Thank you for writing this, Nora. I’ve been unable to interact with anything to do with martial arts for well over a decade, and have struggled with all kinds of similar situations, because of this paralysis that always comes over me. I’m so happy to have found a way to stay grounded and engaged, and I’m sure the hypothesis occurred to me because of this post.

If you’re looking for a solid take from the Israeli side regarding the current conflict, I would start here:

https://​​www.jewishagency.org/​​special-briefing-05-2021/​​

It was cut short because the air-raid sirens started going about 22 minutes in. As far as I know everyone on the video is safe at the moment.

I was in Israel in February of 2020, just slightly before the world changed, and had an opportunity to meet Michael Oren (I suppose you could say I had some bias because my wife is related to him in a roundabout way… which I suppose means I’m related to him in a… circuitous way?). I don’t agree with him on everything, but I can speak to quality of his character.

Since traveling there, my take on all things Israeli-Palestinian conflict—if the person talking about it hasn’t spent time in Israeli and Palestine, seen the area, and talked to people there… I hate to sound close-minded, but I’m likely to discount what they say, especially since taking one side or the other in the US is sometimes more about virtue signaling than genuine concern.

Trust me, I had plenty of opinions on the matter and was well aware of the counter points to them that are common in the US. Every idea I had fell flat when I encountered the on-the-ground complexity.

The larger problem—even for the people who are genuinely concerned—the conflict there is not sufficiently analogous to other historical situations (or current and better understood political conflicts) that I’m aware of. If you’re in North America and have never been there, whatever model you have of it in your mind almost certainly has significant gaps.

I left with some vague ideas of different ways of looking at it that might help, but was more or less thinking “wow… I’m not a stupid person, but this situation is really complicated and I have no idea what the next best move is. I’m scared to even hazard a guess because I’d hate for anyone to misconstrue it as advice and share some burden for things potentially going sideways.”

Episode 47: Heidegger

So last time we were trying to understand Heidegger’s work as a prophet (in the Old Testament sense) of the meaning crisis. We took a look at this notion of “the thing beyond itself” and “realness” as simultaneously the shining into our framing and the withdrawing beyond our framing in a deeply interpenetrating manner. We took a look at this deeper notion of truth—not truth as correctness, but truth as aletheia, that which grounds the agent-arena relationship in attunement and allows us the potential to remember being by getting into an attunement with its simultaneous disclosure and withdrawal.

But we can forget that; we can get into a profound kind of modal confusion and this is the history of metaphysics as the emergence of nihilism. We can forget the being mode, we can get trapped into the having mode in which the metaphysics is a propositional project of trying to just use truth as correctness, and we misunderstand being as a particular being. We try to capture the unlimitedness aspect of being, but we only do it at the limit (which Heidegger is deeply critical of). So we understand being in terms of a Supreme Being, a being at the limit, and beyond the limit. This is ontotheology; we understand God as the Supreme Being and this is deeply enmeshed (for Heidegger) with nihilism, because this ontotheology, this version of theology from classical traditional theism, this way of understanding being gets us into the deep forgetfulness and modal confusion that is the hallmark of nihilism.

Of course, we could perhaps remember the being mode, and this is what Corbin (following Heidegger) talks about as gnosis.

Episode 46: Conclusion and the Prophets of the Meaning Crisis

Last time I finished the discussion of wisdom and connected it to enlightenment and argued for the wise cultivation of enlightenment as our deepest kind of existential response to the meaning crisis, a way in which we can awaken from the meaning crisis. I then wanted to put that scientific model of spirituality (for lack of a better phrase) into discourse with some of the central prophets of the meaning crisis. I’m using the word prophet as it’s used in the Old Testament; I’m talking about individuals who were crucial for articulating the advent and helping to propose or promise a response to the meaning crisis. I put a diagram on the board in which Heidegger played a central role; there’s many connections in there that I’ll point out that I will not be able to fully address, because the people are there insofar as they help us articulate the response, not to be examined for their own sake.

I mentioned the work of Nishida and Nishitani in the Kyoto School; I will talk briefly about Nishitani here but I won’t be able to go into that in depth. I do intend to pursue this later in another series I’m putting together (I’m putting together a couple of series to follow this one) and I would like to do a series that will include work on the Kyoto School that I’ve entitled The God Beyond God, in which we look at all of these great non-theistic thinkers within both Eastern and Western traditions, and things like the Kyoto School that tried to bridge between them. So I will have to neglect (to some degree) the Kyoto School in this series but I promise to follow it up more deeply in another series.

The first 45 lectures have been, to some extent, “how did we get here, and where is here anyway?”, and these remaining five lectures are something like “what do other people think about being here?” This episode mostly touches on Husserl (who doesn’t really make it into the summary at the beginning of the next episode).

One operationalization of “know” in this case is being able to accurately predict every move of the Go AI. This is a useful framing, because instead of a hard pass/​fail criterion, we can focus on improving our calibration.

Now the success criterion might be:

• You have to be able to attain a Brier score of 0 in predicting the moves of the best go bot that you have access to.

What’s missing are some necessary constraints.

Most likely, you want to prohibit the following strategies:

1. Running a second instance of the Go AI on the same position, and using as your prediction the move that instance #2 makes.

2. Manually tracing through the source code to determine what the output would be if it was run.

3. Memorizing the source code and tracing through it in your head.

4. Constraining the input moves to ones where every Go program would make the same move, then using the output of one of a different Go program as your input.

   1. Corollary: you can’t use any automation whatsoever to determine what move to make. Any automated system that can allow you to make accurate predictions is effectively a Go program.

Overall, then you might just want to prohibit the use of Turing machines. However, my understanding is that this results in a ban on algorithms. I don’t have enough CS to say what’s left to us if we’re denied algorithms.

Here’s a second operationalization of “know.” You’re allowed to train up using all the computerized help you want. But then, to prove your ability, you have to perfectly predict the output of the Go program on a set of randomly generated board positions, using only the power of your own brain. A softer criterion is to organize a competition, where participants are ranked by Brier score on this challenge.

However, this version of the success criterion is just a harder version of being an inhumanly good Go player. Not only do you have to play as well as the best Go program, you have to match its play. It’s the difference between being a basketball player with stats as good as Michael Jordan’s, and literally being able to copy his every move in novel situations indefinitely.

Neither of these versions of the success criterion operationalization seems particularly interesting. Both are too restrictive to be relevant to AI safety.

Did you have a different operationalization in mind?

This is an interesting direction to explore but as is I don’t have any idea what you mean by understand the go bot and I fear figuring that out would itself require answering more than you want to ask.

For instance, what if I just memorize the source code. I can slowly apply each step on paper and as the adversarial training process has no training data or human expert input if I know the rules of go I can, Chinese room style, fully replicate the best go bot using my knowledge given enough time.

But if that doesn’t count and you don’t just mean be better than them at go then you must have in mind that I’d somehow have the same ‘insights’ as the program. But now to state the challenge we need a precise (mathematical) definition that specifies the insights contained in a trained ML model which means we’ve already solved the problem.

An even more pointed example: chess endgame tables. What does it mean to ‘fully understand’ it beyond understanding the algorithms which construct them, and is it a reasonable goal to attempt to play chess endgames as well as the tables?

The Art of War is written in Classical Chinese. Google Translate uses modern Chinese. Google Translate will spit out garbage if you attempt to translate directly from the original text. You have to translate it twice: once from Classical Chinese into modern Chinese and then from modern Chinese into English.

You can get a translation from Classical Chinese into modern Chinese by going here, selecting the chapter you want to read and then clicking the “原文/​译文” button in the upper right corner. The Classical Chinese will remain black. A modern Chinese translation will appear in gold. Copy the gold text into Google Translate to get a better English translation.

Example classical Chinese:

孙子曰：凡用兵之法，驰车千驷，革车千乘，带甲十万，千里馈粮，则内外之费，宾客之用，胶漆之材，车甲之奉，日费千金，然后十万之师举矣。

Example modern Chinese:

孙子说：用兵作战，一般要出动战车千乘，运输车千辆，集结士兵十万，沿途千里转运粮草。这样国内外的各种费用开支，招待使节、策士的用度，维修作战器械的胶、漆等材料费用，保养战车、甲胄的支出等，每天都要耗费数额巨大的资金。作好了这些准备后，十万大军才能出动啊！

Google Translate from modern Chinese into English:

Sun Tzu said: In combat with soldiers, you usually have to dispatch thousands of chariots, thousands of transport vehicles, and gather 100,000 soldiers to transport grain and grass thousands of miles along the way. In this way, all kinds of domestic and foreign expenses, the cost of entertaining envoys and strategists, the cost of repairing materials such as glue and paint for combat equipment, and the maintenance of tanks and armor, all cost a huge amount of money every day. After these preparations are made, the 100,000 army can be dispatched!

To expand on this a bit, I think that people with working partners would be the group most likely to be deterred from working at MIRI if it was in either Bellingham or Peekskill. The two-body problem can be a serious constraint, and large metro areas tend to be much easier to find two jobs in. That may be getting better with the rise of remote work, but I do think it’s worth keeping in mind.

If one of your goals is to play as well as the best go bot and/​or write a program that plays equally well from scratch, it seems like it’s probably impossible. A lot of the go bot’s ‘knowledge’ could well be things like “here’s a linear combination of 20000 features of the board predictive of winning”. There’s no reason for the coefficients of that linear combination to be compressible in any way; it’s just a mathematical fact that these particular coefficients happen to be the best at predicting winning. If you accepted “here the model is taking a giant linear combination of features” as “understanding” it might be more doable.

Two things:

ONE: I love how “should I learn to drive for this trip right here?” cascades into this vast set of questions about possible future history, and AGI, and so on <3

Another great place for linking “right now practical” questions with “long term civilizational” questions is retirement. If you have no cached thoughts on retirement, you might profitably apply the same techniques used for car stuff to “being rich if or when the singularity happens” and see if either thought changes the other?

TWO: I used to think “I want to live this year”, “If I want to live in year Y then I will also want to live in year Y+1″. Then by induction: “I will want to live forever”.

However, then I noticed that this model wasn’t probabilistic, and was flinching from possible the deepest practical question in philosophy, which is “suicide”. Figuring out the causes and probabilities of people changing from “I do NOT want to kill myself in year Y” to “I DO want to kill myself in year Y+1” suggests a target for modeling? Which would end up probabilistic?

Occam (applied to modeling) says that the simplest possible model is univariate, so like maybe there is some value P which is the annual probability of “decaying into suicidalness that year”? I do mean decay here, sadly. Tragically, it looks to me like suicide goes up late in life… and also suicides might be hiding in “accidental car deaths” for insurance reasons? So maybe the right thing is not just a univariate model but a model where the probability goes up the older you get?

This approach, for me, put bounds on the value of my life (lowering the expected value of cryonics, for example) and caused me to be interested in authentic durable happiness, in general, in humans, and also a subject I invented for myself that I call “gerontopsychology” (then it turned out other people thought of the same coinage, but they aren’t focused on the generalizable causes of suicidal ideation among the elderly the way I am).

Ok three things...

THREE: I drive <3

If covalent vs. noncovalent bonds are something you’re not familiar with, it sounds like you’d benefit from reading the chapter(s) on chemical bonding (every gen chem textbook should have one). I’d also infer from that that you won’t have much of a background in thermodynamics, which rears its head when you try to understand the energy-storing and energy-releasing reactions of metabolism.

I want to learn biochemistry so I can reason about stuff that goes on in the body! I’ve started Lehninger’s Principles of Biochemistry and I mostly get it, but some of the stuff it assumes (e.g. covalent vs. noncovalent bonds) I’m not familiar with.

Boy, the historyofvaccines.org page on LYMErix is pretty brutal.
https://​​www.historyofvaccines.org/​​index.php/​​content/​​articles/​​history-lyme-disease-vaccine

Looks like a French company has another candidate, VLA15 that is heading (headed? as of this past March) into Phase II trials (with some help from Pfizer) for ages 5 and up.

https://​​valneva.com/​​press-release/​​valneva-and-pfizer-announce-initiation-of-phase-2-study-for-lyme-disease-vaccine-candidate/​​

Does anyone know how to find out where these trials happen? Would love to sign up if it’s not too late.

This chapter felt more allusive than the earlier ones. It kind of reminds me of sword manuals where the writer assumes the reader already has had F2F training and is “pointing” instead of “explaining”.

I looked up “the nine heavens” and it didn’t help much. Maybe the phrase changed meaning over time? It seems to have been connected to the planets after Ptolemy’s model reached China, but I don’t know of Sunzi wrote before or after that. If after, I wondered if maybe were some sort of planets/​pantheon/​astrology linkages, that would mean something to someone who caught more references than I do?

I see that The Art Of War has 13 chapters. If you had done them all, I’d have probably binged the whole sequence this evening :-)

Thx for the comment.

There’s a scientific-consensus answer to how many calories people typically need, the Harris-Benedict formula.

Out of interest, I checked wikipedia, which gives three different equations (Harris-Benedict, Mifflin St Jeor, and Katch-McArdle), saying “Historically, the most notable formula was the Harris–Benedict equation”.

I just put in my details on them, and my daily caloric burn varied over 1000 calories (one was 500 less than the other which was 500 less than the other). One of them was pretty close to 2000. /​shrug

Ok. It’s falsifiable but requires expensive evidence you don’t have an easy way to collect. What do you propose we do? Like you said, the economics currently reward sites for being as addictive as possible. Moreover, better ai methods will allow generation of even more addictive sites. Or just optimization—compare tik tok to YouTube. Tiktok is a set of very short clips that have become super clickbait through careful winnowing. Every single clip has someone in the top 1 percent of attractiveness… maybe higher than that...or something immediately violent or dramatic or funny.

It would be like the kind of junk food you would get if you had realtime feedback to what people were eating and had millions of samples to pick from.

That’s what we need for LW: BibTeX support :P

I’m curious if I’m an outlier here—have you really never tried to relate some joke or funny story and then cracked up before you could finish it? I can’t tickle myself, but I can easily make myself laugh.

Anyhow, I think this is to some extent a low-dimensional analogy for a high-dimensional world. When the world is complicated, trying to do something new can result in finding connections to something you already know about, and studying something familiar can uncover the surprising. This is for exactly the same reason that a 1D patch on a string is close to fewer neighbors than a 3D section of space, which in turn has fewer neighbors than a drug molecule has in the space of possible chemicals. In high dimensional problems, both connections and surprises are so common as to be unavoidable. But if we’re just walking around on the 2D surface of the earth, we’ll probably run into connections and surprises at about the rate we’d expect from our stories.

Reading this, I wonder how much of the quality is related to your own wisdom, and how much can be attributed to the original author?

The original:

国之贫于师者远输，远输则百姓贫；

What google translate makes of it:

If the country is poor than the teacher, those who lose far away will lose, and the people who lose far will be poor;

Your version

Resupplying an army over long distances impoverishes a country.

The last one seems to me like pragmatic direct clean common sense. If someone didn’t know it, I would think they were not right for military planning.

If every planned military campaign was to be checked against a checklist of common sense objections (that someone might have forgotten mid-planning based on groupthink or enthusiasm or something) then having that line in the checklist makes sense to me. Useful! Practical!

Google’s translation has no such virtues and I can’t read Chinese. Other lines have a similar pattern of “quality”?

Maybe… I’m reading (and appreciating!) “Lsusr’s Steelman Of Sunzi” rather than “the original itself”??

Maybe there’s a gap in my cross-cultural literary interpretation capacities (and/​or Google is still terrible at Chinese translation) and I’m expecting to be spoon fed, and any normally capable classical Chinese reader would have “basically gotten the same thing you did” from the original, but then perhaps admire specifically your skill at translating it to a new domain (English) with different literary standards (pop culture infused midwits like me)?

Another theory might be that in the original, all the lines ARE vague. Perhaps they are elliptically written to avoid political attacks in Sunzi’s own time? Or just honestly lacking the punch and wisdom? Still, all of them together might cause someone who reads Chinese to “read between the lines” and be able to “get” each line?

One test might be to round trip your translation back through Google, and back to Chinese, and see if it “sounds way more sane and direct and useful” in the very last stage still?

Some lines from the original:

故兵闻拙速，未睹巧之久也。夫兵久而国利者，未之有也。

国之贫于师者远输，远输则百姓贫；

故兵贵胜，不贵久。

Google’s version:

Therefore, the soldiers heard the clumsy speed, and they have not seen the cleverness for a long time. There is no such thing as a husband and a soldier who benefits the country for a long time.

If the country is poor than the teacher, those who lose far away will lose, and the people who lose far will be poor;

Therefore, a soldier is expensive to win, not for a long time.

Your take:

There is no such thing as a beneficial protracted war.

Resupplying an army over long distances impoverishes a country.

A valuable victory is a quick victory.

Then back through Google?

没有所谓的持久战。

长途补给军队会使一个国家贫穷。

宝贵的胜利是快速的胜利。

So, the final Chinese looks pretty different in terms of characters and density and so on, but I can’t read or judge it as language. Your English seems way better than Google’s English. Is there a similar jump in quality from Chinese at the beginning as compared to the Chinese at the end?