For now, here is an unsatisfactory response that will be very rambly and probably off topic.
(For what it’s worth, I found it quite helpful to see these motivations laid out like this, and am glad that you Logan wrote this comment and that you Raemon asked the question that provoked it.)
but the parent comment felt like it was too focused in on math
er, sorry, too focused in on math for it to help me with the thing i’m trying to figure out, in a way i was quickly able to recognize, i meant. i didn’t mean to assert that it was just too focused in on math for a comment, in some generic purpose-independent way! 😛
Where does ‘interaction’ fit in all of this anyway?
Logan:
it somehow fits into the heart of deep mastery [from “Knowing”]
Ooh huh hmmm!
I had missed this before, but… I think achieving deep mastery is actually not the goal of {the part of my work I consider most important}. Or, to be more precise, it’s not the job of this part of my work to produce deep mastery. I think.
(The Knowing article describes deep mastery as “extensive familiarity, lots of factual knowledge, rich predictive and explanatory models, and also practical mastery in a wide variety of situations”.)
The job of this part of my work is to make contact at all, and to nurture this contact just enough that it becomes possible to deepen that contact with more ordinary methods, like actual mathematical models. (Which are also an important part of my work, but do not as much seem like the bottleneck.) This part of my work isn’t supposed to produce extensive familiarity, lots of factual knowledge, etc.
Metaphorically, it’s like an expedition that travels deep into a jungle trying to find a viable route for a road, or something. They never see the actual road–once they’ve just marked off where the road may one day go, they move on to the next project. Their work is actually quite different from that of the people coming in after, who cut the trees and build the bridges and pave the road. Those latter people always have the road behind them which connects them to civilization, so they can truck in supplies and it’s basically a normal construction job, if one at the frontier. The expedition people are on their own, and can’t carry enough food to last the whole expedition, so they need to live off the jungle.
That… probably explains some of my confusion.
the conversation with robin you quoted did feel relevant, but the parent comment felt like it was too focused in on math and thereby somewhat orthogonal to or missing the point of what i was trying to figure out. (the real thing i’m interested in isn’t even about math but about philosophical intuitions.)
this made me want to try to say the thing differently, this time using the concept of gears-level models:
https://www.lesswrong.com/posts/B7P97C27rvHPz3s9B/gears-in-understanding
(maybe everything i’m saying below is obvious already, but then again, maybe it’ll help.)
suppose that you are looking at a drawing of a simple mechanism, like the second image in the article above, which i’ll try to reproduce here:
if the drawing is detailed enough and you are able to understand the mechanism well enough, then you can reason out how the mechanism will behave in different circumstances; in the example, you can figure out that if the gear on the top-left turns counter-clockwise, the gear on the bottom-right will turn counter-clockwise as well.
but you haven’t interacted with the mechanism at all! everything you’ve done has happened inside your map!
nevertheless, if you understand how gears work and you look at the drawing and think in detail about how each gear will turn, your model resists the idea that the bottom-right gear can turn clockwise while the top-left one turns counter-clockwise. it might be that your model of how gears work is wrong, or it might be that the drawing doesn’t accurately represent how the mechanism works, or you might be misunderstanding it, or you might be making a mistake while thinking about it. but if none of these is true, and the top-left gear turns counter-clockwise, then the bottom-right gear has to turn counter-clockwise as well.
when you work out from the drawing how the bottom-right gear has to turn, are you in contact, not just with any part of the territory, but with the part of the territory that is the actual physical gears? even though you are not physically interacting with the actual gears at all, just thinking about a map of the gears?
the way i’m thinking about it in the top-level comment is that since this process is able to resist misconceptions you may have, and is thereby able to bring your {anticipations / implicit models} about the physical gears more in line with reality, therefore yes it is “contact with that part of the territory” in the sense that is relevant to “knowledge of the territory takes patient and direct contact with the territory.”
“a map that clings super tightly to the territory” is a phrase from duncan’s reply to my comment on “The Territory” which seems to me to describe gears-level models well.
– # –
i should note that the thing i’m ultimately interested in, namely the way i use philosophical intuitions in my work on agi alignment, isn’t even anywhere as detailed as a gears-level model. nevertheless, i still think that these intuitions cling sufficiently tightly to the territory that this work is well worth doing. in the ontology of my top-level comment, my work is betting on these intuitions being good enough to be able to resist and correct my implicit models of agi alignment, and to therefore constitute significant contact with this region of the territory.
something i don’t know how to reflect well in a comment like this, and think i should say explicitly, is that the game i’m playing here is not just to find a version of logan’s sentence that covers the kind of work i do. it is to find a version that does that and additionally does not lose what i thought i understood when i was taking “contact with the territory” to be the opposite of “it all happens in your map”, and therefore would have taken {thinking about a drawing of the mechanism} as not being in contact with the territory, since it consists entirely of thinking about a map.
for some reason i haven’t really figured out yet, it seemed really important for this to say that in order to be “contact with the territory”, an experience has to be able to resist and correct my {anticipations / implicit models}, not just my explicit verbal models.
(i tried to say some more things about this here, but apparently it needs to gestate more first. it wouldn’t be very surprising if i ended up later disagreeing with my current model, since it’s not even particularly clear to me yet.)
Given our discussion on the “territory” essay about how the “in contact with the territory vs. all in the map” distinction has been confusing me, I’ve been trying to find a way to think about the “observing vs. merely seeing” distinction without identifying it with the other one.
{My first attempt to phrase it that seems to be actually at all helping with my confusion} is this: “Observation (in the relevant sense) is bringing my {anticipations / implicit models} in contact with something that might {contradict / resist / collide with / set / correct / change} them in a way that would make them better reflect the territory.” (Where by “might” I mean something like “has a reasonably good chance to”.)
Thus under this attempted definition, doing math about, say, star formation could be “observation” in the relevant sense (about the stars, not just about math) even though it doesn’t involve directly getting data from the stars, because it can collide with a person’s implicit models of how star formation works in a way that would tend to cause them to reflect reality better. And, of course, directly recording data from a telescope and using it to test hypotheses would be observation.
On the other hand, repeatedly walking up the stairs without paying attention would not be (much) observation of the stairs, because it would be very unlikely to change my anticipations about things like how many steps there are. And moreover, counting how many numbers there are on my bedroom clock would also not be much observation (of what the clock is like), even though it does involve getting data directly from the clock, because it would also be very unlikely to change my anticipations about the clock (because I am very confident that I know the answer).
I’m not sure whether “observation” is actually a good handle for the cluster I’m drawing here, but I think I probably do think that the cluster I’m drawing here helps me with cashing out the phrase “contact with the territory” in “knowing the territory takes patient and direct contact with the territory” in a way that isn’t based on the “in contact with the territory vs. all in the map” dichotomy.
I keep wanting to add something to this comment about how this attempted definition might apply to something like Googling or asking an expert, but I think I actually still have too much confusion there, and I guess the comment seems sufficiently worthwhile to me even without that.
[meta note: what i’m doing above is trying to find articulations of the intuitive picture that is emerging in my mind, which is hopefully in the vicinity of what you Logan meant to communicate but might not be]
This comment is about something I’m confused about, and I’m sufficiently confused about it that I can’t write it as a clearly-articulated question or statement. Its current state is more like a confused question that my brain in trying to untangle as I’m reading this sequence. So I’ll probably meander, and the meandering probably won’t come together into a clear satisfying thing by the end of this comment.
A big reason I’m interested in Logan-style naturalism is that you (Logan) frequently say things about it that resonate with ways in which I approach my own work. The most salient instance is your concept of “pre-conceptual intimacy”:
In pre-conceptual intimacy, they’re making a lot of fascinating observations and surprisingly quick improvements to relevant parts of life. But they’re also feeling very confused and disoriented, because their pre-existing concepts around the problem just don’t seem to make sense anymore, and they don’t have new stories about what’s going on yet either. They tend to utter phrases like, “Is memory even a thing?”, “How could I ever have thought that?”, and “I really have no idea what’s going on with this, and it turns out I never have.”
https://www.lesswrong.com/posts/EKc4RfKhPRmnLtRXn/research-facilitation-invitation
There is a particular mode that I’m in when I do what I think of as my best work, and this paragraph reminds me of it. Although, hm, now that I actually have it in front of me, most of the details don’t quite match… but I think that’s mostly because most of the words in this paragraph are about pre-existing concepts, and when I’m in this mode, I mostly just don’t pay attention to any concepts that don’t currently fit.
–I think I maybe only feel particularly disoriented when I have pre-existing concepts that don’t fit, but still feel like they encapsulate something important that I don’t want to lose sight of?–
Anyway, when I try to correct for these things, “pre-conceptual intimacy” seems to resonate a lot with this mode that I sometimes work in.
However, an aspect of the description above that still doesn’t match is that, when I’m working in this mode, I’m not in any very obvious way making any observations. It seems like most of what I’m doing is that I’m having a vague confused intuition, and I, uh. I bump them around in my head until they maybe turn into concepts that make sense, or are at least a little more like that? (Turns out that I don’t actually particularly know how to describe the thing.)
It wouldn’t be incorrect, I think, to say that I’m looking at two parts of my map that are inconsistent with each other, and I’m trying to make them consistent, or I’m looking at one part of my map (my confused intuition) and I try to use it to fill in a different, blank part of my map.
There are ways to increase some kinds of knowledge that largely involve staring at maps. Perhaps your own map is not clearly labeled in places, or it’s somehow inconsistent with itself, or it doesn’t match the map of an expert.
[...]
But the main thing a cartographer ought to be focused on, the vast majority of the time, is the world itself.
This seems obviously right for literal cartography. I want to talk, as part of my meandering, about how it might apply to math. I don’t actually feel confused about math, I just find it a helpful example.
It seems to me that, on the one hand, (most) math research could be described as being all about staring at some parts of your map and trying to use them to fill in other parts of your map. But–
Now I mean something like, “The thing that is made of something other than my own perceptions and interpretations. The thing that resists my expectations, according to its own rules. The thing that does not care what I think, or what I have happened to imagine.”
–but on the other hand, math certainly resists my expectations according to its own rules. Moreover, when I try to do prove a math result, I am in contact with that resistance: I get feedback from the territory (of math), even though it seems like in some sense it wouldn’t be incorrect to say that all I am doing is to stare at different parts of my map.
I think this is somehow an important node in my confusion (though, again, I don’t actually feel confused about math): When reading your posts, I seem to have formed a, uh, story? frame?, that says that {getting feedback from the territory} is important and that it is sort of the opposite of {merely staring at your map}. So if I think of doing math as both “getting feedback from the territory” and “nothing but staring at your map”, that breaks that model.
Maybe this just means that I should not think of math proofs as happening all in the map; maybe I should say that because doing math proofs gives me feedback about the way that math resists my expectations, therefore by definition it is not happening all in my map.
…I feel uncomfortable and kind of dirty about the previous paragraph, and that’s after working on it for a while trying to make it less bad. As written, it seems to be saying: I have formed a picture of a constellation in my head, and now I’m looking at the sky and there is no star where my mental picture says there should be one; and I now want to know whether the real constellation instead has this nearby star or that one. Sometimes it genuinely makes sense to ask “does this way of thinking about things feel more revealing, or that one?” But in this case, it just feels like a wrong question. The real thing inside me I would like to convey is that I’m mentally lightly touching on each of the two pictures I could draw, getting in touch with how both feel somewhat right but fairly wrong, because touching on what the world looks like from these two wrong perspectives jiggers something around in my head that makes me feel that I’m a little closer to resolving my confusion.
Maybe this just means that I should not think of math proofs as happening all in the map; maybe I should say that because doing math proofs gives me feedback about the way that math resists my expectations, therefore by definition it is not happening all in my map.
But suppose I’m a physicist. I spend some of my time doing experiments, and I spend some of my time thinking about physics and about my experiments, and as I do the latter, I frequently do math. I’m not interested in studying math, the thing I want to study is physics, but math is an important tool for doing so. And when I use this tool, it resists my expectations just as it does when I do math for its own sake; once I have a formal model of some physical phenomenon, math tells me something about what to expect from my physical observations in a way that does not care what I think, or what I happened to imagine. But it feels like in the case of physics, something important is captured by saying that my experiments are making direct contact with the territory of physics, whereas the math I do is all in my map.
Worse, consider Einstein when he said that if Eddington’s attempt to verify General Relativity had failed, “Then I would have felt sorry for the dear Lord. The theory is correct.” [1] At this point, he had obviously done a lot of math about GR, but the math couldn’t have given him that confidence that the theory was correct; given how little empirical observation [2] it was based on, it must have been {philosophical arguments slash his sense of how physics worked} that allowed him to come to this conclusion. And for him to predict so well on so little data, his philosophical intuitions must somehow have had the power to resist expectations according to their own rules, not in the way physics experiments do, but kind of close to the way mathematical calculations do. But if we tried to say that Einstein’s philosophical intuitions didn’t happen in the map, then… would any sort of thinking that actually does anything useful count as “happening in the map”?
[1] Each source I’ve checked seems to give a slightly different quote and story, which, uuuh, but anyway.
[2] (Empirical observation distinguishing it from Newtonian mechanics, I mean.)
I think that I personally have a tendency to spend a lot of time thinking about my map, and that I could, in many domains I care about, benefit from noticing a bunch of low-hanging fruit in making more direct observations. But I don’t actually think that {what I consider to be my highest-quality work} to be an example of this. I mean, that work is certainly informed by intuitions I’ve formed in contact with present day machine-learning systems, or by doing math, or by watching my own thinking. But it’s not, I think, made of contact with these things, and my contact with the territory (AGI alignment) seems to me about as tenuous as Einstein’s with his. I think that the best work I do is in fact made up of thinking about what some existing parts of my map can tell me about what should be in other parts of my map.
So what am I to make of naturalism, or “Knowing the territory takes patient and direct observation”?
One story I could try to tell is that naturalism won’t have much to tell me about the part of my work I consider most important. I could either say that “Knowing the territory takes patient and direct observation” is false because sometimes you can come to deep understanding of the territory just by thinking about it; or I could perhaps say that it’s true, and that just thinking is useful but isn’t enough to give you deep mastery of the subject; or I guess I could say that my approach to my work is just fundamentally doomed.
This story may be true. But it doesn’t currently ring true, because it doesn’t explain why numerous things you’ve said about naturalism have had such a strong resonance with my model of my work process.
When I do my thing that is in fact mostly “just thinking about it”, I have a distinction in my head that I track with my felt sense, which feels as if it’s tracking “whether I’m interacting with the real thing, or merely making up stories about it”. In the first case, I feel like I am in contact with something that can resist my expectations (although in truth this thing is itself made of anticipations).
I very much have habits for this kind of work that rhyme with patience: I approach it with a frame of mind that “isn’t expecting to find answers today”, that is looking at the feeling of the thought on the tip of my tongue and pokes around in that vicinity but isn’t expecting to be able to articulate that thought by any particular point in time. I look at the problem from this perspective and from that perspective, and feel successful if this shifts something in my felt sense that makes me feel a little bit less confused.
And there is an experience and sensation there that, as a felt sense, very much resonates with the idea of peeling back interpretative layers and increasing sensation at the point of contact, and with the metaphor of being “naked” in contact with the thing.
Maybe all of this is me missing you and interpreting your words in terms of something I’m familiar with. But what I actually think is going on is that your words are painting a picture of a constellation in my head that sort-of-but-not-quite matches the stars I see in the night sky; and that there is some nearby picture that does make sense of what I’m seeing; but I, like, just really don’t know what that picture looks like, yet. (And so I look at it from this wrong perspective and from that one, and notice how that shifts my felt sense of it a little and makes me feel a little less confused–which is why I had to write a long meandering semi-essay in order to be able to say anything detailed about it at all.)
So I feel like the internet has made people think there are no good people to look up to, and this makes it harder to trust new people.
This strongly clicked for me. It feels like there is more to say around this (and I don’t know what / don’t know how to say it yet), but this feels like part of the puzzle.
[Added:] Actually, perhaps it seems even more central to me that it feels like the same thing that has made people think there are no good people to look up to also has made have a decreased sense of looking up to institutional cultures. Like, my inner simulator imagines that people joining the NYT look up less to the existing institutional culture than in previous generations, in ways that are bound up with looking up less to the existing staff.
jsalvatier’s answer also clicks and feels relevant.
I have an anticipation whereby if you want to be part of the popular discourse yet not simply ‘play your role’, you have to walk through fields of people saying awful things about you. [...] I feel like Musk does it constantly, and I think that Musk not letting this get to him this is part of what allows his basic successes with Tesla and SpaceX to be part of the discussion.
By contrast, this feels to me like a different question: I don’t think the stable, cooperative institutions of old were all that good at “not simply playing your role”. It would be great to have a new kind of institution that is good at both of these, and it seems conceivable that this is part of the puzzle about how to build stable cooperative institutions at all in our times, but my guess is that it’s not a big part of the answer to where these institutions used to come from.
FWIW, https://www.worldometers.info/coronavirus/ counts all confirmed cases and has a table by country, which lists the Diamond Princess separately (“international conveyance”). It doesn’t distinguish asymptomatic from mild, but does separate out “serious, critical” cases, which stand at 36/705 (plus 7/705 deaths and 100/705 recovered).
Not quite the same, but similar mood: since there is no evidence, we thought we’d check if we could reinterpret the evidence to not be evidence.
The reference for “no evidence” was a CDC page that stated that “[t]here is no evidence showing that people can get infected by breathing in the virus” without discussing any evidence (the archive date on this link is within a week from when the article says they accessed the page):
http://web.archive.org/web/20170902234244/https://www.cdc.gov/norovirus/hcp/clinical-overview.html
The original article:
Xiao, Shenglan, Julian W. Tang, and Yuguo Li. “Airborne or fomite transmission for norovirus? A case study revisited.” International journal of environmental research and public health 14.12 (2017): 1571.
https://www.mdpi.com/1660-4601/14/12/1571/htm