Review
I’m not sure to what degree this post can stand alone apart from the whole sequence. But in general I found the naturalism method a useful tool for understanding the world. (Is there reason that the Lesswrong review isn’t also done at the sequence level in addition to the post level)
Within the sequence, this post in particular stood out. Many people describe the importance of sitting through a period of being stuck without abandoning a project since often that is a stage towards clarity. This post pointed to a general actionable strategy for doing so which I haven't seen elsewhere. It went beyond that by showing some concrete expressions of that strategy and attitude. For example: a. reconnecting with the felt sense of the topic and direction. b. Returning to a more fluid approach if the investigation has become distorted by something like looking over ones shoulder at what others might think, or an over-commitment to the explicit structure of ones method (but checking afterward if ones method needs a reformulation or whether it was intuitively driving you and the problem was trying to legibly meet the standard)
One thing I didn’t like about it was the chatGPT dialogue, and think that might be worth skipping
This is the fifth post in a sequence that demonstrates a complete naturalist study, specifically a study of query hugging (sort of), as described in The Nuts and Bolts of Naturalism. This one continues my demo of phases one and two: Locating Fulcrum Experiences and Getting Your Eyes On. For context on this sequence, see the intro post. Reminder that this is meant as reference material.
Before throwing in the towel, I thought I might as well try talking things through with ChatGPT.
Why? I think it was just something I was doing a lot at the time. “Not sure how to proceed? Throw GPT4 at it for a while and see what happens.” I didn’t expect anything in particular from the activity.
As it turns out, I can probably count on one hand the number of work-related discussions that have been as useful to me as this one—which probably says as much about my social deficits as it does about GPT4's abilities. It talked me around, and I un-gave-up.
The most important part was using GPT as a Focusing partner. I did a (mostly) canonical version of Gendlin's Focusing, and came to understand that I was struggling with the balancing act between authenticity and legibility. I decided to almost completely abandon legibility.
The relevant section of chat log:
(Note: Often, ChatGPT suddenly turns into a mini encyclopedia entry. I fondly imagine an excited nerdy 12-year-old trying to get full marks on an essay at these moments. Despite considering them extraneous to me personally, since it’s giving me background on terms I have chosen myself, I’ve left these parts in as potentially valuable context for onlookers, but I've italicized them so you can skip them if you'd prefer.)
I decided to stick with “hug the query” after all. The main thing that seemed off about it was the stiltedness, which appeared to come from the pressure I felt to be legible. The change I made was to almost completely abandon my attempts to be legible, even to abandon my attempts to follow my own instructions, relying only on my intuitions in the moment. I also quit sharing log excerpts in a semi-private slack channel, since I suspected that the live blogging was similarly hampering my fluidity.
Farther From the Issue
I was able to mostly skip fulcrum location this time, because it turned out I’d been on the right track with “crucial” as a conceptual pointer. “Closeness to the issue” was more a reframing of “crucial”, another way of understanding the correlated experience “what everything turns on”, than a new topic. I’d therefore done most of the early work already.
Terminology reminder:
To get my eyes on in the new frame, I began with an exploratory exercise.
I sure did a lot of "lab work" in this study. According to my memories, I don't usually do this much of it; often I rely almost exclusively on field work. (See my study of Spaciousness In Partner Dance, for example, which involved no lab work at all.) I think the focus on lab work might be partially an artifact of the context of this study: when I’m trying to demonstrate things, I tend to be more deliberate and controlled than otherwise.
Regardless, this is certainly one way a naturalist study can go. So let's talk a little about what I was doing with all this lab work.
For the most part, I was "playing with toys", as described here. "A toy is a manipulable low-stakes situation that reliably yields desired experiences as you interact with it." The "desired experiences" were ones related to my concepts of "distraction", "concentration", "crucial", and "closeness to the issue". I manipulated the low-stakes situation to observe experiences related to those concepts.
Since I was using toys in the early stages of study rather than as part of "collection" or "experimentation", I wasn’t aimed at very specific experiences. Rather, I was casting a pretty broad net, trying stuff that might entangle my observations with regions of territory I was interested in so that I could find out what experiences those regions afford. During candle gazing, for instance, there was no particular collection of sensations I hoped to cause; rather, I had some vague intuition that staring at a candle for an hour would involve "concentration, whatever that is". This is what toys are like in the context of something between “locating fulcrum experiences” and "getting your eyes on". Lab work looks a little different in each phase of a naturalist study.
My design process for all of these toys was minimal. I never went through all five of the questions I lay out in "Collection"; in fact I was never reflectively aware of designing anything at all. From the inside, it was entirely improvisation.
Nevertheless, as I look back over the five toy design questions I suggest in Collection, I see that all of my lab work really did meet each criterion. Each scenario I set up was meant to draw out a particular experience, the stakes were always low, the situation was under my control, and I left a bunch of space for reflection through note-taking either during the exercise or immediately afterward.
If I’d designed these by following my own directions instead of improvising, I’d have accomplished that by asking myself,
For the sake of illustration, I’ll describe this next bit of lab work in terms of that structure, even though I wasn’t following it explicitly in the moment.
In my exploratory exercise, I wanted to cause experiences related to “closeness to the issue” (so I could get my eyes on about those experiences). I wasn’t quite sure what those experiences would be, but my earlier investigations suggested there might be a feeling that “things can turn on this”, and some kind of yearning in my chest[1].
I suspected experiences like this were to be found near evaluation, decision making, updating, and learning. One domain that involves several of those activities is design, and I happened to have a design project on my to-do list: Reworking the interior design of my cabin. There would be a lot of decisions to make, and I suspected that some kind of yearning in my chest is a major part of any design process for me; that’s the engine that makes design go. So I chose to play with the interior design of my cabin as a toy.
I think there were two key moments in this session of lab work. One was when, after ten or fifteen minutes of pretty systematic list-making, I felt that I might be getting “farther from the issue”, rather than closer. I stopped to adjust my approach, and then asked, "What caused me to pause and evaluate?"
I went in with a concept of "closeness to the issue"—a conceptual pointer, a dot on my map—but it was here that I began to make contact with the reality of the thing phenomenologically.
In this case, I was actually observing the experience of moving farther away from what mattered. I paused to evaluate because I could feel myself moving "farther from the issue", and then I asked myself how I could tell. What did I feel? What sensations tipped me off? The answer (as closely as I could observe it at the time) was, “bogged down”, “veering off”, and “sinking into a pattern that feels nice only in virtue of being a pattern”.
The other key moment in this session was when I deliberately attempted to identify "the issue" to "stay close to".
In the moments recorded by that excerpt, I was indeed adopting something I now recognize as the posture I would come to learn well, the one that might be called “query hugging” (although my own words for it are a little different).
I’d like to point out, though, that this is not much of a phenomenological snapshot. It’s progress toward the ability to take such a snapshot; but the excerpt is full of conceptualization and pattern recognition. There’s almost no fine-grained phenomenological detail. What was it like to make that move? What was my mind doing between each of those verbalizations?
I don’t know. As yet, the grain of my phenomenological access to this fulcrum experience was still quite coarse.
Close Embrace
Next, I repeated this exercise, but while learning real analysis instead of redesigning my living space.
(Reminder: The exercise was, “Try an activity related to evaluation, decision making, updating, or learning; as you go, watch for experiences related to “closeness to the issue”, and try to take phenomenological snapshots in those moments.”)
My real analysis note-taking practice began as such an exercise, but a somewhat different activity grew out of it. For one thing, I did not only take notes on experiences that smelled like “closeness to the issue” or “what everything turns on”; instead I took notes almost constantly, on my textbook and my engagement with its exercises, with occasional elaboration on observations that reminded me of the concepts and experiences I was after.
Secondly, I did not stop at a single session. I found this new activity so valuable that it became my default use of brain-time for an entire month.
From a naturalist perspective, the point of all this observation was to help my attention worm its way into details of my experience that were previously so subtle as to be invisible. But in addition to sensitizing myself to experiences around “closeness to the issue”, I think I learned a lot about how to learn math. These two skill sets seem highly complementary, to me.
In this excerpt, you might notice me shifting away from thinking in terms of an action to take, and toward thinking in terms of a mental posture to continuously hold. I regard this as one of the most important moments in my study. I began to understand that the foundational skill “Hug the Query” points to is not a discrete action, begun in one moment and ended in another, like the sort of hug you might give in greeting. It’s more like a disposition that can be cultivated, or a habitually held posture.
For this reason, I’ve come to think of it not as hugging, but as “dancing in close embrace”.
“Close embrace” is one of the ways two people can physically connect during partner dance. It is basically a hug, with the partners touching at thigh, hip, stomach, chest, shoulders, and arms; but entire dances are performed this way, without the dancers breaking apart. Most of Argentine tango is danced in a version of close embrace, for example, as is much of zouk.
If you’re not a partner dancer, then “close embrace” may not be a useful handle for you. Some other associations that come to mind: A rock-climber keeping their hips near the cliff face, the speed with which embers fluctuate in brightness according to the changing wind, sunflowers following the sun across the sky, a jazz musician keeping the groove across improvised key changes, a snowboarder constantly shifting their weight to balance.
POU Loops
Up to this point, I'd been feeling my way around the conceptual pointer "closeness to the issue", but I had not really gotten the concrete experiences in focus yet. I had found some relevant thought processes; I had noticed intuitions involving it; I had made use of the concept, and of glimmers of correlated experiences out of the corner of my eye, to navigate.
But I had not really keyed into the phenomenology beneath it. I had not precisely located a candidate for a true fulcrum experience, though I’d laid the groundwork for doing so.
Things changed the moment that I began to practice Predict, Observe, Update loops.
I studied math for about an hour, then wrote
I think that I chose the word “luster” because I happened to be learning the poem “Ulalume” at the time, which includes the lines,
At the end of our path a liquescent
And nebulous lustre was born
The feeling those lines cause for me, when I sink into them, is very similar to the feeling I tend to get when “moving toward the issue”.
I want to point out that “luster” or “lustrousness”, here, is a focusing handle. Focusing handles for bodily felt senses do not make sense to everybody.
Am I really talking about a physical sensation located in my chest, the way that the rhythmic pressure of my heartbeat is such a sensation? In part, yes; there really is a component that is, as far as I can tell, “coming from my chest”.
But I think there are more components than that. It’s a cluster of sensations, some of which seem more psychological, such as a shift in attentional direction. If I described only the most physiological component sensations, I think I’d say something like “sharpness” or “ intensity”, rather than “luster”. The chest sensation is like an indicator for the larger experience; it increases when the rest increases, decreases when the rest decreases.
There are a lot of ways to describe the overall experience. “Luster” just happened to be what fit for me at the time.
One of the barriers to fulcrum location is insisting that your handles make sense, or be justifiable or externally legible, that they be the “correct” descriptive words in some sort of objective sense. In the context of naturalism, none of that matters at all. Words might even get in your way; if you’re especially visual, for example, an image might be a more appropriate handle for organizing your observations. In another month I would most likely have chosen a word other than "luster", and it would have been just as accurate and worked just as well, because in that month I myself would have been different.
The pace of my study picked up dramatically after this point. I think this small handful of iterations of the POU loop, combined with my earlier insight about "constant attunement to the relationships among the ideas", unlocked a new mental posture for me. It unlocked “dancing in close embrace with the world”.
My experience of studying math changed after that also. It was more fluid, less choppy, more satisfying. It was also somehow simpler, as though there was only one thing I was doing the whole time, a steady march in one direction.
Shortly after this, I was finally ready to move to the next stage of my study: collection. It was time to zoom out again, to start collecting experiences of chest luster, observing the patterns over time.
By the way: It's as yet unclear to me how critical bodily felt senses are to this method. If you are not currently aware of having them, it might be possible to learn to be aware. I think a really important question is, "What exactly are bodily felt senses doing for the people who can use them like this, and do you have other mechanisms for achieving the same ends?"
My favorite resource on felt senses, bodily or otherwise, is "The Felt Sense: What, Why, and How" by Kaj Sotala.