Direct Observation
post by LoganStrohl (BrienneYudkowsky) · 2022-02-21T20:51:45.060Z · LW · GW · 7 commentsContents
7 comments
My dad’s a retired science teacher, so my little brother and I began playing the “I’m not touching you!” game unusually early.
In case you’re not familiar with “I’m not touching you!”, here’s how it works.
Step 1: Ask your parents (or the internet or whatever) why you don’t fall through the floor.
Step 2: Learn about electrostatic repulsion, and conclude that things never “really touch”.
Step 3: Poke your sibling.
Step 4: When they say “Stop touching me!”, insist that “I’m not touching you!”
Step 5: Find out whether your parents consider “technically correct” to be an adequate defense. (They don’t.)
It is true that things do not touch in the way we naively imagine. Their component atoms’ electrons repel each other. “Perfectly direct contact” does not ordinarily happen.
Nonetheless, there is a big difference between the “not touching” that’s happening when you poke your sibling, and the “not touching” that’s happening when you point at your sibling from three meters away.
Similarly, there is no unmediated subjective experience of the territory. When you poke your little brother’s arm, your brain does something with those nerve signals to turn them into phenomena, into the subjective experience of tactile sensations. The sensations are not the nerve signals. There is always processing distance between your subjective experiences and whatever is happening to your sense organs.
However, some processing distances are smaller than others.
Consider the processing distance between [the electrostatic repulsion at your fingertip] and [your plan to test how upset you can make your brother before your parents intervene]. Although your plan exists mostly inside your head, it definitely has something to do with that electrostatic repulsion, in about the same way that a pointing finger has something to do with whatever it’s pointing at. A plan is made of anticipations, imaginings, desires, and so forth; yet it isn’t completely out of contact with the territory.
But consider the distance between [the electrostatic repulsion at your fingertip] and [the immediate tactile sensations of your finger]. The subjective sensations still exist only inside your head, but the processing distance is much smaller. Those sensations put you in closer contact with the underlying reality of electrostatic repulsion. Compared to imaginings or desires, they are much more tightly entangled with what’s happening out there.
Here’s how I think contact with the territory works.
For me to make contact with my brother’s arm, there are three conditions that must be met, three entities that must coincide.
- Presence. An arm must be present. There must be a region of territory available to be contacted. I can’t make much contact with my brother’s arm if he’s in another city.
- Personhood. There must be a contactor, a mind participating in the experience of contact, a place where attending occurs. I can’t make contact in the relevant sense if I’m under general anesthesia, even if my brother’s arm happens to brush my finger while I’m under.
- Sensation. There must be an experience of pressure on the skin of my hand as it presses up against my brother’s arm. There has to be some kind of sensation I might become aware of. I can’t make contact if I’ve lost all the nerves in my fingers to frostbite.
Presence, personhood, and sensation. Lose any of the three of these, and there’s no longer contact in the sense I mean.
But note that each of these conditions can obtain to a greater or lesser extent, depending on the situation. It isn’t all or nothing.
In the context of studying arms, a memory of an arm involves greater presence than does the description of a fictional arm in a novel. A photograph has more presence than a memory.
Presence of arms: actual arm > photograph of an arm > memory of an arm > fictional description of an arm
Personhood is similarly variable. A person’s body can be physically present while their mind is otherwise occupied. They can selectively disengage parts of themselves when encountering a topic they’d prefer not to think about, letting experiences “bounce off” or relegating them to a “sandbox” for safety. It’s possible to limit the mental space in which attending occurs, without eliminating it entirely—or to expand that space beyond what’s available by default. This is what phrases like “showing up”, “checking out”, and “blankface” refer to.
Finally, just as a sighted person can wear sunglasses or even a blindfold, sensation isn’t all or nothing. It’s possible to be more or less numb to sensations at the point of contact, even when all your nerves are in working order.
Numbness often happens when top-down processes overwrite bottom-up processes, as in the Ponzo illusion:
If the top blue line looks longer to you at first, it’s probably because you’re interpreting the visual sensations through an expectation that things near the bottom of the image are “closer”, while things near the top are “farther away”. If the top line were farther away, then it would have to be longer than the bottom line to produce the visual stimuli you encounter.
This interpretation probably happens so quickly that your raw visual sensations are overwritten by it. To see the blue lines as the same length, if you even can, you have to deliberately back up and peel off a thick layer of perceptual interpretation, excavating sensations that correspond more directly to the lines themselves.
But if you do peel back that interpretive layer, then you increase sensation at the point of contact with this image. It’s like removing your gloves and actually touching the snow. You are feeling more of the image itself, rather than feeling your own processing systems.
There is a gradient from “seeing” (in the Holmesian sense) to “observation”, and it is identical to the gradient from low to high contact with the territory. Dial up presence, personhood, or sensation, and contact with the territory gets more direct, more naked. Gloves off, touching the snow. Dial them way down, and you’re floating miles above the real world.
(The “how” is tricky. That’s a whole different sequence. But even before figuring out how to do it, it’s worth considering that there may exist the possibility of movement toward more direct contact. It can help a lot to merely be aware that your experiences exist somewhere along the gradient from low to high contact with the territory.)
In the sentence “Knowing the territory takes patient and direct observation”, this is what I mean by “direct”. I mean showing up as a whole person, in the presence of real-world data, and getting your preconceptions out of the way so you can be sensitive to what is there.
Knowing the territory takes patient and naked observation.
7 comments
Comments sorted by top scores.
comment by Alex_Altair · 2022-04-12T21:08:25.296Z · LW(p) · GW(p)
But if you do peel back that interpretive layer, then you increase sensation at the point of contact with this image.
Interestingly, the initial way I managed to perceive the lines as equal length is by splitting the image into the "railroad" part and the "lines" part, and trying to ignore the "railroad" part and just noticing the lines. I think this substantial reduces my overall contact with the image!
I know what you mean though, and another way I can perceive the lines as equal length is by treating the image as a 2D layout of blobs of color, which definitely increases my contact with the image.
Replies from: Korz↑ comment by Mart_Korz (Korz) · 2022-08-05T07:26:31.804Z · LW(p) · GW(p)
A re-interpretation that makes me perceive the lines as equal length without any problems is to label them as "some kind of magnets that hover above each other"
As this is somewhat realistic, while letting the lines integrate well with the rest of the railroad (maybe this is an artistic impression for Maglev trains?), this does not break immersion for me.
comment by London L. (london-l) · 2022-02-28T22:58:18.689Z · LW(p) · GW(p)
This is my first comment on this site, so if I'm missing particular norms, please let me know.
I understand why this was done, but it is amusing to me that in order to describe what "contact with the territory" looks like, you must use map-like terms such as "arm", "finger", and "hand".
I was going to recommend that you could qualify all of these terms, but I realized that this would likely not be needed for most readers; hopefully most people will understand, after having reading the previous essays, that by "hand", you're taking advantage of our preconceptions and built-in maps in order to describe your thought.
(I also realize that this has been done elsewhere out of necessity in the essays as well - it just particularly stood out to me in this one section.)
Replies from: BrienneYudkowsky↑ comment by LoganStrohl (BrienneYudkowsky) · 2022-03-04T06:53:10.176Z · LW(p) · GW(p)
it sounds like you understand a core component of why this sequence was so bloody difficult to write.
comment by LoganStrohl (BrienneYudkowsky) · 2022-10-04T20:18:15.801Z · LW(p) · GW(p)
>(The “how” is tricky. That’s a whole different sequence. But even before figuring out how to do it, it’s worth considering that there may exist the possibility of movement toward more direct contact. It can help a lot to merely be aware that your experiences exist somewhere along the gradient from low to high contact with the territory.)
Update: This post [LW · GW] gets into a lot of the "how".
comment by Shiroe · 2022-10-01T14:45:13.909Z · LW(p) · GW(p)
Quick note on the Ponzo illusion: In my view, seeing the top bar as longer is actually a more primitive, fundamental observation. The idea that the bars ought to appear as the same length is an additional interpretative layer thrown on top of this, justified by geometric principles and theories about human visual perception. The direct (or "raw") observation, however, is that the top bar appears longer.
comment by DaystarEld · 2022-03-15T18:03:09.172Z · LW(p) · GW(p)
These categories also feel super useful and worth distinguishing for a number of circumstances, such as one layer of trying to understand why two people could ostensibly experience the same thing, such as eyewitness a robbery, but remember it very differently.