The three existing ways of explaining the three characteristics of existence

post by Kaj_Sotala · 2021-03-07T18:20:24.298Z · LW · GW · 1 comments

Contents

  On the nature of insights and explanations
    The problem with intellectual explanations
    The problem with no-explanation explanations
    A middle way
None
1 comment

The following was originally intended to be part of the introductory post to my series on the three characteristics of existence [LW · GW], explaining how I disliked most existing explanations to the 3Cs and what my series was trying to do differently. I ended up cutting most of this excerpt in order to make the post shorter, but have since then occasionally found myself wanting to refer back to this explanation. So I'm posting it here now.

Some of the content in this post may be previously familiar, because I ended up re-using some of the paragraphs here to explain slightly different points in the actual final introduction.

On the nature of insights and explanations

Most explanations of the three characteristics that I have seen have fallen into one of two categories, both of which I am unhappy with: either purely intellectual explanations, or no-explanation explanations.

The problem with intellectual explanations

The first style attempts to explain the 3Cs on the level of intellectual theory. This wouldn’t necessarily be bad by itself: after all, my article is also an intellectual theory. The problem arises when they are discussed purely as theories, without making the connection to human experience.

For example, a common way of describing one aspect of no-self goes along the lines of “You are not your thoughts, because you can observe your thoughts, and you can’t be something that you can observe”. If this is viewed purely as an intellectual position, it is easy to think of a counterargument for it. For example, we can write a computer program which happens to internally assign information about its memory use to a variable called memory_use, and then monitors the content of that variable to make sure that it is not using too much memory. According to the argument we just heard, the information that the program stored in the variable memory_use wouldn’t be a part of the program, because the program can monitor memory_use. But this is clearly incorrect, so the argument is false: there can be parts of you which observe other parts of you, while all being “you”.

The problem here is that the 3Cs are not abstract intellectual claims. Any written explanations of them are trying to point at things which are true about our minds, but which violate some of the implicit assumptions that most human brains tend to run on. When a writer says that “you can’t be something that you can observe”, this should not be taken as a general claim about the nature of equivalence relationships.

How should it be interpreted, then? There are many ways of reading this, and I don’t want to claim to have the definite interpretation, but here’s one take on it.

In an earlier article [LW · GW] trying to explain insight meditation, I mentioned that I had once had the thought that I could never be happy. When I paid closer attention to why I thought that, I noticed that my mental image of a happy person included strong extraversion, which conflicted with the self-image that I had of myself as an introvert. After I noticed the happiness-extraversion connection, it became apparent that I could be happy even as an introvert, and the original thought disappeared.

Essentially, I had originally believed that “I can never be happy”, and this belief about me didn’t feel like a “belief”. It felt like a basic truth of what I was, the kind of truth that you just know by virtue of having direct access to it - in the same way that you might look at an apple and just know that you are having the experience of seeing an apple, by virtue of having that experience.

But when I investigated the details of that experience, I realized that this wasn’t actually a fact about me, but rather a model of myself that my brain had constructed - and a false one at that. I had been observing a model, and that model was just a model, as opposed to being me.

In How An Algorithm Feels From Inside [LW · GW], Eliezer Yudkowsky writes:

… we don't instinctively see our intuitions as "intuitions", we just see them as the world.  When you look at a green cup, you don't think of yourself as seeing a picture reconstructed in your visual cortex—although that is what you are seeing—you just see a green cup.  You think, "Why, look, this cup is green," not, "The picture in my visual cortex of this cup is green."

And in the same way, when people argue over whether the falling tree makes a sound, or whether Pluto is a planet, they don't see themselves as arguing over whether a categorization should be active in their neural networks.  It seems like either the tree makes a sound, or not. [...]

Before you can question your intuitions, you have to realize that what your mind's eye is looking at is an intuition—some cognitive algorithm, as seen from the inside—rather than a direct perception of the Way Things Really Are.

This is the same insight as the one that “anything that you can observe is not you” is pointing at: anything that appears in your consciousness as “you” is part of your map about yourself, rather than being the territory. 

Of course, this is not entirely true. The map is being drawn by your brain itself, and is contained within the brain, so it is also a part of the territory - this is the “variable memory_use” objection. So it is true that, in the sense of being physically located within the brain, your thoughts about you are you. But that is not the sense of “being you” that we are interested in. 

What is more relevant is that anything which appears in your consciousness has been constructed by the brain to represent some aspect of itself or its environment, just as the contents of the variable memory_use are computed and determined by the program in which it is embedded.

Now, someone might ask, if that was the point, why say it in such a weird way? Why not just say “everything in your consciousness is constructed by the brain”? And furthermore, why are people giving the impression that this is something revolutionary and mind-blowing, when it is just basic science? Are the Buddhists just stuck repeating things that were revolutionary 2500 years ago?

Well, imagine that you find me looking upset. You ask what is going on, and I tell you that I feel bad because I can never be happy. You tell me “oh, don’t worry about that, all of your thoughts about yourself are just models constructed by the brain”. Instead of feeling better, I throw a rock at you and tell you to go away.

I already knew about science and the fact that I can’t perceive reality directly. Telling me that wouldn’t have helped. What did help was paying attention to the details of my conscious experience, noticing things about the belief which didn’t make sense, and thus realizing that it cannot actually be true. If I had not thought to do that myself, an outside observer might have been able to guide my attention to my own experience so that I could notice these things.

Likewise, “you are not that which you can observe” is intended to guide the reader’s attention to their own experience. Suppose that someone tells me that “you are not your thoughts, for you can observe your thoughts”. This is often followed by a number of other prompts, such as “you are not your body, for you can observe your body” and “you are not your emotions, for you can observe your emotions”. What this instruction is intended to do is to make me scan through everything in my consciousness, and hopefully notice that I can’t find anything that wouldn’t be some kind of a mental sensation. 

Anything that I could potentially associate as being “me” is a sensation in my mind, rather than being the thing which actually produces the sensation… and this is true for anything that I can find. Everything that I observe, is just models. Actually noticing this, can help loosen implicit mental models which interpret things in my experience as making contact with direct truths of myself.

Note that this is not saying that all of my models of myself would be wrong, necessarily. The map not being the territory does not mean that the map would have to be incorrect. But until I see the map as a map, I cannot even ask the question of whether it corresponds with the territory or not.

Now one might see why trying to explain the three characteristics on an overly intellectual level doesn’t really work. The point of the explanations is to get you to pay attention to aspects of your experience, in order to notice where your experience contradicts some of your schemas which are so deep as to be invisible to you. But at the same time, many of the characteristics are things that we already know - at least on an intellectual level. So if someone comes and says something like “you are not your thoughts” and sells this as a deep insight, a typical reaction goes something along the lines of 

So clearly presenting this content framed as a conventional intellectual theory easily goes wrong. People will treat it as an intellectual theory to be analyzed and assume that they already understand what it's trying to say, rather than treating it as a set of hints about things in their experience that could be valuable to pay attention to. 

In the worst case, someone offers a philosophical argument for why a claim about the 3Cs is wrong ("there is a self because..." "just because I can observe my thoughts doesn't mean that they are me..." "things are actually permanent because..."), someone else responds with why that argument is wrong, and then people get stuck in a debate that's entirely missing the point.

The problem with no-explanation explanations

Intellectual explanations often being misinterpreted leads many people to the opposite strategy: do not explain anything. Since the point is to get you to pay attention to your subjective experience, and intellectual explanations provide no benefit unless they get you to actually do the practices, the teacher refuses to give you any intellectual explanation. Instead, the teacher just gives you pointers to what you should be looking at [LW · GW], tells you to do it and stop asking stupid questions, and whacks you with a stick until you do what they tell you.

This is the other way of explaining the 3Cs which I am unhappy with. Now, it certainly has its advantages: if your audience trusts the teacher’s authority, then they will just shut up and do the practices, and maybe learn. 

Another advantage of the no-explanation explanation is that when you are trying to pay attention to the structure of your thoughts, anything that might get you lost in content risks leading you astray. We can roughly say that you can investigate your own mind on two different levels: in terms of the content in your mind, and in terms of what kind of structure underlies all of that content:

When I noticed that my belief about myself was wrong, I did it by noticing a piece of structure which led me to content - I realized that I had a particular mental image about happy people, and then started thinking about whether or not it is actually true that only extroverted people can be happy. 

If I had been focused on investigating structure, I would not have dived into content by asking “is this true”. Rather, when I saw the assumption that I had made and saw counterexamples arise in my mind - memories of people who were introverted and happy - I might have done something like noting that the counterexamples were also in the form of mental images. I could then have done something like dissecting any arising content into smaller components, such as separating them into mental images, mental speech, and physical sensation.

Now, suppose that you are trying to specifically investigate structure. Someone has explained to you the theory of how everything is just mental sensations. Then you practice observing this in practice. At some point you might notice something which supports this view, and start thinking “oh great, I am really seeing how everything is mental sensations!”. 

The moment that you start doing that, you are thinking about a piece of content: instead of your attention being on the structure of your thoughts, your attention is on the intellectual theory of how everything is sensations (which is a piece of content), or maybe on your self-image at being good at meditating and noticing these kinds of things (ditto). 

Sometimes thinking about theory while meditating is useful, but frequently it would be better to just take the thought as another piece of mental content to investigate the structure of. The more theory you have, the easier it is to get distracted by it: no-explanation explanations avoid this problem by giving you minimal theory.

Daniel Ingram describes a minimalist approach to theory in the following anecdote (“noting” is a specific style of insight meditation, where you make a note of everything that appears in your consciousness):

One of my best insight meditation teachers, Venerable Sayadaw U Rajinda, would hold interviews every two days while I was on my third retreat at a beautiful center in Penang, Malaysia, that was very conducive to practice. I would come in and describe all sorts of experiences that I was all excited about, and he would listen calmly to me go on and on and then finally ask, “Did you note it?” That was almost all he ever said.

It was amazing how easy it was to forget that simple instruction, and equally amazing how extremely useful it was when I remembered to follow it. He didn’t seem to care about anything other than that I grew to know my reality as it was with great precision and consistency. I knew very little theory then, but during those two weeks I practiced noting quickly all day long and made the fastest progress I have ever made in my life, getting all the way to the very brink of first awakening in a mere fourteen-day retreat. Since that time, I have been a big fan of this particularly direct and down-to-earth method.

If you want to do noting practice, then you just focus on doing noting practice. If this produces what feels like an important insight about the nature of the mind, then you note that you've had what feels like an important insight, and continue noting anything else that comes up. If it produces the most mind-blowing experience you've ever had, then you note the fact that you just had what felt like the most mind-blowing experience you've ever had, and continue noting anything else that comes up. And so on. The better you are at ignoring all the content that comes up, the deeper you can go into seeing the structure behind it - and the less your mind is burdened with thoughts about theory, the easier those thoughts are to ignore.

(Note that I am not claiming that it is actually always a good idea to only focus on structure. Only focusing on structure is good, if you want to specifically investigate structure. Focusing too much on this can lead to suppressing your ordinary emotions and psychological content that would actually be a good idea to deal with, and lead to all kinds of unhealthy behavior. I do think that there's tremendous value to be had in deep investigations to structure, but that there's just as much if not more value in investigating your content. If someone wanted to improve their practical day-to-day well-being, I would probably recommend putting more effort into learning content-focused practices such as Focusing [LW · GW], Internal Family Systems [LW · GW] or Coherence Therapy [LW · GW] than in insight practices, if they had to choose just one.)

Another strength of the “no explanations, just practice” method is that describing what you should expect to find may by itself lead you astray. On some occasions, I read a description about some meditative insight. I then imagined that “oh, directly experiencing this must feel like X”, and went looking for something that would feel like X. When I did eventually experience the insight, I recognized it from the description but also realized that I had been imagining it entirely wrong. It might have been better if I had never read of it, and just went in blind instead.

Romeo Stevens notes that:

… a lot of people seem to take a paint by numbers approach to these practices that then don't really work for them. The investigation you do into your own experience has to be a real investigation, and not one in which you are highly confident about what there is to find. i.e. people feel like they're meditating wrong if their experience doesn't seem to match whichever map they are using. [...]

I do think that this has resulted in ambient memetic immunity of the same type hypothesized by Scott Alexander and others about why new psychotherapy methods work for a while and then seem to stop working. People get some sort of idea of what all these experiences are supposed to be and as a result ignore actual moment to moment sensation. This happened to me with piti (a jhana factor). I realized I had been keeping my eyes to the horizon looking for some sort of special spiritual sensation instead of noticing what was actually happening in my body.

It is very easy to keep thinking about how everything is mental sensations, instead of actually looking at your experience. Much of insight meditation involves noticing the conflict between what you expect to see, and what you actually see. If you have a strong preconception of what you should see, then this will get in the way of actually seeing what’s there [LW · GW].

But for all of its benefits, the no-explanation style of explanation has enormous downsides. Most notably, it will completely fail if you have no particular reason to trust the person’s authority: they are telling you to do this thing which they claim to lead to deep insights, while also claiming that it can’t be explained or that it is pointless to explain? Why should you give their words any weight, and how do you know that they are not just peddling a set of practices that will make you go insane rather than giving you any real insight?

Here is a common kind of reaction [LW(p) · GW(p)]:

"I have this great insight, but I not only can't explain it to you, but I'm going to spend the balance of my time explaining why you couldn't understand it if I tried to explain it" sounds awfully close to bulveristic stories like, "If only you weren't blinded by sin, you too would see the glory of the coming of the lord".

The no-explanation explanation depends completely on the reader already having trust in meditation being something that is worth doing. In the absence of that trust, it fails completely.

Furthermore, I do think that it is useful to have some idea of what you are looking for. Now, most such explanations do talk about what you should be paying attention to… but since they do not provide the generator of that advice, you don’t know how and when it should be modified to suit individual circumstances. If you are under the close supervision of a teacher or in standardized circumstances - such as in a monastery - this is not a problem. Outside that traditional setting, it is.

A middle way

Here [LW(p) · GW(p)] is another skeptical comment, which I agree with myself:

I think that the only coherent way to convince us that Enlightenment is real is to provide a model from a 3rd party perspective. [...] The model doesn't have to be fully mathematically rigorous: as always, it can be a little fuzzy and informal. However, it must be precise enough in order to (i) correctly capture the essentials and (ii) be interpretable more or less unambiguously by the sufficiently educated reader.

Now, having such a model doesn't mean you can actually reproduce Enlightenment itself. [...] However, producing such a model would give us the enormous advantages of (i) being able to come up with experimental tests for the model (ii) understanding what sort of advantages we would gain by reaching Enlightenment (iii) being sure that your are talking about something that is at least a coherent possible world even if we are still unsure whether you are describing the actual world.

This is what I am hoping to do: explain the three characteristics in a way which combines the first- and third-person explanations, saying both things that someone can realize while meditating, and how those experiences can be understood from a naturalistic, reductionist perspective. 

I gave an example of this when I elaborated on the “what you cannot observe is not you” thing: I noted that "your thoughts are not you, your body is not you" etc. is intended to direct your attention through all the objects that you might have in consciousness, so as to lead you to the realization that you can't find anything in your consciousness that wouldn't be a mental sensation... and that this is then an experiential realization of the intellectually-understood fact that everything in your consciousness is something that the brain computes, rather than representing raw reality.

Back when I was initially getting interested in meditation myself, the scientific evidence about what exactly it was doing was even more scarce than it is today. Yet there were already some preliminary claims suggesting that meditation was understandable in third-party terms, as the brain getting more access into earlier stages of its own processing. Now, given what I know about the replication crisis and scientific evidence today, I would not find those early results compelling if you showed them to me now… but it is still true that I would have been much less likely to engage in meditation if those results had not existed. I want to write something which might be similarly compelling to the me of today, if there was a me of today who had never read about or tried meditation before.

Now, the question of “what kinds of information does meditation produce” is related but distinct from “should I engage in meditation”. This is especially so given that various schools of meditation make use of that information in different ways, and thus get different results. I have thus attempted to structure my explanation so that I cover various concrete claims about the human mind. These claims should hopefully be interesting regardless of whether you intend to engage in meditation or not.

1 comments

Comments sorted by top scores.

comment by Chris_Leong · 2021-03-08T01:05:38.033Z · LW(p) · GW(p)

“You are not your thoughts, because you can observe your thoughts, and you can’t be something that you can observe”. If this is viewed purely as an intellectual position, it is easy to think of a counterargument for it... For example, we can write a computer program which happens to internally assign information about its memory use to a variable called memoryuse, and then monitors the content of that variable to make sure that it is not using too much memory

Yeah, I've always found that argument confusing for that reason. Thanks for clarifying.