Disclaimer: I'm writing this post because I'm about to go on a 10-day meditation retreat and I wanted to both explain to people why I'm doing that and also put down my thoughts about meditation now so that I can compare them to my thoughts when I get back. I originally titled the post "In Defense of Meditation as a "First Person Science,"" assuming that I was in fact defending. However, given the number of reasons in the post I have for not treating meditation as "scientific," and given that I haven't given any time to trying to think of other ways to approach the problem of collecting data for "first person science", and given that I am biased towards wanting to believe that meditation is useful because I'm about to spend a lot of time doing it, I decided to change the title to reflect that at this point, it is more of an introduction to an interesting point of view than a properly defended argument. Also, I got the idea from my friend David, who has a blog post about the same idea from a somewhat different perspective.

There are things going on inside your head. These things exist in some sense- clearly they do because you're sitting there observing them. Since these things, these subjective sensations, exist, we should be able to apply some sort of a scientific (or "scientific") approach to finding out how they work in a subjective sense. By this I mean a "first-person" as opposed to a "third-person" approach: in the "first-person" view, the most basic thing you can talk about are base sensations. In the "third-person" view, the most basic things you can talk about are neurons (or quantum fields in curved spacetime, depending on just how far down you want to go).

If you are a reductionist like myself, the assumption is that the "first-person" approach should be able to be entirely explained by the "third-person" approach. However, even if this is true, we should still be able to make statements about the way subjective experiences seem to work on the subjective level. Namely, you have some sensations and they're not completely random. How (and to what degree) can we explain any regularities or structure to these sensations within the context of the way sensations actually seem to us?

Now, there are obvious problems with doing any sort of investigation on the self in an attempt to find a "scientific" answer- it's obviously impossible to do anything like a double-blind randomized controlled study with a large sample size, and it's hard to obtain anything resembling quantitative data about yourself, or really anything that's terribly amenable to concrete analysis. (How many times have you been asked to rate some experience of yours on a scale of 1-10 and had essentially no idea what to put down? This happens to me pretty much any time I am given a questionnaire.) But let's say that one wanted to apply a stripped-down version of the scientific approach to internal experience. At the bare minimum, you need to acquire observations about your internal experience. How would one do this? Well, sit down and observe your thoughts. It's that simple.

"Ooh, ooh, pick me!" says the philosopher in the crowd. "I already do that! That's what I get paid for! I'm a first person scientist! Yay!" 

I would disagree. What philosophers do is not typically observing their thoughts, but rather thinking about their thoughts. This is akin to mistaking a theoretician for an experimentalist.

Who is an experimentalist in this metaphor? A meditator! Experienced meditators will sit for hours every day with the goal of being able to observe every subjective sensation that occurs at the moment that it occurs.
For anyone who's done data analysis on any reasonably complicated experiment before, the analogy should be clear (the analogy also extends to the experience of having a vague notion of how to be doing things and what your results should probably look like, but the process of getting to the actual results being deeply frustrating.)

I don't mean any of this to claim that meditative practices have anywhere near the sort of rigor that (ideally) goes into scientific practices, or that theories derived from observations made during meditative practice should be the last word on internal experience. All I mean is that, when you want to get more information about your own experience, you cannot simply assume that you have already observed it, you have to actually sit down and do the observation.

(Note: From here on out, my ideas about Buddhism and meditative practices are largely sourced and reinterpreted from Daniel Ingram's "Mastering the Core Teachings of the Buddha: An Unusually Hardcore Dharma Book" , which disagrees deeply and at great length with many western Buddhist traditions. I consider my conflict of interest declared.)

Most people imagine that meditation is essentially about calming down the mind or maybe figuring out what you're thinking about and trying not to think too hard about things that make you anxious. This can be very useful in the context of mindfulness-based cognitive behavioral therapy, but this is not the sort of meditation I'm referring to. That is all content-focused. What I care about is looking past the content to see how subjective experience itself actually works. Namely, it doesn't matter whether you're thinking an angry or a neutral or a depressed thought, or if you're feeling tense or relaxed- all that matters is that you examine the sensations closely enough to figure out how they work. If you have a thought, do you notice it arriving? How does it arrive? How long does it stay? Do you notice it going away? If there's a sensation on your body, does it stay in one place if you focus on it, or does it move around? Can you be aware of more than one sensation at a time? Do sensations come continuously or discretely? Can you notice if there is a separate sensation for seeing something than there is for reacting to that sight? Can you observe any moments in which you aren't experiencing any sensations?

The state of enlightenment, viewed in this framework, is the state in which you become completely aware of the actual structure of all your subjective experience at all times. Assuming this is something you can actually do by mediating (which is the claim made in Buddhism, and I recognize that this is extremely bold and therefore should be highly suspect), then why wouldn't you want to meditate?

Note that I define enlightenment in terms of structure and not content- contrary to popular notions, enlightenment is not about being permanently blissful, calm, radiant, and perfectly ethical. All that stuff is content. It seems to be the case that knowledge of structure contributes to producing better content- for instance, if you see that the seemingly permanent anger that you have is in fact transient, you're less likely to act rashly on an angry impulse- but the two aren't necessarily linked, and it certainly isn't the case that you cease having sensations of anger as an experienced meditator.

Now, meditative practices also come with some presuppositions about what the fundamental aspects of all sensory experience are (note that this is also a point against describing meditation as "scientific"- the theoretical judgments may deeply affect our capacity to get unbiased data. Then again, this is a general problem even in the harder sciences, so maybe we shouldn't be too hard on ourselves.) Here they are, in order of the sense they currently make to me.

1) Impermanence
Sensations show up, and then they go away. They don't stick around forever. That's all this means. It's seems so obvious to me that it's hardly worth trying to explain in more depth, except to say that focusing on the showing up and the going away constitutes the vast majority of meditative practice.

2) Dissatisfaction
Nothing permanently satisfies you- if you have a desire for food that you sate by eating food, it won't be too long before you want more food again. If you shift in your chair, eventually you'll want to shift again. Or maybe you'll want a more comfortable chair, or if you're fine with the chair then you'll find something else to desire. This cycle of desire never ends. This much, on a general sort of level, seems pretty obvious to me. In Buddhism they make the more radical claim that this pattern exists in every single sensation you have. Now, this sounds like a bit of a downer, but note that this is supposed to be a statement about structure, not about content. Both good and bad, pleasurable and painful thoughts are dissatisfactory.

3) No-Self
This one is the trickiest of the three- the essential claim is that our naive intuition about selfhood is deeply flawed, and that one can see that this is the case by noticing that sensations seem to happen "on their own." Thoughts bubble up into your head without "you" making them do so.

On closer inspection, every individual thing that you observe doesn't seem to be your "self." Is that twinge on your shoulder your "self"? What about the sensation of being located behind your eyes? Is whatever thought you just had about the previous sentence your "self"? The idea is that by following these lines of questioning you will find that there is not a self to be found anywhere, and then something about your understanding shifts. Honestly, the descriptions of no-self seem to me sufficiently vague that I can't determine how plausible the Buddhist solution to the notion of self is.

However, there is very clearly something wrong with ordinary self. How many times have you felt like you are two or more people talking to each other? How stable is your sense of personal identity? Mine changes extremely rapidly. These things all present severe challenges to naive notions of selfhood. Maybe this alternative approach is right.

And that's it! The Three Characteristics. (They get capital letters because they are a Big Deal.) The idea is that by honing your skills at observing these characteristics in all sensations, all the time, you can break free of ordinary patterns of thought and see sensations in their truest form, while on the way shedding attachment to things and to your notion of what it means to have a "self". Pretty cool. Time to see if it works.