Making Psychology a Science

…by borrowing from how alchemy turned into chemistry.

I don’t think psychology is a science yet. Science is quite hard to do well. It’s even harder when the topic isn’t objective. Psyches are extremely reflexive, which is to say, they change based on how they’re viewed. And even worse, they change based on how they’re viewed by psyches. Including themselves.

This whole picture makes objective scientific methods woefully inadequate here. It’s hard to even define what psyches are because they’re so reflexive!

So we end up with what amounts to psychological theories as cargo culting science. They sound astonishingly like humorism (the main medical theory before we figured out that germs are what make people sick). There’s even an analog to bloodletting: psychological problems are supposedly due to an imbalance of humors psychoemotional energies, so to heal the trauma that causes the imbalance we have to release the excess blood trapped survival energy.

(Not to say there’s no value to this view! There was also some value to bloodletting. I just think we can do much, much better.)

I think this cargo culting happens because objective things are easier to study and easier for the modern world to take seriously and thus to fund. But psychology really can’t be an objective science. It’s necessarily a subjective science. So if we’re going to make real scientific progress on understanding the human condition in a meaningful way, we absolutely have to develop good subjective scientific methods.

In particular, the core of good science best as I can tell amounts to finding and conducting crucial experiments. The problem is that that “finding” part is usually extremely difficult. Physics is an absurdly easy case: off the top of my head I can list at least a dozen that define the history of the field. But it’s much, much harder to list those for chemistry or biology — and in my experience, most reflexive sciences are very much like biology.

I keep finding I run into problems when trying to give folk crucial experiments in subjective science. There are a few I can point at, like dabbling with “occlumency” in light of the hostile telepaths problem. But even that isn’t really sharply crucial the way that, say, the Michelson-Morley experiment decisively killed the luminiferous ether hypothesis and paved the way to Einstein’s relativity theories.

So given all this… how do we make real progress on psychology¹ as a proper subjective science?

Relatedly: how do we find which crucial experiments in psychology are worth doing? We could have worked on finding ways to falsify humorism, for instance, but that’s actually quite hard and wouldn’t have told us much once it failed. In some sense we’re trying to replace psychological humorism with real mental medicine. Maybe we can just draw from the history of medical science here?

Borrowing from chemistry

I feel inspired by the history of chemistry right now. Early chemistry struggled to differentiate itself from alchemy. They didn’t have any underlying theory other than the alchemical ones, most of which couldn’t be falsified. There was an idea, for instance, that all metals are some mixture of sulphur and mercury, and that gold was defined by them being in perfect balance. So if you could somehow separate and cleverly recombine those two key components, you could convert any base metal into gold. But failure to do so wasn’t disproof of the sulfur/mercury idea: you could always attribute any failed attempt to an imperfection in the process or a lack of skill or having done it under an unfavorable condition of the stars or whatever.

Once the idea of scientific scrutiny came about, it was vividly unclear where to start in alchemy. Do we try to force the sulphur/mercury thing to make more falsifiable predictions and test those? If they fail, what then? If they succeed, what else does that imply? It wasn’t clear where the key leverage points were. Bear in mind that what we’d recognize today as atomic theory hadn’t yet been thought of (although its seeds had indeed been planted in ancient Greece, which is why we use the Greek word “atom”, meaning “uncuttable”).

I poked around the history of this development and asked Claude to summarize how this transition worked. Here’s my summary of Claude’s summary:

1. The scientific revolution in chemistry started by making chemical reactions reliably reproducible. Alchemists had been able to get some fascinating reactions to happen sometimes, but it was haphazard, and they didn’t know why it worked or didn’t. They’d record info that we now know is irrelevant, like the tide phase and the time of day and the astrological positions of the planets. Early chemists dropped the effort to achieve certain reactions and instead explored what’s involved in making reactions succeed or fail. As a result, chemists learned from every experiment, even if an alchemist would view the effort as a failure: the fact that a reaction didn’t happen would reveal something about the conditions under which it would. So the growth of chemical insight became relentless even if alchemical ideas couldn’t change much.

2. This “work out how to make reactions reliable” effort caused them to notice the importance of measurement and precision. It started becoming clear that the ratio of different substances were relevant for making specific reactions happen, for instance, and they started being able to describe those reactions mathematically.

3. There was a shift from goal-orientation (turn lead into gold, create “the elixir of life”, etc.) to discovering the composition and properties of various substances. E.g., there’s a substance called “aqua regia” that can dissolve gold. Alchemists viewed this as a possible tool in their attempted Great Work. Early chemists used it as a starting point for curiosity: what exactly is aqua regia made of? What’s key to its ability to dissolve gold? What exactly happens when the gold dissolves? That curiosity eventually let them figure out ways of recovering the gold. (There’s a fun story of a Hungarian chemist in fact doing this to hide Nobel prizes made of gold from the Nazis during World War II.)

All of this was doable before there was any deep underlying theory. We could start converting alchemy into a proper science by just looking at what actually happens and trying to replicate it. Viewing every effort as a learning opportunity regardless of what happens. Getting really curious about why the reactions happen under these conditions and not those ones. Then explanations sort of emerge from real phenomena.

In particular, alchemical ideas (like the sulphur/mercury thing) didn’t matter all that much. They might inspire ideas about what exploratory experiments to try. But it was totally fine that we couldn’t tell whether or not those old ideas were accurate. We were just gathering a much more precise sense of what there is to explain via looking at what actually happens.

Applying it to subjective science

I’m not sure we can just port the analogy straight over to developing a science of psyches. But I have hope! I think we can draw inspiration from it.

Chemistry started by looking at chemical reactions, which is to say, how they changed under different conditions. That focus invited us to work out which properties of the various chemicals were relevant for those reactions. We could do something very similar for subjective structures: what exactly is involved in creating state shifts? There are lots of states like flow, depression, hypnotic trance, etc. We have a bunch of tools for inducing or avoiding or maintaining those states. Sure, I personally don’t like getting depressed… but what precisely are the processes that would let me create it, and un-create it? Kind of like dissolving and un-dissolving gold from aqua regia.

I think for this to work, we need to adopt the “drop goals and just learn from every exploration” attitude. Yes, Buddhism might have some great insights, but it’s pretty singularly focused on ending suffering. But what exactly is the functional structure of suffering? What precisely does it take to increase it in my own subjective experience? Obviously I don’t want that! But I do want to know how suffering works. In this sense Buddhism is a kind of subjective engineering endeavor (the way alchemy was a chemical engineering effort): it might have helpful things to suggest, but it’s not yet a science. And we should be able to make it much more effective at what it does by developing a real science here.

That said, things get tricky when it comes to measurement. Measurement goes hand in hand with objectivity for the most part. We generally deal with reflexive situations (like, say, an animal’s life cycle) by standing outside that situation and describing the reflexive thing as a whole, objectively. That approach could work if we wanted to develop a science of how others’ subjectivities appear to us. But if we want a science of how our own subjectivities appear to us (which I think is utterly critical for having a science of spirituality and wisdom), we cannot pretend we can treat everything objectively. We have to swallow that we’re examining inherently reflexive situations. Which is to say, we must develop the analog of measurement for systems that we can only see from the inside.

At this point it’s a bit funny to talk about what “we” could do to explore individual subjectivities scientifically. “We” can’t. But you can explore yours, and I can explore mine, and we can compare notes and swap experiments and learn things about each other and ourselves.

I think that’s actually quite exciting. It means that subjective science is open source. It doesn’t make hardly any sense for someone to come along and proclaim expertise on how your subjectivity works. All they can do is suggest what you might find if you were to try certain experiments. Everyone can engage in subjective science; no one can really own it.

But reflexivity is tricky

However, there’s something to be aware of here in terms of reflexivity. Let me give an example.

I like the Enneagram. I find it useful. Sometimes someone’s Enneagram type sort of punches me in the face so hard I can’t fail to see it in them. And generally, when I use that type to ask them about their subjective experience, they’re often stunned by how much I get about them. I’ve sometimes accidentally given the impression that I’m psychic as a result.

But it’s actually unclear how much Enneagram type is inherent to a person versus something they conform to because the type is presented to them. When someone is super obviously (say) a Nine to me, is that because they “really are” a Nine (i.e., their subjective structure is objectively of type Nine such that we could reliably detect it as a Nine architecture given the right tools)? Or is it more like a self-fulfilling prophecy where they sometimes act a little Nine-like and people respond to them in kind, which reinforces that way of operating, etc.? In which case they might have a totally different Enneagram type (or none at all!) in a different situation.

How would you tell the difference? Given that the very act of trying to detect someone’s Enneagram type might cause their subjectivity to take on a specific type?

I’m reminded of how in math graduate school I once got treated like the class dunce. I went in for an oral exam and the professor started talking more slowly to me, like I was particularly thick-headed and couldn’t think straight about even quite obvious things. I recognized both his way of interacting with me and my own defeated body language from my own students. I was acting like the kids who were struggling to pass with a C-. He was talking to me like I would talk to those kids! I graduated as the valedictorian of my undergrad college, and quite a few of my fellow graduate students in my math program were quite sure I would most definitely pass the qualifying exams to move to doctoral level work. And yet, given the right social context, I suddenly reliably became a complete idiot who could barely remember what a number was.

I think subjective structures respond a lot to how they’re viewed. And that effect gets used internally: we can change how we’re subjectively arranged by changing our own self-image.

This just isn’t something chemistry ever had to deal with. The chemist is implicitly outside of the chemical reaction they’re examining. Chemistry is intrinsically an objective science. So measurement is pretty straightforward.

I run into this issue a lot when trying to explain what kind of foundation I think memetics needs as a science. There’s a clear objective component: you can look at the flow of memetic influences in culture, or observe their effects on other people. But what it feels like on the inside to be affected by a meme looks quite different! It typically doesn’t look like “a memetic influence”. It looks like realizing something. Getting caught up in a Flat Earth conspiracy meme feels like “realizing” that the Earth really is flat, that there really is some kind of conspiracy, etc. Converting to Catholicism in a subjectively meaningful way is less “I now believe in Christ” and more like a deep transformation in how the world seems to you. Et cetera.

So I think a big component of developing subjective science involves stepping into and out of memetic lenses. I think it’s helpful to be able to become Catholic or MAGA or woke or whatever, and then also to stop. There’s a real skill here, because these memes usually want to stay in residence dominating your subjectivity: it’s really hard to fake faith in the Lord in a subjectively convincing way, and once you’re doing it for real it’s very hard to deconvert. But I really do think there’s a skill here that lets you switch back and forth pretty fluently. And I think you can’t really understand how things affect your subjectivity if you basically always get stuck in (or out of) a given memetic view.

I also think this approach is kind of hazardous. It’s a bit like researching psychoactive substances by taking them. Some of them are very likely to have permanent effects on you. It’s hard to predict ahead of time what those effects are, since that’s precisely what you’re trying to explore.

I’m particularly reminded of a friend who committed suicide earlier this year, basically in the wake of “trying” meth. It was like a demon that possessed her. She eventually got off of it, but it wrecked her sense of meaning and identity while she was on it. She was something like a year sober when she took her life.

This stuff is real. It’s extremely potent. I think there’s cause for concern in doing these explorations. Ideologies like to possess people and stay in control of their bodies and minds. And on the inside it almost never feels like possession! It feels like clarity, or ease, or simplicity, or rightness. Some of them seriously want to eat your soul, for lack of a better term.

Meth ruined my friend’s psyche by giving her an experience of power and capability without any cares at all. Once she knew what that was like, her sober state of being was subjectively agonizing in comparison, even if the meth state was objectively deeply dysfunctional and damaging to things she cared about. There was always this relentless whisper: “You could make all that pain go away if you just take another hit.” And it was right. It’s a horrid nihilistic truth that points away from wholesome outcomes. One of those examples that makes me think “That which can be destroyed by the truth should be” is dangerously context numb.

I suspect there’s a potent meta-memetic skill here that most people don’t develop but that many people could. Basically the ability to fully dive into an influence without letting it keep hold of them. If subjective science were a mature discipline, developing this skill would be a core part of the intro training, I think. Along with some kind of recommendations about which subjective experiments to try first, and which ones to avoid. (E.g., definitely do not start by taking a hit of meth. Maybe never ever touch it. We can hopefully understand whatever’s important there without losing any more souls to demons like that one.)

“But isn’t this just Buddhism?”

A lot of spirituality already explores this area. I think we’ve been wrestling with a need for a science of subjectivity for at least tens of thousands of years, and we’ve developed some methods for approaching parts of it. It’s the subjective equivalent of the accumulated lore that created alchemy, or medical humorism.

Buddhism in particular is super popular in my social circles for this kind of topic. “But Buddhism already addresses this with XYZ” is a pretty common refrain.

I want to emphasize something here to make clear what I’m seeing. I really don’t think Buddhism solves the need for science. It’s kind of subjective alchemy. It has a goal, and a bunch of claimed methods for getting there, and a bunch of people who say they’ve achieved whatever the thing is. Awakening or whatever. Enlightenment is kind of Buddhism’s philosopher’s stone.

The whole thing is based on solving a problem via ancient methods and frameworks. Much of which has to be transmitted. Failure to get the tools to work is a failure on the part of the student, and rarely if ever casts meaningful doubt on the tradition. The analogy to alchemy is really quite striking.

That isn’t to say it’s useless or bad. Just that it’s not science. It doesn’t have science’s relentless power of inevitable insight. There was a brilliant gift that graced the world about 400 years ago that gave us an unparalleled mastery of the objective world. I’m pretty sure we can apply that gift to subjectivity, and to other reflexive domains as we expand our clarity of vision. We can draw inspiration from wisdom traditions just as chemistry drew inspiration from alchemy. But the same way that chemistry completed the objective project of alchemy by proving we cannot convert lead to gold via chemical reactions, and physics learned how to actually create gold via nuclear reactions, I think we can complete meaningful parts of the wisdom traditions. I think we can develop incredibly potent psychotechnologies that outdo what the Buddha and his lineage of students were able to create, in part by building on their insights but also by honoring that science really is exactly this kind of potent.

With all that said, I do think that some Buddhist methods are great for exploring subjectivity. Just sitting and watching your thoughts and examining their structure is an incredible tool, and it seems to develop some ability to see more. (Although, again, reflexivity: how much of what we come to see forms because we’re looking as opposed to having already been there and we’re just noticing?) There are maps of subjective states along the path that I think are pretty relevant to consider and take seriously (as in: do they happen? Why? How? Can we induce them? Can we reverse the order? How self-fulfilling is the map? Etc.). There’s a ton of other stuff that I think should be sources of serious inspiration.

…just like alchemists figured out a whole lot of really curious chemical reactions and methods that formed the basis of chemistry. But chemistry still did something really meaningfully new and potent that alchemy was just never, ever going to accomplish without the core insight of science.

What now?

I’m not sure chemistry’s history is the right thing to draw from here, or a complete vision. But I think it has some promise, and at least lets us get started.

Eventually I bet we can start doing really crucial experiments in subjective science. But first: let’s explore!

One of the things I’m personally most excited about is developing that meta-memetic skill I gestured at earlier. I might write about it in its own post. I think it’s key to navigating psychedelic spaces well, and to expanding our capacity for deep empathy without bricking ourselves. I also think that political discourse has a lot more hope to do something sane and kind if more people can see that memes are speaking through us, and just stop taking their urges so seriously.

In particular, I think there’s something exciting about getting very good at creating the “chemical reaction” of donning a memetic lens, or taking it off. It’s gonna be different for different memes. But I bet there are general rules. I’ve noticed quite a few in my own explorations. Again, I might write about this sometime in its own post.

But quite importantly, this isn’t just my research program. It’s necessarily open-source and very individual. We can say some general things about how subjectivities seem to work when seen from the outside (which is to say, how others’ subjectivities appear to us). But developing a science of first-person perspective really is something that I think we understand by each one of us learning how to play with it.

So: what do you see as possible here? What are you playing with? What “reactions” are you able to reliably induce that seem significant, and how do you do it? Do you think you could describe it in enough detail that someone else might be able to try? What are the parameters such that tiny changes create big yet predictable differences?

And where do you see me being confused here? What am I missing?

How might you, and we, actually do good subjective science?

The field is wide open!

1. You might fairly ask “Why focus on psychology?” Well, honestly, it’s because it’s the form of subjective science I’m personally most excited about and interested in. I like looking at how I and others tick. But perhaps with a heavy dose of bias, I also think that developing a real science of psychology might be one of a few really key pieces for encouraging a globally sane and kind dynamically stable state for the human species. Sure, there are cultural and economic design questions we need to answer, most of which are deeply reflexive. But even if we figure out some ideal form of government or whatever, we still have to understand how individuals can fit into such a system in a way that’s healthy and good. Otherwise it’s just an abstract ideal for abstract pseudo-humans. I also suspect that it’s easier to do subjective experiments on individuals than on groups, and that the laws governing intelligent systems are highly scalable and largely apply just as well to nations as they do to individuals. So I kind of suspect that psychology is to subjective science what physics is to objective science.