In this post I present a model of human psychological development based on increases in the structural complexity of ontology. Or more simply, it’s a model about how human minds grow in terms of how complex their models of the world are. It breaks down development into 8 stages that build on each other, and spans from birth to what, for lack of a better term, we might call enlightenment. I think you’ll find it useful if you’re interested in better understanding how human thinking changes and grows throughout a lifetime, in developing yourself, or if you notice some confusion that humans seem to change in meaningful ways as they age but don’t already have a strong gears-level model of how that works.
The model rests on several claims about the world:
Humans engage in psychological development throughout their lives.
This psychological development is marked by phases that are different from each other in kind (a difference in “type signature”) along certain key dimensions.
This difference in kind is due to increasing structural complexity of one’s models (ontology) in reference to the reality (the ontic) being modeled.
One can progress towards greater complexity.
This greater complexity confers the disposition to live a better life, contingent on integration of the insights this complexity offers.
For this post I assume you already agree with Claim 1 as it’s not very controversial. For evidence of Claim 2 and Claim 4 I defer to both content linked in the Background section and the writing of David Chapman. For Claim 3 I defer to my previous post, “Phenomenological Complexity Classes”. And for Claim 5 I present myself (you can see some of my personal story around this in the “Phenomenological Complexity Classes” post) and basically everyone else who has improved their lives through self-help, positive psychology, and Buddhist practices leading to awakening.
With these claims in mind, I then present my current best general model of what this psychological development looks like. If you’re familiar with the “Phenomenological Complexity Classes” post (just “PCC” hereon), this is basically version 2 of that.
Background
Presenting the model in full detail with everything justified within the text to the best of my ability would require giving a lot of background, probably something book length. I’m definitely not going to do that, so with my apologies here are a bunch of links that set up the knowledge base out of which the model is formed.
I wrote PCC in the early months of 2017. In it I lay out a theory that generalizes Robert Kegan’s constructive developmental framework in terms of the structructure of the experience of ontology. This turned out to be generally convergent with Michael Commons’ model of hierarchical complexity, although it takes a somewhat different approach to how the structure is defined, informed by my philosophical leanings (more on those shortly). This shouldn’t be too surprising, though, as Ken Wilber, whatever you think of him more generally, has done a lot of work exploring how various models of psychological development are correlated and has created possibly the most complete (if insufficiently detailed) model of developmental psychology, so it would have been a problem if my model didn’t correlate with existing ones (more on this later).
During this same time I was investigating the general patterns that had caused me to “move up” according to these models of psychological development, mostly out of a kind of frustration that it had required so much work for me to do it and a desire to share what I learned from that effort so that others might not have to struggle as much. I first gave some specific advice about what I think helped me achieve this: “act into fear and abandon all hope”. I also figured out that psychological development is nonmonotonic (things get worse when you move away from a local maxima before they get better by moving to a higher maxima). It was also around this time period that I made my way to Zen practice, finding self-help and positive psychology techniques were no longer enough on their own to further improve my life.
These investigations led me to spend a lot of time considering philosophical issues, both prior to formulating the model in PCC and after, partly in pursuit of questions around AI alignment but also with questions about developmental psychology in mind. Because my thinking is heavily influenced by things like Hegelian dialect, Husserl’s phenomenology and transcendental idealism, Heidegger’s notions of caring and being, Sartre’s explorations of meaning and meaninglessness, Madhyamaka notions like dependent origination and non-dualism, and a strong background in mathematics, science, engineering, and LessWrong-style rationality, the model is also infused with a worldview that includes them. I additionally have a higher-order theory of consciousness built around nested feedback loops which seeks to explain something of what I was fundamentally pointing at in PCC and, by extension, in the current model, but I don’t believe anything in the model actually hinges on this.
Before we get to the heart of the post, I want to set expectations.
My attempts to model this started six years ago when I learned about Kegan at a house party. His ideas excited me. They accurately accounted for a large body of evidence I had about humans, including myself and changes I had undergone, that I had previously been unable to explain with a unified model. But as I soon discovered, much of psychology is more like alchemy than chemistry, and so there are many possible perspectives on the ground truth of what’s happening with human minds because there are many fuzzy layers of inference between the physical activity of the brain and the human-level conception of it as mind. Since we continue to have only fairly primitive methods for correlating our intuitive experience of the mind with the physical phenomena arising within the brain, we’re left to do our best to piece together theories that are instrumentally useful to us even if they are all, in a certain sense, fake.
Thus it has been quite useful to me, and hopefully others, to “say wrong things” about psychological development as part of a process of incrementally searching model space for useful ideas and perspectives that may help us in various ways even if it’s hard to nail down how strongly correlated these ideas and perspectives are to the real operation of the brain. Maybe one day we’ll have molecularly precise brain imaging and we can develop theories in a more rigorous way, but for the time being we’re left to explore the space like we are philosophers in the ancient world trying to understand physics. It’s in this spirit I’m offering my updated model of human psychological development.
Hopefully that conveys the right epistemic tone for this post. I’m not trying to say I have the explanation of what’s going on with developmental psychology and how it relates to all human endeavors. Instead I’m offering up the best gears-level model I’ve been able to construct by trying to infer what’s going on given very noisy data. Other models are possible and useful, and you shouldn’t discount them. I just think mine is useful, too.
The Model
I call this a model of ontological development because it’s a model of developmental psychology focused on changes in ontology (i.e. the map), specifically around the phenomenological structure of ontology, viz. the way in which reality is experienced. It’s a type-theory-like model in that it takes basic constructs and builds them up into more and more complex “types” of thought that the mind works with to explain the progression of adult psychological development.
It’s clearly “wrong” in the sense that the mind almost certainly is not actually operating directly on constructs of the sort described (although I think there’s a story where it’s closer to the truth than it might otherwise seem), yet I think it is useful in that it lumps together details in a way that makes them possible to work with and, importantly, helps you make predictions about actions you might take that will help you progress from one stage to the next. I’ll offer some evidence that these stages exist, but mostly by reference to other models. This isn’t a rigorous, scientific model, but a useful, “alchemical” model that has helped me and might help you.
On that point, the primary use I have for this model is to better model people by accounting for their stage of development, or even just the possibility that different people are at different stages of development and need different things to psychologically accommodate them. It helps me think about many kinds of human interactions and how to navigate them, since it’s rare I find myself in a scenario full of likeminded folks all at the same developmental stage and with similar traumas. And since a common failure mode of psychological development is to typical mind fallacy other people to your current stage of development and forget about when you yourself were at an earlier stage (e.g. the common trope of adults not being able to relate to what it’s like to be a kid despite having been one themselves) or that you yourself might not be at the final stage of development (e.g. the “arrogant teenager” trope), having a developmental model can go a long ways towards empathizing with folks at other stages.
The model is broken into stages, each stage separated by a different ontological structure. These differences are arranged in a type hierarchy, with each encapsulating and extending upon the previous one to ultimately transcend what was conceivable within the previous stage. This is very similar to the type hierarchy in both PCC and Commons’ model of hierarchical complexity if you want to see additional examples. I’ll try to explain how the type is built up at each stage without getting overly technical and bogged down in notation, since although that’s fun I don’t think it’s actually that useful for anything (if you want notation, see PCC, which you can easily extend for yourself to make sense in this post).
Since each stage represents a type change, moving from one stage to the next involves a moment of deep insight. In Zen terminology we’d call these satori or kensho (a deep understanding of some aspect of reality); you might also reasonably call them awakenings. In Ingram’s model of awakening there’s something like a “rearranging” that happens in the mind as it “shifts gears” into the new way of being. And in Commons’ model these are type changes, as they are here, which Commons describes by analogy to the nascent type theory of abstract algebra.
This means that ultimately this deep change in how the mind constructs itself via a change in the structural complexity of thought is the diagnostic measure of what stage a person is in, which is annoying because it can only be inferred from observable behavior, and we can be mistaken about what stage someone is at, including being mistaken about the stage a person assesses themselves to be at! This is importantly different from how scientific models of psychological development are defined, because they need to rely on easily observable markers that reliably identify stages. This means things might not quite line up with other developmental theories in expected ways because they will tend to either rank people at higher stages than this model will due to the more stringent requirement to be counted in a stage or at lower stages because a person may have the key insight for a stage but not have integrated it to an extent that it shows up in their observed behavior.
Because the preceding paragraph is important, I’ll say the same thing again another way. Each stage is defined by a particular “insight” or type change I’ve expressed in some particular way below, but it should be understood that these insights are wordless and have multiple expressions, even if I think they have a common structure which I will explicate, but the change in structure is generally not how the change between stages is directly experienced. The definitive characteristic of a stage is grokking the insight that defines it, so a person isn’t at a stage until they’ve grokked it so thoroughly that they could never not realize it. This is importantly different than an intellectual understanding; you might say it’s something like a stage change happens at the System 1 level.
Also, a person can be in a stage but not have fully integrated the insight that puts them in that stage, so outward behavior might not reliably indicate stage. This is because developmental stages, as presented here, are statistical patterns over thoughts, and it’s each thought that has a certain type signature that signifies a stage, and we say a person is at a stage when they reliably can think with a certain type signature. So importantly a person’s stage is not an essence or property of them, but something they continually express via the structure of their thoughts. This naturally means that a person can fail to integrate an insight into a higher stage they “have” into other thoughts via compartmentalization (cf. multi-agent models of mind). This is why you might hear spiritual folks talk about “integration work” to take the insights gained during practices like meditation, prayer, and psychedelic use and apply them to real life. The same holds for this theory.
This theory also proposes rather strongly that there’s no skipping stages since that would be like skipping a type in a type hierarchy. If you look at someone and it seems like they didn’t pass through a stage, that’s a high prior that you’ve misassessed them and they are actually at a lower stage than you initially thought. This is possible because even at some particular stage N, it’s possible to work that stage real hard to think what could be effortlessly thought in stage N+1 or N+2, sort of like how it’s possible to solve some algebra problems without algebra and some calculus problems without calculus by using clever tricks. We’ll explore this phenomenon in some more detail as we walk through the stages.
With all that settled, at last I’m ready to present the model’s stages.
Stage 0: Ontological Emptiness
This is the stage we all start out at before we are even a someone. This is the pre-ontological, empty stage. There’s no feedback, so there’s no information, so there’s no thought or ontology, only non-differentiation within the ontic. This is the stage of rocks, feed-forward circuits, and zygotes.
It’s tempting to just say that and move on, but I think it’s important to note that part of us is always operating at this stage. That is, just because our minds are capable of thought doesn’t mean they stop doing things that don’t demand it. The tricky part is that there is nothing it is like to have stage 0 “thought”, so it’s somewhat tricky to reason about as being part of the self. Not that we can’t think about it, mind you, but rather what’s tricky is holding this understanding of what it is “like” to be a physical process while also being one that is self-aware and not accidentally mixing up something of what it is like to be something rather than nothing.
This aspect of nothingness or emptiness or formlessness will turn out to be important in stages 4 onwards, but for now we can leave it be.
Stage 1: Ontological Birth
This is the stage where self-awareness appears. We go from being nothing to something and someone to ourselves. Feedback loops start creating information, that information is ontology, and it starts to build up. This happens pretty early on for humans, before we are even born, although it’s hard to pin down the exact moment when it starts, though I don’t think that’s an impossible task.
This stage is typified by the ability to reify things as things, i.e. to perceive the world as made up of objects rather than just undifferentiated stuff. We take the world into ourselves and start creating our private, subjective maps, built in the territory but not of it. These objects are “simple” in that they have no internal structure, but may have properties or “essences”, like a ball that is round and red and bouncy, not by virtue of its place in reality, but because the model simply says that the ball directly is irreducibly those things.
What’s interesting about this stage is that although we can think in terms of objects, that’s the only structure permitted by the type signature of the stage, so there’s no way to connect objects together, thus there’s no way to do things like connect these objects to each other or, importantly, back to some internalized sense of what’s real, thus the objects can spin a bit freely from external reality. Not that they are disconnected from reality, just not self-aware of the connection.
By the time kids are talking they’re already out of this stage. It roughly matches to Kegan’s stage 0 and Commons’ stages 1 through 4. Note that if you’re familiar with PCC, this is more conservative than what I claimed for PCC’s stage 1 because I was mistaken about how much you can do with just bare reification into simple objects.
Stage 2: Ontological Correlation
This is the stage where ontology becomes aware of itself (we become self-self-aware) because the type signature moves up to see that objects can be connected to each other. Notions like causation aren’t fully developed yet because there is no field over which things can operate on other things, but there’s enough to correlate together objects and understand they are related to each other, even if in ways that would seem confused to an adult. Children move into this stage when they are pretty young, almost always before language is developed, and they tend to be well integrated into it by the time they are 3 or 4.
Unlike with the first two stages, which are entered without anything it is like to be oneself, this is the first stage that can be characterized by an insight, which might be experienced as something like “things are related to other things”, but this is speculation because it happens prior to the ability to form episodic memory that might carry forward an impression of the moment of insight.
As stated, there is only fully recognition of correlation and not causation at this stage. This leads to stuff like thinking you can change reality by changing your thoughts. That is, if you believe something hard enough, because reality and your model of it are correlated, changing the model can change reality. That’s kind of true in a convoluted way (e.g. you work to make reality the way you want it to be), but at the level of understanding here leads to stuff like thinking if you will it to happen then it will happen without doing anything else.
None of this is to say that causation can’t and isn’t modeled in stage 2, only that any model of it is necessarily incomplete because there’s nowhere in the type signature to hold the causal structure. Thus causal models developed at this stage tend to be brittle (and endlessly amusing to adults), but struggling to work out causality is an important developmental activity that will lead towards the insight for stage 3.
This stage roughly maps to a combination of Kegan’s stages 1 and 2, and it’s a stage humans can reach and then stay in their whole lives, although that’s not very common. What’s more common is that people have the insight for stage 3 but then never fully integrate it and so have lots of lingering stage 2 thought patterns. Also matches with Commons’ stages 5 through 9.
Stage 3: Ontological Causation
This is the typical “adult” stage, where it’s realized that not only are things related to other things, but that there is a system in which or a field over which things interact and that this system or field is itself a higher-order thing that exists only by virtue of being made up of first-order things. Seeing that there is a container holding all the things and their relationships allows considering everything together as a whole while also considering it to be made of parts at the same time. The age at which transition into this stage varies, but it usually happens somewhere around puberty, but can happen later or possibly never.
The key insight here is something like “everything has to fit together”, in that beliefs about the world don’t exist in isolation and they are all tied together and have to reflect one another. An ego-centric version of the insight might be something like “I have to act to make things happen”.
Note that this stage is a bit weird in that it’s possible to put off moving into it for a long time by being really good at working with stage 2 type thinking. That is, for most purposes, it’s possible for someone to stay in stage 2 and not make the jump to stage 3 ever or until later than normal because they have the option to just be really, really good at understanding how objects are related to the reification of the world, not as a system made up of things, but as a separate thing related to all things. The difference is subtle but important, because not having the stage 3 insight or not having it fully integrated continues to enable thinking about the world in a disjoint, childlike way where the map and territory can get confused on an everyday level.
This stage is also important because has a kind of natural balance to it. Although it’s not complete, it’s historically sufficient for getting on with most human endeavors, and it’s only those trying especially hard to understand the world and their experience of it or those living in complex modern societies that place multicultural demands on people who find it’s not enough. Most humans alive now and in the past never got past this stage, and that’s been mostly okay because as long as a person never needs to grapple with more than one consensus about how the world is they’ll likely never run into challenges it can’t handle.
This stage basically maps to Kegan’s stage 3, although as presented in The Evolving Self and not in later works where the stage was somewhat deflated; in more recent work you might say it can map all the way up to Kegan’s stage 5 because, just as it is possible to work stage 2 type thinking really hard to achieve outcomes similar to stage 3, it’s possible to work stage 3 really hard to achieve outcomes similar to higher stages. Also lines up with Commons’ stages 10 through 12.
Stage 4: Ontological Separation
This is the first of the so-called post-conventional stages because it’s the first one beyond where most adults do most of their thinking. This is the stage where the world is no longer the whole thing, but there can be multiple models of the world and those models can be in relation with one another. That is, the system or field talked about in stage 3 becomes itself an object that can be related to other systems or fields which hold different objects. If you like, this is like having thoughts that permit the existence of morphisms between different ways of thinking about the world.
The insight for this stage looks a bit different to different people. I might describe it as something like “things are not the things themselves”. A personal version of this would be something like “I’m not me”. I’m hesitant to phrase this as “the map is not the territory” because that’s an idea lots of people understand but don’t grok, but it’s got the right idea, so I mention it only because if you think about this stage for 5 minutes you’d realize this way of phrasing the insight yourself but I wouldn’t get to provide a caveat about it.
Another way I think about it is that the stage 4 insight involves realizing there’s a space (or emptiness) between the model of the world and the world itself. They’re related, of course, and causally connected, but there’s also a gap. It becomes possible to grok things like intentionality and the subject-object distinction such that your experience of the world is infused with realizing you’re experiencing things indirectly.
As I say, at the personal level this is something like realizing “I’m not me” or, put more formally, that the subject experiencing reality is not fully identified with itself, so that it’s possible to simultaneously find some aspect of yourself being both the subject and object of a thought. This is enabled by being able to hold two copies of the self, modeled as a system, at the same time and tie them together in a relationship. As a consequence, it’s no longer obvious that one “can’t” do things just because the model says they can’t be done because the model is not exactly the world and it might be wrong.
There’s two different ways this is approached in stage 3, so let me talk about them to make it clear how it’s different. One is to take the self and compress it down into a simple object with no internal structure beyond whatever properties or essences it is assigned (a stage 1 type thing), and then a similarly structured thought with less complexity can be had. The other is to keep the complexity but project it out into the world so it’s disidentified with but via dissociation rather than objectification.
For what it’s worth, I also consider this stage to be where the Buddhist notion of stream entry happens. I bring this up because it points to this stage being one that is both an important transformation and one that is unstable in a way that it causes the person who thinks in this way to be put on a path where they must now come to terms with intentionality, the subject-object distinction, and the space that separates the ontological from the ontic.
This stage roughly matches up with Kegan’s original stage 4 or his later stage 5 and with Commons’ stage 13.
Stage 5: Ontological Flexibility
Building on the previous stage by analogy to the way stage 3 builds on stage 2, this stage sees the ontological realization of a meta-system in which to think about systems in relation to each other, like going from having the tools to work with mathematical categories to having a coherent system of systems worthy of the name category theory. Structurally this is similar to stage 3, but with the objects replaced by systems so that a full additional level of complexity can be held within a thought without having to compress it down into simple objects.
The insight for this stage is something like “things can be more than one way”, i.e. there’s more than one model that “correctly” describes the world because the method used to construct the model contains one or more free variables, hence even the idea of “correctness” becomes conditional on what was assumed. You might describe this as grokking the existence of a hyperprior (or universal prior) necessary to believe anything.
On a personal level this can be expressed as “I can be anyone”, by which I mean identity is not fixed by instead flexible and can be swapped out like a mask, and not just put on as if acting but totally replacing the mask that was previously being worn, because there is always some mask in place. At this stage belief also becomes more flexible, and beliefs become open to direct manipulation (as opposed to the indirect manipulation possible at earlier stages) because the network of beliefs can finally be adequately modeled within a single thought.
Like with stage 4, it’s possible to think one is thinking at stage 5 while really at stage 3 or stage 4 by similar mechanisms to the ones we’ve discussed around stages 3 and 4. The big confusions are mistaking being able to compress systems into objects for being able to think in terms of systems of systems without flattening them down, as in stage 3, and mistaking thinking of systems as being in relation to one another for having a coherent system of systems model into which everything fits, as in stage 4.
This stage lines up with Kegan’s stage 5 and Commons’ stages 14 and 15. Since I said stage 4 lines up with stream entry and thereby first path, stage 5 also matches up, in my estimation, with second path.
Stage 6: Ontological Fracture
At this point you may have already noticed the tik-tok pattern of the stages: the odd-numbered stages are about creating new structures to hold more complexity, and the even-numbers ones are about putting the previous structures into relationship with one another. So it is with stage 6, putting meta-systems into relationship with one another. In some ways this is less revolutionary than stages 2 and 4 because there’s already a lot of complexity that can be modeled in a single thought at this stage, but it can only go 2 levels deep right now. Stage 6 is the start of being able to have 3 levels of structure which means we’re starting to approach thought that can handle recursion without resorting to compression.
The insight for this stage is something like “everything’s connected”, but as stated that sounds a lot like the insights for stages 2 and 3. A better way to put it might be something like grokking that wherever you go, whatever you do, you’re always embedded in the system and that any appearance of Cartesian dualism or any distinction of any kind is an artifact of the map rather than the territory. I titled this stage ontological fracture because it takes the insight of stage 5 that there’s a free variable that enables flexibility and takes it one step further to crack apart the very idea that ontology is how the world is experienced.
At this point we’re well past the point where it’s easy to confuse this stage for an earlier one, so there’s lots of “near enemies” or “false friends” when I try to explain this in everyday terms without pointing to the type change already discussed. There’s are very stage 2 and 3 ways of talking that sound superficially similar to how someone at stage 6 might talk (e.g. “it’s all connected”, “interdependence”, “karma”, etc.), with the difference being that the stage 2 and 3 versions of the same stage 6 thought involve cached, flat, gearless models, but this presents a tricky verification problem because it’s going to be hard for someone at, say, stage 3 to tell apart someone at stage 3, 4, or 5 and someone at stage 6 saying superficially similar things.
This gets at a problem with developmental psychology that’s been popping up along with every stage, but becomes a bigger issue with each stage, which is that to someone at stage N, the wisdom of stage N+2 is nearly indistinguishable from someone spouting nonsense who is pretending to have wisdom. The only difference is that the stage N person can nail down the stage N+2 person to say coherent things in patches if you press them hard enough whereas the person spouting nonsense will keep spouting nonsense, but at the level of 5 or 10 words it’s going to be hard to tell these folks apart, and even with more words it still might be hard to tell them apart unless the stage N person is pushing for specific answers they can verify. The greater the gap between two people, the harder this verification problem becomes.
The standard solution is for the stage N+M person to shut up, and luckily stage 6 is the point where shutting up actually starts to make sense, because it involves thinking things that no longer neatly fit in language and instead can only be talked around and pointed at but can’t be communicated directly because there’s no distinction to be made, since stage 6 is the stage where the space between subject and object, emptiness, or the ontic start to fit into experience rather than being something that is there but can’t be experienced.
A personal expression of this stage’s insight might be “I just am” because there’s a first-hand experience of reality as it is outside the ontology necessary to reach it. Confusingly, this same sort of direct experiencing can happen at earlier stages without it pushing a person into this stage, and for a person below this stage that kind of experience will seem more like something special than ordinary. So this a kind of nothing-is-special, I-just-am thought that is only possible when the type signature of thought makes space for both the ontological and the ontic emptiness in which it arises.
Stage 6 also sees an increased ability to manipulate one’s own preferences as a result of being to model experientially what it is like to “choose” to do something. I put “choose” in quotes because doing this involves realizing there are actually no free choices, but rather that what we choose is a product of the mechanisms of the mind and body. Plenty of people understand this intellectually, so again this is about grokking this and then being able to work with preferences within your mind such that you might go straight towards doing things that will change them rather than trying to reshape them indirectly. Not everyone may hone this ability, though, and Cook-Greuter’s model (more on this model later) suggests some people at this stage will explicitly not have this ability because that model proposes that this stage has two different expressions, an idea I’m skeptical of but open to given the limited data we have about people at this stage.
Kegan doesn’t have this stage, and it partly matches up with Commons’ stage 16 although I think we could also read Commons’ stage 16 as overlapping with the high end of my stage 5. The only thing this really maps to in existing systems with Cook-Greuter’s stage 5⁄6 and Buddhist third path.
Stage 7: Ontological Fusion
This stage completes the model by expanding the type of thought to hold a system containing systems of systems. This grants three full levels of structural modeling allowing both things, systems, and systems of systems to be completely contained within a single thought and provide a surrounding meta-meta-system in which they be operated on. This is significant in that it provides enough structure to iteratively mentally model recursion over a series of thoughts without having to either compress a system into a simple thing or have the system being considered take up the entirety of the type, leaving no room for operating on the considered system itself.
The insight might be something like “everything’s perfect” or “nothing was ever separate to begin with”. A personal version of the insight might be something like “I am the world, yet the world is not me”. This leads to the stage often being expressed in terms of something like non-dualism, which here means fully realizing in every moment that the map is not the territory and that the map is embedded in the territory and never getting confused and thinking as if there were some Cartesian split between the model and the modeled. This enables a kind of perspective flexibility similar to but much expanded upon from the stage 5 level of flexibility, like going from being able to wear different masks to being able to wear every mask at once.
Because I believe this stage has a type signature that permits unrestricted iteration of recursive ontology, this would appear to be a reasonable final stage in that higher stages are theoretically possible but not necessary, their only benefit being able to do something like consider an extra level of recursion within the same thought, which is nice but not strictly necessary, since three “registers” is enough. However, I made a similar claim in PCC about stage 5 because at the time I didn’t have a grasp on what stage 6 and 7 would look like or that they even existed, so it’s possible I will turn out to be wrong here, but I will say I’m at least more convinced of my own argument for stage 7 being enough to think with arbitrarily complex ontology than I was by my arguments in PCC for stage 5 doing the same.
This stage is off-the-map for both Kegan and Commons, but lines up with Cook-Greuter’s stage 6 and Buddhist fourth path.
On this point about fourth path, that should mean that shifting into this stage constitutes enlightenment. I’m unsure. It’s possible, but I think it’s more likely there’s two things going on when we talk about enlightenment: persistent non-symbolic experiences (PNSE) and stage 7 thinking. You might call these the narrow and broad aspects of enlightenment, respectively, or not-self consciousness and cosmic consciousness (“cosmos” being from Greek “κόσμος” meaning order of the world). However, I’m to some degree speculating here because I’m not at stage 7 or enlightened, so although I think there’s a body of evidence pointing to there being these two aspects to what is traditionally thought of as enlightenment and stage 7 describes one of them, I’m not altogether convinced, and I could easily be misunderstanding, with stage 7 either being the same as PNSE or being totally unrelated to enlightenment.
Similar Models
I’ve already discussed some of the correlations to other models along the way of describing the stages, but I have a few additional things to say on the subject.
At a high level, you might say my model is something like an extension of Kegan’s model. At least, that is if you look at Kegan’s model as presented in The Evolving Self and not at the versions with deflated stages present in his later work with Lahey. Unfortunately Kegan’s model doesn’t really have gears, which was, in hindsight, much of what I was trying to provide when I wrote PCC. After writing PCC it was pointed out to me I had reinvented something similar to but definitely different from Commons’ model in that we use somewhat different gears but have a similar tact for finding them.
Both the Spiral Dynamics model and Wilber’s extension of it offer alternative developmental models that extend beyond Kegan and Commons. Unfortunately these models are somewhat vague and don’t really have the kind of gears I want, but they do basically line up with my model stretching from infrared to ultraviolet.
While writing this post, I learned someone else extended Kegan’s model to the same point I have, although she did it somewhat differently than I have and without gears. Thus my model also roughly aligns with Cook-Greuter’s Ego Development Theory, especially onward from Cook-Greuter stage 4⁄5, which mostly corresponds to my stage 4. Below this stage we differ somewhat, with Cook-Greuter making more distinctions than I do and being, in my opinion, overly specific (and therefore wrong by failing to account for a wider range of human experiences) about what is common at each of the preconventional and conventional stages.
As mentioned a few times, I think there’s also an alignment with the Buddhist four-path model of enlightenment, and, although not mentioned above, with the five ranks model. Whether or not this is true, though, depends a lot on how you choose to interpret those models. In particular the five ranks system is explicitly not a stage model (“rank” is a terrible translation, but it’s standard) but instead describes different perspectives or places from which reality can be experienced, and the only sense in which we can get stages out of it is by considering which of the ranks a person is able to see the world from. Despite this, I actually think there is something there, and thinking of this stage model as correlating with enlightenment has been helpful to my practice thus far, so I think there can be something skillful in making this connection.
There are many other developmental psychology models I could try to relate my model to, but I’ve limited myself to the ones I’m most familiar with. If there’s another one you know and love, feel free to try to work out the correlations and post them in the comments.
What Next?
So, we have this model. What to do about it?
As I say, I’ve mainly found it useful for modeling people better, but it’s also useful as a kind of rough guide to what is possible as a human. If you look at this or similar developmental models and place yourself around, say, my model’s stage 4, then you have some hints about what’s next for you if you apply some effort to psychological development and seek out practices that may help you work towards the next stage. Although you can’t go directly towards each stage (a longer discussion to perhaps have another time), knowing a next stage exists can motivate effort and keep you from becoming complacent with what you have already accomplished.
As for what’s next for this model itself, I don’t know. It’s possible I’ll want to come back in 3 to 4 years and refine it more again, so it’s definitely provisional in that sense, but not so provisional that I don’t think it’s useful.
As a general point, I think what’s important more so than even my specific model, which you may or may not find useful or understandable, is that you consider the value of a developmental model of mind that can move from less to more complex stages. There’s many models out there, and although they all try to describe roughly the same territory, they do it in different ways that may click for different people better than others or may be more useful in some contexts than others.
Thanks
If you read this all, thanks for putting up with it! I’d love to put in the time to explicate it more, but this is my 80⁄20 attempt to share the model so others who find it useful can know about it without putting in the hundreds of hours of work it would take to produce a polished work on the subject. I’m sorry that it was neither shorter nor longer.
I’ll do my best to clear up confusions or questions in the comments. I’m mainly interested in discussing the theory itself rather than developmental psychology in general, but I realize this post may be too tempting a target to not want to discuss the broader class of models in the comments.
A Model of Ontological Development
Introduction
In this post I present a model of human psychological development based on increases in the structural complexity of ontology. Or more simply, it’s a model about how human minds grow in terms of how complex their models of the world are. It breaks down development into 8 stages that build on each other, and spans from birth to what, for lack of a better term, we might call enlightenment. I think you’ll find it useful if you’re interested in better understanding how human thinking changes and grows throughout a lifetime, in developing yourself, or if you notice some confusion that humans seem to change in meaningful ways as they age but don’t already have a strong gears-level model of how that works.
The model rests on several claims about the world:
Humans engage in psychological development throughout their lives.
This psychological development is marked by phases that are different from each other in kind (a difference in “type signature”) along certain key dimensions.
This difference in kind is due to increasing structural complexity of one’s models (ontology) in reference to the reality (the ontic) being modeled.
One can progress towards greater complexity.
This greater complexity confers the disposition to live a better life, contingent on integration of the insights this complexity offers.
For this post I assume you already agree with Claim 1 as it’s not very controversial. For evidence of Claim 2 and Claim 4 I defer to both content linked in the Background section and the writing of David Chapman. For Claim 3 I defer to my previous post, “Phenomenological Complexity Classes”. And for Claim 5 I present myself (you can see some of my personal story around this in the “Phenomenological Complexity Classes” post) and basically everyone else who has improved their lives through self-help, positive psychology, and Buddhist practices leading to awakening.
With these claims in mind, I then present my current best general model of what this psychological development looks like. If you’re familiar with the “Phenomenological Complexity Classes” post (just “PCC” hereon), this is basically version 2 of that.
Background
Presenting the model in full detail with everything justified within the text to the best of my ability would require giving a lot of background, probably something book length. I’m definitely not going to do that, so with my apologies here are a bunch of links that set up the knowledge base out of which the model is formed.
I wrote PCC in the early months of 2017. In it I lay out a theory that generalizes Robert Kegan’s constructive developmental framework in terms of the structructure of the experience of ontology. This turned out to be generally convergent with Michael Commons’ model of hierarchical complexity, although it takes a somewhat different approach to how the structure is defined, informed by my philosophical leanings (more on those shortly). This shouldn’t be too surprising, though, as Ken Wilber, whatever you think of him more generally, has done a lot of work exploring how various models of psychological development are correlated and has created possibly the most complete (if insufficiently detailed) model of developmental psychology, so it would have been a problem if my model didn’t correlate with existing ones (more on this later).
During this same time I was investigating the general patterns that had caused me to “move up” according to these models of psychological development, mostly out of a kind of frustration that it had required so much work for me to do it and a desire to share what I learned from that effort so that others might not have to struggle as much. I first gave some specific advice about what I think helped me achieve this: “act into fear and abandon all hope”. I also figured out that psychological development is nonmonotonic (things get worse when you move away from a local maxima before they get better by moving to a higher maxima). It was also around this time period that I made my way to Zen practice, finding self-help and positive psychology techniques were no longer enough on their own to further improve my life.
These investigations led me to spend a lot of time considering philosophical issues, both prior to formulating the model in PCC and after, partly in pursuit of questions around AI alignment but also with questions about developmental psychology in mind. Because my thinking is heavily influenced by things like Hegelian dialect, Husserl’s phenomenology and transcendental idealism, Heidegger’s notions of caring and being, Sartre’s explorations of meaning and meaninglessness, Madhyamaka notions like dependent origination and non-dualism, and a strong background in mathematics, science, engineering, and LessWrong-style rationality, the model is also infused with a worldview that includes them. I additionally have a higher-order theory of consciousness built around nested feedback loops which seeks to explain something of what I was fundamentally pointing at in PCC and, by extension, in the current model, but I don’t believe anything in the model actually hinges on this.
The final piece of the puzzle came from studying Buddhist models of psychological development. Daniel Ingram’s Mastering the Core Teachings of the Buddha gave me some useful models to work with, which helped me better understand the four-stage model of enlightenment and the ox-herding pictures, and Dongshan’s five ranks put it all together for me, rounding out my current model.
Caveat Emptor
Before we get to the heart of the post, I want to set expectations.
My attempts to model this started six years ago when I learned about Kegan at a house party. His ideas excited me. They accurately accounted for a large body of evidence I had about humans, including myself and changes I had undergone, that I had previously been unable to explain with a unified model. But as I soon discovered, much of psychology is more like alchemy than chemistry, and so there are many possible perspectives on the ground truth of what’s happening with human minds because there are many fuzzy layers of inference between the physical activity of the brain and the human-level conception of it as mind. Since we continue to have only fairly primitive methods for correlating our intuitive experience of the mind with the physical phenomena arising within the brain, we’re left to do our best to piece together theories that are instrumentally useful to us even if they are all, in a certain sense, fake.
Thus it has been quite useful to me, and hopefully others, to “say wrong things” about psychological development as part of a process of incrementally searching model space for useful ideas and perspectives that may help us in various ways even if it’s hard to nail down how strongly correlated these ideas and perspectives are to the real operation of the brain. Maybe one day we’ll have molecularly precise brain imaging and we can develop theories in a more rigorous way, but for the time being we’re left to explore the space like we are philosophers in the ancient world trying to understand physics. It’s in this spirit I’m offering my updated model of human psychological development.
Hopefully that conveys the right epistemic tone for this post. I’m not trying to say I have the explanation of what’s going on with developmental psychology and how it relates to all human endeavors. Instead I’m offering up the best gears-level model I’ve been able to construct by trying to infer what’s going on given very noisy data. Other models are possible and useful, and you shouldn’t discount them. I just think mine is useful, too.
The Model
I call this a model of ontological development because it’s a model of developmental psychology focused on changes in ontology (i.e. the map), specifically around the phenomenological structure of ontology, viz. the way in which reality is experienced. It’s a type-theory-like model in that it takes basic constructs and builds them up into more and more complex “types” of thought that the mind works with to explain the progression of adult psychological development.
It’s clearly “wrong” in the sense that the mind almost certainly is not actually operating directly on constructs of the sort described (although I think there’s a story where it’s closer to the truth than it might otherwise seem), yet I think it is useful in that it lumps together details in a way that makes them possible to work with and, importantly, helps you make predictions about actions you might take that will help you progress from one stage to the next. I’ll offer some evidence that these stages exist, but mostly by reference to other models. This isn’t a rigorous, scientific model, but a useful, “alchemical” model that has helped me and might help you.
On that point, the primary use I have for this model is to better model people by accounting for their stage of development, or even just the possibility that different people are at different stages of development and need different things to psychologically accommodate them. It helps me think about many kinds of human interactions and how to navigate them, since it’s rare I find myself in a scenario full of likeminded folks all at the same developmental stage and with similar traumas. And since a common failure mode of psychological development is to typical mind fallacy other people to your current stage of development and forget about when you yourself were at an earlier stage (e.g. the common trope of adults not being able to relate to what it’s like to be a kid despite having been one themselves) or that you yourself might not be at the final stage of development (e.g. the “arrogant teenager” trope), having a developmental model can go a long ways towards empathizing with folks at other stages.
The model is broken into stages, each stage separated by a different ontological structure. These differences are arranged in a type hierarchy, with each encapsulating and extending upon the previous one to ultimately transcend what was conceivable within the previous stage. This is very similar to the type hierarchy in both PCC and Commons’ model of hierarchical complexity if you want to see additional examples. I’ll try to explain how the type is built up at each stage without getting overly technical and bogged down in notation, since although that’s fun I don’t think it’s actually that useful for anything (if you want notation, see PCC, which you can easily extend for yourself to make sense in this post).
Since each stage represents a type change, moving from one stage to the next involves a moment of deep insight. In Zen terminology we’d call these satori or kensho (a deep understanding of some aspect of reality); you might also reasonably call them awakenings. In Ingram’s model of awakening there’s something like a “rearranging” that happens in the mind as it “shifts gears” into the new way of being. And in Commons’ model these are type changes, as they are here, which Commons describes by analogy to the nascent type theory of abstract algebra.
This means that ultimately this deep change in how the mind constructs itself via a change in the structural complexity of thought is the diagnostic measure of what stage a person is in, which is annoying because it can only be inferred from observable behavior, and we can be mistaken about what stage someone is at, including being mistaken about the stage a person assesses themselves to be at! This is importantly different from how scientific models of psychological development are defined, because they need to rely on easily observable markers that reliably identify stages. This means things might not quite line up with other developmental theories in expected ways because they will tend to either rank people at higher stages than this model will due to the more stringent requirement to be counted in a stage or at lower stages because a person may have the key insight for a stage but not have integrated it to an extent that it shows up in their observed behavior.
Because the preceding paragraph is important, I’ll say the same thing again another way. Each stage is defined by a particular “insight” or type change I’ve expressed in some particular way below, but it should be understood that these insights are wordless and have multiple expressions, even if I think they have a common structure which I will explicate, but the change in structure is generally not how the change between stages is directly experienced. The definitive characteristic of a stage is grokking the insight that defines it, so a person isn’t at a stage until they’ve grokked it so thoroughly that they could never not realize it. This is importantly different than an intellectual understanding; you might say it’s something like a stage change happens at the System 1 level.
Also, a person can be in a stage but not have fully integrated the insight that puts them in that stage, so outward behavior might not reliably indicate stage. This is because developmental stages, as presented here, are statistical patterns over thoughts, and it’s each thought that has a certain type signature that signifies a stage, and we say a person is at a stage when they reliably can think with a certain type signature. So importantly a person’s stage is not an essence or property of them, but something they continually express via the structure of their thoughts. This naturally means that a person can fail to integrate an insight into a higher stage they “have” into other thoughts via compartmentalization (cf. multi-agent models of mind). This is why you might hear spiritual folks talk about “integration work” to take the insights gained during practices like meditation, prayer, and psychedelic use and apply them to real life. The same holds for this theory.
This theory also proposes rather strongly that there’s no skipping stages since that would be like skipping a type in a type hierarchy. If you look at someone and it seems like they didn’t pass through a stage, that’s a high prior that you’ve misassessed them and they are actually at a lower stage than you initially thought. This is possible because even at some particular stage N, it’s possible to work that stage real hard to think what could be effortlessly thought in stage N+1 or N+2, sort of like how it’s possible to solve some algebra problems without algebra and some calculus problems without calculus by using clever tricks. We’ll explore this phenomenon in some more detail as we walk through the stages.
With all that settled, at last I’m ready to present the model’s stages.
Stage 0: Ontological Emptiness
This is the stage we all start out at before we are even a someone. This is the pre-ontological, empty stage. There’s no feedback, so there’s no information, so there’s no thought or ontology, only non-differentiation within the ontic. This is the stage of rocks, feed-forward circuits, and zygotes.
It’s tempting to just say that and move on, but I think it’s important to note that part of us is always operating at this stage. That is, just because our minds are capable of thought doesn’t mean they stop doing things that don’t demand it. The tricky part is that there is nothing it is like to have stage 0 “thought”, so it’s somewhat tricky to reason about as being part of the self. Not that we can’t think about it, mind you, but rather what’s tricky is holding this understanding of what it is “like” to be a physical process while also being one that is self-aware and not accidentally mixing up something of what it is like to be something rather than nothing.
This aspect of nothingness or emptiness or formlessness will turn out to be important in stages 4 onwards, but for now we can leave it be.
Stage 1: Ontological Birth
This is the stage where self-awareness appears. We go from being nothing to something and someone to ourselves. Feedback loops start creating information, that information is ontology, and it starts to build up. This happens pretty early on for humans, before we are even born, although it’s hard to pin down the exact moment when it starts, though I don’t think that’s an impossible task.
This stage is typified by the ability to reify things as things, i.e. to perceive the world as made up of objects rather than just undifferentiated stuff. We take the world into ourselves and start creating our private, subjective maps, built in the territory but not of it. These objects are “simple” in that they have no internal structure, but may have properties or “essences”, like a ball that is round and red and bouncy, not by virtue of its place in reality, but because the model simply says that the ball directly is irreducibly those things.
What’s interesting about this stage is that although we can think in terms of objects, that’s the only structure permitted by the type signature of the stage, so there’s no way to connect objects together, thus there’s no way to do things like connect these objects to each other or, importantly, back to some internalized sense of what’s real, thus the objects can spin a bit freely from external reality. Not that they are disconnected from reality, just not self-aware of the connection.
By the time kids are talking they’re already out of this stage. It roughly matches to Kegan’s stage 0 and Commons’ stages 1 through 4. Note that if you’re familiar with PCC, this is more conservative than what I claimed for PCC’s stage 1 because I was mistaken about how much you can do with just bare reification into simple objects.
Stage 2: Ontological Correlation
This is the stage where ontology becomes aware of itself (we become self-self-aware) because the type signature moves up to see that objects can be connected to each other. Notions like causation aren’t fully developed yet because there is no field over which things can operate on other things, but there’s enough to correlate together objects and understand they are related to each other, even if in ways that would seem confused to an adult. Children move into this stage when they are pretty young, almost always before language is developed, and they tend to be well integrated into it by the time they are 3 or 4.
Unlike with the first two stages, which are entered without anything it is like to be oneself, this is the first stage that can be characterized by an insight, which might be experienced as something like “things are related to other things”, but this is speculation because it happens prior to the ability to form episodic memory that might carry forward an impression of the moment of insight.
As stated, there is only fully recognition of correlation and not causation at this stage. This leads to stuff like thinking you can change reality by changing your thoughts. That is, if you believe something hard enough, because reality and your model of it are correlated, changing the model can change reality. That’s kind of true in a convoluted way (e.g. you work to make reality the way you want it to be), but at the level of understanding here leads to stuff like thinking if you will it to happen then it will happen without doing anything else.
None of this is to say that causation can’t and isn’t modeled in stage 2, only that any model of it is necessarily incomplete because there’s nowhere in the type signature to hold the causal structure. Thus causal models developed at this stage tend to be brittle (and endlessly amusing to adults), but struggling to work out causality is an important developmental activity that will lead towards the insight for stage 3.
This stage roughly maps to a combination of Kegan’s stages 1 and 2, and it’s a stage humans can reach and then stay in their whole lives, although that’s not very common. What’s more common is that people have the insight for stage 3 but then never fully integrate it and so have lots of lingering stage 2 thought patterns. Also matches with Commons’ stages 5 through 9.
Stage 3: Ontological Causation
This is the typical “adult” stage, where it’s realized that not only are things related to other things, but that there is a system in which or a field over which things interact and that this system or field is itself a higher-order thing that exists only by virtue of being made up of first-order things. Seeing that there is a container holding all the things and their relationships allows considering everything together as a whole while also considering it to be made of parts at the same time. The age at which transition into this stage varies, but it usually happens somewhere around puberty, but can happen later or possibly never.
The key insight here is something like “everything has to fit together”, in that beliefs about the world don’t exist in isolation and they are all tied together and have to reflect one another. An ego-centric version of the insight might be something like “I have to act to make things happen”.
Note that this stage is a bit weird in that it’s possible to put off moving into it for a long time by being really good at working with stage 2 type thinking. That is, for most purposes, it’s possible for someone to stay in stage 2 and not make the jump to stage 3 ever or until later than normal because they have the option to just be really, really good at understanding how objects are related to the reification of the world, not as a system made up of things, but as a separate thing related to all things. The difference is subtle but important, because not having the stage 3 insight or not having it fully integrated continues to enable thinking about the world in a disjoint, childlike way where the map and territory can get confused on an everyday level.
This stage is also important because has a kind of natural balance to it. Although it’s not complete, it’s historically sufficient for getting on with most human endeavors, and it’s only those trying especially hard to understand the world and their experience of it or those living in complex modern societies that place multicultural demands on people who find it’s not enough. Most humans alive now and in the past never got past this stage, and that’s been mostly okay because as long as a person never needs to grapple with more than one consensus about how the world is they’ll likely never run into challenges it can’t handle.
This stage basically maps to Kegan’s stage 3, although as presented in The Evolving Self and not in later works where the stage was somewhat deflated; in more recent work you might say it can map all the way up to Kegan’s stage 5 because, just as it is possible to work stage 2 type thinking really hard to achieve outcomes similar to stage 3, it’s possible to work stage 3 really hard to achieve outcomes similar to higher stages. Also lines up with Commons’ stages 10 through 12.
Stage 4: Ontological Separation
This is the first of the so-called post-conventional stages because it’s the first one beyond where most adults do most of their thinking. This is the stage where the world is no longer the whole thing, but there can be multiple models of the world and those models can be in relation with one another. That is, the system or field talked about in stage 3 becomes itself an object that can be related to other systems or fields which hold different objects. If you like, this is like having thoughts that permit the existence of morphisms between different ways of thinking about the world.
The insight for this stage looks a bit different to different people. I might describe it as something like “things are not the things themselves”. A personal version of this would be something like “I’m not me”. I’m hesitant to phrase this as “the map is not the territory” because that’s an idea lots of people understand but don’t grok, but it’s got the right idea, so I mention it only because if you think about this stage for 5 minutes you’d realize this way of phrasing the insight yourself but I wouldn’t get to provide a caveat about it.
Another way I think about it is that the stage 4 insight involves realizing there’s a space (or emptiness) between the model of the world and the world itself. They’re related, of course, and causally connected, but there’s also a gap. It becomes possible to grok things like intentionality and the subject-object distinction such that your experience of the world is infused with realizing you’re experiencing things indirectly.
As I say, at the personal level this is something like realizing “I’m not me” or, put more formally, that the subject experiencing reality is not fully identified with itself, so that it’s possible to simultaneously find some aspect of yourself being both the subject and object of a thought. This is enabled by being able to hold two copies of the self, modeled as a system, at the same time and tie them together in a relationship. As a consequence, it’s no longer obvious that one “can’t” do things just because the model says they can’t be done because the model is not exactly the world and it might be wrong.
There’s two different ways this is approached in stage 3, so let me talk about them to make it clear how it’s different. One is to take the self and compress it down into a simple object with no internal structure beyond whatever properties or essences it is assigned (a stage 1 type thing), and then a similarly structured thought with less complexity can be had. The other is to keep the complexity but project it out into the world so it’s disidentified with but via dissociation rather than objectification.
For what it’s worth, I also consider this stage to be where the Buddhist notion of stream entry happens. I bring this up because it points to this stage being one that is both an important transformation and one that is unstable in a way that it causes the person who thinks in this way to be put on a path where they must now come to terms with intentionality, the subject-object distinction, and the space that separates the ontological from the ontic.
This stage roughly matches up with Kegan’s original stage 4 or his later stage 5 and with Commons’ stage 13.
Stage 5: Ontological Flexibility
Building on the previous stage by analogy to the way stage 3 builds on stage 2, this stage sees the ontological realization of a meta-system in which to think about systems in relation to each other, like going from having the tools to work with mathematical categories to having a coherent system of systems worthy of the name category theory. Structurally this is similar to stage 3, but with the objects replaced by systems so that a full additional level of complexity can be held within a thought without having to compress it down into simple objects.
The insight for this stage is something like “things can be more than one way”, i.e. there’s more than one model that “correctly” describes the world because the method used to construct the model contains one or more free variables, hence even the idea of “correctness” becomes conditional on what was assumed. You might describe this as grokking the existence of a hyperprior (or universal prior) necessary to believe anything.
On a personal level this can be expressed as “I can be anyone”, by which I mean identity is not fixed by instead flexible and can be swapped out like a mask, and not just put on as if acting but totally replacing the mask that was previously being worn, because there is always some mask in place. At this stage belief also becomes more flexible, and beliefs become open to direct manipulation (as opposed to the indirect manipulation possible at earlier stages) because the network of beliefs can finally be adequately modeled within a single thought.
Like with stage 4, it’s possible to think one is thinking at stage 5 while really at stage 3 or stage 4 by similar mechanisms to the ones we’ve discussed around stages 3 and 4. The big confusions are mistaking being able to compress systems into objects for being able to think in terms of systems of systems without flattening them down, as in stage 3, and mistaking thinking of systems as being in relation to one another for having a coherent system of systems model into which everything fits, as in stage 4.
This stage lines up with Kegan’s stage 5 and Commons’ stages 14 and 15. Since I said stage 4 lines up with stream entry and thereby first path, stage 5 also matches up, in my estimation, with second path.
Stage 6: Ontological Fracture
At this point you may have already noticed the tik-tok pattern of the stages: the odd-numbered stages are about creating new structures to hold more complexity, and the even-numbers ones are about putting the previous structures into relationship with one another. So it is with stage 6, putting meta-systems into relationship with one another. In some ways this is less revolutionary than stages 2 and 4 because there’s already a lot of complexity that can be modeled in a single thought at this stage, but it can only go 2 levels deep right now. Stage 6 is the start of being able to have 3 levels of structure which means we’re starting to approach thought that can handle recursion without resorting to compression.
The insight for this stage is something like “everything’s connected”, but as stated that sounds a lot like the insights for stages 2 and 3. A better way to put it might be something like grokking that wherever you go, whatever you do, you’re always embedded in the system and that any appearance of Cartesian dualism or any distinction of any kind is an artifact of the map rather than the territory. I titled this stage ontological fracture because it takes the insight of stage 5 that there’s a free variable that enables flexibility and takes it one step further to crack apart the very idea that ontology is how the world is experienced.
At this point we’re well past the point where it’s easy to confuse this stage for an earlier one, so there’s lots of “near enemies” or “false friends” when I try to explain this in everyday terms without pointing to the type change already discussed. There’s are very stage 2 and 3 ways of talking that sound superficially similar to how someone at stage 6 might talk (e.g. “it’s all connected”, “interdependence”, “karma”, etc.), with the difference being that the stage 2 and 3 versions of the same stage 6 thought involve cached, flat, gearless models, but this presents a tricky verification problem because it’s going to be hard for someone at, say, stage 3 to tell apart someone at stage 3, 4, or 5 and someone at stage 6 saying superficially similar things.
This gets at a problem with developmental psychology that’s been popping up along with every stage, but becomes a bigger issue with each stage, which is that to someone at stage N, the wisdom of stage N+2 is nearly indistinguishable from someone spouting nonsense who is pretending to have wisdom. The only difference is that the stage N person can nail down the stage N+2 person to say coherent things in patches if you press them hard enough whereas the person spouting nonsense will keep spouting nonsense, but at the level of 5 or 10 words it’s going to be hard to tell these folks apart, and even with more words it still might be hard to tell them apart unless the stage N person is pushing for specific answers they can verify. The greater the gap between two people, the harder this verification problem becomes.
The standard solution is for the stage N+M person to shut up, and luckily stage 6 is the point where shutting up actually starts to make sense, because it involves thinking things that no longer neatly fit in language and instead can only be talked around and pointed at but can’t be communicated directly because there’s no distinction to be made, since stage 6 is the stage where the space between subject and object, emptiness, or the ontic start to fit into experience rather than being something that is there but can’t be experienced.
A personal expression of this stage’s insight might be “I just am” because there’s a first-hand experience of reality as it is outside the ontology necessary to reach it. Confusingly, this same sort of direct experiencing can happen at earlier stages without it pushing a person into this stage, and for a person below this stage that kind of experience will seem more like something special than ordinary. So this a kind of nothing-is-special, I-just-am thought that is only possible when the type signature of thought makes space for both the ontological and the ontic emptiness in which it arises.
Stage 6 also sees an increased ability to manipulate one’s own preferences as a result of being to model experientially what it is like to “choose” to do something. I put “choose” in quotes because doing this involves realizing there are actually no free choices, but rather that what we choose is a product of the mechanisms of the mind and body. Plenty of people understand this intellectually, so again this is about grokking this and then being able to work with preferences within your mind such that you might go straight towards doing things that will change them rather than trying to reshape them indirectly. Not everyone may hone this ability, though, and Cook-Greuter’s model (more on this model later) suggests some people at this stage will explicitly not have this ability because that model proposes that this stage has two different expressions, an idea I’m skeptical of but open to given the limited data we have about people at this stage.
Kegan doesn’t have this stage, and it partly matches up with Commons’ stage 16 although I think we could also read Commons’ stage 16 as overlapping with the high end of my stage 5. The only thing this really maps to in existing systems with Cook-Greuter’s stage 5⁄6 and Buddhist third path.
Stage 7: Ontological Fusion
This stage completes the model by expanding the type of thought to hold a system containing systems of systems. This grants three full levels of structural modeling allowing both things, systems, and systems of systems to be completely contained within a single thought and provide a surrounding meta-meta-system in which they be operated on. This is significant in that it provides enough structure to iteratively mentally model recursion over a series of thoughts without having to either compress a system into a simple thing or have the system being considered take up the entirety of the type, leaving no room for operating on the considered system itself.
The insight might be something like “everything’s perfect” or “nothing was ever separate to begin with”. A personal version of the insight might be something like “I am the world, yet the world is not me”. This leads to the stage often being expressed in terms of something like non-dualism, which here means fully realizing in every moment that the map is not the territory and that the map is embedded in the territory and never getting confused and thinking as if there were some Cartesian split between the model and the modeled. This enables a kind of perspective flexibility similar to but much expanded upon from the stage 5 level of flexibility, like going from being able to wear different masks to being able to wear every mask at once.
Because I believe this stage has a type signature that permits unrestricted iteration of recursive ontology, this would appear to be a reasonable final stage in that higher stages are theoretically possible but not necessary, their only benefit being able to do something like consider an extra level of recursion within the same thought, which is nice but not strictly necessary, since three “registers” is enough. However, I made a similar claim in PCC about stage 5 because at the time I didn’t have a grasp on what stage 6 and 7 would look like or that they even existed, so it’s possible I will turn out to be wrong here, but I will say I’m at least more convinced of my own argument for stage 7 being enough to think with arbitrarily complex ontology than I was by my arguments in PCC for stage 5 doing the same.
This stage is off-the-map for both Kegan and Commons, but lines up with Cook-Greuter’s stage 6 and Buddhist fourth path.
On this point about fourth path, that should mean that shifting into this stage constitutes enlightenment. I’m unsure. It’s possible, but I think it’s more likely there’s two things going on when we talk about enlightenment: persistent non-symbolic experiences (PNSE) and stage 7 thinking. You might call these the narrow and broad aspects of enlightenment, respectively, or not-self consciousness and cosmic consciousness (“cosmos” being from Greek “κόσμος” meaning order of the world). However, I’m to some degree speculating here because I’m not at stage 7 or enlightened, so although I think there’s a body of evidence pointing to there being these two aspects to what is traditionally thought of as enlightenment and stage 7 describes one of them, I’m not altogether convinced, and I could easily be misunderstanding, with stage 7 either being the same as PNSE or being totally unrelated to enlightenment.
Similar Models
I’ve already discussed some of the correlations to other models along the way of describing the stages, but I have a few additional things to say on the subject.
At a high level, you might say my model is something like an extension of Kegan’s model. At least, that is if you look at Kegan’s model as presented in The Evolving Self and not at the versions with deflated stages present in his later work with Lahey. Unfortunately Kegan’s model doesn’t really have gears, which was, in hindsight, much of what I was trying to provide when I wrote PCC. After writing PCC it was pointed out to me I had reinvented something similar to but definitely different from Commons’ model in that we use somewhat different gears but have a similar tact for finding them.
Both the Spiral Dynamics model and Wilber’s extension of it offer alternative developmental models that extend beyond Kegan and Commons. Unfortunately these models are somewhat vague and don’t really have the kind of gears I want, but they do basically line up with my model stretching from infrared to ultraviolet.
While writing this post, I learned someone else extended Kegan’s model to the same point I have, although she did it somewhat differently than I have and without gears. Thus my model also roughly aligns with Cook-Greuter’s Ego Development Theory, especially onward from Cook-Greuter stage 4⁄5, which mostly corresponds to my stage 4. Below this stage we differ somewhat, with Cook-Greuter making more distinctions than I do and being, in my opinion, overly specific (and therefore wrong by failing to account for a wider range of human experiences) about what is common at each of the preconventional and conventional stages.
As mentioned a few times, I think there’s also an alignment with the Buddhist four-path model of enlightenment, and, although not mentioned above, with the five ranks model. Whether or not this is true, though, depends a lot on how you choose to interpret those models. In particular the five ranks system is explicitly not a stage model (“rank” is a terrible translation, but it’s standard) but instead describes different perspectives or places from which reality can be experienced, and the only sense in which we can get stages out of it is by considering which of the ranks a person is able to see the world from. Despite this, I actually think there is something there, and thinking of this stage model as correlating with enlightenment has been helpful to my practice thus far, so I think there can be something skillful in making this connection.
There are many other developmental psychology models I could try to relate my model to, but I’ve limited myself to the ones I’m most familiar with. If there’s another one you know and love, feel free to try to work out the correlations and post them in the comments.
What Next?
So, we have this model. What to do about it?
As I say, I’ve mainly found it useful for modeling people better, but it’s also useful as a kind of rough guide to what is possible as a human. If you look at this or similar developmental models and place yourself around, say, my model’s stage 4, then you have some hints about what’s next for you if you apply some effort to psychological development and seek out practices that may help you work towards the next stage. Although you can’t go directly towards each stage (a longer discussion to perhaps have another time), knowing a next stage exists can motivate effort and keep you from becoming complacent with what you have already accomplished.
As for what’s next for this model itself, I don’t know. It’s possible I’ll want to come back in 3 to 4 years and refine it more again, so it’s definitely provisional in that sense, but not so provisional that I don’t think it’s useful.
As a general point, I think what’s important more so than even my specific model, which you may or may not find useful or understandable, is that you consider the value of a developmental model of mind that can move from less to more complex stages. There’s many models out there, and although they all try to describe roughly the same territory, they do it in different ways that may click for different people better than others or may be more useful in some contexts than others.
Thanks
If you read this all, thanks for putting up with it! I’d love to put in the time to explicate it more, but this is my 80⁄20 attempt to share the model so others who find it useful can know about it without putting in the hundreds of hours of work it would take to produce a polished work on the subject. I’m sorry that it was neither shorter nor longer.
I’ll do my best to clear up confusions or questions in the comments. I’m mainly interested in discussing the theory itself rather than developmental psychology in general, but I realize this post may be too tempting a target to not want to discuss the broader class of models in the comments.