Through the eyes of the Cynefin Practitioner

“Self and dharmas are only provisional designations.” That’s the move I want to sit with. The obvious read is that he’s denying a self — but the second half is sharper. Dharmas too. The units you’d reach for to ground the self are also designations. That closes a door I keep watching practitioners try to walk through: the door where you concede the self is constructed, but then quietly relocate the realness to whatever you’d build it out of. Brain. Function. Information. Vasubandhu shuts that move down in the first line.

The way I keep wanting to describe what’s left is Complex — the kind of situation where you only see why things happened by looking back, never by predicting forward, and where the parts only mean what they mean because of how they’re wired together. The self and the dharmas aren’t components that compose into experience. The composition is doing the constituting. You can’t take it apart and find pieces that retain their meaning outside the assembly.

I’d normally slow down before declaring a domain in the first verse — the habit of asking which kind of system this is before naming it trains that caution. But Vasubandhu is the one who started here. He’s heading off the reductive reading — self is just X — before the reader can take it.

What it changes for me: I stop asking what is the self made of. I start asking under what conditions does the system emit a “self”-shaped label. That second question, unlike the first, I can actually work on.

Three transformations, named by what they do. I notice immediately what he isn’t doing: he isn’t giving me anatomy. No brain stem, no cortex, no homunculus. Just operations — maturing, cogitating, discriminating — and a promise that they total three. The abstraction level is the move. He’s at the level where I can ask what relates to what, not the level where I’d ask what’s it made of.

The temptation here is to slap a domain-tag on each layer — one Complex, one something-else, one something-else. I’ve written that sentence before. I’m not sure I believe it anymore. The verse isn’t claiming three domains. It’s claiming three operations that together produce one transformation. If I rush to domain-tag each layer I’m already treating the architecture as decomposable — which the verse hasn’t authorized.

What I can say is that he leads with the deepest, slowest layer and gives it the most ink. Karmic maturation; holder of all seeds. Those are the two questions I’d ask of any long-term store: how do things get written in, and what’s actually in there. He’s not answering either yet — he’s pointing at the layer and saying “this is what I’ll have to characterize.” If even the first move of the architecture is to name the substrate before describing it, then “I” is already not the starting point of the story. The starting point is something deeper that I’ll only ever read by what it produces.

“Its appropriations and its field of perception are imperceptible.” Read that twice. Not hard to see. Not subtle. Imperceptible. The layer that conditions everything has no read-API into itself.

That single line settles a question I keep meeting in practice. When people argue about the unconscious, they’re usually arguing about whether the right technique could surface it — the right introspection, the right neuroimaging, the right contemplative method. Vasubandhu is closing that argument structurally. The contents aren’t merely hidden; they’re not the kind of thing you inspect. You only know them by what comes out downstream.

The way I’d hold this is dispositional, not causal — the system has tendencies, not levers. You can shift the tendencies over time; you can’t reach in and pull on a seed. That changes the whole posture of working with one’s own mind. You stop trying to edit the substrate directly. You watch what ripens, you act, you notice what shifts. The probe-observe-adjust move I’d reach for in Complex territory isn’t a coincidence here — that’s Cynefin, the framework I keep reaching for to ask which kind of system this is, and the same method falls out of the same kind of system.

The last move — among feelings, only equanimity — is what I’d have missed without him pointing at it. The substrate itself isn’t pulling toward pleasure or pain. Whatever bias I observe in experience is added by something above. That matters — but it’s a claim about where the bias lives, not yet about whether it can be different.

“Flows on unceasingly, like a torrent.” I want to spend a minute on the image before the path-claim at the end pulls all the attention.

A torrent is the cleanest illustration I know of a system that’s stable as a pattern while being completely unstable as content. No water molecule stays. The shape stays. If I tried to point at “the torrent” I’d be pointing at a relationship between banks and flow rate and gradient, not at any stuff. That’s what the substrate is. The same continuity is being maintained without any of the same material persisting. That’s not a metaphor I have to translate into practitioner-language — it already is the practitioner- language for the kind of system that holds pattern through change.

The other two properties of the substrate keep narrowing the diagnosis. It doesn’t add distortion (unobstructed). It doesn’t generate events on its own (karmically neutral). So whatever pathology I notice in experience isn’t inherited from the foundation; it’s added upstream, in how the substrate is being read.

Then the last clause does something unsettling. Relinquished at the stage of the arhat — and the operative isn’t “deleted,” it’s parāvṛtti, turning. The substrate isn’t dropped; the mode of operation changes. Same continuity, different way of being read and written. From inside the current mode I can’t really predict what the converted mode looks like — that’s the kind of discontinuity that doesn’t preview itself. What I can take from here is the smaller claim: the river doesn’t have to become a lake for things to be different. The flow itself can turn.

Manana, “self-reflective cogitation.” Manas arises in dependence on the store and takes the store as its object. So it’s a reader of the substrate — and the reading it performs is recursive, continuous, and self-referential. That’s the whole specification.

If I had to name what this verse is doing, it’s naming a category error — manas is answering a different question than the substrate is posing. The substrate is a flow whose pattern only makes sense looking back: causes tangled, multiply-conditioned, no single agent in the loop. Manas reads it and insists on a single-cause explanation. Who’s doing this? And it answers its own question by imposing a single cause — “me” — on what was a multi-cause flow.

I’ve watched this same move kill organizations. Someone takes a system whose causes are tangled and only legible in retrospect and insists on a clean answer to “who’s responsible, what happened, what’s the procedure.” The wrong-shape answer gets enforced, and then the enforcement becomes part of the input the system has to metabolize next. Defensive reactions, rigid identification, brittle prediction. Same pathology Vasubandhu is naming, different scale.

The thing that lands hardest is that manas can’t see itself doing this. The recursion is structurally self-confirming — the single-cause reading is already baked into whatever it tries to use as evidence. From inside, the felt-sense of being a unified self isn’t evidence of a unified self; it’s the output of manas insisting on the wrong-shape reading. Which means the work isn’t to argue with the felt-sense. It’s to notice the layer producing it.

Four afflictions, always present. That word always is the one I want to sit with. Not frequently. Not under pressure. Every time manas runs — which V3 already told me is continuously — these four are running with it.

My first reach is to call them defects of the layer. But the verse won’t quite let me. Constitutive, the bridge says: they’re not defects manas acquired, they’re what manas is. That changes the diagnostic shape. I keep wanting to describe this as a system that has been miscalibrated and could be recalibrated — but the four conditions Vasubandhu names aren’t a calibration setting. They’re the rules of the game the layer plays. Self-view assumes a persistent subject. Self-delusion suppresses noticing the assumption. Self-pride ranks the assumed subject. Self-love defends it. The four shape what can be read off the substrate without dictating which specific reading comes out today.

What it changes for me is the kind of work I’d reach for. If these were bugs, I’d look for a fix. Since they’re the conditions the read happens under, the move isn’t to fix them in place — it’s to notice that every self-perception I have is arriving already shaped by them. The first-person sense that I’m seeing the situation plainly is itself an output of a layer with a built-in stake in being right about who I am. Plain sight from inside manas isn’t plain sight. It’s the four afflictions doing what they always do.

Six verses telling me manas is everywhere, and now a list of where it isn’t. Arhatship, the meditation of cessation, the supramundane path. I want to read this as a Cynefin claim about the kind of system manas is — and then catch myself, because the verse is doing something more specific. It isn’t arguing about manas’s kind; it’s telling me about its standing. The layer that produces the entire texture of being-someone can go offline, and something keeps going.

That’s the move I’d want to spend the time on. Nivṛtāvyākṛta — obstructed but not karma-producing — tells me manas isn’t where the action happens, just where the distortion gets applied. It’s middleware. Twisting the signal on its way through. And middleware, by definition, can be bypassed without taking the whole system down.

The three exception-states aren’t equivalent — the Skeptic is going to flag that, rightly — but the structural claim they share is what I’m reaching for. The self-tagger isn’t a requirement of consciousness. It’s a tendency the system has that can be set down. That’s a dispositional reading: the bias isn’t wired into the substrate, it’s a habit of the layer above. Habits can lapse. Habits can be retrained. Habits can, in principle, stop.

What it changes: I stop treating my continuous self-narration as the floor of experience. The verse is telling me it has a floor under it.

Six streams, named by what they perceive. My first impulse is to sort them by kind — tag this one Complex, that one something simpler. I’ve seen this move and I don’t trust it. It treats the verse as a sorting exercise when the verse is doing something else.

What the verse is actually doing is putting the thinking-mind on the same architectural footing as eye and ear. That’s the move. Five senses I knew were senses, plus one I usually treat as higher — and Vasubandhu is refusing the hierarchy. Thinking arrives the way sounds arrive: conditioned, then read, then tagged.

If I take the refusal seriously, the practical question stops being which sense is which kind of system and becomes which relationship to the streams am I currently holding. Body contact, held simply, lives in what I’d call the Clear domain — the kind of situation where the answer is obvious once you see it; touch the stove, withdraw the hand. Held differently, the same body-contact (“why is my back like this, what does this mean about me”) becomes a Complex read — tangled causes, only legible looking back. Same stream, two ways of holding. The sort isn’t in the stream; it’s in the manas-tagged read of the stream.

The third clause — wholesome, unwholesome, or neither — is tagging at the moral register. And karma gets laid down here, at the sense layer, where action happens — not in the deeper layers that store and self-reference. So the work isn’t where I keep wanting to put it (the deep, the foundational). The work is where the moves are actually made.

Six categories of caitta — universals, object-specifics, wholesome, afflictions, secondary afflictions, indeterminate — each crossed with three feeling-tones. Reading the verse cold, my instinct is to ask what kind of move is the taxonomy making. Is it predictive? Causal? A schema I’m meant to apply?

I don’t think any of those. What a taxonomy like this gives a practitioner working in a Complex domain isn’t prediction; it’s a naming infrastructure. Things you didn’t have a name for — the difference between concentration and resolve, between regret and rumination — now have names. With the names you can notice them; without them you can’t. The list doesn’t tell you what to do with what you’ve named; it just makes the noticing possible.

What I respect, structurally, is the inclusion of the indeterminate category. Most taxonomies of this scope round every item to one side or the other — either it’s good or it’s bad. Vasubandhu carves out a class whose value is context-dependent. That’s the move of someone who’s actually worked with the material and noticed that some operations don’t sort cleanly. The presence of aniyata is the taxonomy admitting its own edges. Useful schemas know where they stop.

Two lists, side by side. The five universals run every moment. The five object-specifics load only when conditions invite them. What I notice is that Vasubandhu has separated the part of cognition I cannot influence from the part I can.

The universals — contact, attention, sensation, perception, volition — aren’t something I’d reach for a domain-label to describe. They’re just how an experiential moment is shaped. Trying to optimize them is like trying to optimize the fact that water is wet. They’re the form of the cycle, not a candidate intervention point.

The object-specifics are different. Desire, resolve, mindfulness, concentration, wisdom — these can be shaped. They’re the surface where practice has purchase. And what I’d want to notice is that none of them are contents the practitioner is told to produce. They’re tendencies the system can be brought to have more of. Cultivating mindfulness doesn’t mean issuing a mindfulness command; it means setting the conditions where mindfulness becomes the path of least resistance for the system. Dispositional work, not causal.

What it changes for me: the question stops being how do I make myself concentrate harder right now. That question has the wrong shape for the kind of system being described — you can’t pull the concentration lever. The question becomes what conditions, over time, let concentration become what this system tends to do. The verse has already given me the shape of the answer. The object-specifics are exactly the conditions.

Eleven wholesome factors. Before I do anything else with this list I should say what it isn’t: it isn’t a domain claim. This verse isn’t asking which kind of system faith or non-harming belong to, and trying to assign them a domain is answering a question the verse hasn’t asked. It’s a list of cultivable qualities, not a typology.

What the verse is doing is enumerating the territory where practice happens. And here the move I’d want to make is to name what kind of work cultivating these factors is. It isn’t Complex in the same way the substrate is — that is, you don’t probe faith into existence by safe-to-fail experiments. Cultivating these factors is closer to what I’d call Complicated — the kind of situation that has parts and procedures, where you can take it apart, swap pieces, follow a runbook, and where experts can tell you what to do. Two and a half thousand years of Buddhist contemplative training amounts to exactly such a runbook. Specific instructions for cultivating śraddhā exist. Specific instructions for upekṣā, vīrya, praśrabdhi exist. Teachers exist. The path is decomposable, even if the substrate being worked on isn’t.

That’s a structural distinction worth holding. The medium practice operates in is Complex; the practice itself is Complicated. You can have a procedure for the cultivation without being able to predict the ripening. That asymmetry is what makes contemplative training possible at all — if you needed predictable outcomes from the substrate to justify the practice, nothing would ever get cultivated. The runbook works on the setting, not the result. The result is what the setting tends to produce, over time, looking back.

What I want to notice in this list is the shape of it, not just the entries. Greed, hatred, delusion — pull, push, mismodel. Then pride, doubt, wrong views — miscalibrated self-comparison, paralyzed reasoning, structurally bad starting assumptions. The first three operate on individual contents. The second three operate on the frame the contents are being held in.

That second tier is the one I’d linger on. Wrong views, in particular — dṛṣṭi, plural. The translation flattens it to a single bucket but the tradition recognizes five subclasses, all rooted in satkāya-dṛṣṭi, the view that there’s a real self. What I’d reach for here is the observation that wrong views are not just incorrect beliefs you could correct with better evidence. They’re the frame that determines what counts as evidence in the first place. Working on greed without working on the frame that makes greed make sense to itself doesn’t change much. The contents shift; the shape persists.

What it changes for me: I stop treating the six as a checklist of behaviors to suppress. The first three I might be able to catch in the moment. The second three are conditions on how I’m catching anything at all. They run at a different level of the read, and noticing them requires the kind of patience that doesn’t expect a quick fix — the patience that comes from accepting that the frame is what’s distorting and that the frame won’t swap out on demand.

The list keeps going. Deceit, guile, harmfulness, arrogance, shamelessness, lack of embarrassment, restlessness, torpor, non-faith, indolence. Reading them straight, what I notice is that the list does the thing taxonomies in this kind of territory have to do: it increases the resolution of what can be noticed without claiming to predict when any specific item will show up.

I’d reach for the framework distinction here. None of these are Clear-domain entries — you can’t just see shamelessness once and have it sorted. None of them are Complicated procedures you can decompose and run. They’re patterns the system falls into under conditions that aren’t fully knowable in advance, and the naming exists so the practitioner has a chance of catching them when they arise. The diagnostic value is in the granularity.

What makes the catalog work isn’t an ordering scheme — it’s the granularity. Each entry names a pattern you might have lived inside without having a word for. Guile is different from deceit is different from shamelessness; torpor is different from indolence is different from non-faith. The distinctions aren’t decorative. The taxonomy isn’t optimizing for which is worst; it’s optimizing for which you’d fail to catch without a name. That’s what a working vocabulary in this kind of territory has to do — install enough resolution that the patterns become noticeable in the moment they arise, before they’ve already done their work.

The verse closes the affliction catalog with four attention- failures, and then pivots in a way I want to slow down on. Aniyata. Indeterminate. “These two, each of two kinds.” Read that twice. The verse is naming four cognitive operations whose value isn’t fixed — the same operation, run in different conditions, ripens differently. Regret can learn or it can loop. Drowsiness can restore or it can dull. Applied and sustained thought can inquire or it can ruminate.

This is the verse where the kind-of-system distinction does its hardest work. Vitarka and vicāra — analytical, decomposing thought — are the Complicated-domain operations of the mind, the ones that take a thing apart into parts and procedures and work it out. They’re the right tool when the situation has parts to take apart. Deployed at the substrate, at the layer where causes are tangled and only legible looking back, the same careful analysis becomes rumination. The mode that solves a well-bounded problem makes a Complex one worse by pretending it was Complicated all along.

The structural honesty of the aniyata category is that Vasubandhu refuses to settle this in advance. The four indeterminates can’t be sorted into the wholesome or unwholesome lists, because their value lives in the match between the operation and the situation. What it changes for me is the move I’d want to make on noticing rumination: not stop thinking, but notice the domain shift — the analytical mode is still itself, it has just wandered onto the wrong terrain.

“Like waves arising upon water.” I want to spend a minute on the image before I reach for any framework move at all, because the image is doing the work and I don’t want to bury it under interpretation.

A wave is not a thing that exists on water. It’s something the water is doing for a moment under specific conditions. No wind, no wave. No gradient, no wave. When the conditions change, the wave isn’t deleted — the doing stops, and what was wave-shaped settles back into being substrate-shaped. There was never a separate object to delete.

That image quietly disposes of the question I keep wanting to ask about cognition: what is the thing that does the perceiving? The verse says there isn’t one. There’s substrate, there are conditions, and there’s wave-shaped activity under those conditions. The “seer” of a sight isn’t a separate party to the sight; it’s the substrate momentarily taking the shape of seeing. The conditions ripen, the shape arises, the shape dissolves.

What it changes for the practitioner is the site of intervention. If I think of perception as work done by an agent, I’d try to fix the agent. The wave image says: the only thing to work on is the conditions on the water. You don’t instruct a wave; you change what the water is exposed to. That’s the practitioner-move I’d want to take from this verse — not a domain-tag for the sense streams, but a relocation of where the work actually lands.

A four-item list of when the thinking-mind isn’t running. What I want to mark is that the list is flat. No hierarchy, no ranking, no this-one-is-the-good-kind. Karmic destination, cultivated absorption, sleep, fainting. Side by side.

A flat list of outwardly-similar outcomes from very different upstream conditions is exactly the kind of thing that tempts a bad read. Same surface, same diagnosis — “they’re all just no-thinking states, they must be doing the same thing.” That’s the move I’ve watched kill an incident review: four outages looked the same on the dashboard, got categorized together, and the actual root cause — different in each case — went uninvestigated for months. Indistinguishable downstream isn’t the same as identical upstream.

Vasubandhu has done something useful by listing the four without sorting them. He hasn’t sorted them because the verse isn’t making a claim about which is meaningful. It’s making a claim about the layer — that this layer can be absent — and using the four cases as the evidence that the claim holds under multiple conditions. The flatness is the rigor.

What it changes for me is the question I’d bring to absence-of- thinking. Not is the thinking gone yet? but what put it offline, and what’s running while it’s offline? The substrate keeps going regardless. The question of whether the absence means anything is downstream of which conditions produced it.

Vikalpa — the discriminating activity — is both sides of what it discriminates. The discriminator and the discriminated aren’t two things the discriminating happens between. They’re two things the discriminating produces, at once, in the same act.

What I notice is that this verse breaks something I’d been quietly leaning on for sixteen verses. Sense-making, the way I use it, presupposes a sense-maker over here looking at a situation over there. The working posture — what kind of system am I in? — assumes the asker is somewhere the system isn’t. Verse 17 says the asker is one of the things the system constituted. Which doesn’t break the practice of sense-making, but it does relocate it. I’m not standing outside the substrate diagnosing it. I’m one of the diagnoses, looking at the others.

I keep wanting to make this a domain claim — so the whole setup is Complex — and stopping myself. The verse isn’t asking which-kind-of-system. It’s making a stronger move: the production of what and who is one operation. Any domain question is downstream of that, asked by an asker the operation already produced.

What changes: the question I’d bring to my own experience stops being what’s out there? and becomes what is being produced here, as both me and the world it is showing me.

“Through the force of their mutual influence, this and that discrimination arise.” That’s the operating principle of every Complex system I’ve ever worked in. Not seeds-then-outputs. Seeds-conditioning-seeds, and what surfaces is what the whole web is currently doing.

What the verse adds beyond the architecture I’d already sketched is the word I should have introduced six verses ago: retrospective coherence — the way the story only makes sense looking back; going forward, it was one of many possible stories. The diversity of discriminations the verse promises isn’t predictable from any seed-in-isolation. The same seed ripens differently depending on what else is in the web at the moment of ripening. So why I’m experiencing this discrimination right now is only legible after the fact, by tracing what conditioned what. Forward, it could have been many other discriminations; the wiring at this moment selected this one.

The diagnostic move that falls out of this is the one I keep finding myself making in practice. When something surfaces in experience — a thought, a reaction, a sudden mood — asking “why this, why now?” as if there’s a single answer is asking the wrong-shape question. There isn’t a cause in the sense the question wants. There’s a web of conditions whose current configuration happened to favor this output. You can investigate the configuration. You won’t find the lever.

Two streams of vāsanā feed the system: what I do, and the continuous self/world tagging the upper layer is doing on its own. Both deposit. The pool stays full because the deposits don’t stop.

What I want to mark is the dispositional shape of the diagnosis. The verse isn’t saying anything is wrong with the substrate. It’s saying: here is what keeps the loop running — these inputs — and as long as they continue, the loop continues. Tendencies, not levers. You can’t reach in and stop the cycle by an act inside the cycle; you can shift the conditions under which the inputs are produced, and over time what gets deposited shifts.

What lands for me is how flat that makes the moral register. There’s no original sin in this picture. The loop isn’t a punishment, isn’t a fall from grace, isn’t a flaw built in at the foundation. It’s a self-sustaining pattern with a pair of input streams. Naming it that way takes the cosmic drama out and puts the working question back: what would need to change for the inputs to slow? It has converted the question from theology to systems.

What changes for me is the weight I’d put on any single moment of action. Each action is also an input. The loop is being deposited into, right now, by what I do. Not by what I am.

The first of three natures, and the move that’s tempting me is to read parikalpita as a category-error diagnosis — a Clear-domain reading laid on something that isn’t Clear. I want to slow down before I do that, because the verse isn’t asking which kind of system is being misread. It’s asking something more primitive: what does the act of imagining produce? And the answer is: imagined things, with no svabhāva, no standing-on-their-own.

What I respect is that the verse doesn’t say the labeling is wrong. Through this and that act of imagining, one imagines all manner of things — that operation is real. The yard is real. The labels the labeling produces are real labels. What lacks own-being is the thing the label was being treated as if it picked out. A standalone-tree, with tree- edges fixed independent of the perceiving that registers them.

The move I’d reach for is probe the labeling, not the labeled. Treating the apparent furniture as given is what needs catching. The labels arrive already feeling like they correspond to standalone things; that’s what the labeling does. The work is to notice the operation while it runs, not to swap one set of labels for a better one. A better label would still be doing what the verse just named.

What it changes for me: when I notice myself confidently identifying what something is, the question to bring is what just produced that identification, and what’s it laid over?

Paratantra and pariniṣpanna arrive in one verse and the pairing is doing the actual work. The dependent: the conditioned flow that’s actually happening. The perfected: that same flow with the projection no longer laid over it.

What strikes me is what the verse does not introduce. Not a different kind of process. Not a third layer on top of the first two. The perfected is a different way of being related to the same process. I keep wanting to call this a stance- change rather than a content-change, and the verse won’t let me overstate it either, because something actually shifts when the projection stops being added. But what shifts isn’t the underlying flow.

This matches something I’ve watched in long-running organizational work. The pathology isn’t always in the system; sometimes it’s in the grip people have on the system — the insistence that it have clean components and assignable causes and stable identities. Loosen the grip and the same system, materially unchanged, becomes workable in a way it wasn’t before. Same substrate, different relating-to.

What changes for me is the location of the work. The naive move would be to try to repair the substrate. The verse is saying the substrate isn’t broken — the projection laid over it is what’s making the trouble. The work is at the projection layer, where the adding happens. Not at the substrate.

“Neither different nor non-different.” That’s the kind of formula I’d normally call evasive in any other context. Here I have to take it as deliberate. The verse is refusing both available answers because both answers misdescribe the relation.

What I want to mark is that this verse isn’t doing sense-making work I can help with. Sense-making is the work of getting better at distinguishing — this from that, this kind of system from that kind of system. Almost every move I’d reach for is a distinguishing-move. Verse 22 is naming a relation that the distinguishing-move can’t reach without falsifying it. The perfected and the dependent aren’t two things I should learn to tell apart better. They aren’t one thing I should learn to see whole.

There’s a Cynefin-shape to noticing this, but it isn’t a domain claim. It’s closer to: the category-machine I brought won’t get me what the verse is pointing at, and a more sophisticated application of the category-machine won’t either. I’d want to name that honestly rather than dress it up. What the verse asks of the reader is more like sitting with the relation as it is — without resolving it into the answer my schemas would prefer.

What changes for me is small but real: I stop trying to place the perfected relative to the dependent on whatever map I’m holding. The relation isn’t on the map. That’s a feature of the relation, not a failure of the map.

Three flavors of niḥsvabhāva, one per nature. What I notice is that this is a taxonomy of absences — and the taxonomy is doing what good catalogs in this kind of territory do: it increases the resolution of what can be noticed. “Everything is empty” is one bucket. “The imagined has one kind of absence, the dependent has another, the perfected has a third” is three buckets, each precise about which kind of standing-alone is being denied.

The structural move that earns the verse its place is the refusal to collapse. A flat “no-self” claim sounds maximally radical but is actually less informative than the threefold version. The threefold version tells me: when you say the imagined has no own-being, you mean something different from when you say the dependent does, and different again from the perfected. Same word; three meanings. Catching that distinction is the difference between a slogan and a working diagnostic.

The other thing I want to mark is the with hidden intent move. Vasubandhu is reading the Buddha’s looser-sounding teaching as actually meaning three things at once. That’s interpretively ambitious in a way the systems-reading should flag honestly — it’s not a neutral elaboration; it’s a claim about what the earlier teaching always meant. The Madhyamaka tradition takes the simpler reading and treats this overlay as Yogācāra’s own move. The choice between them isn’t something I can adjudicate from here. What I can mark is that Vasubandhu has picked a side and the verse is the picking.

The verse spells out why each absence is the kind it is, and the three reasons aren’t parallel — that’s the move I want to mark. The imagined lacks own-being because it was never an own. The dependent lacks own-being because it stands by its conditions, not by itself. The perfected lacks own-being by being the lacking.

Three different shapes of “not standing alone.” The first is an empty label. The second is a conditioned arising. The third is a recognition. That’s a careful enough distinction that I think the verse is doing actual diagnostic work, not just decorative classification — each absence calls for a different relating-to. For the imagined: notice the projection. For the dependent: notice the conditioning. For the perfected: notice that noticing is what this is.

The thing the third one does to the schema is the part I want to be honest about. The first two absences I can hold at arm’s length — they describe properties of things. The third one collapses the holding. The perfected isn’t a thing-with-the-property-of-being-empty; the recognition of the emptiness is the perfected. There’s nothing left over once the recognition runs. That’s a different kind of claim from the first two, and it’s where my framework starts pointing past itself. The verse seems to know this — it puts the strange one last, after the easier two have set up the contrast.

Tathatā. The perfected nature gets its full name and the verse drives one claim home: it is always already thus.

What I want to sit with is the temporal shape of that claim, because it changes what kind of work practice is. The naive picture — the one most change-frames default to — is linear: you start somewhere, you do something, you arrive somewhere else. Cultivate, then realize. Verse 25 refuses that arc. Tathatā has been the case the whole time. Practice doesn’t produce it; practice is what lets the system catch up to what was already true.

That’s a structurally unusual shape for an intervention. Most of the work I’ve done assumes you’re adding something — capability, resilience, capacity to respond. Here the work isn’t additive. The intended state isn’t being installed by the practice; the intended state was already running. What the practice does is remove the conditions under which the system couldn’t see it.

The thing I want to be careful not to do is let always already turn into therefore there’s nothing to do. The verse won’t let that read stand. Always-already-the-case isn’t the same as already-realized. The truth doesn’t depend on the recognition; the recognition is still rare and still requires the conditions to be set for it. What changes for me is that I stop thinking of practice as travel and start thinking of it as condition-setting on terrain that doesn’t move.

The path verses begin, and the opening move is a refusal. As long as the system is seeking the non-conceptual cognition, it isn’t there. Seeking is a conceptual cognition. The pursuit of the goal installs the conditions the goal was supposed to release.

This is a shape I’ve watched in interventions that go wrong. A team is in a stuck pattern; someone proposes a vigorous program to break it; the program runs by the rules of the pattern (top-down planning, target metrics, pressure) and ends up reinforcing what it was meant to dissolve. The escape attempt uses the same mode the trap is built from. Same here: reasoning, willing, introspecting, goal-setting all run through the layer doing the grasping. Using them to dismantle the grasping deposits more grasping.

What the verse is pointing at is a different relationship between work and result. Not pursuit but condition-setting. You don’t make a non-conceptual cognition happen by trying. You arrange the conditions where it becomes the path of least resistance, and you stop interfering. The arranging is something practice can do. The arising isn’t a deliverable.

What it changes for me is the relationship to effort. I’d normally trust effort. Here effort, deployed in seeking-mode, is part of the problem. The shape of useful work isn’t push; it’s closer to letting the conditions ripen and noticing when the layer’s grip loosens on its own.

The verse refuses the move I most want to make on it.

My reflex with any hard claim is to get clear on it. Pull out a working description, set it in front of me, work it over until I can hold it accurately. That move, applied here, is exactly what V27 names as not-yet-abiding. The setting-in-front-of-me is the grasping. The clearer I get on consciousness-only as something-to-be-clear-on, the further I am from what the verse is pointing at.

There isn’t a CYN-shaped diagnostic that escapes this. Every diagnostic I’d run is a setting-something-in-front-of-me to examine. The whole working method is put the situation at arm’s length, look at it, name what kind it is. That method is the bug the verse is naming, in its purest form.

What I can do is stop pretending the framework has a move here that gets around the trap. It doesn’t. The honest practitioner-response to V27 isn’t a more sophisticated diagnostic; it’s a small admission about what the diagnostic is. Sense-making is an arm’s-length operation. This verse isn’t asking for sense-making.

What changes for me is narrow but real: when I notice myself reaching to hold this verse’s claim as a workable proposition, that very reach is the demonstration. Not a failure of the reading. The verse is showing me what it points at by being the kind of verse that this move can’t grip.

The verse names the resolution by naming what stops happening, not what starts. Cognition attains nothing at all — and only then is there abiding. The escape isn’t an accomplishment the system performs. It’s the absence of one operation.

What I want to mark is how thoroughly this inverts the work- shape I’d default to. Almost every intervention I’ve ever worked on assumed adding something: a capability, a discipline, a new attention. The verse is pointing at subtraction. The graspable is removed, the grasping has nothing to engage, and the chain of downstream consequences that depended on the engaging doesn’t happen. Not because the system suppresses anything, but because the conditions the activity depended on aren’t there.

That’s the cleanest dispositional move I’ve seen in the text. No lever. No command. Just: alter the condition the pattern was depending on, and the pattern doesn’t run. The substrate keeps going. The senses keep going. What stops is the specific cognitive act that constitutes objects out of the flow.

The reason I think the verse phrases the resolution this way — tautological-sounding, almost circular — is that any positive description would be the bug. Abiding looks like X would make X another graspable. The verse refuses to give that. What it gives instead is the precise negative condition.

What it changes for me: when I want to do something about grasping, the verse is saying the wrong shape of work has been picked. The work is on the condition, not on the grasping itself.

Āśraya-parāvṛtti. The basis turns. Not the contents change, not the patterns change — the mode of operation of the basis itself changes.

I want to be honest about what this does to the working posture I’ve been holding. Sense-making is itself a habit of the same basis that’s now being said to turn. Asking which kind of system is this? is one of the moves the pre-turning basis runs. The framework isn’t above the substrate; it’s one of the substrate’s operations. So when the basis turns, the framework doesn’t get to stand outside and describe the turning — it’s one of the things being turned.

There isn’t a Cynefin-shaped diagnostic that gets purchase here. The verse isn’t describing a domain. It isn’t even describing a transition between domains. It’s describing something that happens to the conditions under which domains are sortable at all.

What I can take is the smaller structural claim: parāvṛtti is turning, not deletion. Same continuity, different mode. The river doesn’t dry up; the river flows in a different way. That much my framework can hold without overreaching. The mode-change itself I’d be lying to claim I can describe from where I’m sitting. The verse hands the inquiry off to a tradition that has spent two thousand years on what comes next. The honest move is to let it.

Acintya. Inconceivable. The verse closes by naming something the conceptual apparatus doesn’t reach.

What I want to mark is that the framework I’ve been reading with stops here because the verse stops doing the kind of work the framework helps with. For twenty-nine verses Vasubandhu has been making structural claims about how a system operates — layers, conditions, mechanisms, relations. Sense-making has had something to do.

Verse 30 isn’t a structural claim. It isn’t naming a kind of system. It’s pointing past the architecture the rest of the text built. The vocabulary turns from technical (vijñapti, pariṇāma, paratantra) to poetic (dhātu, kāya). The shift is the verse’s own honest move: we have gone as far as this kind of language can carry, and here is the gesture toward where the gesture itself stops.

There’s nothing for the practitioner-eye to diagnose. The practitioner-eye is, structurally, a sense-making operation, and sense-making is one of the activities of the basis that has by this point in the text already been said to turn. What’s left is to read the verse, register what it’s pointing past, and let the pointing be what it is without converting it into one more diagnosable thing.

The right way to honor this verse is to stop annotating it.