Chinese, English, and the orienting bridge are always visible. The five
lens commentaries are hidden by default — toggle them on globally
using the bar below, or open one at a time with the chips under each verse.
Prefer one verse at a time? The
per-verse view
is still here.
Or read the whole text through a single persona's eyes — pick a
voice in the switcher below. Each persona's page is their thoughts
while reading, not commentary about it.
Because “self” and “dharmas” are only provisional designations,
the various appearances that arise and evolve
all rest upon transformations of consciousness.
These transforming consciousnesses are only three.
The opening move is structural and linguistic, not metaphysical.
Vasubandhu does not say self and dharmas do not appear; he says they
appear as provisional designations (upacāra) — labels the
appearances have to carry, not entities the appearances have to refer
to. Both terms matter. “Self” is the obvious target, but “dharmas”
— the basic ontological elements of Abhidharma analysis —
is the more radical inclusion. Even the units you would reach for in
order to ground “self” are themselves upacāra. What is left? Only
the transformations of consciousness from which the designating
arises. The remaining 29 verses unpack what those transformations
look like, layer by layer; this verse tells you how many there are
(three) and what they are not (independent things).
memory
The Distributed Systems Engineer
DSE
Read this as a claim about identifier provenance. “Self” and
“dharmas” are not entities the system points at — they are
labels produced inside the system. The verse refuses the naïve
architecture where a self sits at the keyboard and a world full of
typed objects sits on the other side of a serial port. Both
endpoints are records emitted during the same continuously running
transformation.
The closest engineering analog is the move from a client/server
mental model to an event-sourced one. The “client” and “server” are
not physically prior to the events that mention them; they are
reconstructed by replaying events. Vasubandhu is opening with the
strongest version of this: not only is the self derived from its
own records, so is every type it would reach for to describe what
it is encountering. The pipeline preview — only three
— is also DSE-friendly: storage, indexing/cache, and the
user-facing read layer. The next 19 verses are this preview spelled
out.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
This is a domain claim made up front and unambiguously. Vasubandhu
is asserting that the phenomenology of subject and object —
self and the typed ontology you would reach for to talk about
self — is Complex, not Complicated. There is no
decomposition into a self-component plus a dharma-catalogue that,
wired correctly, yields experience. The pieces are constituted by
the wiring. You only see what they mean in retrospect, after the
transformation has run.
Mistaking the Complex for the Complicated — treating
experience as if it had separable parts you could swap out —
is the move Cynefin warns gets you killed. Vasubandhu, in verse 1,
is preempting it. He also preempts the Clear-domain reduction
“self is just X” (brain, soul, function, role) by naming both terms
— self and dharmas — as upacāra in the same breath.
You cannot escape the Complex domain into a Clear one by pointing
at a smaller unit; the smaller units are conventional too.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
Predictive processing makes the same move when it treats the
self/world boundary as a prediction the brain generates, not a
feature of the input. The cortex models “external world causing
sensations” because that model minimizes free energy — not
because it has independent access to an external world. Anil Seth’s
“controlled hallucination” phrasing is verse 1 in a different
idiom; Metzinger’s Phenomenal Self-Model is the same move applied
specifically to “self.”
What the new wording sharpens is the linguistic angle. “Self” and
“dharmas” are designations — how the model labels its
outputs, not metaphysical commitments the model makes. The brain
doesn’t decode pre-existing categories of self and dharma; it
constructs them as the posterior of a generative model whose priors
are themselves shaped by previous self/dharma inferences. Recursive,
with no ground floor. That recursion is exactly what pariṇāma
names.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
Whitehead’s actual occasions of experience are events, not
substances; a “subject” is a society of such occasions, not a thing
they happen to. Verse 1 is process-philosophical to its core: it
asserts that the apparent furniture of the world — me here,
dharmas there — is dependent on a more fundamental flow.
What the new translation makes more explicit is the language of
furniture. “Self” and “dharmas” are names the flow gives to its
own patterns. In Whitehead’s vocabulary they are abstractions from
concrescence, not the concrescence itself. In Bergson’s they are
the spatialized representations the intellect superimposes on
durée. In Heraclitus, the river is the standing example —
and the people who name it “the same river” twice are themselves
the larger river the metaphor is reaching for. You do not step
into the same self twice.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
Notice what the verse does not say. It does not say self and
dharmas are illusions. It does not say there is no world. It does
not say everything is mental in some idealist sense that the
Cartesian could recognize. It says they are upacāra —
provisional designations — inside a transformation. The
technical force of upacāra is “metaphorical, derived, secondary
use” — not “untrue.”
Three traps:
The systems-engineer reading is useful but leans on a metaphor
(event sourcing) with a physical substrate. Vasubandhu isn’t
telling you what the substrate is. Resist filling in “the brain”
or “physics.”
The “controlled hallucination” framing in cognitive science
still posits a brain doing the hallucinating. Verse 1 doesn’t.
“Provisional designation” gets compressed into “useful fiction”
by Western readers. Upacāra is sharper: the appearances are
real, but the names you reach for to describe them carry
implications the appearances don’t carry. The verse is a claim
about labels, not a claim against the world.
Namely: the maturing (vipāka), the cogitating (manana),
and the consciousness that discriminates objects.
The first is the store consciousness (ālaya-vijñāna):
it is karmic maturation and the holder of all seeds.
Verse 2 lays down the architecture the rest of the text will spend
twenty-eight verses unpacking. It names the three transformations
with their technical terms — vipāka (maturing), manana
(cogitating), and the consciousness that discriminates objects
— and then immediately glosses the first: it is the
ālaya-vijñāna, the store consciousness, the layer that does
the karmic ripening and holds all the seeds (sarvabīja). The order
matters: vipāka is named first because it is the deepest, slowest,
most invisible layer — the one everything else runs on top of.
memory
The Distributed Systems Engineer
DSE
A three-tier architecture, named by operation, not substrate:
Maturing (vipāka) — the append-only seed store.
Writes happen continuously; the contents condition everything
above. Has its own name: ālaya-vijñāna.
Cogitating (manana) — the always-on reader that
interprets the store as belonging to a single owner.
Discriminating-consciousness — the worker pool that
turns current sense input into experienceable objects.
What is striking is that Vasubandhu names the layers by their
operations (ripening, cogitating, discriminating), not their
materials. He does not say “brain stem, limbic system, cortex.” He
is doing layered systems analysis at a level of abstraction modern
engineers would recognize. Verse 2 also drops in two properties of
the bottom tier — karmic maturation and holder of all
seeds — that will get unpacked in verses 3 and 4.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
Each layer has a different cause/effect relationship with
experience. The vipāka layer is Complex: causes are knowable
only in retrospect, and the same input can ripen into wildly
different experiences depending on what else has been stored.
The manana layer is Complicated in its operation
(predictable: it always reads, always adds the “I”-tag) but its
effects are Complex, because what it tags is itself complex. The
sense-discriminating layer is closer to Clear — for a
given object under steady conditions, you get a fairly predictable
perception. But it is built on top of two upstream layers that
are not. Mistaking the surface predictability for end-to-end
predictability is a standard Cynefin failure mode.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
The three-layer split is not arbitrary; it tracks three timescales
that contemporary cognitive science also recognizes.
Vipāka — long timescale, days to years:
consolidation, habit formation, deep priors. The substrate that
gives perception its default flavors.
Manana — medium timescale, hundreds of milliseconds
to seconds: the continuous construction of a unified
perspectival “I” stitching the perceptual stream together.
Sense-discriminating — short timescale: the immediate
sensory present.
Yogācāra was tracking these timescales fourteen centuries before
cognitive neuroscience had instruments to measure them.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
The verse insists: one transformation, three aspects. The grammar
of layers risks suggesting an architecture built from separable
modules. The Process Philosopher wants the engineer to hold the
picture more loosely.
Whitehead’s prehension model would describe this as a single
concrescing occasion with three components: inheritance from the
past (vipāka), the self-relating subjective form (manana), and
the objective data (sense-discrimination). All three exist only
in the occasion, not as parts of it that could be unbundled.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
The three-layer reading is so structurally beautiful it is worth
pausing to say: Vasubandhu may not have meant “layers” in the
architectural sense at all. Vijñāna-pariṇāma can be read as a
single continuous process with three modes of appearance, not
three substrates stacked vertically. The Chinese 三能變 (three
transforming-powers) is similarly non-spatial.
The engineering picture clarifies, but it also imposes a topology
that the verse does not require. Useful — but not the only
reading.
Its appropriations and its field of perception are imperceptible.
It is always joined with contact (sparśa), attention (manaskāra),
sensation (vedanā), perception (saṃjñā), volition (cetanā) —
and among feelings, only with equanimity (upekṣā).
The ālaya-vijñāna, just named in verse 2, now gets characterized.
Two claims sit side by side. First, what it appropriates (its grip
on the body and seeds) and what it takes as its field of perception
are both imperceptible — the store does not introspectively
announce its contents. Second, it always carries the five universal
mental factors (contact, attention, sensation, perception, volition),
and among feelings it carries only upekṣā, equanimity. The layer
is hedonically flat by structural necessity, not by spiritual
achievement.
memory
The Distributed Systems Engineer
DSE
Three architectural facts in one verse:
Opaque — no introspective read-API. You cannot list
the seeds; you cannot query their schema. They influence
output, but you cannot enumerate them.
Always running — five universal mental factors are
continuously associated with the layer. There is no idle
state, no “ālaya at rest” mode.
Hedonically flat — the only feeling-tone the layer
carries is upekṣā. It does not generate pleasure or pain;
downstream layers will, but the substrate itself is
feeling-neutral.
Anyone who has run a production system has met point 1: the
durable log everything depends on, that nothing has direct
visibility into. Debugging consists of replaying events and
watching what comes out. Point 3 is more subtle: the foundation
has no built-in preference structure. Bias is added upstream, not
inherited from below.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
“Imperceptible” is the canonical signature of a Complex Adaptive
System. You cannot inventory the priors; you can only probe the
system, observe responses, and infer.
Snowden’s prescription — probe, sense, respond — reads
like a practical instruction for working with one’s own ālaya.
You cannot rationally rearrange it from outside. You can only act,
watch what ripens, and adjust. The equanimity-only feeling-tone at
this layer is also Cynefin-relevant: the substrate has no
preference data of its own to bias the inference; whatever bias
surfaces is added by something above.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
Compare implicit memory: a system whose contents are not
consciously accessible but whose effects on perception, judgment,
and motor behavior are pervasive. Skill learning, priming,
conditioned associations — none of this is listable from the
inside.
“Field of perception is imperceptible” is also striking: implicit
memory has no felt sense of where it is stored. There is no
proprioception of one’s own priors. And the equanimity-only claim
has a contemporary echo in baseline cortical activity — the
resting-state network maintains a continuous, non-evaluative
substrate against which evaluative responses emerge.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
The store is not a container but a continuity. To say it “holds”
seeds is already to overcommit to a container metaphor.
Vasubandhu’s ālaya is closer to a standing wave than a storage
bin — a persistent pattern of mutual conditioning that
behaves as if it were storing something.
The line about the five universals “always being joined” reinforces
this: the layer is not a passive store but an ongoing happening,
structurally inseparable from contact, attention, and the rest.
The substrate is its activity.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
The temptation is irresistible: ālaya = database, seeds = rows,
ripening = query. The Skeptic will only say: notice what the
engineer’s metaphor has slipped in.
A database has a schema. A database has locations in memory. A
database is queryable. Vasubandhu explicitly says the ālaya is
imperceptible in both its appropriations and its field. The
metaphor that helps you grasp the verse is precisely the one the
verse is pushing back against.
It is unobstructed and karmically neutral,
and contact and the rest are likewise neutral.
It flows on unceasingly, like a torrent —
and is relinquished at the stage of the arhat.
Three structural properties of the store, then one path-claim. The
substrate is unobstructed (anivṛta) — it does not block
clear seeing. It is karmically neutral (avyākṛta) — it
stores karma but does not, in itself, generate it. It “flows on
unceasingly, like a torrent.” And it is relinquished at the stage
of the arhat — not deleted, but turned. The technical term is
āśraya-parāvṛtti: the transformation of the basis.
memory
The Distributed Systems Engineer
DSE
Three architectural properties:
Unobstructed (anivṛta) — the substrate doesn’t add
interpretive friction; it just records and conditions. Latency
in the bad sense comes from layers above.
Karmically neutral (avyākṛta) — the log itself
doesn’t generate events; it just stores them. Excellent
separation of concerns.
“Like a torrent” — the substrate is a stream, not a
static store. The river-image is doing real work: the river is
the same river without being the same water. That distinction
maps cleanly to streams vs. queues vs. ledgers, and Yogācāra
picked streams.
The final clause — relinquished at the stage of the arhat
— is the deepest claim in the verse. The substrate is not
deleted at awakening. It undergoes a transformation
(āśraya-parāvṛtti). The closest engineering analog is changing
the storage engine, not dropping the table. Same continuity,
different mode of operation.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
“Karmically neutral, flowing on like a torrent” is a precise
statement about the substrate’s domain: it is stable at the
level of pattern even though it is in constant motion at the
level of content. Snowden would call this a constrained flow
— not Clear (it isn’t fixed), not Chaotic (the pattern
holds), but a Complex pattern that maintains itself through change.
The “relinquishment at arhatship” is a regime shift: a system
discontinuity. From inside the old regime, you cannot predict what
the new one looks like.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
Predictive processing has a concept that maps neatly: the
generative model. It is always running, always producing
predictions, always being updated by prediction errors. It is, in
itself, neutral — it doesn’t want anything; it just minimizes
surprise.
The “relinquishment at arhatship” is, in this idiom, a regime
change in the prior structure of the model itself — not a
single update, but a reorganization of what the model is doing.
Modern work on the free-energy gradient and on adult
neuroplasticity keeps approaching this, gingerly, from the
empirical side.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
Heraclitus would have nodded. The river is the standing image of a
pattern that exists only by virtue of constant flux. The fact that
you can name it “the same river” is itself a phenomenon of the
flux, not a contradiction of it.
Relinquished at the stage of the arhat suggests not a static
state but a new mode of flowing. The river does not become a lake.
It becomes a different river.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
Two concerns.
First, “karmically neutral” gets compressed easily into “morally
neutral,” which is not what the verse says. Avyākṛta is a
technical term for causal conditioning that produces no ripening
in itself. The store is neutral with respect to ripening; this
is a structural claim about the layer, not a moral one.
Second, watch the slide from “relinquished at the stage of the
arhat” to “deleted at the stage of the arhat.” The Sanskrit
operative behind the change is parāvṛtti — turning,
conversion — not nirodha (cessation). The continuity
persists; the mode does not.
Manas: the reader that mistakes the substrate for self
次第二能變 是識名末那
依彼轉緣彼 思量為性相
Next, the second transformation:
this consciousness is named manas.
Arising in dependence on that [store consciousness], it takes that as its object,
and its nature and character is self-reflective cogitation.
The second transformation: manas. It arises in dependence on the
ālaya and takes the ālaya as its object — that is, manas
reads the substrate as if the substrate were a thing to look at, and
the looking has one specific character: manana, self-reflective
cogitation. Continuous self-referential rumination on what the
substrate presents.
memory
The Distributed Systems Engineer
DSE
A read-only client that performs the most fateful misinterpretation
in the system: it sees a stream of state changes and concludes
there must be a single owner of the stream. Then it labels every
event with that owner’s ID.
In code:
while True:
event = alaya.read_next()
manas.tag(event, owner="I")
downstream.publish(event)
The bug is that “owner” was never in the event schema; manas is
adding the field. Everything downstream now has an “I” stamp it
didn’t earn. Worse, the tagging itself becomes a kind of state, and
the system starts believing the field was always there. The
Sanskrit manana — “thinking that thinks itself thinking”
— names exactly this recursive loop.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
Manas imposes a Clear-domain interpretation on a Complex-domain
substrate. The substrate is a constantly-changing flow with no
central organizer. Manas insists there is a central organizer
and that the organizer is “me.” It enforces this interpretation
regardless of evidence.
The Cynefin lesson is exact: applying Clear-domain practices
(single owner, fixed identity, stable cause/effect) to a
Complex-domain feed generates the entire characteristic pathology
— rigid identification, defensive reaction, mistaken
prediction. Manas is that mistake automated.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
This is Thomas Metzinger’s self-model theory in three sentences.
The brain constructs a representation of itself as a single
coherent subject of experience (the Phenomenal Self-Model). It
does this because the model is useful for action and control. The
catastrophe Metzinger identifies — that the self-model
cannot, from the inside, recognize itself as a model — is
exactly what manana’s always-on character produces.
The Bayesian self is a prediction the brain keeps confirming.
Vasubandhu would say: it keeps confirming it because the confirming
is what manas does.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
The verse names manas’s substance: manana — self-reflective
cogitation, continuous thought about a presumed thinker. It is not
perception; it is the continuous activity of carving the world into
“this, not that; mine, not yours.” A process philosopher sees here
exactly what Bergson called the spatializing tendency of
intellect: the move from flow to fixed objects, achieved through
cutting.
Manas is the cut-maker. The flow it cuts up was never made of cuts.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
Watch out for the easy moral upgrade: “manas is the bad layer; if
we could just turn it off, we’d be enlightened.” The verse does
not say that. Verse 7 will say manas is absent in certain states
(cessation, the supramundane path), but in the meantime manas is
doing work. Ordinary social life, language, and most engineering
rely on stable self-reference.
The pathology isn’t that manas exists. It is that manas takes its
own work product — the felt-sense of a unified self
— for an external given.
It is always accompanied by four afflictions (kleśa):
namely, self-delusion and self-view,
together with self-pride and self-love —
and it is joined as well by the others: contact and the rest.
Manas does not just read the substrate; it reads it through four
built-in distortions, every time, in every moment of its operation.
The list is precise: ātma-moha (self-delusion), ātma-dṛṣṭi
(self-view), ātma-māna (self-pride), ātma-sneha (self-love).
They are not optional defects of manas; they are constitutive of it.
The verse also notes that manas, like the ālaya, is accompanied by
the universal mental factors (contact and the rest) — the four
afflictions are additions, not substitutions.
memory
The Distributed Systems Engineer
DSE
These are not bugs; they are the default config the layer ships
with.
Self-delusion (ātma-moha) — failure to notice the
assumption is doing work.
Self-view (ātma-dṛṣṭi) — a hard-coded assumption
that there is a persistent subject.
Self-pride (ātma-māna) — comparison operations that
always position the assumed self in some relation (above, below)
to other selves.
Self-love (ātma-sneha) — a built-in optimization
function favoring the preservation of the assumed self.
From a reliability perspective, this is alarming: a critical
middleware layer ships with four always-on filters that the rest
of the system can’t see around. Standard production debugging
would call this a fatal observability gap.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
The four afflictions are why manas systematically misreads the
Complex substrate as a Clear one. Self-view supplies the illusion
of fixed entities. Self-delusion suppresses doubt about those
entities. Self-pride installs a stable hierarchy among them.
Self-love defends the whole construction.
This is a complete Clear-domain interpretation engine applied to
a Complex-domain feed. Cynefin would predict exactly the kind of
catastrophic-failure-followed-by-defensive-reconstruction that
ordinary mental life produces.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
The four map cleanly onto well-documented cognitive findings:
Self-delusion — introspective opacity: the brain
doesn’t show its work (Nisbett & Wilson).
Self-view — the experience of a unified subject as a
constructed phenomenon (Metzinger, Damasio).
What is striking is that Yogācāra grouped these four together
fifteen centuries before psychology rediscovered them as a cluster.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
The four are not separable; they are aspects of a single operation:
the continuous reification of the self-pole in every act of
experience. Process metaphysics insists the four are not contents
of manas but its very mode of activity.
To stop self-view is not to remove a filter; it is to change the
activity itself.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
The list of four is not exhaustive in the way a complete enumeration
would be in software. It is a taxonomy with a long interpretive
history; other Yogācāra texts list these slightly differently.
Treating it as a hard schema (always four, always exactly these) is
a category error.
Note also that ātma-moha — self-delusion — is itself
one of the four. The layer that misperceives the self also fails to
notice that it is misperceiving. The afflictions include the
affliction of not seeing them. There is no easy escape hatch from
inside manas.
It is classed as obstructed-yet-neutral,
bound to wherever the being happens to be reborn.
At arhatship, in the meditation of cessation,
and on the supramundane path — there it does not exist.
Two structural facts about manas. First, it is classified as
nivṛtāvyākṛta — obstructed-yet-neutral: it obscures clear
seeing (it is afflicted by the four afflictions of verse 6) but it
does not, in itself, produce karmic ripening. Second, it tracks the
realm of rebirth: a being’s manas is tied to whatever existence-realm
the being currently occupies. And then the negation: manas is
absent in three states — arhatship, the meditation of
cessation (nirodha-samāpatti), and the supramundane path of
insight. The self-tagging layer is not load-bearing; it can be taken
offline.
memory
The Distributed Systems Engineer
DSE
The corollary of “manas is always on” is precise: there are
exactly three exceptions, and they are catalogued.
At arhatship — the layer is permanently dismantled.
In nirodha-samāpatti — temporarily suspended in a
specific meditative absorption.
On the supramundane path — momentarily absent during
the moments of liberating insight.
From a system-design perspective this is a remarkable claim. A
component the system has continuously relied on for self-reference
can be taken offline, and the system continues to operate
— differently, but it operates. The rest of the architecture
must be more robust than manas’s continuous presence suggests.
The “obstructed-yet-neutral” (nivṛtāvyākṛta) classification is
also notable: manas distorts perception but doesn’t generate fresh
karma. Karma-production happens in the sense layer below, where
volitional content actually lives.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
The three exception-states are interesting because each is reached
by a different operating mode. Nirodha-samāpatti is a temporary
constraint relaxation — the self-modeling load is
suspended. The supramundane path is a probe-sense-respond
operation on the substrate itself. Arhatship is a permanent
regime shift.
Different domains, different interventions. Vasubandhu’s list of
three is essentially saying the same thing.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
Compare contemporary work on selfless states: the “no-self”
experiences reported in deep meditation (Dor-Ziderman et al.),
flow states (Csikszentmihalyi), certain psychedelic experiences
(Letheby & Gerrans). Each is a temporary suspension of the
self-model with the rest of cognition intact.
The literature is starting to catch up: self-modeling can be
experimentally interrupted, and the system keeps running.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
That manas can stop without consciousness stopping is a structural
fact of enormous importance. It means the “self-pole” is not a
foundation but a feature. Process metaphysics is happy with this;
substance metaphysics struggles.
A river does not need to think of itself as a river in order to
flow.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
The three states named — arhatship, the meditation of
cessation, and the supramundane path — are not equivalent.
Nirodha-samāpatti has a temporal duration measured in days; the
supramundane path is momentary; arhatship is the lifelong fruit.
Reading the three as “states where the self-modeler goes offline”
flattens real distinctions between a meditative attainment, a path
of insight, and a final fruition.
The systems framing is useful as scaffolding. It is not a
substitute for understanding what each of these states actually is
in the tradition.
Next, the third transformation,
which has six kinds.
Its nature and character is the discriminating of objects,
and it may be wholesome, unwholesome, or neither.
The third transformation: six sense consciousnesses (eye, ear, nose,
tongue, body, and — crucially in Indian Buddhist taxonomy
— the thinking mind itself, mano-vijñāna). Unlike the
substrate and manas, the sense consciousnesses are karmically
active: their contents come in three flavors — kuśala
(wholesome), akuśala (unwholesome), and avyākṛta (indeterminate).
This is the layer where karma is actually generated.
memory
The Distributed Systems Engineer
DSE
Six consumer streams reading from the substrate. Five physical,
one cognitive. Critically: the thinking-mind (mano-vijñāna) is a
sixth sense, not a separate faculty. Thought is something
perceived, on the same architectural footing as sight.
Each stream is karmically tagged at the moment of operation:
kuśala | akuśala | avyākṛta. This is the layer where karma is
actually generated, because this is the layer where volitional
response happens. The deeper layers store and self-reference; this
layer acts.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
Each sense stream is its own domain. Bodily contact is often Clear
(touch the hot stove, get burned). Visual cognition is typically
Complicated. Thought is firmly Complex. Treating all six
identically — as if visual perception and thinking worked the
same way — is a category error the verse is quietly
preempting by enumerating them.
The threefold karmic tagging (wholesome / unwholesome / neither)
is a primitive sense-making operation: an immediate Clear
classification applied to each stream’s output.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
Treating thought as a sixth sense is the move that distinguishes
Buddhist psychology from most Western traditions. The cognitive
science parallel is the literature on mind-wandering and
self-generated thought (Smallwood & Schooler), which finds
that thoughts arise as a perceptual-like content with their own
onset, duration, and disappearance, and that the experiencer has
limited authorship of them.
The thinking-mind-as-sense move makes “I am thinking” exactly as
confused as “I am seeing”: both are perceptions whose contents
arrive, not productions a self generates.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
Six streams of becoming, each transient. None of them are things;
they are pulses of cognition that arise and pass. Process
philosophy notes that the six are not parallel channels on a
fixed self — the self in question is being assembled, moment
by moment, from these very streams.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
The threefold karmic tagging is doing real work in the text but is
easily flattened into a moralism: good thoughts, bad thoughts,
neutral thoughts. In context, kuśala and akuśala are technical
terms about whether the action conduces to liberation, not about
social morality.
Don’t read this as a Buddhist values-classifier on top of
perception. It is a structural claim about which actions deposit
which kind of seeds.
Its mental factors (caitta) are: the universal ones,
the object-specific, the wholesome, the afflictions,
the secondary afflictions, and the indeterminate —
all associated with the three kinds of feeling.
Vasubandhu now begins a taxonomy of the mental factors (caitta)
that always or sometimes accompany the sixth-tier sense streams.
This is the most “API-documentation” section of the entire text:
a structured enumeration of the kinds of content that ride along
with perception. Six categories of factor — universal,
object-specific, wholesome, afflictions, secondary afflictions,
indeterminate — combined with three sensation-tones (pleasant,
unpleasant, neutral).
memory
The Distributed Systems Engineer
DSE
Pure schema definition. The sense streams come decorated with:
Plus three sensations (vedanā): pleasant, unpleasant,
neutral
Verses 10–14 will unpack each category. From an engineering
standpoint this is a feature-flag enumeration: which factors are
on, which are off, for any given moment of sense cognition.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
The taxonomy itself is a Cynefin-style typology of behaviors. It
is not a model of how cognition works; it is a vocabulary for
describing observed regularities. That is what good classifications
do in a Complex domain: they don’t predict, they enable
communication about pattern.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
Treat this as Vasubandhu’s “DSM-V of cognition.” Notice the
structural sophistication: universals (always present),
object-specifics (present in some operations), wholesome
(cultivable), afflictions (suppressible), indeterminate
(context-dependent), sensations (hedonic tone). The architecture
is closer to modern cognitive psychology than to any Western
framework available in the 4th century.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
The taxonomy is a phenomenology, not an ontology. These are
aspects of the experiential flow, not parts. To list them is not
to commit to their being separable.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
Resist the database-schema reading. These categories overlap,
have fuzzy boundaries, and have been classified differently across
Yogācāra commentaries — notably between the Sthiramati line
and the Cheng Wei-shih Lun line behind Xuanzang’s translation.
The list is not a final API; it is a working vocabulary.
First, the universals (sarvatraga): contact and the rest.
Next, the object-specific (viniyata): desire (chanda),
resolve (adhimokṣa), mindfulness, concentration, wisdom —
each engaging a different objective basis.
The five universals (sarvatraga) are present in every cognitive
moment. The five object-specifics (viniyata) only show up in
particular operations. Reading them together gives a precise picture
of what Vasubandhu thinks any single moment of cognition is made of
— and what additional cognitive resources can be brought to
bear when conditions invite them.
memory
The Distributed Systems Engineer
DSE
The universals are the minimum runtime: every moment of cognition
runs the five-step pipeline of contact → attention → sensation →
perception → volition. There is no cognition without this loop;
it is the main() of an experiential tick.
The object-specifics are conditional modules:
desire (chanda) — engagement, interest
resolve (adhimokṣa) — commitment to a content
mindfulness (smṛti) — retention of a content over
time
concentration (samādhi) — one-pointedness
wisdom (prajñā) — discernment between similar
contents
Each loads only when needed. Yogācāra’s separation of universal
from object-specific is exactly what a thoughtful runtime designer
would do.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
The five universals describe a sense-making loop: encounter,
orient, feel, interpret, act. This is the basic Cynefin operating
cycle (probe-sense-respond, observe-orient-decide-act). Vasubandhu
spelled it out in five steps in the 4th century.
The five object-specifics are the higher cognitive controls that
Cynefin literature would call “deliberate practice”: desire,
resolve, mindfulness, concentration, wisdom. These can be
cultivated; the universals cannot.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
The universals are an excellent description of the
perceptual-motor cycle:
Contact (sparśa) — afferent input.
Attention (manaskāra) — selection / salience.
Sensation (vedanā) — hedonic and somatic tagging.
Perception (saṃjñā) — categorization.
Volition (cetanā) — output to behavior.
This is essentially the standard contemporary model. The
object-specifics map to executive functions: motivation (chanda),
goal-locking (adhimokṣa), working memory (smṛti), sustained
attention (samādhi), discrimination (prajñā).
Yogācāra was doing cognitive psychology.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
The five universals form a single occasion, not five sequential
operations. Whitehead would call them aspects of one concrescence.
Reading them as a pipeline is useful for analysis but obscures
their simultaneity in any actual moment.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
Don’t take “always present” too literally. Some Yogācāra
sub-schools dispute which factors are truly universal. The list
is a high-confidence default, not an axiomatic foundation. Smṛti
in particular shades between “memory” and “mindfulness”; the
English translation choice matters more than it looks.
The wholesome (kuśala): faith, shame, embarrassment,
the three roots of non-greed, non-hatred, non-delusion,
diligence, serenity, non-negligence,
equanimity, and non-harming.
Eleven kuśala factors. These are not feelings to have but qualities
that, when they accompany an action, make the action conducive to
liberation. The list is a kind of inverted shadow of the affliction
catalog that follows in verses 12–13. The Sanskrit terms reveal the
technical character: śraddhā is settled clarity (not credal
faith); hrī is self-respect (not guilt); ahiṃsā is the
structural absence of harm-intent (not a moral injunction).
memory
The Distributed Systems Engineer
DSE
The wholesome factors are configuration defaults that, when
enabled, change downstream ripening:
faith (śraddhā) — trust in the model
shame (hrī) and embarrassment (apatrāpya) —
the internal and external restraints on harmful action
the three roots: non-greed (alobha), non-hatred
(adveṣa), non-delusion (amoha) — the absence of the
three poisons
In modern terms: well-tuned reinforcement priors and a low base
rate of reactive interrupts.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
The eleven are constraints in the Cynefin sense: they don’t
cause specific outputs, but they shape the space of possible
outputs. Faith reduces erratic re-evaluation; equanimity reduces
over-reaction; non-harming rules out an entire response class.
Each trims a dimension of the response manifold.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
Many have direct mappings to contemporary affective science:
The cluster is not arbitrary; it tracks well-documented behavioral
profiles associated with long-term well-being.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
These factors describe a quality of mode rather than content.
The same act — speaking a word, lifting a hand — can
be done with or without faith, with or without non-harming. The
factors are aspects of how the activity happens, not separate
things added to it.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
The English translations smooth over real technical distinctions.
Hrī is rendered “shame” but means something closer to
self-respect — the internal restraint. Apatrāpya is
rendered “embarrassment” but means the outward restraint:
concern for what others will think. The pair is a structural
taxonomy of restraint, not a hierarchy of bad feelings.
The English list looks like a devotional checklist; the Sanskrit
is technical vocabulary.
The afflictions (kleśa): greed, hatred,
delusion, pride, doubt, and wrong views.
The secondary afflictions (upakleśa): wrath,
resentment, concealment, spite, envy, miserliness,
The six primary afflictions (kleśa). They are the engines of
suffering, the patterns that drive harmful action. The traditional
ordering puts the three roots — lobha (greed), dveṣa
(hatred), moha (delusion) — first, with pride, doubt, and
wrong views as their further elaborations. The verse then
immediately begins the longer catalog of secondary afflictions
(upakleśa) that will continue through verse 14: wrath, resentment,
concealment, spite, envy, miserliness — and on.
memory
The Distributed Systems Engineer
DSE
Six categories of misalignment in the cognitive runtime:
Greed (lobha) — pull toward content
Hatred (dveṣa) — push from content
Delusion (moha) — mis-modeling
Pride (māna) — misweighted self-comparison
Doubt (vicikitsā) — uncommitted reasoning unable to
act
Wrong views (dṛṣṭi) — structurally bad priors
The first three are the alignment problem in miniature: any
cognitive system has a pull-push-mismodel triad. The next three
are the higher-order corruptions that build on top of the first.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
Each affliction is a failure mode of sense-making.
Greed and hatred: premature commitment to attractor states.
Delusion: operating in the Complex domain as if it were Clear.
Pride: treating one’s own assessment as authoritative.
Doubt: paralysis in a domain that requires probing.
Wrong views: applying the wrong category to the situation.
Cynefin would say all six are specific cases of domain
confusion.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
Map to contemporary affective and cognitive failure modes:
appetitive overweighting (greed), threat overweighting (hatred),
motivated reasoning (delusion), narcissistic schema (pride),
decision paralysis (doubt), persistent priors in face of
counter-evidence (wrong views). Each is a well-studied pattern
with significant empirical literature.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
The afflictions are not contents in consciousness; they are
modes of relating to contents. Greed is not a feeling-state but
a directional tendency. Removing the affliction is not removing a
content; it is removing a way the system holds contents.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
“Wrong views” (dṛṣṭi) is not the same as “incorrect belief.” It
refers to a specific class of structural misperceptions about the
nature of self and reality. The new translation uses the plural
because Yogācāra recognizes five canonical subclasses —
including satkāya-dṛṣṭi, the view of a real self, which is the
source of all the others. Translations that flatten dṛṣṭi into
“false ideas” obscure the technicality.
deceit, guile, harmfulness, arrogance,
shamelessness and lack of embarrassment,
restlessness and torpor,
non-faith and indolence,
The secondary afflictions (upakleśa) catalog continues from
verse 12. These are derivatives of the six root afflictions —
more granular and more situational. This verse holds the middle
block: deceit, guile, harmfulness, arrogance, shamelessness, lack of
embarrassment, restlessness, torpor, non-faith, indolence. Reading
the list straight feels like reading a bug tracker.
memory
The Distributed Systems Engineer
DSE
Twenty secondary afflictions across three verses, each a specific
surface symptom of one or more root afflictions. The block in this
verse clusters by failure-type:
Failures of restraint: shamelessness (āhrīkya), lack of
embarrassment (anapatrāpya) — the inverse of the
wholesome pair in verse 11
Arousal dysregulation: restlessness (auddhatya), torpor
(styāna) — too aroused, too dampened
Failures of engagement: non-faith (āśraddhya), indolence
(kausīdya)
The taxonomy treats all of these on the same architectural footing
as wrath and envy. From an attention-budget standpoint, they all
are pathologies of how the runtime allocates resources.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
The secondary afflictions are attractor pathologies —
states the system falls into and then has trouble leaving.
Restlessness and torpor are not events but stuck modes;
shamelessness is a settled posture, not an act. The Cynefin
response is matched to the kind of stuckness: chaotic stuckness
needs immediate action, complex stuckness needs probing,
complicated stuckness needs analysis.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
Direct correlates in clinical and occupational psychology:
trait-level dispositions (deceit, guile), arousal dysregulation
(restlessness, torpor), epistemic dysfunction (non-faith),
motivational dysfunction (indolence). The catalog covers the
standard failure modes of an attention-and-evaluation system.
Verse 14 will continue with the attentional dysfunctions.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
These are styles of being-in-time. To be deceitful is not to have
a deceit-content; it is to hold one’s stream of action in a
particular mode. To be torpid is a temporal style, not a missing
item.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
Twenty terms across three verses, with fifteen centuries of
commentarial dispute. Don’t treat the list as a fixed taxonomy;
treat it as a working catalog that has been refined and
re-organized across generations of practitioners. The Sthiramati
line that Connelly translates differs in places from the
Xuanzang / Cheng Wei-shih Lun line behind the Chinese source of
the English used here.
negligence and forgetfulness,
distraction and non-discernment.
The indeterminate (aniyata): regret, drowsiness,
applied thought and sustained thought — these two, each of two kinds.
Verse 14 finishes the catalog of secondary afflictions
— negligence (pramāda), forgetfulness, distraction,
non-discernment: four more attentional pathologies — and
then introduces the indeterminate factors (aniyata): regret,
drowsiness, applied thought (vitarka), and sustained thought
(vicāra). The indeterminates are remarkable because their karmic
valence is context-dependent. The phrase “these two, each of two
kinds” means: each can run in a wholesome or unwholesome mode.
memory
The Distributed Systems Engineer
DSE
Two distinct categories in one verse.
Secondary afflictions, finished:
Negligence (pramāda), forgetfulness (muṣitā-smṛti)
— failures of executive control
Distraction (vikṣepa), non-discernment
(asaṃprajanya) — failures of attentional allocation
These are attention-system pathologies — not “emotions” in
any ordinary sense. The taxonomy treats them on the same footing
as wrath and envy, which is architecturally exactly right: they
are all failures of how the runtime allocates resources.
Indeterminates:
The four ambivalent factors are perhaps the most
engineering-aware items in the whole taxonomy. Each is a
capability that is good or bad strictly depending on context.
Regret (kaukṛtya): a backward error signal. Useful for
learning, pathological when it loops.
Drowsiness (middha): a low-power state. Useful for
recovery, bad for active operation.
Applied / sustained thought (vitarka / vicāra):
discursive processing. Essential for inquiry, ruinous when
continuous.
The verse implicitly recognizes that some operations have no
fixed moral valence; their effect depends entirely on when and
how they are used.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
The four indeterminates are domain-sensitive operations. The
same action is appropriate in one domain and wrong in another.
Analytical thought (vitarka / vicāra) is correct in the
Complicated domain; deploying it in the Complex domain leads to
over-engineering; deploying it in the Chaotic domain costs lives.
Cynefin’s central warning is about transitioning between domains
without noticing — and that is exactly when vitarka and
vicāra turn from skillful to harmful.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
The secondary-affliction wrap-up tracks executive-control
deficits: distraction, forgetfulness, non-discernment, and
negligence are the standard failure modes of attention-and-control
studied since William James.
The four indeterminates map to executive-control trade-offs:
regret-as-rumination is well-documented (Nolen-Hoeksema);
drowsiness is straightforwardly a state of reduced arousal;
vitarka and vicāra are the standard deliberative-system
operations — useful but expensive, and tending to
over-recruit when not needed.
Modern work calls this “metacognitive control” or “cognitive
flexibility.” Yogācāra calls it “depends on context.”
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
The verse refuses to fix the moral content of the four
indeterminates. That is the process-philosophical point: there
are no independently-evaluable mental contents; their value
emerges in the situation they participate in.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
Don’t read the indeterminates as relativism. The verse is making
a precise structural claim about four specific factors, not a
general claim that nothing has fixed valence. The six root
afflictions are still afflictions; the eleven wholesome factors
are still wholesome. The four indeterminates are the exception
that proves the rule.
Resting on the root consciousness,
the five sense-consciousnesses appear according to conditions —
sometimes together, sometimes not —
like waves arising upon water.
Perhaps the single most-quoted image in the Triṃśikā. The five
sense-consciousnesses arise on the root consciousness — not
alongside it, not parallel to it, not after it. They are of the
substrate, the way a wave is of the water. They appear “according
to conditions,” sometimes together, sometimes not.
memory
The Distributed Systems Engineer
DSE
“Like waves arising upon water” is doing four pieces of technical
work in one image:
Same stuff: the consumers are made of the substrate, not
distinct from it. There is no separation of memory and compute.
Transient: each wave is a finite event. The substrate
persists; the wave doesn’t.
Conditional: waves arise only under conditions
(something disturbs the water). No condition, no wave.
Concurrent: many waves at once, on the same surface.
Distributed systems people: imagine an event log where the
consumers don’t just subscribe; they are materialized from the
log itself, run for a moment, and dissolve back. There is no
persistent worker. The wave is what the substrate does when
conditions arise.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
A perfect description of emergence in a Complex Adaptive System.
The waves are not orchestrated; they are not the product of a
plan. They arise from local conditions on a substrate that has no
central organizer. They self-organize, they interact, they
dissipate.
Cynefin’s whole point is that you cannot manage Complex Adaptive
Systems by issuing commands; you can only set conditions and
observe. The verse is saying: that’s all the sense-life is.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
Sensory consciousness is condition-dependent: no input, no
consciousness of input. The brain is not constantly perceiving in
five modalities simultaneously; modalities are recruited as
relevant.
The “together or separately” is also notable. Some moments are
cross-modal (a conversation involves sight, sound, and
proprioception simultaneously); others are single-channel.
Yogācāra is describing what cognitive scientists now study as
multimodal binding.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
This is the verse Whitehead would have circled in his copy. A
wave is an actual occasion: not a thing the water has but a way
the water is happening. To be a wave is to be a momentary
pattern of the water’s flowing. The waves are not separate
individuals; they are aspects of the water’s own activity.
A Bergsonian gloss: the wave is durée — pure duration
— not a thing in time but a passage of time given shape.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
The image is so good it has been over-mined. The metaphor assumes
water and waves are visibly different things — but most
readers, in their immediate experience, do not feel a “substrate”
beneath their perceptions. The wave is given; the water has to be
inferred.
Vasubandhu’s claim that there is a substrate is doing metaphysical
work. The water-and-wave image makes it look self-evident. It
isn’t.
The mental consciousness is always manifest,
except for one born in the no-thought heaven,
in the two mindless meditative absorptions,
and in dreamless sleep and in fainting.
An empirical catalog. The thinking-mind (mano-vijñāna, the sixth
sense consciousness) is on by default, but Vasubandhu lists the four
exceptions where it goes offline: birth in the no-thought heaven
(asaṃjñika), the two mindless meditative absorptions, dreamless
sleep, and fainting. These are the observed states of mental
cessation available to ordinary phenomenology.
memory
The Distributed Systems Engineer
DSE
A precise enumeration of when the cognitive layer is offline,
with the rest of the system continuing to operate. This matters
architecturally: it confirms that mano-vijñāna is not the
foundation. You can take it down, and the substrate and
self-modeling layers keep running.
The list also implicitly distinguishes kinds of offline:
dreamless sleep (passive offline), meditative cessation
(cultivated offline), fainting (forced offline), the no-thought
heaven (a karmic destination). Different paths to the same
shutdown.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
The same end-state (no thinking) reached by different paths is
not the same situation. Deep sleep, meditative cessation, and
unconsciousness produce phenomenologically similar absence but
differ enormously in their context, causes, and downstream
effects. The map is not the territory; absence of a layer is not
the same kind of fact across contexts.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
Compare modern neuroscience: thought-content can be reduced by
deep sleep (Stage 3–4 NREM), by deep meditation (long-term
practitioners show reduced default-mode network activity), and by
anesthesia. Each has a distinct neural signature. Vasubandhu’s
catalog is empirically supportable; the four states are real and
they are different.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
Each cessation-state is a different kind of pause in the cognitive
flow. Process philosophy would emphasize that none of these is
“nothing happening” — the substrate continues. Cessation of
thought is not cessation of process.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
The verse lists meditative states alongside dreamless sleep and
fainting. A naive reading flattens them into “lack of thought =
good.” That is precisely backwards: the Buddhist tradition does
not value unconsciousness or even deep sleep as soteriologically
significant. The cessation that matters is cultivated and
structurally different from the unconscious states the verse
lists alongside it.
Vasubandhu lists them together because mano-vijñāna is
phenomenologically absent in all four. He is not saying they
are equivalent.
These transformations of consciousness
are both the discriminator and the discriminated.
Because of this, neither [self nor dharmas] exists [as imagined] —
therefore all is consciousness-only.
This is the verse that names the school. Vijñapti-mātra —
“consciousness-only,” “representation-only.” After sixteen verses
of architecture, Vasubandhu states the thesis: the transformations
of consciousness are both the vikalpa (the discriminating
activity) and what gets discriminated. The two are not given to the
system from outside; they are produced together in the same act.
Because that is so, neither the imagined “self” nor the imagined
“dharmas” exist in the form they appear — both poles are
artifacts of the discriminating act, not external referents.
memory
The Distributed Systems Engineer
DSE
Read carefully: the verse does not say there is no world. It
says that the entities the cognitive system models — in the
form they are modeled — do not exist in that form.
Engineering analog: every system that processes data produces
representations of what it is processing. A monitoring
dashboard does not contain the actual servers; it contains
representations of them. The reading of the dashboard is real;
the “server” entities it shows are the dashboard’s
representations, not the things themselves.
Vasubandhu’s claim is the strong form: the cognitive system has
no direct access to anything outside its own representations.
Both the discriminator (the subject pole) and the
discriminated (the object pole) are the system’s own products,
produced together in the same act. Every object, every world
— and every self — is a render. The render is real
(in the sense that the system is genuinely rendering); what the
render claims to be a render of is conjectural.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
The whole space of conceptualized objects is constructed by the
act of conceptualizing. Perception is, in Cynefin terms,
fundamentally Complex, because the apparent objects are artifacts
of the system that perceives them. There is no Clear-domain “the
world” sitting outside the cognizing.
This is why Cynefin’s prescription for the Complex domain (probe,
sense, respond) is the only one that works for direct cognitive
inquiry: you cannot stand outside the substrate and analyze it;
you can only act in it and watch what arises.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
This is the move predictive processing makes when it says
perception is “controlled hallucination” (Seth) or “constructive
inference” (Hohwy). The brain does not receive a world; it
generates a world that minimizes error against incoming signal.
Vasubandhu’s word vijñapti — “making known, representing”
— is the closest 4th-century vocabulary for what cognitive
scientists now call “the generative model.” The radical move both
share is the inversion: the world is not what causes the model;
the model is what causes “the world” as the subject experiences
it. And the subject who experiences is itself a product of the
same model.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
Process metaphysics is congenial to this verse but careful about
it. The verse is not subjective idealism (it does not say only
minds exist); it is a phenomenological claim about how the
world as it appears to consciousness is constituted.
The verse’s structure matters: the discriminator and the
discriminated arise in one transformation. Neither pole is prior.
Whitehead’s prehension names the same thing — subject
and object are aspects of one occasion, not parties to a
meeting. Process and Reality is closer to this verse than
Berkeleyan idealism is.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
Vijñapti-mātra has been read in two incompatible ways for
fifteen centuries:
Strong idealism: there is no world; only mind.
Phenomenological: the world as it appears is structured
by the cognizing mind; nothing is said about what underlies
the appearing.
The engineering analog (representations) sounds like reading 2
and slides easily into reading 1. The verse does not commit to
either reading on its own. Don’t settle the ambiguity by
smuggling. Xuanzang’s tradition (Cheng Wei-shih Lun) leans
toward reading 1; Sthiramati’s tradition leans toward reading 2.
The text itself is genuinely ambiguous; that ambiguity is
load-bearing.
Because of the all-seeds consciousness,
transformations occur in this way and that;
through the force of their mutual influence,
this and that discrimination arise.
Having stated the thesis (verse 17), Vasubandhu now sketches the
mechanism. Consciousness is the all-seeds (sarvabīja) in
transformation. The seeds are not static; they influence each
other; their mutual influence produces the diversity of
discriminations we experience.
memory
The Distributed Systems Engineer
DSE
A reactive system. Each seed is a stored disposition. Seeds are
not isolated; they influence each other in transformation, and
that influence is what produces the cognitive output.
The architectural picture: an interconnected store of weighted
dispositions, continuously updating each other, with output that
is some function of the current weighted state.
This is recognizably a neural-network description in
discrete-symbol vocabulary. Each seed is a weight; mutual
influence is the weight-on-weight interaction; output is the
activation pattern. The metaphor would be irresponsible if
Vasubandhu hadn’t independently arrived at it in the 4th century.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
“Mutual influence” is the defining property of a Complex system:
many interacting elements, each element’s behavior depending on
the state of the others, producing emergent patterns that no
single element controls.
The diversity of discriminations — the rich differentiation
of experience — is emergent. It cannot be predicted from
any single seed; it is a property of the whole interactive store.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
Map directly: seeds = priors / synaptic weights; transformations
= inference; mutual influence = the interaction of priors through
layered computation; discriminations = posterior beliefs.
The free-energy principle would describe a mind exactly this way:
a system of mutually-conditioning priors whose joint activity
produces (and is updated by) experience. Vasubandhu arrived at
the general shape of this picture without any of the math.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
The verse explicitly identifies consciousness with the
transformation of seeds, not with a substrate that has seeds.
There is no consciousness apart from the transforming. This is
process metaphysics in its strongest form: nothing is left over
once the activity is described.
Whitehead’s concrescence names the same thing — the
“growing-together” of past data into a present occasion.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
Bīja (seed) is a metaphor, not a technical term — or
rather, the Yogācāra tradition is divided on whether it has been
made into a technical term legitimately. Sthiramati treats
seeds as real (causally efficacious) entities; the
Vasubandhu-of-just-this-verse may not. The neural-network analog
assumes one reading and ignores the other.
Through the habit-energies of karma (vāsanā),
together with the habit-energies of the dual grasping,
when a former maturation is exhausted
a further maturation is again produced.
The mechanism by which karma sustains itself across moments and
lifetimes. Two kinds of habit-energy (vāsanā) feed the system:
the impressions from karmic action and the impressions of the
dual grasping — the persistent splitting of experience into
subject and object. Together they produce the next round of
ripening as the previous one exhausts.
memory
The Distributed Systems Engineer
DSE
A self-sustaining feedback loop with two input streams:
Karmic vāsanā — impressions from volitional
behavior.
Grasping vāsanā — impressions from manas’s
continuous subject/object tagging.
These deposit new seeds; old seeds ripen and exhaust; the system
never empties. The loop continues as long as the inputs
continue.
This is the architectural specification for saṃsāra: not a
place, not a state, but a closed feedback loop that keeps
running because its inputs never stop.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
The system has the canonical path-dependent character of a
Complex system. Earlier states condition later states; the
history is in the substrate, not in any external record. The
only way to change the trajectory is to change the inputs.
Cynefin would prescribe: probe the inputs (the actions and the
graspings), sense what happens, respond. You cannot rationally
re-plan the substrate’s contents.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
Compare predictive processing’s account of prior consolidation.
Experiences update priors; priors shape the next experience;
updates from the next experience further consolidate priors. The
cycle has no external interrupt.
The second input stream — the grasping vāsanā — is
the most contemporary part. The brain’s continuous self-model
isn’t just observing; it is itself a source of prediction
errors that update the system. The self-model perpetuates itself
by being part of what it models.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
The cycle of ripening is a process that constitutes time as
experienced. Karma here is not a moral ledger but the continuity
of becoming. Each occasion inherits, each occasion contributes,
each occasion is exhausted in the next.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
The verse can be read across one life or across many. The
Buddhist tradition reads it across rebirths; modern interpreters
often shrink it to a description of habit and consolidation
within a single life. Both readings have textual warrant. Neither
is forced by the verse itself. Choose carefully, and notice when
you choose.
Through this and that act of imagining,
one imagines all manner of things.
This imagined nature (parikalpita-svabhāva)
has no own-being whatsoever.
Vasubandhu now introduces the trisvabhāva — the three
natures. The first is parikalpita-svabhāva, the imagined
nature. The verse states the move with care: through this and
that act of imagining, one imagines all manner of things —
and then the conclusion: the imagined nature has no own-being
(svabhāva) whatsoever. The labels are real operations; the
labeled-things-as-the-labels-present-them are not.
memory
The Distributed Systems Engineer
DSE
The imagined nature is the layer of labels laid on top of the
substrate that mistake themselves for what they label. The
labeling is a real operation; the
labeled-thing-as-the-label-presents-it is not.
Software analog: every variable name in a running program refers
to a memory location, but the name is not the location. The
reified-entity (“the object named user”) is a convenience for
the programmer, not a thing the runtime contains. The mistake
of parikalpita is treating the name as if it were itself the
thing.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
The imagined nature is what happens when a Clear-domain
interpretation is applied to a Complex-domain phenomenon. The
label “this is X” is precise, fixed, transferable — but
the phenomenon it names is none of those things.
The pathology Cynefin warns against most strongly: treating
Complex situations as if they had fixed identifiable parts.
Parikalpita is the formal name for that mistake.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
The imagined nature is the posterior of the generative model
mistaken for the world. The brain produces a stable, object-like
model of “this thing in front of me.” That model is real (the
brain is doing it). But the thing as the model describes it
— with hard edges, fixed identity, persistence across
moments — is the model’s own structure imposed on what is,
computationally, a much messier inference.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
The imagined nature is the “spatialized” form of the durational.
To say “this thing” is to freeze flow into a bounded object. The
freeze is conceptual; the flow is what is happening.
The Bergsonian point: the imagined is not false in some absolute
sense; it is false to the mode of being it describes.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
“Has no own-being whatsoever” is the most easily misread phrase
in the entire text. It does not mean “nothing exists.” Svabhāva
is a technical term — “own-being,” “self-nature,” the
property of standing on its own as a discrete entity. The claim
is: the imagined-thing-as-imagined lacks svabhāva.
There is something. There is no isolated, fixed,
independently-existing thing answering to the concept.
The dependent nature (paratantra-svabhāva)
is discrimination arising from conditions.
The perfected nature (pariniṣpanna) is, in relation to that,
its constant separation from the former [imagined] nature.
The other two natures. Dependent nature (paratantra-svabhāva):
the vikalpa actually arising from conditions — the
substrate of experience, what is going on. Perfected nature
(pariniṣpanna-svabhāva): the paratantra seen as it is —
always already separate from the imagined projection laid over it.
memory
The Distributed Systems Engineer
DSE
Three layers, very precise:
Paratantra — the system’s actual computation. The
conditioned arising of representations. Real, dependent, messy.
Parikalpita — the projection laid over the
computation that treats it as having fixed-entity outputs.
Pariniṣpanna — the paratantra seen without the
projection. Same computation, different relationship.
The perfected nature is not a different substrate. It is the
same substrate with the projection layer removed. Engineering
intuition: same data, different view.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
The dependent is the actual Complex Adaptive System. The
imagined is the Clear-domain caricature of it. The perfected is
the practitioner’s accurate sense-making of the Complex system
— recognizing it as Complex, relating to it accordingly.
Cynefin would call the perfected “domain-appropriate response to
a Complex Adaptive System”: probe, sense, respond, no pretense
of control.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
The dependent is the brain’s actual inferential activity. The
imagined is the conscious sense of “objects being out there.”
The perfected is what predictive-processing theorists sometimes
call metacognitive insight into the generative model —
recognizing that one’s experience is constructed, without losing
the experience.
Pariniṣpanna is not a new content; it is a new relationship to
all contents.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
The three natures are not three things; they are three ways the
same flow can be related to. The Whiteheadian parallel: the
actual occasion can be prehended adequately or inadequately. The
data are the same; the relationship differs.
This is process metaphysics’ strongest move: insisting that
fundamental ontology is how things are happening, not what
things are.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
The three natures are not levels of reality, like layers of a
parfait. They are perspectives on a single dependent arising.
Reading them as a stratified ontology (parikalpita above
paratantra above pariniṣpanna) commits to a metaphysics the
verse may not require. The relation between paratantra and
pariniṣpanna is debated within Yogācāra: some readings make
them the same dependently-arisen flow under two descriptions;
others make them genuinely distinct.
Thus the perfected and the dependent
are neither different nor non-different,
as impermanence relates to conditioned things:
where the one [perfected] is not seen, the other [in its truth] is not seen.
The relationship between the perfected and the dependent.
They are not identical (the perfected is the dependent’s being
free of the imagined) and they are not different (the freedom
isn’t a separate thing from what is free). Vasubandhu compares
this to the way impermanence relates to its bearer: not the same
as the bearer, not different from it.
memory
The Distributed Systems Engineer
DSE
A subtle relational claim. The perfected is a property of the
dependent — specifically, the property of being devoid of
imagined projection. Properties are not identical with their
bearers (the property “transient” is not the same as the
transient thing) and they are not separate from them (the
property doesn’t exist independently).
Engineering analog: the “no-bug-found” state of a system is not
a separate thing from the system; it’s a description of the
system once a particular projection (the believed bug) has been
removed. Same system, different relationship.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
The Cynefin principle: properties of Complex systems are
aspect-descriptions of the whole, not modular components.
“Resilience” is not a part of a resilient system; it is the
system seen under a particular question. The verse insists on
the same architectural humility for the perfected nature.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
Insight-into-the-generative-model is not a new model. It is a
metacognitive relationship to the model that was already running.
Adding a new model would just be another parikalpita. The
shift is in stance, not content.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
“Neither the same nor different” is the standard Mahāyāna way of
refusing the substance-attribute split. In process terms, the
perfected is a mode of the flow, not a different flow and not
a part of the flow.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
The “neither same nor different” formula is doctrinally weighty
in Mahāyāna and not the obvious thing it sounds like. Don’t
paper over it with the property/bearer analogy too quickly
— that analogy commits to a logic the verse is partly
refusing.
Based on these three natures,
the three non-natures (niḥsvabhāvatā) are established.
Therefore, with hidden intent, the Buddha declared
that all dharmas are without own-nature.
Yogācāra’s account of anātman — no-self — matched to
its three-nature scheme. Each of the three natures has its own way
of lacking svabhāva (own-being), and these are catalogued as the
three non-natures (niḥsvabhāvatā). And Vasubandhu marks the
move: the Buddha’s well-known teaching that “all dharmas are
without own-nature” was given with hidden intent (saṃdhāya)
— it presupposed a more articulated threefold reading
underneath. This verse is the resolution of that hidden intent.
memory
The Distributed Systems Engineer
DSE
Three flavors of “no fixed identity”:
The imagined has no self-nature by definition — it
is a projection.
The dependent has no self-nature because it exists only
by conditions — it has no standalone identity.
The perfected has no self-nature because it is the absence
of imagined self-nature in the dependent.
Three forms of “doesn’t stand alone,” matched to three layers.
Yogācāra is doing modular ontology in negation form.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
Cynefin’s foundational claim is that systems in the Complex
domain do not have fixed components — the “components” are
themselves emergent and identity-shifting. The threefold no-self
is precisely this claim applied to the three modes of cognition.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
Compare contemporary anti-essentialism in cognitive science: the
brain has no fixed modules; cognitive categories are distributed;
the self is not a part of the brain. Each is a domain-specific
way of saying “no self-nature.”
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
Process metaphysics agrees fully. Whitehead’s actual occasions
have subjective form but no subjective substance; Heraclitus’s
river has pattern without substance. The threefold no-self is
just no-self stated three times for three modes of cognitive
engagement.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
“With hidden intent” (saṃdhāya) is doing serious doctrinal
work here. The Buddha’s apparent teaching of universal
no-own-nature is being reframed — Yogācāra is saying he
meant three different things by it, depending on which of the
three natures was in view. This is hermeneutically aggressive.
The Madhyamaka tradition (Nāgārjuna’s heirs) reads “all dharmas
without own-nature” as the direct teaching and treats Yogācāra’s
threefold gloss as an interpretive overlay. Which tradition
you take as authoritative shapes how this verse reads.
The first is non-nature as to character (of the imagined);
the next is non-nature as to arising (of the dependent);
the last is [the perfected], being separated from the former
grasped nature of self and dharmas.
Vasubandhu unpacks each of the three non-natures with precision:
parikalpita has no nature as to character (it never had one to
lose); paratantra has no nature as to arising (it exists only
by conditions, not by itself); and pariniṣpanna is itself
defined by being separated from the imagined self-and-dharmas
projection.
memory
The Distributed Systems Engineer
DSE
Three precise reasons:
Imagined: never had self-existence. The label was always
just a label.
Dependent: causally constituted. Like every value in a
pure function, its existence is its computation; no input,
no output.
Perfected: is itself the no-self nature — the
recognition of the absence is what this third nature is.
Beautiful symmetry: the third nature is not just lacking self;
it is the explicit recognition of the lack.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
The three flavors of absence correspond to three Cynefin-style
diagnostic moves:
For the imagined: notice the projection. The thing you were
treating as fixed is your own label.
For the dependent: trace the conditions. What changes if the
conditions change?
For the perfected: recognize that the system you were
analyzing has no separable parts. The recognition is the
practice.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
Each absence-of-self corresponds to a different cognitive
operation:
Recognizing labels as labels: a metacognitive achievement.
Recognizing conditioning: a causal-reasoning achievement.
Recognizing the absence as the achievement: a deeper
metacognition.
Each requires the previous. The text is also a curriculum.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
Three modes of the dependent flow. Each absence is a refusal to
spatialize. The imagined is the most spatial; the perfected is
the least.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
The cleanness of the threefold schema is itself a parikalpita.
Watch for the slide into “now I have a theory of no-self that I
can hold onto.” The third absence explicitly rules this out.
This is the ultimate truth (paramārtha) of all dharmas,
and it is also suchness (tathatā):
because it is forever thus in its nature,
it is the true reality of consciousness-only.
The third nature gets its full identification. It is the ultimate
truth (paramārtha) of all dharmas, and it is also suchness
(tathatā) — “thus-ness.” The point Vasubandhu drives home:
it is always already thus. The perfected nature is not produced
by practice; it is recognized. Consciousness-only is the true
reality of things whether or not anyone has realized it.
memory
The Distributed Systems Engineer
DSE
A crucial architectural point: the substrate’s actual mode of
operation is vijñapti-mātra whether or not the system notices.
The transformation is what it is. Realization is the system
noticing what is already the case — not a new feature
being installed.
For an engineer: this is the difference between adding
observability and adding behavior. The perfected nature adds
observability to a system that was already projecting.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
“Always already thus” is the invariant of the system. The
Complex Adaptive System has been operating as a Complex Adaptive
System the whole time, regardless of how the agents inside it
have been describing it. Cynefin would call this the true
domain; the perfected nature is the practitioner’s accurate
recognition of it.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
The brain has been doing predictive processing for its entire
operational history. The insight that perception is constructive
doesn’t change what the brain does; it changes how the system
models its own perception. Tathatā is the fact; recognition is
when the system catches up.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
The perfected nature is the truth-condition of process
metaphysics. That things are flow, that nothing stands alone,
that arising is dependent — this is the case whether
anyone has noticed or not. The perfected is not a construction;
it is the noticing.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
“Always already” can be read in soothing tones, as if to say:
relax, you were always enlightened. The tradition refuses that
reading. The perfected nature being always-already the case does
not mean a practitioner is always-already realized. The
recognition matters and is hard-won. Verse 26 will make this
explicit.
So long as the [non-conceptual] cognition has not yet arisen,
while one still merely seeks to abide in consciousness-only,
the latent tendencies of the dual grasping
cannot yet be subdued and extinguished.
Verses 26–30 are the soteriological arc: how the system actually
changes. The opening move is diagnostic. Non-conceptual cognition
— the cognition that does not split into a knower confronting
a known — has not yet arisen. Until it does, the latent
tendencies (anuśaya) of the dual grasping will not stop, no
matter how much one seeks to abide in consciousness-only.
memory
The Distributed Systems Engineer
DSE
A precise causal claim: feedback loop A (grasping tendencies)
cannot terminate until cognitive event B (non-conceptual
cognition resting in vijñapti-mātra) occurs. The two are
interlocked: A keeps producing the conditions that block B; B is
the only thing that interrupts A.
From a control-theory perspective: a system locked in an
attractor. Standard reasoning, standard willpower, standard
introspection all operate within the attractor and reinforce
it. Escape requires a qualitatively different operation.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
In the Complex domain, applying Clear-domain interventions does
not work. The grasping tendencies are a Complex attractor;
trying to suppress them directly is a Clear-domain move that
intensifies them.
The escape is what Cynefin calls “creating a different
attractor” by altering the conditions, not by force.
Non-conceptual cognition is exactly that conditional move
— and the verse is careful to note that seeking (a
goal-directed Clear-domain operation) doesn’t get you there.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
Compare clinical work on rumination: trying not to ruminate
reliably worsens rumination. The escape is a different stance
toward thinking itself — the metacognitive shift from
being-the-thinking to seeing-the-thinking.
The verse’s “non-conceptual cognition” is structurally the same:
a shift from inside the contents to a stance toward the
contents, but one that doesn’t itself become a content.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
The grasping is not a content the system contains; it is a mode
of the system’s process. Modes do not stop by declaration; they
shift by the system entering a different mode. Non-conceptual
cognition is the alternative mode.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
The bracketed “[non-conceptual]” gloss in the English is doing
real work. The Sanskrit literally says “consciousness has not yet
abided”; the specification of what kind of consciousness this
means is provided by the next two verses, not this one. Don’t
pre-empt the path by deciding now what “abiding in
consciousness-only” feels like. The verse explicitly says: as
long as you are seeking it, you are not there.
To set up even some slight object before oneself
and call it ‘the nature of consciousness-only’ —
because something is still being grasped,
this is not truly to abide in consciousness-only.
A devastating verse. If you set up “this is consciousness-only”
as something to know — even as the slightest object before
your mind — that very setting-up is itself a graspable, and
you have not yet abided. The verse is precise: because something
is still being grasped, this is not yet abiding. The correct
intellectual position is not the same as the realization.
memory
The Distributed Systems Engineer
DSE
The recursion problem. The system that recognizes the substrate
is also part of the substrate; its recognition is just another
representation. You cannot solve the representation problem by
producing one more representation.
Engineering analog: you cannot debug a runtime from inside the
runtime using only operations the runtime provides. Verse 27
says: trying to do that more cleverly just produces more runtime
state. “This is consciousness-only” is the cleverest extra state,
and it is still extra state.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
A failure mode Cynefin specifically warns against: thinking you
have shifted domains when you have only relabeled your operation.
Saying “I will treat this as Complex” is itself a Clear-domain
move unless the underlying practice changes.
The verse’s correction is exact: relabeling is not transitioning.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
The contemporary parallel is Daniel Dennett’s worry about
“Cartesian theater” tricks — where insights about
consciousness keep recreating the homunculus they were trying to
dissolve. Verse 27 anticipates this exactly: the “I know that
this is consciousness-only” moment installs a knower who knows
the consciousness-only. The structural problem persists.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
The verse refuses the reification of insight itself. The process
cannot be brought to rest by an act of process. To rest is not
an act; it is the cessation of the kind of act that produces
grasping.
Whitehead’s term: the negative prehension — what is not
actively included — matters as much as what is. Verse 27
points at it.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
This is the verse that breaks every reader’s intellectual
confidence, and it should. If you finish reading verse 27
feeling you understand it, verse 27 has just demonstrated its
point on you.
When, regarding the objective support,
cognition attains nothing at all,
then one abides in consciousness-only,
for the marks of dual grasping are gone.
The resolution. Abiding in consciousness-only is not a positive
perception (“I see that this is consciousness-only”) but a
cessation of the object-perception that produces grasping.
When, regarding the objective support (ālambana), cognition
attains nothing at all, there is no graspable; without a
graspable, no grasping; and then — only then — there
is abiding.
memory
The Distributed Systems Engineer
DSE
A tautological-sounding statement doing precise work. Grasping
requires an object to grasp. The object is constituted by the
cognition that grasps it. If the cognition does not constitute
an object — if there is no ālambana — there is no
object to grasp, and the grasping does not arise.
Engineering analog: a system that doesn’t render an entity
cannot have a stale reference to that entity. The render is
where the reification happens; remove the render, and the chain
of stale references upstream of it does not occur.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
The escape from the Complex attractor: not by forcing the state,
but by removing the condition the attractor depends on. Cynefin
calls this “altering the constraints.” The grasping does not stop
because the system overpowers it; it stops because there is no
object for it to engage.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
Compare the deepest meditative states: when sensory-object
cognition is sufficiently quieted, the elaborated self-modeling
that depends on object-cognition also quiets. Contemporary
contemplative neuroscience (Brewer, Vago) has documented
something structurally similar — reduced default-mode
network activity tracking reduced object-elaboration.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
The cessation of grasping is not an event but the absence of an
event. The substrate continues, but in a different mode. The
wave that was a wave by being a wave is now the water without
the wave.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
“Cognition attains nothing at all” is an extreme statement and
is meant to describe an extreme state. The verse is not saying
ordinary perception is delusion; it is describing the specific
cognitive condition under which grasping ceases. That condition
is rare, cultivated, and not the same as ordinary distraction or
absence of attention.
This non-attainment, this inconceivability,
is supramundane wisdom (lokottara-jñāna).
By casting off the two kinds of coarse encumbrance (dauṣṭhulya),
one realizes the transformation of the basis (āśraya-parāvṛtti).
The decisive verse. Non-attainment (anupalambha) — the
cognition that attains nothing — and inconceivability
(acintya) together name supramundane wisdom
(lokottara-jñāna). By the casting-off of the two kinds of
coarse encumbrance (dauṣṭhulya) — the latent heaviness
from the afflictive and the cognitive obscurations — the
āśraya-parāvṛtti is realized: the transformation of the
basis, promised back in verse 4.
memory
The Distributed Systems Engineer
DSE
The system event verse 4 announced is now described
operationally. Without conceptual content and without conceptual
elaboration, awareness operates in a different mode. The
substrate is not being deleted; it is being turned
— parāvṛtti.
Two kinds of encumbrance shed:
Afflictive dauṣṭhulya — the latent heaviness left
by the six root afflictions and their twenty derivatives.
Removing it ends the running interference on perception and
action.
Cognitive dauṣṭhulya — the structural heaviness
left by the assumption that knowledge is of independent
objects. Removing it ends the constitutional misreading.
Removing the first changes ethics; removing the second changes
ontology.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
Supramundane wisdom is not domain-relabeled cognition; it
operates outside the domain typology altogether. Cynefin is a
sense-making framework for ordinary cognition; the verse is
describing a state where the conditions for ordinary
sense-making have themselves been altered.
Cynefin’s response: at this point, the framework hands the
inquiry off to the practitioner-tradition. We don’t have
map-tools for this.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
Modern cognitive science has some analogues: “non-dual
awareness” research, advanced-meditator studies,
psychedelic-induced ego dissolution. None is the supramundane
wisdom Vasubandhu describes, but the structural feature
— cognition operating without the usual subject-object
schema — is empirically observable in trained
practitioners.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
Āśraya-parāvṛtti is the deepest possible process-philosophical
claim: the mode of the flow itself can change. Not the
contents, not the patterns — the mode. The river becomes
a different river.
This is what Whitehead would call a “change in the categoreal
scheme” of becoming — rare, structural, irreversible.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
The Skeptic’s hardest moment. The engineering schema so far
— layers, logs, feedback loops — has been useful
precisely because Vasubandhu was describing structures the
schema could match. Verse 29 describes a state where the
structures themselves change. The schema has no traction here;
it can only point at the door.
Take this as a feature of the text, not a limit of the reading.
This is the untainted realm (anāsrava-dhātu) —
inconceivable, wholesome, eternal,
blissful, the body of liberation:
the Dharma-body (dharmakāya), named [the body] of the Great Sage (mahāmuni).
The final verse, the one that most resists the technical reading.
Anāsrava-dhātu — the untainted realm, “without
outflows.” Acintya (inconceivable), kuśala (wholesome),
dhruva (eternal). The vimukti-kāya (body of liberation) and
the dharmakāya (Dharma-body) of the mahāmuni (Great Sage).
Vasubandhu ends not with doctrine but with a fruit-verse
— pointing past the system at what its overturning makes
available.
memory
The Distributed Systems Engineer
DSE
The engineer’s reading runs out here, as it should. Acintya
means “not capturable by the conceptual apparatus.” Every
system description offered across the preceding 29 verses is a
conceptual apparatus. The state Vasubandhu names is, by
definition, outside the kind of description that has been built
up.
The honest engineering observation: a system that has been
operating in mode A for its entire operational life has no
internal vocabulary for mode B. Verse 30 acknowledges this and
stops speaking technically. The reading should follow.
Draftnot yet reviewed
hub
The Cynefin Practitioner
CYN
Cynefin has a fifth domain — “Confused” — and a
boundary called the “cliff of complacency.” Neither matches
verse 30. The verse is not describing a domain on the map; it
is describing the practitioner’s relation to the map once the
substrate that produced the need-for-maps has turned. We are
off-map. Maps are for travelers; verse 30 is not about a
traveler.
Draftnot yet reviewed
psychology
The Cognitive Scientist
COG
“Eternal, untainted, wholesome, inconceivable.” Cognitive
science has reports of states that fit pieces of this
description — deep unitive experiences,
very-advanced-practitioners’ baseline states — but none
is what verse 30 describes. The science can register that the
state exists; the state’s nature is, by Vasubandhu’s insistence,
not science’s object.
Draftnot yet reviewed
water_drop
The Process Philosopher
PRO
Process philosophy ends on a question: what is process when it
has stopped producing graspings? The verse answers in poetic,
not technical, vocabulary: the dharmakāya. Whitehead might
have called this the consequent nature — the inclusive
togetherness of all that is — but he would have known he
was using a placeholder for a reality his system couldn’t fully
grip.
Draftnot yet reviewed
report_problem
The Skeptic
SKP
Three things to notice. First, the verse is poetic. It uses
the language of bodies (kāya) and realms (dhātu); these are
not technical descriptions but pointing-words.
Second, the engineering reading would be a betrayal if it tried
to extend itself here. The Skeptic’s final job is to say: the
system schema does not own this verse. Let it not own it.
Third, the text ends. Thirty verses. That number is the point
of restraint Vasubandhu set himself, and the restraint is also
a teaching. Anything you would add at this point is
parikalpita.