二識

*A Yogācāra key for a mind that has forgotten it is one mind*

這個概念在說什麼

In the Yogācāra (瑜伽行派) / Consciousness-Only (唯識) tradition, vijñāna (識) — what we loosely translate "consciousness" — is never left as a single undifferentiated blob. The school is famous for its eight-consciousness system (八識): the five senses, the mental factor, the afflicted mind (末那識), and the storehouse (阿賴耶識). But alongside that taxonomic project, the tradition also performed a more functional analysis. It asked, *in any given moment of knowing, what is the mind actually doing?*

The 《攝大乘論釋（無性釋）》(*Vijñaptimātratā-vyākhyā* commentary by Wu-xing, T1598), the Indian commentary translated by Xuanzang 玄奘, offers one of its most surgical answers in a four-line verse:

「一則名緣識，第二名受者，此中能受用，分別推心法。」

*One is called 緣識 (yuán-shí, "grasping/conditioned consciousness"); the second is called 受者 (shòu-zhě, "the enjoyer"). In these, the one that can partake-and-use is the enjoyer; the one that discriminates-and-infers mental phenomena is the grasping consciousness.*

Two consciousnesses, distinguished not by their *contents* but by their *functions*. Let me unfold each carefully, because the modern mind, used to neuroscience and computer metaphors, is almost guaranteed to map them wrong.

1. 緣識 (Yuán-shí) — The Grasping/Discriminating Consciousness

The character 緣 in Buddhist Chinese is technical. It means "to take as condition," "to grasp toward," "to engage with objects." You see it in 攀緣 (climbing toward conditions) and 緣起 (conditioned arising). So 緣識 is not just "consciousness that thinks" in any vague sense — it is consciousness *in its grasping mode*, the mode in which mind reaches out, grasps an object, *and goes to work on it*.

What work? The verse says 分別推求心法 — *to discriminate (分別), infer (推), and seek out (求) mental phenomena (心法)*. This is a tightly packed verb triad, and each word adds something:

- 分別 — to differentiate, to parse, to cut reality into categories. "This is a face. This is a face I recognize. This face belongs to my mother. Her expression looks worried. Worried means…" - 推 — to infer, to draw inferences, to operate on what has been parsed. "If she looks worried, perhaps she is upset with me. If she is upset with me, perhaps it is about yesterday's call." - 求 — to seek, to actively probe, to go looking for further specification. "What exactly is the worry? Is it financial? Relational? Health?"

So 緣識 is the mind's investigative arm — the part of consciousness that *treats the world as a problem to be solved*. It is the function that, given any phenomenon, immediately begins to break it into parts, classify it, name it, infer from it, and probe further. It is precisely what contemporary psychology calls *cognitive appraisal*, *conceptual processing*, or — in the AI literature — the *inference engine*.

This is also why it is called *grasp*ing consciousness: it is the function that *reaches toward* objects, that cannot rest until it has pinned each thing down. It is the hungry, classifying mind.

2. 受者 (Shòu-zhě) — The Enjoyer / The Partaker

The second consciousness is named not by what it does *to* objects but by what it does *with* them. 能受用 — *to receive (受) and use/enjoy (用)*. The Sanskrit term behind this in the Yogācāra sources is *bhoktṛ* — "the enjoyer" or "the one who partakes." And "enjoyment" here must be taken in its widest possible sense. It includes:

- Enjoying a meal, a sunset, a piece of music. - Suffering a headache, a grief, a humiliation. - Being neutral, registering an event without strong affect. - Appropriating an object — taking it in, making it mine, integrating it into "my experience."

受者 is the mind's *receptive-and-appropriative arm*. It is the function that, when consciousness has been "grasped" by some phenomenon, then actually *partakes of* it. If 緣識 is the hand that reaches out and grabs, 受者 is the mouth that tastes — though "tastes" must be understood to include bitter tastes too.

The crucial point: 受者 is not identical with the Buddhist term 受 (vedanā, "feeling," one of the five aggregates). 受 in the five-aggregate scheme is specifically *the affective tone of experience* — pleasant, unpleasant, neutral. 受者 is broader: it is the entire function of *taking in and using* objects, including but not limited to the affective dimension. So mapping 受者 → "the feeling aggregate" would be a flattening of the concept, and one I will return to in the misreadings section.

3. The Functional Distinction, In One Image

Picture two moments at a wine-tasting:

- 緣識 at work: "Pinot Noir. 2019. Burgundy. Notes of cherry, forest floor, light tannin. Slightly oxidized at the rim." - 受者 at work: *the actual sip, the warmth, the slight bitterness, the smile, the swallow, the "mmm."*

Notice these are not two different minds or two different people. They are two distinguishable *functions* of the same stream of knowing. The first *unpacks* the object into conceptual content; the second *takes in* the object as lived experience. Both are needed; neither alone makes the tasting complete.

4. The Crucial Constraint: 二識不並生

Here is where the Yogācāra analysis becomes rigorous in a way that contemporary metaphors often miss. The same text issues a hard limit:

「無處無容，非前非後，同身同類，二識並生。」

*There is no place, no occasion, no [state of being] "before" or "after," in which two consciousnesses of the same body and the same type arise together.*

This is a strong claim. Read it slowly. Two vijñānas — of the *same body* and the *same type* — cannot arise simultaneously, nor immediately in succession as if simultaneous. The commentator Wu-xing is putting a boundary on the simultaneity thesis: mind does not run two of these functional streams in parallel; it *alternates* between them, even when the alternation is so fast that ordinary introspection cannot catch the seam.

Why does this matter for a contemporary reader? Because the modern assumption — quite plausibly from the standpoint of cognitive science — is that analysis and feeling are *parallel processes*, simultaneously running, the brain doing both at once. The Yogācāra position here is subtler: even if functionally distinct, they are *serial*, not parallel. They are two modes of one stream. And this serial structure has implications for meditation and self-knowledge that we'll come to.

(One small caution on the wording: the classical Chinese allows a reading in which the "no" might distribute only over "together," so that "no place/time makes them arise as if together" is the cleanest translation. This reading is consistent with the broader Buddhist moment-to-moment analysis in which mental events are discrete but contiguous. Either way, the operative phrase is *not simultaneously*.)

5. Why This Is *Not* the Same as 二性 (相、見)

A reader of the same text will immediately encounter a parallel "two" that is easy to confuse:

「由二性者，謂相及見。於一識中有相有見二分俱轉，相見二分不即不離。」

*By "two natures" (二性) is meant the object-aspect (相, *nimitta/ākāra*) and the perceiving-aspect (見, *darśana*). Within one consciousness, an object-aspect and a perceiving-aspect run together; they are neither identical nor separate.*

This is a *structural* analysis within *one* consciousness: every act of knowing has a "shape it presents" (相) and a "side that knows" (見), and these two run *together* (俱轉). The Chinese commentators unpack it as the *相分* (the aspect-as-part — how the object appears) and the *見分* (the seeing-as-part — how it is known).

But — and this is the precise point the text insists on — *this is not the same "two" as 二識*. The 相/見 二性 is "the two within one consciousness, running together." The 緣識/受者 二識 is "two consciousnesses distinguished by function, not arising together." They sound alike, both are "二" something, both involve the analysis of vijñāna, but they are doing different work:

| | 二性 (相/見) | 二識 (緣識/受者) | |---|---|---| | Level | Inside one consciousness | Between two consciousnesses | | Relationship | Simultaneous | Non-simultaneous (in same body, same type) | | Question answered | *How* does one consciousness know? | *What kind* of knowing is happening? | | Function | Subject-side / object-side structure | Discriminating / enjoying |

Conflating these is one of the most common errors I see when contemporary readers first encounter Yogācāra, and I will return to it below.

6. The Alternative Voice: 一意識

The same text records that not every teacher in the tradition agreed to the 二識 split:

「又於此中有一類師說一意識，彼彼依轉得彼彼名，如意思業名身語業。」

*There is a class of teachers who assert only one consciousness (一意識). It runs depending on various supports and thereby acquires various names — just as a single intention, when manifested through the body, is called bodily action, and through speech, verbal action.*

This is the *one-consciousness* thesis: there's really only one stream of vijñāna, and the names "eye-consciousness," "ear-consciousness," "grasping consciousness," etc., are labels based on which *āśraya* (support, base) the stream is currently flowing through — just as one intention gets called by different names depending on whether it issues in body or speech.

Wu-xing does not adjudicate between this view and the 二識 view. He records both as live options within the tradition. For our purposes, the important thing is this: *whether* you split vijñāna into two, into eight, or leave it as one, the *functional* distinction between grasping-and-discriminating and receiving-and-using remains descriptively useful. The 二識 classification names a real contrast in how mind works, regardless of one's metaphysical commitments about the *number* of consciousnesses.

生活裡的走查

Let's walk this distinction through a few scenes that contemporary readers will recognize, and watch carefully where the mapping is tight and where it tends to slip.

Scene 1: The Inbox (a Tuesday morning email)

You open your laptop. A new message arrives. Watch the sequence:

1. *Sender's name appears.* — 緣識 immediately fires: "Ah, the director. From the People team. Subject line contains 're: schedule.'" Names, categories, hierarchies, implications — all unpacked before you've read a single word. 2. *You read the body.* — 緣識 continues: "She is asking me to shift Friday's meeting to Thursday because… the visiting partner can only make Thursday… she wants to know by end of day." 3. *The body receives the impact.* — Now 受者 engages: there's a small contraction in the chest (Friday was the morning you'd booked for focused work), a mild irritation (you've already planned around Friday), then the faint beginning of resignation. *That* — the felt quality of being pushed around by someone else's schedule — is 受者 at work. 4. *Then 緣識 re-engages.* — "If I say yes, I'll have to move X. If I say no, she'll think I'm inflexible. If I counter-offer Wednesday, that requires…"

Notice the *seriality*. Even in a thirty-second reaction, the mind has already alternated: 緣識 → 受者 → 緣識 → 受者. They are not running in parallel. The felt irritation (受者) modifies what the next inference (緣識) will count as relevant. The next inference modulates whether the next felt quality is guilt or defiance.

Where contemporary readers often go wrong: assuming the felt irritation (受者) is somehow "after" or "downstream" of the cognitive assessment (緣識) in a simple causal sequence — "I evaluated the threat, then felt bad." The Yogācāra model here is more like a *coupled oscillation* than a single arrow, with the strict constraint that any single *moment* of vijñāna is one or the other, never both at once. This is a subtle empirical claim that doesn't quite map onto the brain-science picture of two hemispheres or two parallel systems "doing both at once." It is closer to a description of how attention works: at any given micro-moment, attention is on *this* (the inference) or on *that* (the feeling), and the seam is just very fast.

Scene 2: Eating lunch alone

You sit down with a bowl of noodles.

1. *緣識* at the door: "Looks like dan dan noodles. Sesame paste. Scallions. A bit too much chili oil. Vendor near the office, ¥28." 2. *受者* at the first bite: heat, salt, the chew of the noodles, a sudden wave of comfort, a memory of your grandmother's kitchen flickers and is gone. 3. *緣識* sneaks back in: "Hmm, that's interesting, I was just thinking about her. Why now? Maybe because the texture. The texture is similar to what she used to make." 4. *受者*: the memory brings a small warmth, then a small ache.

Again, alternating. And here is the contemporary trap: most of us are trained to live almost entirely in the 緣識 phase even while eating. We eat *while reading*, *while on a call*, *while planning*, *while answering messages*. The 受者 phase is starved. We taste almost nothing — not because our palate is broken, but because we never let consciousness *be in the receiving mode* for more than a flicker. The food becomes fuel for the analytical machine.

Where the mapping slips: it is tempting to think "受者 = sensory / 緣識 = conceptual." But this is too narrow. 受者 includes the *felt sense of meaning*, the implicit wisdom of the body. 緣識 includes *recognition*, *categorization*, even some kinds of pattern-recognition that feel intuitive. The distinction is functional, not sensory-versus-conceptual.

Scene 3: An argument with someone you love

It begins — as most arguments do — not with a feeling, but with 緣識 firing fast and dirty:

1. *緣識*: "She just implied I don't care about her family. That sentence has the structure of accusation. It tracks back to last Tuesday's comment about the call I missed. She's accumulating. She is doing that thing where she…" 2. *受者*: a flush of shame, an urge to defend, a tightness in the throat. 3. *緣識*: "If I apologize now, it will be interpreted as a pattern. If I explain, she'll escalate. If I go silent, she'll escalate more. The optimal move is to…"

You already see the trap: the 緣識 of "what is the optimal move" *is itself a way of avoiding 受者*. The felt shame, the throat-tightness — that is the part that actually has something to teach. By staying in 緣識, we keep strategizing, never arriving at the feeling that is asking to be received.

In a more skillful version of the same argument, what happens?

1. *緣識*: "She is upset about last Tuesday." 2. *受者*: a slow, honest wave of hurt arrives. Noticed. Not acted on. 3. Then — only then — 緣識: "Given that I actually do feel hurt, what would be a true thing to say?" 4. Then a word is chosen — and only that word, because the analytic phase now has different data to work with.

This is the practical relevance of the *non-simultaneous arising*: if both ran in parallel, you couldn't *choose* to be in 受者 first. The whole possibility of contemplative practice rests on the fact that mind is serial, that you can put attention on one mode and let the other wait. (One of the most striking empirical findings in modern mindfulness research is exactly this — that affect-labeling and body-sensing actually *change* the cognitive appraisals that follow. The neuroscience is correlational; the Buddhist claim is constitutive. Don't mistake one for the other.)

Scene 4: Scrolling before sleep

This one I want to walk slowly, because it is where contemporary life has produced something almost unprecedented in human history.

You lie in bed. Phone in hand. The sequence of micro-events:

1. Each new image triggers a 緣識 burst: a face, an outfit, a status, a political stance, a meme — labeled, categorized, judged, in less than a second. 2. Each judgment triggers a micro 受者: a tiny envy, a tiny contempt, a tiny relief, a tiny hope, a tiny despair. So fast you don't notice. 3. The next 緣識 arrives before the previous 受者 has been received. The previous 受者 is *interrupted*, not felt. 4. After forty minutes, you put the phone down and feel vaguely bad, vaguely anxious, vaguely empty. You can't quite say why.

What's happened, on the Yogācāra analysis? *受者 has been almost entirely starved*. The analytical machine has run continuously, with the receiving function being interrupted before it could complete its work. So all those micro-feelings are still there — unfinished — accumulating in a kind of affective sediment. This is *not* a metaphor for what neuroscience calls "limbic saturation" — that is a parallel finding, and I should be careful: neuroscience describes correlations and patterns; the Yogācāra model describes a functional structure. They rhyme, but they aren't the same claim.

The classical model would say: the symptom you feel — the vague badness — is the consequence of 二識不並生 being violated in practice. You tried to make them run in parallel. The mind cannot. So 受者 silently *queued up* unprocessed material, which then leaks out as undifferentiated unease.

當代人為什麼需要它

Why does this rather technical distinction from a 7th-century Indian commentary deserve a place in a contemporary reader's mental furniture? Three reasons.

1. It names a real split you can feel

A lot of modern malaise has the shape of a *dissociation between analyzing and receiving*. Therapy culture has many names for it — overthinking, alexithymia, emotional unavailability, intellectualization as defense, "in my head" versus "in my body." The Yogācāra diagnosis, two thousand years before those terms, said: this is because your 緣識 is running hot and your 受者 is running starved, or alternately, your 受者 is overwhelmed and your 緣識 is offering rationalizations instead of inquiry. The 二識 framework gives you a vocabulary that is *functional* (not psychodynamic, not neurochemical) and that maps the split without collapsing either side.

2. It contradicts the prestige hierarchy of the analytical mind

Contemporary culture idolizes 緣識. To be "smart" is to be quick at parsing, classifying, inferring, naming. The receiving function — "I just sat with this and let it in" — is treated as passive, soft, even regressive. The 二識 framework says no: *both are consciousness*. Both are vijñāna. Both are active modes of knowing. To refuse to receive is not to transcend emotion — it is to leave half your cognition unemployed.

The Buddhist critique of the pride of analysis is ancient: the analytic mind *in itself* does not liberate; it only liberates when joined with the receptive function that lets it *see* what is actually present, rather than only what it has classified. This is why the tradition places vipaśyanā (analytic insight) and samatha (calm abiding, the receptive function in concentrated form) as a *pair*, never as alternatives.

3. It points to a practice, not just a theory

The most useful thing about the non-simultaneous constraint — 「非前非後，同身同類，二識並生」— is that it is also an instruction. *If you are stuck in 緣識, you must wait for 受者 to arise in its own time; you cannot force it.* And vice versa: if you are stuck in raw 受者 (overwhelm, undifferentiated feeling), you must invite 緣識 in to *parse* what is being received. Practice in this tradition is precisely the cultivation of the alternation, the willingness to *be in one mode at a time* and let the other come.

Modern mindfulness instructions often say "notice what you are feeling" or "label the thought." The Yogācāra model gives those instructions a metaphysical backbone: noticing is shifting toward 受者; labeling is shifting toward 緣識; the practice is to do each one cleanly, in turn, without smuggling the other in. This is a more demanding version of the popular version, but also more honest.

(One caveat, important to honor: the broader Buddhist tradition contains other models — the five aggregates (五蘊), the twelve ayatanas (十二處), the eighteen dhātus (十八界) — each with its own practical implications. The 二識 distinction is one tool among many. It is not meant to replace the others, and a sophisticated practitioner would not treat it as the final word.)

常見誤讀與澄清

Misreading 1: "二識 = 理智 vs 情感" (Reason vs. Emotion)

The most common modern gloss is: 緣識 = the rational/intellectual function, 受者 = the emotional/feeling function. This is approximately true, but it flattens what is actually a more interesting distinction.

Why it fails: 緣識 includes non-rational pattern-recognition, intuitive categorizations, and even quasi-perceptual recognitions (this is "mother," this is "threatening," this is "familiar"). 受者 includes not just emotion but also the felt sense of meaning, the implicit bodily wisdom, the sense of an object as *mattering* before any emotion is named. The map "head / heart" — while seductive — pushes too much into 緣識 (most "thinking" has affective valence) and too little into 受者 (most "feeling" has structural content).

The classical analysis is functional, not ontological: 緣識 is whatever is *reaching-toward-and-differentiating*; 受者 is whatever is *taking-in-and-using*. Each act of consciousness can be analyzed as one or the other; some acts are mixed in close succession.

Misreading 2: "二識 = 主體 / 客體" (Subject vs. Object)

This is the philosophical gloss. It maps 緣識 to "the cognizing subject" and 受者 to "the object being cognized." It is a misreading encouraged by familiarity with Kant and modern epistemology.

Why it fails: 受者 is not the object; it is *consciousness in its receiving mode*. The enjoyer of the wine is not the wine; it is the knower when the knower is partaking. The relation is internal to vijñāna, not between vijñāna and a separate object. The Yogācāra system is explicit on this: there is no knower-out-there, only consciousness knowing-as-this-or-that. The 相/見 二性 already covers the subject/object structure (perceiving-aspect / object-aspect, *both within one consciousness*); 緣識/受者 is a different question.

Misreading 3: "二識 = 二性" (Conflating 二識 with 相/見)

Because both involve the word "two," both are introduced in the same text, and both are about vijñāna, many readers slide them together. I have watched this happen in study groups many times.

Why it fails: they live at different levels and have different structures (see the table above). 二性 is *simultaneous within one consciousness*. 二識 is *non-simultaneous across two consciousnesses of the same type in the same body*. If you collapse them, you lose the central feature of each. The 相/見 pair runs together; the 緣識/受者 pair alternates.

Misreading 4: "受者 = 受蘊 (the feeling aggregate)"

The Chinese character 受 is shared, but the terms are not interchangeable.

- 受蘊 (vedanā-skandha) is the affective tone of experience — pleasant, unpleasant, or neutral — and is one of the five aggregates (五蘊). The famous Heart Sutra line — 「照見五蘊皆空，度一切苦厄」 (*the Heart Sutra*: "having seen the five aggregates as empty, [one is freed] from all suffering") — sits in this conceptual neighborhood, but is about the five aggregates as a whole. - 受者 (bhoktṛ) is a *consciousness* (vijñāna) in its receiving-applying mode, broader than affective tone, encompassing all "taking-in-and-using" of objects.

Conflating them will lead you to think the 二識 framework is about feelings, when in fact it is about *how knowing works*.

Misreading 5: "This is just modern neuroscience with old words"

It is tempting, and partly true, to see parallels. Cognitive appraisal (Lazarus), affect labeling (Lieberman), the predictive-processing literature, even the dual-process accounts (Kahneman's System 1 / System 2) all sound at moments like restatements of 緣識/受者. Some contemporary Buddhist-adjacent writers enthusiastically announce that "neuroscience has confirmed the Buddha."

It hasn't, and it can't — not in the strong sense. Neuroscience describes *correlates* and *patterns* in a nervous system; Yogācāra describes *functions of consciousness* in a metaphysical framework that includes rebirth, storehouse consciousness, and the non-duality of perceiver and perceived. Where they rhyme, the rhyme is informative. Where they diverge, neither is refuted. A careful contemporary reader holds both as separate maps of overlapping territory, not as one map with two names.

Misreading 6: "二識 means there are two selves"

The opposite, actually. The whole point of the Yogācāra analysis — and the broader 唯識 framework — is that there is no enduring self behind these functions. 緣識 is not "your intellect," 受者 is not "your soul." They are *functions of the stream of consciousness*, with no fixed owner. The 二識 distinction is a tool for *deconstructing* the sense of a unified self, not for finding one. Even when a moment feels like "I, the knower, am enjoying this wine," the analysis says: that "I" is just the seam where receiving and discriminating are meeting. Pull the seam apart; no owner remains.

A final word.

The reason a four-line verse from a 7th-century Indian commentary still has traction is not that it has been confirmed by fMRI. It is that *the distinction it names is in your experience right now*, and the misuses it diagnoses — over-analysis, starved reception, the prestige of the head, the refusal of the heart, the illusion that both can run at once — are among the most characteristic distortions of contemporary consciousness.

The verse is asking you to look:

When you know, are you reaching out and naming? — that is 緣識. When you know, are you receiving and partaking? — that is 受者. They cannot both be happening *now*. So which one is?

That question, asked honestly and repeatedly, is the beginning of the practice that the commentarial tradition is pointing at. Everything else is commentary.