What this concept is actually saying
If you have ever tried to introduce a friend and realized that you cannot point to "the person you mean" without also saying *who they are, what they look like, what they do*, you have already stumbled onto the heart of what 能相 (lakṣaka, "that which characterizes") is about. It is a term from the *Mahāyāna-sūtrālaṃkāra* (《大乘莊嚴經論》, T1604), one of the foundational Yogācāra texts, and it is a paired concept: 能相 only exists in relation to 所相 (lakṣya, "that which is to be characterized"), and vice versa. The two together exhaust the category of ākāra, "characteristic" or "aspect." The text opens by saying the Buddhas teach the difference between 所相 and 能相 "in order to gather in and benefit sentient beings," which is a quiet way of saying: *if you confuse the two, you'll misunderstand how knowing works.*
So before we go further, the relation: 所相 is what gets pointed to — the thing, the phenomenon, the dharma. 能相 is the act of pointing-to — the characterizing, distinguishing, displaying function that takes the 所相 as its object and renders it knowable as *this* rather than *that*. Take away either side and the other collapses. Without a 所相, the 能相 has nothing to characterize; without the 能相, the 所相 remains undifferentiated. The two are correlative.
First: the 所相, the things being characterized
The text lays out five categories of 所相, and notes that these are the Buddha's *brief* teaching — expanded, the list is unlimited. The verse-summary gives them memorable one-word handles:
1. 色法 (rūpa-dharma) — material phenomena. The verse-name is 「共」 ("shared") because form is what multiple senses can share and what is accessible to others. 2. 心法 (citta-dharma) — mind as such. Verse-name: 「心」. 3. 心數法 (caitta-dharma) — mental factors (those that arise *with* mind and shape its character — feeling, perception, intention, attention, and so on). Verse-name: 「見」 ("seeing," signaling their role in cognition). 4. 不相應法 (viprayukta-saṃskāra) — "non-associated formations," phenomena that don't fit neatly into matter, mind, or mental factors (time, motion, becoming, aggregation, and the like). Verse-name: 「位」 ("position," pointing to their abstract, positional nature). 5. 無為法 (asaṃskṛta-dharma) — the unconditioned. Verse-name: 「不轉」 ("non-turning," marking what does not arise and pass away).
These five give 能相 its five kinds of *objects*. Now we move to what 能相 itself is.
Second: the three 能相, three ways of characterizing
The text groups 能相 under three major headings — and one of them gets unpacked in fascinating detail:
(1) 分別相 (vikalpita-lakṣaṇa) — the discriminating/imaginal characteristic. This is the most elaborated of the three. It corresponds to what Yogācāra elsewhere calls 遍計所執性, the imagined nature. Importantly, all three sub-forms below are *declared "non-true discrimination"* (非真分別). They are not bad — they are how ordinary knowing works — but they are not wisdom either. They are the conceptual layer.
The text distinguishes three modes:
- 有覺分別相 — discriminating-with-awareness. A mental factor called 想 (*saṃjñā*, "perception"/"apprehension") seizes upon an object of meaning (義境), and from this grasping there arise all sorts of "mental-verbal understandings" (意言解). Think: you see a face, you recognize it, and immediately inner speech lights up — "that's Maya, she looks tired, she's wearing that green coat." There is conscious apprehension and there is verbalization. Awareness is on.
- 無覺分別相 — discriminating-without-awareness. Here, "seeds of mental-speech" (意言種子) directly give rise to a "light of meaning" (義光), but they do not yet produce the varied mental-verbal elaborations. Think: you walk into a room and something feels wrong *before* you know what; you catch a familiar smell and recognition dawns before the name does; you get a knot in your stomach before the worry has a sentence. The meaning is illuminated; the words haven't formed.
- 相因分別相 — discriminating-by-mutual-causation. Based on names, light of meaning arises; based on meaning, light of name arises. The technical phrase is 「名義互光起」 — *name and meaning mutually light each other up*. You cannot entertain the concept *democracy* without lighting up certain clusters of meaning; you cannot grasp a particular kind of fairness without summoning words. Names and meanings form a self-arousing circuit. This is how conceptual networks actually function — there is no clean boundary between the word and what it points to.
All three of these are called 非真分別, non-true discrimination. That does not mean they are useless. They are exactly how ordinary cognition works. But they are *not* the wisdom that sees things as they are.
(2) 依他相 (paratantra-lakṣaṇa) — the dependently-arisen characteristic. This is the mode in which dharmas actually function — arising from conditions, sustained by conditions, ceasing when conditions fall away. It is the "how things actually work" face of 能相. The source text does not elaborate it further in our excerpt, so I will resist the temptation to fill in.
(3) 真實相 (pariniṣpanna-lakṣaṇa) — the perfected/reality characteristic. This is the mode in which dharmas are known as they truly are — free of the conceptual overlay of 分別相, yet not divorced from 依他相. Again, the source is silent on the details here, and so will I be.
The three together are a single teaching, not three separate topics: things are characterized in three ways — through our conceptual overlay, through their dependent arising, and through their actual nature.
Third: the five 學境 (gocara), the practice face of 能相
This is where 能相 stops being abstract. The text says the 能相 has *five fields of learning* — five places where the practitioner actually engages with the characterizing function:
1. 能持 (that-which-holds). The Buddha's correct teaching (正法). It "holds" because it can sustain the practitioner and can serve as an object of mind. 2. 所持 (that-which-is-held). Correct recollection (正憶念), the mental hold of what has been heard. It is "held" because it is upheld by the teaching above it. 3. 鏡像 (mirror-image). Samādhi-mind is the mirror; the dharma-realm (法界) is the image. As concentration matures, the mind settles into and reflects the dharma-realm. 4. 明悟 (penetrating understanding). Supramundane wisdom (出世間慧) that sees "existence" and "non-existence" as they really are. "Existence" here means dharma-anātman (法無我 — the selflessness of phenomena); "non-existence" means the absence of a grasping subject and grasped object (能取、所取). This is the eye of wisdom opening. 5. 轉依 (āśraya-parāvṛtti — transformation of the basis). The fifth item, named without further elaboration in our excerpt. This is the deepest one — the moment when the basic locus of experience actually shifts, rather than being merely understood.
Read as a sequence, this is a teaching path: *the dharma is spoken → it is held in mind → the mind becomes a mirror → wisdom sees truly → the basis is transformed*. The five 學境 are how 能相 manifests along the actual road of practice.
Fourth: the four 量相 (pramāṇa-lakṣaṇa)
The text also gestures at another cut through 能相, calling it the four-measure characteristic: 能詮, 義, 意, 了義 — the explanatory, the meaning, the intention, the definitive. It caps this with the striking line: 了義亦無言 — "the definitive, too, is wordless." Our source confirms only the names and this line; we won't try to extract more than that.
Life walk-through: 能相 at the kitchen table
Let's take a single ordinary situation and walk it through the framework. You are standing in your kitchen, looking at a notification on your phone that says a colleague has just been promoted to a role you expected to get.
所相 — what is being characterized. What is actually happening in front of you? Five layers, all present at once. The *色法*: a piece of glass, lit pixels, a hand holding the phone. The *心法*: a stream of awareness directed toward that scene. The *心數法*: feeling (hurt, surprise), perception (this *means* something), intention (do I reply? do I pretend?). The *不相應法*: time (it's 2:14 p.m.; your career has been a sequence of moments), motion (the notification arrived; your thumb might move). The *無為法*: the spacious, unconditioned awareness in which all of this is appearing. The five are not separate rooms — they are five aspects of one happening. The conceptual mistake to avoid is collapsing 所相 into just the pixels, or just the emotion, or just the thought of the future. The text insists: what is being characterized is *all of it, in its full layered character*.
Now 能相 — the characterizing function starts running. The text says there are three modes, and you can watch them cascade:
- 有覺分別相 kicks in first. The mental factor of 想 (perception) latches onto the situation: "she got the role." From that grasping, 意言解 (mental-verbal understanding) bursts forth — "they don't value my work," "I should have spoken up in that meeting," "I'm a fraud," "I'll never recover." There is conscious awareness. There are words. There is a whole script. Most of what we call "overthinking" lives in this layer.
- 無覺分別相 is already running underneath. Before you have any sentence, your stomach drops, your chest tightens, your face changes. 義光 — meaning-light — is on. You already know it matters, before you know *why* it matters. The body has lit up the meaning without words.
- 相因分別相 is the engine keeping them coupled. The name "promotion" lights up a network of meanings — fairness, status, identity. The meaning of "fairness" summons the word "injustice." You cannot pull one without the other; they are mutually igniting. This is why arguing yourself out of the feeling with words ("it's fine, it's fine") often doesn't reach it: you are trying to use the same 名義互光起 circuit to dismantle itself.
The text's quietly devastating point: *all three of these are 能相 — yes, even the most articulate, the most articulate-looking, the most "rational."* And *all three are declared 非真分別* — non-true discrimination. They are real, they are functioning, they are doing their job. But they are not the wisdom that sees the situation as it actually is. They are the conceptual overlay, the way the situation has been *characterized* for you by an old set of habits. That is a *description of the mechanism*, not a verdict on you.
This is where the 學境 matters. The kitchen scene doesn't have to end in the spiral.
- 能持: the dharma is something you have actually heard — maybe a teacher said, "feeling bad is not a verdict; it's a feeling"; maybe a passage landed years ago; maybe just the simple teaching that *arising is dependent*. That teaching can hold you. - 所持: you hold it in mind — *正憶念*. You remember. You don't have to invent it on the spot. - 鏡像: you sit, even for one breath. The samādhi-mind becomes a mirror. The whole scene — the phone, the hurt, the thought, the body — is reflected in it *without being grasped*. The mirror does not judge the image; it shows it. This is exactly the move the text is pointing at: 鏡像以定心為鏡、法界為像. - 明悟: if the practice deepens, a wisdom opens that sees "existence" and "non-existence" correctly. *Existence*: the dharmas here are empty of self — there is no fixed "I" being slighted. *Non-existence*: there is no fixed person promoting and no fixed person being passed over — only a configuration that arose dependent on causes. The grasping subject (能取) and grasped object (所取) loosen. - 轉依: something in the basis of experience actually shifts. You are not the same configuration after as before. Not because you thought better thoughts, but because the locus itself has changed.
And the four 量相 are quietly present too. There are explanations (能詮). There is meaning (義). There is intention (意). And the deepest thing — the *definitive* (了義) — is wordless. No caption can be placed under it. If you have to put it in language, it has already been thinned.
A second walk-through, in a quieter key
You are on the train. You look out the window and see a tree. Without effort, your mind says, "tree." That is 有覺分別相 at its smallest scale. Now let your eye move to the next tree, and to the next. You start to anticipate the rhythm — tree, gap, tree, gap. Before you have the thought, you *feel* the rhythm in your body. That is 無覺分別相. Now let yourself ask, "Why does that rhythm feel like music?" Try to separate the *meaning* of the rhythm from the *name* "music." You can't. The moment you say "music," the meaning is there; the moment you feel the rhythm, the word is on the way. 相因分別相. Three modes, in a single tree-glance.
And here is the gentle, important point: the text is not asking you to stop seeing the tree. It is asking you to notice the *characterizing* as characterizing, not as the thing itself. The tree stays a tree. The seeing stays seeing. What shifts is the relationship between the two.
Why contemporary people need this
Because we live inside our 能相 and mostly mistake it for reality.
A few specific reasons this framework matters now:
1. The character is on, all the time. Modern life is unusually heavy in the 有覺 and 相因 modes — language-mediated, screen-mediated, concept-mediated experience. We are constantly naming, labeling, captioning, evaluating. The framework gives a way to ask, *Is this the situation, or is it the way I am characterizing the situation?* without falling into the popular mistake of thinking "more mindfulness" means "less thinking." The point is not to silence the 能相; it is to *know it as 能相*.
2. The pre-verbal layer is medically, psychologically, and ethically important. The 無覺 mode is where trauma lives, where intuition lives, where the body's first reading of a situation lives. Modern therapies (somatic experiencing, parts work, affect labeling research) are essentially teaching people to come into relation with this layer. The Buddhist framework gives this an ancient, careful vocabulary: 義光 — meaning-light that hasn't yet taken verbal shape. Treating it as merely "irrational" misses what it is doing; treating it as authoritative misses that it is still 能相, still a characterizing function, still subject to being seen through.
3. The mirror-image is a counterweight to the algorithmic mirror. If you have noticed that your feed has learned to characterize you — to feed back exactly the version of you it has constructed from your clicks — you have met the 鏡像 from the outside. The practice the text is pointing at is to turn that mirror around: *you become the mirror, the dharma-realm becomes the image.* This is not escape from the digital world; it is a different relationship to it.
4. The path is structurally laid out. 能持 → 所持 → 鏡像 → 明悟 → 轉依 is a remarkably practical sequence. It says: the dharma must be heard (or read, or encountered in some trustworthy form); held; stabilized in concentration; opened by wisdom; and allowed to transform the basis. For contemporary practitioners who often pick up meditation without hearing, or hold without stabilizing, or stabilize without opening — this is a useful diagnostic. Where in the sequence am I?
5. The non-true-discrimination verdict is liberating, not depressing. If the framework told you the conceptual layer was the enemy, it would be hostile to being human. Instead it says: this layer is *real and functioning*; it is also *not the final word*. You don't have to crush it; you don't have to be controlled by it. You can know it as the conceptual overlay that it is, while it keeps doing its ordinary work.
Common misreadings and clarifications
Misreading 1: "能相 just means 'the mind' or 'the subject.'" It is more precise than that. 能相 is not the self; it is the *characterizing function*. It can run in many places — in perception, in language, in intuition, in conceptual networks, in concentration, in wisdom. Calling it "the mind" risks turning it into a metaphysical substance, which is exactly what 明悟 (the wisdom of "non-existence" of the grasping subject) is meant to undo.
Misreading 2: "能相 means 'any discrimination.'" Closer, but still wrong. 能相 has three modes, and the first — 分別相 — is itself *declared non-true discrimination*. If you flatten 能相 into "any act of distinguishing," you lose the structure: the framework is precisely saying that *some modes of characterizing (分別相) are conceptual overlays, while others (依他相, 真實相) point at how things actually are and how they are actually known*. The contrast is the point.
Misreading 3: "分別 (discrimination) is bad — we should stop it." This is a very popular contemporary misreading, often imported from a half-remembered Zen sensibility. In this text, 分別相 is described, not condemned. It is how ordinary cognition works. The verdict 非真分別 does not mean "false in the sense of wrong"; it means "not the wisdom that sees ultimate nature." It is like saying that a map is not the territory — accurate, useful, but not what it points at. Stopping all discrimination is not the teaching; knowing the difference between the map and the territory is.
Misreading 4: "能相 can exist independently of 所相." The text explicitly puts 所相 first and only then discusses 能相. This is not a stylistic choice; it is the teaching. There is no characterizable without something to characterize. There is no "pure subjectivity" — there is only the act of distinguishing *this from that*. To seek 能相 without 所相 is to seek a seer without a seen, a knower without a known — which is exactly what 明悟 sees through.
Misreading 5: "鏡像 meditation is about visualizing or imagining." No — the text says 定心 is the mirror and 法界 (dharma-realm) is the image. This is not visualization. It is the mind, settled in concentration, reflecting the actual nature of things as it is. A mirror does not *paint* an image; it *shows* what is in front of it. The whole direction is away from construction and toward reflection.
Misreading 6: "轉依 is just a fancy word for insight." Insight (明悟) is the fourth item. 轉依 is the fifth. The text puts them in sequence deliberately. Insight *sees* transformation; 轉依 *is* the transformation of the basis itself — not a new thought about experience, but a change in the locus of experience. They are not the same thing.
Misreading 7: "The four 量相 are four kinds of speech." The text caps this section with 了義亦無言 — the definitive, too, is wordless. If the four were just four kinds of speech, this last line would make no sense. The most one can responsibly say from our source is that there are four measures (能詮, 義, 意, 了義) and that the final one exceeds speech. Going further is going beyond the text.
A short closing note
What strikes me about 能相 when I sit with it is that it is not a metaphysical position. It is a *functional account* of how knowing works. There is what is being known (所相), and there is how it is being characterized (能相). The how is real but not final. The practice is not to destroy the how, but to know it as how — and through that knowing, to encounter what lies beneath and beyond it. The 鏡像 of concentration, the 明悟 of wisdom, the 轉依 of transformation — these are not stages of getting rid of 能相; they are stages of letting it do its real work.
A friend of mine once described the shift this way: *before, I thought the world was happening to me; then I thought I was happening to the world; now I notice that "happening" itself is something I am doing — and I have a choice about how I do it.* That is a remarkably faithful paraphrase of what 能相 is pointing at, even though my friend had never heard the word.
The conceptual overlay is not your enemy. Mistaking it for the territory is.