What This Concept Is Really Saying
可相 (lakṣya, often rendered in English as "the characterizable," "that which is characterized," or "the characterized object") is the companion term to 相 (lakṣaṇa, "characteristic" or "mark"). Together they form a pair that Buddhist philosophy, especially the Madhyamaka (中觀) tradition, uses as a hinge for understanding how anything ever becomes "a thing" at all — and how, on closer inspection, no thing ever really stands on its own.
The core passage from 《十二門論》 (Śāstra of Twelve Gate-Treatises, T1568) puts it with quiet precision:
「以相故知是事名可相。」 *That which can be known through its characteristics is called 可相.*
And 《中論》 (Mūlamadhyamakakārikā / Treatise on the Middle Way, T1564) gives the working definition in a single phrase:
「可著名物。」 *What can be named is called a "thing."*
Put these two together and you have the architecture: 可相 is the thing-as-known-through-its-marks; 相 are the marks-by-which-a-thing-is-known. Neither one precedes the other in a strict sense — and that is precisely where the philosophical weight falls.
The Mutual Dependence Structure
This is not a one-way relationship where characteristics belong to a pre-existing object like ornaments on a tree. It is a mutual conditioning (因待). The 《中論》 states this bluntly:
「相、可相、常相，因待故。」 *Characteristic, characterizable, and the constant characteristic — all exist through mutual dependence.*
And again:
「因相有可相，因可相有相，輾轉相待。」 *Because of characteristic there is the characterizable; because of the characterizable there is the characteristic — they depend on each other in turn.*
This "in turn" (輾轉) is not a casual word. It signals that if you try to find a starting point — *which came first, the mark or the thing?* — the analysis breaks down on both sides.
The Twin Examples: Cow, Elephant, Horse
The classical texts don't leave this abstract. They give us concrete, almost rustic examples that were meant to stop the philosophizing mind in its tracks and point at something you already do without thinking.
The cow (from 《中論》): > 「有峰角、尾端毛、頸下垂𩑶故名牛。」 > *Because of the peaked horns, the tuft at the tail's end, the dewlap hanging beneath the neck, it is called "cow."*
Take away the horns, the tuft, the dewlap — and the word "cow" has no purchase. Not because the animal vanishes, but because the pointing-out function of the word has lost its anchor.
The elephant and the horse (from 《十二門論》): the elephant has twin tusks, a long trunk, three bumps on the head, scoop-shaped ears, a curved-spine back. The horse has upright ears and four equal hooves. In both cases, *離此諸相更無象可被相所相* — apart from these characteristics there is no elephant that can be characterized. And in the case of a *無相事* (a thing with no characteristics), there is likewise nothing for characteristics to apply to.
This last clause is important: the argument works on both ends. A thing with no characteristics cannot be picked out at all — it drops out of language. And a thing with only characteristics, never unified into a "thing," is equally incoherent.
Mutual Characteristic: Smoke and Fire
One of the loveliest moves in the 《中論》 is the smoke-and-fire example:
「一切法皆攝在相、可相二法中，或相為可相、或可相為相。如火以烟為相，烟亦復以火為相。」 *All dharmas are subsumed under these two — characteristic and characterizable. Sometimes the characteristic is the characterizable, and vice versa. As fire takes smoke as its characteristic, smoke in turn takes fire as its characteristic.*
This demolishes the naïve picture where "characteristic" and "thing" are tidy ontological categories — a mark belongs to a substance, a substance wears marks. Instead, what counts as a "characteristic" and what counts as a "thing" can swap roles depending on context. Fire is the mark by which you know there's smoke; smoke is the mark by which you know there's fire. Neither is prior.
The Double Negation
Now the philosophical knife comes out. The 《中論》 performs a clean symmetrical cancellation:
「相法無有故，可相法亦無；可相法無故，相法亦復無。」 *Because characteristic does not exist in itself, characterizable does not exist; because characterizable does not exist, characteristic also does not exist.*
And the closing verse:
「是故今無相，亦無有可相；離相可相已，更亦無有物。」 *Therefore there is no characteristic, no characterizable; apart from characteristic and characterizable, there is also no "thing."*
The third line is the one to sit with: 離相可相已，更亦無有物 — apart from the pair, there is no "thing" either. "Thing" (物) is itself a designation made possible only by the operation of the pair. Strip the pair away, and the noun "thing" dissolves.
Three Wrong Pictures, Three Refutations
The 《十二門論》 is unusually patient with its opponents. It lists three common ways people try to rescue a more comfortable picture — where either the pair is identical, or utterly different, or partially overlapping — and disposes of each one carefully.
1. 相即可相 (characteristic = characterizable) *E.g., "the characteristic of consciousness is consciousness itself; apart from the consciousness used, there is no consciousness."* This is refuted with an image so immediate it almost hurts: > 「凡物不能自知，如指不能自觸，如眼不能自見。」 > *No thing can know itself. A finger cannot touch itself; an eye cannot see itself.*
If characteristic and characterizable were identical, the thing would have to be its own mark — but nothing can perform that trick. The finger meeting itself requires another finger; the eye seeing itself requires a mirror. The line 「若因果為一，而實因果不一」 adds a second blade: even granting identity, it would collapse cause and effect into one, which observation contradicts.
2. 相異可相 (characteristic ≠ characterizable) *E.g., "the cessation of desire is called nirvāṇa," not "desire is the characteristic of nirvāṇa"; or "faith has three marks… it is a mental factor, included in the aggregates of formation."* If they are utterly different, then you need another characteristic to characterize the characteristic — and another for that — leading to an infinite regress: 「相更復應有相，此事不然」 (a characteristic should again have a characteristic; this does not hold).
3. 可相少分為相 (a portion of the characterizable is the characteristic) *E.g., "right view is a characteristic of the path, and is a portion of the path"; "arising, abiding, ceasing are the characteristics of conditioned things, and are portions of conditioned things."* This view secretly re-imports either identity or difference at a smaller scale — and since those have already been refuted, this one falls with them.
The Lamp Analogy and Its Self-Contradiction
An opponent offers a clever image: "Just as a lamp can illuminate itself and also illuminate another, so a characteristic can characterize itself and another." The 《十二門論》 answers coolly: this analogy has already been broken when analyzing the three marks of conditioned things (生住滅, arising/abiding/ceasing); and besides, the opponent has just contradicted himself — earlier he said characteristic and characterizable are different, now he says the characteristic can characterize itself. You can't have both.
Where This All Lands
The pair 相/可相 is not an end in itself. It is a tool for one specific operation: showing that the way things "are" for us — as identifiable, nameable, characterizable items — has no bottom. 《中論》 closes with:
「可著名物…性空寂滅相。」 *What can be named is called "thing"… characterized by the nature of empty quiescence.*
And 《十二門論》 with:
「相、可相空故，一切法皆空。」 *Because characteristic and characterizable are empty, all dharmas are empty.*
So 可相 is not just one term among many. It is the moment when ordinary recognition — *oh, that's a cow, that's a friend, that's an electron, that's a memory* — is taken apart and shown to be a delicate, mutually-supporting construction with no independent ground under it.
Walking Through It in Contemporary Life
Let's trace 可相 through a single ordinary scene, and then through several others, mapping each step carefully onto the classical structure.
Scene 1: Identifying Someone at a Coffee Shop
You're looking for a friend. You scan the room.
The marks (相): a green jacket, a slight limp, reading a paperback with a yellow spine, a habit of tapping the foot.
The thing-being-identified (可相): "my friend Anna."
Now read the analysis carefully:
1. The "thing" only arises as your friend in virtue of the marks. No green jacket, no limp, no book, no foot-tapping — no "Anna" as the person you're waving at. This is exactly the cow's horns and dewlap. The marks are not decorations on a pre-existing Anna; they are the operational handle by which "Anna" is picked out *as the person you're picking out*.
2. The marks are only "marks of Anna" in virtue of Anna. A green jacket in a department store window is not a mark of Anna. A paperback book on a subway seat is not a mark of Anna. They become her marks *only because they belong to the context of being-her*. Strip the friendship out — out goes the function. This is the mutual conditioning (因待): 因相有可相，因可相有相。
3. The marks can themselves be picked out by further marks — and that process has no end. The "green" of the jacket is established by contrast with the gray wall; the "limp" by comparison to an imagined un-limp; the "yellow spine" against the off-white cover. Each mark needs other marks. This is why 《中論》 refuses to let either pole stand alone.
4. The "self-knowledge" check fails. Can Anna be her own characteristic? She can't walk into the room and see herself-as-Anna-being-picked-out-by-you. She needs your eyes. This is the finger-cannot-touch-itself move.
5. Now the deeper move: ask whether there is an "Anna" you could encounter *without* any marks. The texts say no — not because Anna doesn't exist as a person, but because the operation "Anna as identifiable object of recognition" cannot run without the pair. 離相可相已，更亦無有物.
This is the heart of the teaching, and it is gentle. It is not saying your friend is an illusion. It is saying that the mode of her being-for-you-as-identified-thing is a mutually-supporting construction.
Scene 2: The Email from Your Boss
You receive a message. You identify it as "angry."
Marks (相): short sentences, a period where there would normally be a comma, the absence of greeting, "please" replaced by "kindly."
Characterizable (可相): the email as angry.
The same architecture runs. The "anger" of the email isn't some hidden substance inside it that the marks merely report on. The anger *as anger-for-you* only takes shape via the marks, and the marks only count as marks-of-anger via the larger context of how your boss writes. Take the same words from a different sender, in a different context, and they might register as efficient, or tired, or just clipped. What counts as a mark of anger depends on the characterizable; what counts as an angry email depends on its marks.
Common misstep: people often say "the words are just neutral text; I projected the anger onto it." That's one reading, but it misuses the framework. The point isn't that the words are projections and the anger is fake. It's that *the anger-as-identified-in-the-words* and *the words-as-carrying-anger* arise together — neither is more real than the other.
Scene 3: Your Job Title on LinkedIn
You are listed as "Senior Product Manager."
Marks (相): the words "Senior," "Product," "Manager"; the bullet points of past experience; the endorsements.
Characterizable (可相): you-as-the-Sr-PM.
What is most striking here is how the mark and the thing genuinely swap roles — exactly the smoke-and-fire pattern. Your job title is what you present to the world as your identifying mark (相); but it is also what your actual daily activity instantiates as the thing-being-identified (可相). "Senior PM" is your mark on a business card; "Senior PM" is the role your workdays fill out. 火以烟為相，烟亦復以火為相.
Now, the LinkedIn profile is also an object lesson in mutual conditioning through designation: if you change one bullet, the whole thing shifts; remove the title and the bullets become hard to parse; remove the bullets and the title becomes a hollow sound. The pair holds together or falls together.
Scene 4: The Misclassification Trap
A common mis-mapping happens with physical recognition: you see someone across the street and call out to "Anna" — but it's not Anna. Here the failure is instructive. The marks (green jacket, paperback) were insufficient to establish the characterizable (Anna). They were Anna's marks in *some* contexts but not this one. This shows that the relationship is context-bound — there is no airtight, context-independent proof that a given cluster of marks picks out a given thing. The 《十二門論》's warning about *無相事* (things with no characteristics) is, in modern terms, the worry about under-described or unrecognizable cases.
Where the Walk-Through Goes Wrong If You Don't Watch It
The single most common mistake is to read 可相 as "the thing-in-itself" — some hidden, solid substrate that sits beneath the marks and owns them. This is exactly what the texts are refusing. The cow is not "really a thing underneath that has horns." It is "thing-and-horns together, neither prior." If you let 可相 collapse into a reified essence, you have lost the teaching.
The second mistake is to read it as a kind of "subjective idealism" — *I construct the thing by my labels*. That collapses into the first mistake in reverse. The marks are not yours alone; they include Anna's actual limp, the boss's actual typing rhythm, the actual job responsibilities. The mutual dependence is between mark and thing, not between mind and world.
Why Contemporary People Need This
The teaching of 可相 addresses a very modern condition: the inflation of identity-things. We live in a culture that constantly asks "what is this thing, really?" and treats the answer as if it should be a stable, nameable, ownable core. *What is my job, really? What is this relationship, really? What is my gender, really? What is this country, really? What is this mental state, really?* Each of these is treated as a 可相 — a thing — that should have its characteristics cleanly listed and nailed down.
The Madhyamaka move says something almost radical in its quietness: every such "thing" is held together by the marks you use to pick it out, and the marks are only marks because of the thing you pick out. There is no under-layer to find. This is not a counsel of despair — it is a release from the exhausting project of finding the final definition. Once you see the mutual-dependence structure, the panic to "really know what X is" loses some of its grip. The cow is still a cow. You still milk it. But the metaphysical hunt for the Cow-Itself is recognized as a category error.
There are also specific contemporary resonances:
- In product design and AI, we constantly encounter the smoke-and-fire pattern. A "feature" of a product is what characterizes the product; the product is what makes the feature a feature. Try to design one without the other, and you get vaporware or feature-creep. The mutual-dependence analysis is exactly what product teams often fail to learn, and it shows up as a kind of organizational suffering.
- In identity politics and self-understanding, the pair offers a way to honor both that "I am X" is meaningful (because the marks and the thing genuinely co-arise and have real effects) and that "I am X" is not the end of inquiry (because the X-as-fixed-thing is empty of independent existence). Many contemporary debates go in circles because one side insists on the thing, the other on the marks, and neither sees the mutual dependence.
- In mental health discourse, especially around diagnostic categories: depression, anxiety, ADHD are at one level useful characterizations (相) of patterns of suffering (可相); but the suffering is also what makes the patterns count as those particular patterns. Treating either pole as primary — "you really have this thing" or "it's just labels" — falls into one of the three refuted views. The most workable clinical posture often unconsciously mirrors the Madhyamaka one: hold the diagnosis lightly, work with the marks, keep checking what you're actually dealing with.
- In the practice of meditation, 可相 shows up directly. A meditator might try to find the "self" behind the breath, behind the sensations, behind the thoughts. Each time they look for a characteristic (相) they think will point to a finder (可相), they cannot locate one. The mutual-dependence analysis gives a philosophical map of why this search always comes up empty — and why the emptiness is not a failure but a finding.
Common Misreadings and Clarifications
Misreading 1: "可相 is just 'the object itself' / 'the referent'."
This is the most common Western importation from naive realism. In English philosophy, "the characterized object" sounds like the thing-in-the-world the description refers to. But the 《中論》's closing line — 離相可相已，更亦無有物 — explicitly denies that this referent has any existence apart from the characterizing operation. 可相 is not a substrate; it is the characterizable-as-picked-out.
A helpful analogy that does *not* go too far: in some readings of quantum mechanics, an "observable" has no determinate value until a measurement (which itself depends on a setup, which itself depends on what's being measured). The bare "object as it is before measurement" cannot be cleanly specified. This is *analogous*, not equivalent, and contemporary physics is still contested on exactly how far to press such analogies. The Buddhist point is made on logical grounds, not empirical ones.
Misreading 2: "相 is just 'appearance' and 可相 is 'reality behind appearance'."
This comes from the Platonic-Christian habit of taking "appearance vs. reality" as the master pair. In Buddhist philosophy, that habit is precisely what is being unwound. The 相 is not a veil over a deeper 可相; they co-arise. To treat them as appearance-vs-reality is to sneak in a hidden 可相 behind the 相 — exactly the refuted position of "相即可相" with the added error of hierarchy.
Misreading 3: "This is just nominalism — things are just names."
Not quite. Nominalism says the names are conventions and the things are conventions-of-conventions. The Madhyamaka analysis goes further: it asks what makes a name work as a name of *this* thing rather than *that* thing — and finds mutual dependence at every level. Names are part of the system, but so is what gets named; the analysis refuses to grant priority to either pole.
Misreading 4: "If there's no 可相 standing alone, then nothing exists and the world is an illusion."
This is a misreading of "emptiness." The 《十二門論》's closing line — 相、可相空故，一切法皆空 — does not say "nothing exists." It says "no dharma exists with the kind of self-standing nature we tend to assume." Cow's horns, dewlap, tuft — these still function. Milk still comes out. The emptiness is of the independently-existing characterizable, not of the operational, recognizable, milkable cow. 日常的牛不空；空的是「離相而獨存的牛」。
Misreading 5: "相 is one, 可相 is many — like Form and Particulars in Western metaphysics."
Tempting, but a trap. The 《十二門論》 explicitly refutes the one-many relation between them: 相及與可相，一、異不可得 (one and difference both cannot be established). Importing the universal/particular pair smuggles in a separate Form which is exactly the 常相 (constant characteristic) that 《中論》 lists alongside the pair to say 因待故 (it too is dependently arisen, not a Platonic Form).
Misreading 6: "可相 = the object of cognition, 相 = the act of cognition."
This collapses a structural pair into a cognitive-act pair. 相 is not the *act*; it is the *characteristic*. The finger-cannot-touch-itself argument is precisely aimed against making 相 into a self-acting cognition: it shows that if characteristic = characterizable, the act would have to be its own object, which nothing can do.
Misreading 7: "This is just epistemology — a theory about how we know."
It touches epistemology, but its main work is ontological: it concerns how things *are*, not merely how we grasp them. The double-negation conclusion is about the way things exist (or fail to) in themselves, not about the limits of our knowledge. Saying "we can't know the thing-in-itself" is a Kantian move that still leaves the thing-in-itself as a limiting concept. The Buddhist move refuses even that — there is no thing-in-itself to fail to know.
A Note on the Texts
The two principal sources — 《中論》 attributed to Nāgārjuna, translated by Kumārajīva (鳩摩羅什) around the early fifth century, and 《十二門論》 also attributed to Nāgārjuna, likewise via Kumārajīva — handle 可相 with slightly different vocabulary but identical architecture. 《中論》 uses the cow and the smoke-fire examples; 《十二門論》 uses the elephant and horse and walks through the three refuted positions in more detail. Both arrive at the same place: 相 and 可相 are empty of self-standing nature, and therefore all dharmas are empty.
A well-known later line — though from a different text — that catches the same insight at the level of felt experience is from the 《金剛經》 (Diamond Sūtra):
「凡所有相，皆是虛妄。若見諸相非相，則見如來。」 *All characteristics are illusory. If one sees that characteristics are not characteristics, one sees the Tathāgata.*
The 《金剛經》's 凡所有相 here corresponds to the same 相 the 《中論》 is dismantling. The phrase 若見諸相非相 is precisely the double-negation insight: seeing characteristics as not-characteristics (and, by extension, characterizables as not-characterizables, things as not-things).
This is the deep dignity of 可相 as a concept: it is not an obscure scholastic term. It is the moment of recognition turned back on itself, shown to be a delicate, mutually-supporting, ultimately ungroundable dance — and it is that very ungroundability that the tradition points to as freedom.