The brief definition tells us this is "the wisdom that arises after 根本無分別智 (mūla-nirvikalpa-jñāna, root non-discriminating wisdom), capable of sustaining and bringing things into function, also called 後得智 (pṛṣṭhalabdha-jñāna, subsequently-acquired wisdom)." But the real meaning of this definition is not fully clear until we see the two things it presupposes: what "non-discriminating" actually means in the Yogācāra technical sense, and what it means to come "after" a direct realization. This is one of the most practically important concepts in 唯識 (Yogācāra, "Consciousness-Only") Buddhist philosophy, because it addresses a problem every contemplative tradition eventually runs into: what do you do *after* the deep experience? How does insight get expressed, taught, applied, lived — without either collapsing back into ordinary conceptual grasping, or freezing in the ineffable?
Our main material is the two Chinese translations of the 《攝大乘論釋》 (*Mahāyāna-saṃgraha-bhāṣya*, the Commentary on the *Mahāyāna-saṃgraha*): T1597 (玄奘譯世親釋, Xuanzang's translation of Vasubandhu's commentary) and T1598 (無性釋, attributed to a member of Asanga's lineage). I will preserve the classical similes and taxonomies in their original one-to-one form, not collapsed into a generic "wisdom after meditation."
What this concept is saying
The place of 後得智 in the three-wisdom structure
In both T1597 and T1598, the term *nirvikalpa-jñāna* (無分別智, "non-discriminating wisdom") is divided into three:
1. 加行無分別智 (prayoga-nirvikalpa-jñāna) — "preparatory" non-discriminating wisdom. Also called 尋思慧 (*saṃtāna-prajñā*, investigative wisdom) or 悕求慧 (*āśayā-prajñā*, wisdom of aspiration). It arises from hearing the teaching of the non-discriminating principle, gaining confident understanding, and pushing into it through reasoning.
2. 根本無分別智 (mūla-nirvikalpa-jñāna) — "root" or "fundamental" non-discriminating wisdom. Also called 正證慧 (*sākṣāt-prāpta-prajñā*, wisdom of direct realization) or 內證慧 (*pratyakṣa-prajñā*, inwardly-witnessed wisdom). It is the moment of direct, wordless realization of 真如 (suchness), enabled by the preparatory work.
3. 後得無分別智 (pṛṣṭhalabdha-nirvikalpa-jñāna) — "subsequently-acquired" non-discriminating wisdom. Also called 起用慧 (*prayojaka-prajñā*, wisdom that brings into function) or 攝持慧 (*ādhāra-prajñā*, wisdom that sustains).
The causal logic is strict: 加行智 is the cause of 根本智; 後得智 is the fruit of 根本智. The three are not three different substances; they are three phases of one continuous awakening-process. The mute person who becomes able to speak is the same person; the eyes that open are the same eyes that were closed.
The four classical similes — preserved one-to-one
The texts use four simile-systems. They are not interchangeable. Each has its own pedagogical job. I will map each one to its three wisdoms in full, as the source texts do, and I will not collapse them into a single generic image.
Simile 1 — The mute person receiving an object (T1597, 啞人受境). This simile maps the relationship to *expression* (言說, verbal articulation):
- 加行智 = a mute person seeking the meaning of an object but unable to speak (啞人求受境義而不能言). They sense something to be understood, but cannot formulate it. Words-and-searching without yet reaching. - 根本智 = a mute person directly receiving the meaning, in silence (啞人正受境義, 寂無言說). They have it; the meaning is fully present; no words come. - 後得智 = a non-mute person who has received the meaning and can now speak about it (非啞人已受境義, 能依所受而起言說). The text specifies: "speaking" here means being able to *compose language* about what was received. Note the image: the same person, previously mute, now able to speak — not a different person.
Simile 2 — Studying a treatise (T1597, 學論). This simile maps the relationship to *understanding* (受) and *articulating* (分別):
- 加行智 = someone who has not yet understood the treatise and is seeking to understand it. - 根本智 = someone who has memorized the words and is just receiving the dharma — 溫習文字而但受其法. They know the words but are not yet working with meaning. - 後得智 = one who has understood the treatise — 於法、於義皆能領受. They have both the words/structure (法) and the meaning (義), and can work with both.
This simile highlights a specific feature of 後得智: it can *receive* (受) and *discriminate* (分別) at the same time. It is not a step back into conceptual confusion; it is a higher-order capacity that holds both form and penetration together.
Simile 3 — Eyes closed, eyes opened (both translations, 閉目/開目). This simile maps the relationship to *engagement* with phenomena:
- 根本智 = eyes closed (正閉目). The world of appearances is stilled; there is no discriminating of objects. - 後得智 = eyes opened (復開目). The world of appearances returns, but it is now seen through the lens of root wisdom.
Simile 4 — Empty space and the colored images in it (both translations, 虛空/色像). This simile maps the relationship to *the structure of awareness itself*:
- 根本智 = empty space itself (虛空). Space is everywhere, uncontaminated, neither discriminating nor being discriminated. It is the open, ungraspable ground. - 後得智 = colored images appearing in space (空中所現色像). The images can be discriminated; space permits their appearance without itself being stained.
Crucially, the source text says of 後得智 that it is "是所分別亦能分別" — it is both that-which-is-discriminated *and* that-which-discriminates. Both poles are present and active, yet the wisdom is still called *non-discriminating*. This is the heart of the paradox, and we will return to it.
The paradox: how can it be "non-discriminating" if it discriminates?
This is not a contradiction to be smoothed over; it is the whole teaching. The "non-discriminating" of 無分別 does *not* mean "lacking discrimination" in the sense of being blank, comatose, or undifferentiated. It means: free from *vikalpa* (分別) in the technical Yogācāra sense of conceptual fabrications that superimpose a subject-object split onto phenomena.
Ordinary consciousness, in the Yogācāra diagnosis, is constituted by an *inverted grasping* (顛倒執取) of 能取 (subject, the grasping pole) and 所取 (object, the grasped pole) as if they were independently real. This is the inversion that produces suffering and confusion.
後得智, even while engaging appearances, is "non-discriminating" in this deeper sense: it does not superimpose that false subject-object solidity. It *is* discriminating in the ordinary sense — it perceives distinctions, names things, teaches. But it does so without the ontological inversion that makes ordinary discrimination a source of confusion. The text, in T1598, gives this wisdom a name that captures it precisely: 能發語言世俗淨智 — a *pure worldly wisdom capable of generating language*. It speaks, teaches, and discriminates; the speech and discrimination are uncontaminated (無染污) and non-inverting (無顛倒).
What 後得智 actually does — the fivefold analysis by function
Both T1597 and T1598 analyze 後得智 according to its *functions* (所作差別), arriving at the same five. I will lay these out one-by-one, preserving the source's own structure of 「覺察 / 隨憶念 / 為他說 / 總緣智」 and the unifying term 思擇 (īkṣaṇā, "investigation/considering"):
1. 通達思擇 (nirgamana-īkṣaṇā) — Penetrating investigation. After realizing the truth, internally confirming: "I have penetrated this." 通達 = decisive penetration; 思擇 here takes the form of 覺察, inner awareness of having understood.
2. 隨念思擇 (anusmaraṇa-īkṣaṇā) — Recollective investigation. After exiting the meditation (後時), recollecting one's own penetration of non-discriminating reality. 思擇 here takes the form of 隨憶念, post-meditative memory of the realization.
3. 安立思擇 (vyavasthāpana-īkṣaṇā) — Establishing/explicating investigation. After exiting the meditation, 為他說, teaching others what one has penetrated. The communicative and pedagogical function.
4. 和合思擇 (saṃgṛhīta-īkṣaṇā) — Unifying investigation. Using a 總緣智 (total-perspective wisdom) to view all dharmas as having the same one suchness (皆同一相), and using this either to advance toward 轉依 (*āśrayaparāvṛtti*, the "transformation of the basis" at full buddhahood) or to re-arise in that view after the transformation.
5. 如意思擇 (yathecchita-īkṣaṇā) — As-thought investigation. When wisdom is fully present, whatever one thinks of, whatever one aims to investigate, arises accordingly — yathecchita, "as wished." Mastery: the freedom of an expert.
The text repeatedly uses the term 思擇 — not 觀 (contemplation) or 證 (realization) — to remind us that 後得智 is not passive reception. It is *active* discriminating. But the discriminating of wisdom, not the discriminating of inverted grasping.
The technical function: what 後得智 sees, what it sustains
T1598 gives 後得智 two technical functions that are easy to miss if you do not know the Yogācāra three-nature framework. I will not collapse these into generic "emptiness talk."
On the 依他起性 (paratantra-svabhāva, "dependent nature"): Among all the appearances of discriminating that arise from the 阿賴耶識 (*ālaya-vijñāna*, storehouse-consciousness), 後得智 sees the dependent nature as if it were an illusion (如幻). It does not deny the appearance; it does not take the appearance as ultimately, independently real. It sees the appearance as dependently arisen, like a magic trick.
On the 圓成實性 (pariniṣpanna-svabhāva, "perfected nature" / suchness): 後得智 sees that the self-nature of subject and object is, from the start, non-graspable (能取、所取自性本不可得). This is the seeing of suchness, but now *operative*: it can be applied, spoken, taught, lived.
This double-seeing is what makes 後得智 *non-inverting* (無顛倒). Ordinary consciousness either falls into 遍計所執性 (*parikalpita-svabhāva*, "imagined nature" — taking conceptual fabrications as real) or oscillates between that and nihilistic denial. 後得智 holds both the conventional appearance (依他起性) and the ultimate truth (圓成實性) without collapsing either.
On sustaining bodhisattva deeds: The text states explicitly: "All the bodhisattva deeds that are subsequently attained rely on the non-discriminating subsequent wisdom; this wisdom can sustain and increase all those deeds. Its being able to sustain the deeds is precisely because it is non-inverting" (後所得的種種菩薩行, 皆以無分別後得智為所依; 此智能任持並增長諸行; 其所以能持諸行, 正在於無顛倒).
This is not a footnote. 後得智 is the capacity that allows awakened activity to continue. Without it, root wisdom would be a one-time peak with no ongoing life.
Life Audit — walking the concept through contemporary scenarios
The danger with this concept is that it sounds like a technical list. Let me walk it through three scenes that any contemporary reader will recognize, strictly mapping to the classical definitions above, and then flag where the most common mis-mappings happen.
Scenario A — A deep work session that breaks through
You have been wrestling with a difficult problem for weeks: a research design, a strategic decision, a piece of writing that will not come. One morning, sitting quietly before the work, something happens. The internal arguing stops. The problem is suddenly *transparent* — you can see its structure as a whole, not as a list of pieces. There is no internal "I am solving this." Time feels different. Afterwards, you go back to the work and produce something that flows.
Strictly mapping, using the classical definitions:
- 加行智: the weeks of reading, hypothesizing, talking to colleagues, working with concepts to find a way in. The investigative, aspiration-mode wisdom. - 根本智: the morning of stillness when the structure became transparent. Wordless, no subject-object split between "the solver" and "the problem." The 啞人正受境義 moment. - 後得智: the writing afterwards, the explanation to a colleague, the email that lays out the design. And — the part most people miss — the ability to *stay* in that same seeing even when the phone rings, the meeting begins, the deadline hits. Not because the peak-state returns, but because 後得智 *sustains* the realization into function. This sustained seeing has all five functions operating: 通達 (the inner "yes, I see this now"), 隨念 (remembering the morning clarity), 安立 (writing it up for the team), 和合 (the larger view of how it fits the whole project), 如意 (whatever is needed arises accordingly).
Where the mapping goes wrong: The most common mistake is to call the "aha" the real wisdom, and the writing afterwards "just ordinary work." The classical framework insists the writing is *also* wisdom — specifically, non-inverting wisdom. It has the 啞人→非啞人 structure: a person who was at the silent peak can now compose language about it, without the writing being a fall from the peak.
Scenario B — A difficult conversation with someone you love
You have been preparing for this conversation for days. You have read, you have a sense of what to say, you have intentions. The conversation starts. For a stretch of it — sometimes only minutes, sometimes only seconds — the preparation drops. You are not saying the things you planned to say. You are not *saying things* at all; you are *being with* the other person. The room feels different. Later, you try to remember what you said, and you cannot fully reconstruct it. But the relationship moved.
- 加行智: the reading, the rehearsing, the inner script. The mind working with concepts, pushing toward the encounter. - 根本智: the moment of pure attunement, no inner script, no subject-object split between "me managing this conversation" and "them." - 後得智: the ability afterwards to recall the substance, to talk to a friend about what mattered, to integrate the shift, to talk to the other person later about what happened — and, crucially, to *live differently* the next day because of what was seen. The five functions all show up here too: 通達 (an inner "yes, I see this now about the relationship"), 隨念 (the days afterward, remembering the moment), 安立 (putting it into words, sharing it), 和合 (the larger view — that all relationships share the same suchness), 如意 (the same clarity can be re-invoked when needed).
Where the mapping goes wrong: It is tempting to call the moment of attunement the "real" wisdom, and the talk afterwards "just talk." The text explicitly refuses this. The talk afterwards is contaminated only if you *invert* it — turn it into a possession ("my spiritual moment"), a story about yourself, a credential. Otherwise, it is 後得智 doing its work.
Scenario C — A parent's moment with a child
The child is crying about something that, in adult terms, is small. The parent's first response — irritation, fatigue, "what is this about" — gives way, suddenly, to a quiet holding. The child is just a child. Everything that needs to happen is already happening. The parent is not planning a response; they are *there*. Later, the parent has to do many things: talk to the child, set a limit, talk to the other parent, write a note to the school, decide what to do tomorrow.
- 加行智: the years of parenting books, the inner formation, the desire to be a good parent. The accumulated aspiration-mode wisdom. - 根本智: the moment of presence, the stilling of "me-the-parent-managing-the-situation." - 後得智: the talking, the limiting, the explaining, the deciding, the email to the school — all done without losing the felt sense of what the child is, what the parent is, what this life is. The five functions: 通達 (a settled inner "yes, I see this child"), 隨念 (remembering this moment the next time a similar moment arises), 安立 (explaining to the other parent, articulating what was needed), 和合 (seeing the whole arc of the relationship), 如意 (whatever the next moment needs, that arises).
Where the mapping goes wrong: The most seductive mistake here is to call the moment of presence the "real" wisdom, and the functional engagement afterwards "compromise" or "worldliness." The text explicitly refuses this. The functional engagement, when rooted in non-inverting wisdom, *is* wisdom. The "non-discriminating" of 後得智 does *not* mean "pure contemplation, untouched by practical action." It means: free from subject-object inversion, *while* doing practical action. The hands changing a diaper, the email to the school, the limit set in a tired voice at 9 p.m. — these are 後得智's working conditions.
Mapping check — where people most often mis-map
Across all three scenarios, the most frequent mis-mappings in contemporary discussion are:
1. Mapping "non-discriminating" onto "undifferentiated" or "neutral." People think: if 後得智 is non-discriminating, it must be a kind of mellow, don't-take-sides, no-preferences state. *Wrong.* The text's own definition includes 能分別 (the discriminating function) and 為他說 (the teaching function). The non-discriminating refers to the *quality* of the discriminating (no false subject-object solidity), not its absence.
2. Mapping 後得智 onto "ordinary thinking after meditating." People think: after deep meditation, you go back to being a normal person who thinks, talks, and does things. Yes — but the text insists this return is *itself* a distinct wisdom, transformed by the root. The test is whether the discriminating activity comes with the felt sense of a solid "self" doing it and a solid "object" being done to. If yes: ordinary consciousness. If no: 後得智.
3. Mapping the 根本智 peak onto the goal, and 後得智 onto a falling-away. People think: the real achievement is the peak; the return to function is just what happens when the peak fades. *Wrong.* The text is explicit that 後得智 is the *fruit*, and that the bodhisattva path *depends* on it. The peak without the return is, in this framework, *less* than the peak with the return.
4. Mapping the five functions onto five sequential stages. They are not a ladder; they are five *functions* of one wisdom. A single moment of 後得智 can include 通達, 隨念, 安立, 和合, 如意 all at once.
Why contemporaries need this concept
There are at least three reasons this is not just an archaic scholastic idea.
The problem of the contemplative life in a hyper-active world
The contemporary contemplative landscape has a recurring problem: people have peak experiences — in meditation retreats, in therapy breakthroughs, in moments of beauty or love or insight — and then they return to ordinary life and feel the experience is lost. Sometimes they feel they have "failed" spiritually because the peak cannot be sustained. Sometimes they keep chasing peaks. Sometimes they conclude the whole contemplative project is fragile or theatrical.
The 後得智 framework is, in my reading, the most precise classical tool for this exact problem. It says: the peak is not the whole thing; the return is itself a wisdom; the return has its own structure and dignity. The goal is not to live on the peak forever. The goal is to develop a wisdom that *opens its eyes* and functions in the world with the realization intact. This re-frames the entire project, and it addresses a distress that is increasingly common.
The problem of skill-in-action without depth, and depth without skill
Modern life demands both: we need to be able to do complex work, navigate relationships, teach, decide, communicate. But we also know, from a century of psychological and social research, that purely instrumental skill without inner development tends toward burnout, manipulation, and what the Buddhist tradition would call *inverted grasping* — treating phenomena, including other people, as solid objects to be used.
The Yogācāra framework offers a model in which deep realization (根本智) and skilled engagement (後得智) are *not opposed*. They are sequential phases of the same awakening. The wisdom that has seen through subject-object solidity is precisely the wisdom that can then engage skill-in-action without the grasping that makes such action toxic. In an era where many feel forced to choose between "spiritual depth" and "real-world competence," this is a directly relevant reframing.
The AI-era question — what is human wisdom, and how is it developed?
If we are willing — cautiously, as analogy only — to think alongside AI systems: a contemporary AI model can be "trained" on a vast domain (analogous to a long contemplative development), and then "deployed" for specific functional tasks (analogous to 後得智's functional phase). What the Buddhist framework is pointing at, beyond the analogy, is something the analogy cannot capture: that the "training" of human wisdom is not just data-acquisition but a *qualitative* transformation of the cognizing structure, and that the "deployment" is a wisdom that has been transformed, not a routine re-run of the original program.
There is a more important contemporary question lurking here. When AI can do an increasing amount of functional, discriminating, language-generating work — including work that looks like "wisdom"-related work (counseling, teaching, decision-support) — what is the human *practice* of 後得智 actually cultivating? The classical answer is: a way of *being* in the world with the subject-object inversion undone. This is not reducible to a capability. It is a way of being, which is what the practice cultivates.
A note on the limits of such analogies: Neuroscience and contemplative research have correlated certain meditative states with changes in the default mode network, with shifts in attention and self-referential processing, with altered responses to suffering in self and other. These are interesting and broadly consistent with the Buddhist description, but they do not *prove* the Buddhist claim about the non-inverting character of wisdom. A brain state is not the same as an epistemological achievement. The Buddhist tradition is making a claim about the *structure of cognition*, not about neural correlates. I mention the neuroscience not to dismiss it, but to keep the analogy honest.
Common misreadings and clarifications
Let me gather the most common misreadings into one place, with precise corrections grounded in the source texts.
Misreading 1: "Non-discriminating" means no conceptual thought, no language, no preference
Correction: The text is explicit. 後得智 has 能分別 (the discriminating function) and 為他說 (the teaching function). It is 是所分別亦能分別 — it is both that-which-is-discriminated and that-which-discriminates. The *non-discriminating* refers to the absence of *vikalpa* in the technical sense of superimposing subject-object duality, not the absence of discriminating activity. A 後得智 person can have strong preferences, make sharp distinctions, teach clearly, write precisely. They are not being neutral in the sense of mushy.
Misreading 2: 後得智 is just ordinary consciousness after meditation
Correction: This is the single most common error. The whole point of the classical framework is that 後得智 is a *distinct* wisdom, with its own qualities — especially 無顛倒 (non-inverting) and 無染污 (uncontaminated). Ordinary consciousness after meditation is, by the classical account, *inverted*: it takes subject and object as independently real, and so it constantly generates grasping, aversion, and confusion. 後得智 is the *transformation* of that ordinary consciousness, not its re-occurrence. A useful test: does the discriminating activity come with the felt sense of a solid "self" doing it, and a solid "object" being done to? If yes, that is ordinary consciousness. If no, or "no with effortful maintenance," that is the movement toward 後得智.
Misreading 3: 根本智 is the achievement; 後得智 is the consolation prize
Correction: The text says the opposite. 後得智 is the *fruit* (果) of 根本智, and the bodhisattva path *depends* on 後得智. Without it, the realization cannot be sustained, cannot be taught, cannot grow. The peak without the return is, from the standpoint of the bodhisattva path, *less* than the peak with the return. The Romantic intuition that "the direct experience is the real thing, the rest is commentary" is a Romantic intuition; the classical framework here is a *functional* intuition: the wisdom that enables function is, for the bodhisattva, the more important wisdom to develop.
Misreading 4: You can develop 後得智 directly, without 根本智
Correction: The causal sequence is clear in both T1597 and T1598: 加行智 → 根本智 → 後得智. Trying to develop a "non-inverting" functional wisdom without the root of direct realization is, classically, a category mistake. It would be like trying to learn to speak a language by reading about it, but never having heard it natively. You can acquire a *vocabulary* of non-inverting speech, but the *wisdom* of non-inverting speech is rooted in the realization. This is also why so much of traditional Buddhist practice emphasizes meditation (*śamatha* / *samādhi*) and not just study or ethics.
Misreading 5: The three-nature framework can be flattened into "everything is empty"
Correction: The Yogācāra framework that 後得智 operates within is *three natures*, not one blanket "emptiness." Specifically:
- 遍計所執性 (*parikalpita-svabhāva*, imagined nature): the conceptual fabrications we superimpose. The target of deconstruction. - 依他起性 (*paratantra-svabhāva*, dependent nature): the actual arising of phenomena through conditions. Seen as 如幻 ("like an illusion") — not unreal, not independently real, but dependently arisen. - 圓成實性 (*pariniṣpanna-svabhāva*, perfected nature / suchness): the non-graspability of subject and object, the actualized truth.
後得智 sees 依他起性 as illusory *and* sees 圓成實性 as actually present. To flatten this into "everything is empty, nothing exists" is to commit the very inversion the framework is trying to undo. The 依他起性 is not negated; it is seen as it is — dependently arisen, like an illusion — and precisely *because* it is seen as it is, the wisdom that sees it can work freely within it without being captured by it.
Misreading 6: The "fivefold" analysis means five sequential stages
Correction: 通達, 隨念, 安立, 和合, 如意 are five *functions* of 後得智, not five stages on a ladder. A single moment of 後得智 can include several of these simultaneously. In teaching, for example, you are 安立思擇; but in the same act of teaching, you can also be 和合思擇 (seeing the unifying suchness), 如意思擇 (whatever is needed arises accordingly), and so on. The fivefold analysis is a *taxonomy of function*, not a *chronology of attainment*.
Misreading 7: This concept is only relevant for monastics or advanced practitioners
Correction: The *full* development of 後得智 is, classically, a high-stage bodhisattva achievement. But the *structure* — that realization needs to be brought into function, that function can be done with or without non-inverting awareness, that ordinary action is either inverted or non-inverted — is relevant to every practitioner, and arguably to every thoughtful contemporary person. The vocabulary of 加行 / 根本 / 後得, and the discipline of asking "is this action I am about to take arising from inverted grasping, or from sustained realization?" are usable today, even before full realization.
Misreading 8: 後得智 means you no longer have a sense of self or agency
Correction: The text does not say this. The 圓成實性 that 後得智 sees is precisely the *non-graspability* of subject and object — meaning they were never the solid, independently-existing entities ordinary consciousness takes them to be. It does *not* mean that the conventional, dependently-arisen functional self is annihilated. The 啞人→非啞人 person is still a person; the eyes that open are still eyes; the colored images in space are still images. The non-inverting character of 後得智 is about the *structure of grasping*, not about the disappearance of conventional functioning.
A final note on the relation to other famous teachings
The *Heart Sutra* (《般若波羅蜜多心經》) declares: "照見五蘊皆空, 度一切苦厄" — "fully illuminating that the five aggregates are empty, transcending all suffering." This is the voice of 般若 (*prajñā*, discriminating-awareness-as-wisdom) at its most radical, and it is most precisely read as the 根本智 vision: the five aggregates — form, sensation, perception, mental formations, consciousness — seen as empty of independent self-nature.
The *Diamond Sutra* (《金剛般若波羅蜜多經》) says: "應無所住而生其心" — "one should give rise to a mind that does not abide anywhere." This is also a 根本智 teaching, in a slightly different idiom: the mind is not to fix upon any object as if it were a solid abode.
Both of these are typically quoted in the context of *root* wisdom. The 後得智 teaching, in the *Mahāyāna-saṃgraha-bhāṣya*, is what *comes after* such teachings are realized — when the practitioner must *speak*, *teach*, and *act* in a world of five aggregates, without falling back into the very reification the root wisdom saw through.
If the Heart Sutra is the peak, the *Mahāyāna-saṃgraha-bhāṣya* is the part of the tradition that asks: and what does the wisdom look like, on the way back down, with eyes open?
*This reading draws on the two Chinese translations of the Mahāyāna-saṃgraha-bhāṣya: T1597 (玄奘譯世親釋) and T1598 (無性釋). The analysis of the three wisdoms, the four simile-systems (啞人受境, 學論, 閉目開目, 虛空色像), the fivefold 後得智 (通達, 隨念, 安立, 和合, 如意思擇), and the technical role of 後得智 within the three natures (遍計所執, 依他起, 圓成實) is based on these two texts. The similes are preserved in their original one-to-one correspondence, not collapsed into a generic "wisdom after meditation." The contemporary analogies (work, relationships, parenting, AI-era reflection) are offered as illustrations, not as equivalences — the classical framework has its own integrity that no analogy exhausts.*