唯識二十論 偈十三 破極微三
偈十一、十二拆光了原子的「合」，偈十三拆最根本的一根柱子——方分（原子有沒有方向上的面）。兩難：有方分，原子就是六個方分湊成的，早是「多」，不叫不可再分的一顆，「理不應成一」；沒方分，世親就指著太陽問——太陽剛升起照到原子，這原子背光的那邊怎麼會有影？要有影必須有向光面與背光面之別，原子若無方分就該被整個照透、生不出陰面，「以無餘分光所不及」，可物體明明有影。同一招用在「相障」：無方分就沒有擋人的那一面，原子之間怎麼互相阻擋？不相擋就能疊進同一位置，又縮回「一堆只一顆大」的老毛病。
對手想把影、障推給「聚色」，世親一句頂回：聚不是原子以外的東西（聚不異微），哪來異於原子的聚去發影阻擋？「聚不異無二」，推給聚等於沒推。於是安排成原子也好、聚也好，「俱非一實」，原子論徹底散架。
破完原子，世親立刻把同一把尺量向你眼前那片青色（眼等所對的青等實色），問它是一是多：說多則重蹈原子諸過，說一也不通——具體怎麼不通留給偈十四用走路、象馬、水虫等日常例子說。偈十三是 §參 的鐵錘：一個太陽下的影子、一次尋常的相礙，就敲碎了「無方分原子」這最後的退守。記住那畫面——太陽一照，無方分的原子生不出影。
Verses 11 and 12 took apart the atom's "combining"; Verse 13 dismantles the most fundamental pillar of all — directional parts (方分: whether an atom has faces in a direction). A dilemma: if it has directional parts, then the atom is six directional parts assembled together, already "many," and is not the indivisible single thing — "by reason it should not be one" (理不應成一); if it has no directional parts, Vasubandhu points straight at the sun and asks — the sun has just risen and shines upon an atom; how could the side of that atom turned away from the light have a shadow? For there to be a shadow there must be a distinction between a light-facing face and a face away from the light; if the atom has no directional parts it ought to be shone through whole and give rise to no shaded face — "there being no remaining part the light does not reach" (以無餘分光所不及) — yet things plainly do have shadows. The same move applies to "mutual obstruction": with no directional parts there is no face to block another, so how could atoms obstruct one another? Not obstructing, they could pile into one and the same place, and we slide back into the old ailment of "a whole heap being no bigger than a single one."
The opponent wants to push the shadow and the obstruction onto "aggregate-form," but Vasubandhu turns it back in one stroke: the aggregate is not something apart from the atom (the aggregate is not other than the atom) — where would you get an aggregate distinct from the atom to cast a shadow and obstruct? "The aggregate is not other, so there is no second thing" (聚不異無二); pushing it onto the aggregate is no push at all. And so, arranged as atom or as aggregate, "neither is a real single thing" (俱非一實), and atomism utterly falls apart.
Having finished refuting the atom, Vasubandhu lays the same measuring-rod at once against that patch of blue before your eyes (the real form, such as blue, confronting eye and the rest), asking whether it is one or many: say many, and it repeats all the atom's faults; say one, and that too does not hold — just how it does not hold he leaves to Verse 14, told with everyday examples of walking, elephants and horses, water-creatures. Verse 13 is the iron hammer of §III: with one shadow under the sun and one ordinary case of mutual obstruction, it smashes the last redoubt, "the partless atom." Hold that picture in mind — the sun shines, and the partless atom can give rise to no shadow.
§頌
極微有方分，理不應成一； 無應影障無，聚不異無二。
§論（世親自註）
論曰：以一極微六方分異多分為體，云何成一？若一極微無異方分，日輪纔舉光照觸時，云何餘邊得有影現？以無餘分光所不及。又執極微無方分者，云何此彼展轉相障？以無餘分他所不行，可說此彼展轉相礙。既不相礙，應諸極微展轉處同，則諸色聚同一極微量，過如前說。云何不許影障屬聚不屬極微？豈異極微許有聚色發影為障？不爾，若爾聚應無二。謂若聚色不異極微，影障應成不屬聚色。安布差別立為極微，或立為聚，俱非一實，何用思擇極微聚為？猶未能遮外色等相，此復何相？謂眼等境亦是青等實色等性。應共審思，此眼等境青等實性為一為多？設爾何失？二俱有過，多過如前。一亦非理。頌曰：
The canonical text stands in Chinese only. This edition renders the essence, the plain reading, the exposition and the apparatus into English — never the sūtra itself. Switch to 漢讀本 or 並行雙語 above to read it.
§頌 白話
極微若有方分，照理就不該是「一」（成了多分）； 若沒有方分，那就不該有影、有障（可影障明明有）； 而聚色又不異於極微，所以也沒有「第二個（能發影障的東西）」。
§論 白話
世親解釋這首偈。
如果一顆極微是以「六個方向的部分各不相同」作為它的體，那它怎麼還算「一」呢（已經是多分了）？
反過來，如果一顆極微沒有不同的方分：那麼太陽剛升起、光照到它的時候，它另一邊（背光那邊）怎麼會有影出現呢？因為（它若無方分）就沒有「光照不到的那一分」（也就沒有背光的陰面）。
再說，主張極微沒有方分的人：那此一顆、彼一顆極微，又怎麼會彼此互相阻擋呢？要因為有「別人到不了的那一分」（相對的面），才談得上此彼互相阻礙。既然（無方分而）不互相阻礙，那麼眾極微就應該可以疊在同一個位置上，於是一切色聚就只有一顆極微那麼大——過失和前面（偈十一）說的一樣。
為什麼不允許「影和障是屬於聚色、不屬於極微」呢？難道在極微之外，另有一個「聚色」能發出影、作為障嗎？不是這樣的。如果是這樣（聚異於極微），那聚色就不該有「第二個」（與極微不同的東西）了——也就是說，如果聚色不異於極微，那影和障就成了不屬於聚色（而還是落在極微上，難仍不解）。
所以，把這安排成「極微」，或安排成「聚」，兩者都不是一個實在的單一物，那又何必去思量推求「極微、聚」呢？
但這還不能遮除「外色等相」——這（外色）又是什麼相呢？就是說：眼等所對的境，也就是青等實在的色等體性。我們應該一起仔細思量：這眼等所對的青等實性，到底是「一」還是「多」？假設（隨你選），會有什麼過失？兩種都有過失：「多」的過失如前面（破極微）所說；「一」也不合理。世親用下一首偈來說明。
§Verse — plain reading
If an atom has directional parts (方分), then by rights it should not be "one" (it has become many parts); if it has no directional parts, then there should be no shadow and no obstruction (yet shadow and obstruction plainly exist); and since aggregate-form (聚色) is not other than the atom, there is no "second thing" either (nothing else to cast a shadow or obstruct).
§Auto-commentary — plain reading
Vasubandhu explains this verse.
If a single atom (極微 paramāṇu) has, as its very substance, "six directional parts, each distinct," then how can it still count as "one"? (It is already many parts.)
Conversely, if a single atom has no distinct directional parts: then when the sun has just risen and its light shines upon the atom, how could there come to be a shadow on its other side (the side away from the light)? For (having no directional parts) there is no "part the light does not reach" (and so no shaded face away from the light).
Further, take one who holds that the atom has no directional parts: then how could this atom and that atom ever obstruct one another? Only because there is "a part that the other cannot enter" (a facing side) can one speak of this and that obstructing each other. And since (having no directional parts) they do not obstruct one another, then the many atoms ought to be able to pile up in one and the same location, so that every aggregate of form would be no larger than a single atom — the very fault stated before (in Verse 11).
Why not allow that "shadow and obstruction belong to aggregate-form, not to the atom"? Is it that, apart from the atom, there is some further "aggregate-form" that can cast a shadow and serve as an obstruction? It is not so. If it were so (the aggregate being other than the atom), then aggregate-form ought not to have a "second thing" (something distinct from the atom) — that is to say, if aggregate-form is not other than the atom, then shadow and obstruction turn out not to belong to aggregate-form (and still fall back on the atom, so the objection remains unresolved).
Therefore, whether one arranges this as "atom" or arranges it as "aggregate," neither is a real single thing. So why bother to ponder and investigate "atom" and "aggregate" at all?
But this still does not rule out the "characteristic of external form and the rest." And what is this (external form)? It means: the object confronting eye and the rest — that is, the real substance of form and the rest, such as blue and so on. We should together examine carefully: this real nature of blue and so on, confronting eye and the rest — is it in the end "one" or "many"? Suppose (whichever you choose), what fault follows? Both have a fault: the fault of "many" is as stated before (in the refutation of atoms); and "one" too is unreasonable. Vasubandhu makes this clear with the verse that follows.
最後一根柱子：方分
偈十一、十二把原子的「合」拆光了。偈十三來拆最根本的一根柱子——方分（原子有沒有「方向上的部分」，也就是有沒有上下左右這些面）。世親撂下的兩難是：有，也錯；沒有，也錯。
有方分，就不是「一」了。 原子若有六個方向的面，那它本身就是六個方分湊成的——「六方分異為體」，早已是「多」，哪還叫得不可再分的「一顆原子」？這一頭，自破。
一個影子，問倒無方分的原子
那就說它「沒有方分」吧。世親不跟你繞理論，他指著太陽——
太陽剛升起，光照到一顆原子。 請問：這原子背著太陽的那一邊，怎麼會有影（陰面）？要有影，必須有「向光的一面」和「背光的一面」之別。可你說原子沒有方分、沒有面的分別，那它就該被光「整個」照透，根本生不出一個背光的陰面——「以無餘分光所不及」。可是物體明明有影啊。一個小小的影子，就把「無方分的原子」問倒了。
同一招再用一次：相障。 兩個東西為什麼會彼此擋住（相礙）？因為一個有「這一面」、另一個有「那一面」，面對面，過不去。可原子若沒有方分、沒有面，它就沒有「擋得住別人的那一面」，那原子之間怎麼會互相阻擋？若不互相阻擋，原子們就能疊進同一個位置——又縮回「一堆原子只有一顆大」的老毛病（偈十一已破）。
有方分非一、無方分則連影子和阻擋都沒有——兩頭堵死。
想把影子推給「聚」？聚還是原子
對手最後想賴：那影、那阻擋，算在「聚色」頭上，不算在原子頭上，行不行？世親一句頂回去：聚色不是原子以外的東西（聚不異微），你哪來一個「異於原子的聚」去發影、去阻擋？——「聚不異無二」。推給聚，等於沒推。
於是不管你把這東西安排成「原子」還是「聚」，都不是一個實在的單一物——「俱非一實」。原子論到這裡，徹底散架。
順手把戰場挪到「青」
破完原子，世親立刻轉身，把同一把尺量向更直接的對象——你眼前那片青色（眼等所對的青黃等實色）。他問：這片青，到底是「一」還是「多」？兩條路他都看死了：說它是「多」，那就重蹈原子的多種過失；說它是「一」，也不通（一整片青怎麼會有這邊那邊、走得到走不到的分別？）。具體怎麼不通，他留給偈十四，用走路、象馬、水虫一串日常例子來說。
承衍
偈十三是 §參 的鐵錘：用一個太陽下的影子、一次尋常的相礙，把「無方分原子」這個最後的退守也敲碎，連「推給聚色」的退路一併封死，坐實「原子也好、聚也好，都不是一個實在的單一物」。然後它不停步，把炮口從抽象的「原子」轉向你眼前活生生的「青色」，為偈十四的日常驗證鋪好軌。讀偈十三，記住那個畫面——太陽一照，無方分的原子生不出影：唯識破實色，靠的不是玄談，而是一個你每天都看得見的影子。
The last pillar: directional parts
Verses 11 and 12 took apart the atom's "combining." Verse 13 comes to dismantle the most fundamental pillar of all — directional parts (方分 digbhāga: whether an atom has "parts in a direction," that is, whether it has faces such as up, down, left, right). The dilemma Vasubandhu sets down is: say it does — wrong; say it does not — also wrong.
If it has directional parts, then it is no longer "one." If an atom has faces in six directions, then it is itself made up of six directional parts assembled together — "six directional parts, each distinct, as its substance" (六方分異為體); it is already "many," so how can it still be called "one indivisible atom"? On this horn, it refutes itself.
One shadow puts the partless atom to the question
So let us say it "has no directional parts." Vasubandhu will not wander into theory with you; he points at the sun —
The sun has just risen, and its light shines upon an atom. The question is: how could the side of this atom that is turned away from the sun have a shadow (a shaded face)? For there to be a shadow, there must be a distinction between a "side facing the light" and a "side away from the light." But you say the atom has no directional parts, no distinction of faces — then it ought to be shone through "whole" by the light, and simply cannot give rise to a shaded side away from the light — "there being no remaining part the light does not reach" (以無餘分光所不及). Yet things plainly do have shadows. One small shadow puts "the partless atom" to the question.
The same move, used once more: mutual obstruction. Why do two things block each other (obstruct each other)? Because one has "this face" and the other has "that face," and facing each other, neither can pass through. But if an atom has no directional parts, no faces, then it has no "face that can block another," so how could atoms obstruct one another? And if they do not obstruct one another, the atoms could pile up into one and the same location — and we slide right back into the old ailment of "a whole heap of atoms being no bigger than a single one" (already refuted in Verse 11).
If it has directional parts it is not one; if it has no directional parts there is neither shadow nor obstruction — both horns are sealed off.
Want to push the shadow onto the "aggregate"? The aggregate is still the atom
The opponent makes one last attempt to wriggle out: well then, the shadow and the obstruction — count them against "aggregate-form" (聚色), not against the atom; will that do? Vasubandhu turns it right back in a single stroke: aggregate-form is not something apart from the atom (the aggregate is not other than the atom) — where would you get an "aggregate distinct from the atom" to cast a shadow and to obstruct? — "the aggregate is not other [than the atom], so there is no second thing" (聚不異無二). Pushing it onto the aggregate is no push at all.
And so, whether you arrange this thing as "atom" or as "aggregate," neither is a real single thing — "neither is a real one" (俱非一實). At this point atomism has utterly fallen apart.
In passing, moving the battlefield to "blue"
Having finished refuting the atom, Vasubandhu turns at once and lays the same measuring-rod against a more immediate object — that patch of blue before your eyes (the real form, such as blue and yellow, confronting eye and the rest). He asks: this patch of blue — is it in the end "one" or "many"? He has both roads dead to rights: say it is "many," and it repeats the manifold faults of the atom; say it is "one," and that too does not hold (how could a single whole patch of blue have a here and a there, a reachable and an unreachable distinction?). Just how it does not hold, he leaves to Verse 14, to be made out with a string of everyday examples — walking, elephants and horses, water-creatures.
Carried forward (承衍)
Verse 13 is the iron hammer of §III: with a single shadow under the sun and one ordinary case of mutual obstruction, it smashes even the last redoubt, "the partless atom," and seals off along with it the line of retreat of "pushing it onto aggregate-form," making good that "atom or aggregate, neither is a real single thing." Then it does not halt: it swings the cannon's mouth from the abstract "atom" round to the living "blue" before your eyes, laying the track for the everyday verification of Verse 14. Read Verse 13, and hold that picture in mind — the sun shines, and the partless atom can give rise to no shadow: consciousness-only refutes real form not by abstruse talk, but by a shadow you can see every single day.
考據Apparatus
一、科判定位
| 系統 | 此偈之位 |
|---|---|
| 全論五段 | §參 破極微（偈十–十四）之正破偈（三·方分）。偈十二開「有分無分俱大失」之新軸 → 偈十三正破方分：有方分則非一、無方分則無影障；聚不異微故無別二 → 偈十四 結成唯識、啟現量 |
| 偈十二→偈十三 | 偈十二末問「有分無分俱為大失，所以者何」，偈十三正答此方分兩難，並轉就「眼等境青等實性為一為多」收歸總破 |
| 玄奘段位 | §頌〔p0076a15–16〕· §論自註〔p0076a17–b01〕（§參 中最長之自註） |
二、源頌 · 義理脈絡
§頌兩難一斷：「極微有方分，理不應成一」——有六方分異則體已多分，不成一極微；「無應影障無」——若無方分，則無「日照之影、彼此之障」（皆須有「向光面／背光面」「相對之面」方成）；「聚不異無二」——聚色不異極微，故不得別立「影障屬聚不屬微」之第二者。
§論四節：(一)有方分難——六方分異為體，云何成一。(二)無方分難·影喻——若無異方分，日輪纔舉光照時，餘邊（背光面）何得有影？以無餘分（無背光之分）光無不及。(三)無方分難·障喻——無方分則彼此無「他所不行」之餘分，云何展轉相障？不相礙則諸微處同，聚同一微量（過如偈十一）。(四)破「影障屬聚」之救＋轉軸——聚不異微，影障不得獨屬聚；安布為微為聚俱非一實。末轉問「眼等境青等實性為一為多」，斷「二俱有過（多過如前·一亦非理）」，啟偈十四。
三、經論宗依（經證 → 論本）
- 經證：日影、物障、青等色之可見，皆現量共許之事；世親即以「無方分則無影障」之現量反例，破無方分極微。
- 論本：世親自註（T1590）以方分之有無雙破極微，並以日影·相障二現量事為據；論本之破，繫於對辯宗自設之「無方分極微」，逐難拆解，不增不判。
四、本論釋要（世親自註 T1590）
本偈以方分兩難封死極微，鑰在二現量事：
- 日影：物有影，必因有「向光面」與「背光面」之別。極微若無方分，則無背光之餘分，日一照即全體受光，何來餘邊之影？可影實有，故極微非無方分。
- 相障：物相礙，必因有「此面」「彼面」相對而「他所不行」。極微若無方分，則無相對之餘分，何能彼此相障？若不相障，則諸微可同處，聚不過一微量（重蹈偈十一之失）。 有方分則非一（已成多分）、無方分則無影障（違現量），兩難俱在。對方若救「影障屬聚不屬微」，世親破：聚不異微，影障不得獨繫於聚。故安布為微、為聚，「俱非一實」。至此極微之實有徹底瓦解，世親遂收束轉軸——直問「眼等所對之青等實色，其性為一為多」：多則重蹈極微之多過、一則非理（一整片青不可有漸次方位之別，詳偈十四）。本偈所涉之部派「無方分極微」為論本明標之所破，志其執與破，不判其宗。
五、諸家異釋（古德 · 標宗）
- 護法《成唯識寶生論》T31n1591（唐·義淨譯·漢有傳本）——卷三釋本偈，展開方分兩難、日影相障喻及青等實性之轉問。〔標宗：護法於影障喻及方分難有詳釋；古德讀法之一，此處只記其宗，不判。〕
- Vinītadeva Prakaraṇaviṃśakaṭīkā——藏譯存·漢無傳本，依例僅志其名，內容不入義據。
六、譯名 · 古譯對讀（三漢譯）
| 漢譯 | 譯本 | 對讀 |
|---|---|---|
| ★ 唐·玄奘 T31n1590 | 「極微有方分，理不應成一；無應影障無，聚不異無二」 | 五言四句，方分兩難＋影障喻＋聚不異微之斷；自註以日影·相障二現量事破無方分；以此為底本 |
| 陳·真諦 T31n1589《大乘唯識論》 | 散文連展方分兩難與影障喻（古代漢證·參校） | |
| 後魏·般若流支 T31n1588《唯識論》 | 對應偈頌段以散文出之；逐字以 CBETA 為準（古代漢證·參校） |
七、術語（古德所傳梵 · 發音輔助）
| 漢 | 梵（發音輔助） | 義 |
|---|---|---|
| 方分 | digbhāga | 方向之分（向光／背光、此面／彼面）；有則非一、無則無影障 |
| 影 | chāyā | 日照所成之影；須有背光之分方成 |
| 障（相礙） | āvaraṇa | 彼此相礙；須有相對之分方成 |
| 青等實色 | — | 眼等所對之實色（青黃等）；末轉問其為一為多 |
梵羅馬字僅作發音輔助；無古德音寫可徵者從闕，不據現代校本補。
八、互讀（唯識系次第）
| 百法明門論 | 唯識三十論 | 二十論內部 |
|---|---|---|
| #11 色法——極微·青等實色皆所執實色；破之即破色為外境 ⭐ | 偈一「彼依識所變」——青等亦識變似現 | 偈十二 破極微二——本偈承「有分無分俱大失」正破方分 |
| #1 眼識——「眼等境青等」之能緣 | 偈八 似色現識——青等乃識變 | 偈十四 破極微四——以走路象馬水虫破「青等為一」 |
| #8 阿賴耶識——色等似境之所變 | 偈十七「故一切唯識」——§參 歸結 | 偈十 三難——本偈封死「極微不成」之總因 |
九、校本定位
- 漢文正文（§頌）：T31n1590〔p0076a15–16〕（玄奘譯）
- 漢文自註（§論）：T31n1590〔p0076a17–b01〕（玄奘譯；§參 最長自註）
- 參校漢譯：T31n1589（真諦《大乘唯識論》）· T31n1588（般若流支《唯識論》）
- 古德釋：T31n1591（護法《成唯識寶生論》卷三·漢有傳本）
1. Doctrinal-outline placement
| System | This verse's place |
|---|---|
| The treatise's five parts | The direct-refutation verse (III · directional parts) of §III Refuting Atoms (破極微, vv.10–14). Verse 12 opens the new axis 「有分無分俱大失」 ("having parts or not having parts, both are grave faults") → v.13 directly refutes directional parts: if it has directional parts it is not one; if it has no directional parts there is no shadow or obstruction; the aggregate is not other than the atom, so there is no second thing → v.14 concludes to consciousness-only and opens [the refutation of] direct perception. |
| v.12 → v.13 | At its close v.12 asks 「有分無分俱為大失，所以者何」 ("having parts or not having parts, both are grave faults — why so?"); v.13 directly answers this dilemma of directional parts, and turns to "the real nature of blue and the rest, the object of eye and the rest — is it one or many?" to gather everything into the overall refutation. |
| Xuanzang's textual loci | §Verse 〔p0076a15–16〕· §Auto-commentary 〔p0076a17–b01〕 (the longest auto-commentary within §III) |
2. The source-verse · the doctrinal thread
The §Verse poses two horns and one severance: 「極微有方分，理不應成一」 ("if the atom has directional parts, by reason it should not be one") — if six directional parts are distinct, the substance is already many parts, and it does not amount to one atom; 「無應影障無」 ("if it has none, then there should be no shadow or obstruction") — if there are no directional parts, then there is no "shadow from sunlight, no mutual obstruction" (both requiring a "light-facing face / a face away from the light" and "a facing face" in order to hold); 「聚不異無二」 ("the aggregate is not other [than the atom], so there is no second thing") — aggregate-form is not other than the atom, so one cannot separately set up a "second thing" whereby "shadow and obstruction belong to the aggregate, not to the atom."
The §Auto-commentary has four movements: (i) The objection on having directional parts (有方分難) — with six directional parts distinct as its substance, how does it amount to one? (ii) The objection on having no directional parts · the shadow illustration (無方分難·影喻) — if there are no distinct directional parts, then when the orb of the sun has just risen and at the moment its light shines, how could the remaining side (the side away from the light) come to have a shadow? Since there is no remaining part (no part away from the light), the light reaches everywhere. (iii) The objection on having no directional parts · the obstruction illustration (無方分難·障喻) — without directional parts there is no remaining part "the other cannot enter," so how could [atoms] obstruct one another in turn? Not obstructing each other, the various atoms occupy the same place, and the aggregate is no more than a single atom's measure (the fault as in Verse 11). (iv) Refuting the rescue "shadow and obstruction belong to the aggregate" + the pivot (破「影障屬聚」之救＋轉軸) — the aggregate is not other than the atom, so shadow and obstruction cannot belong to the aggregate alone; arranged as atom or as aggregate, neither is a real single thing. At the close it turns to ask "the real nature of blue and the rest, the object of eye and the rest — is it one or many?" and severs this with "both alternatives have a fault (the fault of many as before · the one too is unreasonable)," opening Verse 14.
3. Scripture and treatise: the grounding (scriptural proof → treatise-root)
- Scriptural proof. The shadow of the sun, the obstruction of things, and the visibility of form such as blue and the rest are all matters jointly admitted in direct perception (現量共許); Vasubandhu uses precisely the direct-perception counter-example "if there are no directional parts there is no shadow or obstruction" to refute the partless atom.
- Treatise-root. Vasubandhu's auto-commentary (T1590) doubly refutes the atom through the presence or absence of directional parts, taking as evidence the two direct-perception facts of the sun's shadow and mutual obstruction; the treatise-root's refutation is fastened to the "partless atom" that the disputing position itself posits, dismantling each objection in turn, adding nothing and adjudicating nothing.
4. The treatise-root in essence (Vasubandhu's auto-commentary, T1590)
This verse seals off the atom by the dilemma of directional parts (方分兩難), the key lying in two direct-perception facts:
- The sun's shadow (日影): a thing has a shadow only because of the distinction between a "light-facing face" and a "face away from the light." If the atom has no directional parts, then there is no remaining part away from the light; once the sun shines, the whole body receives light — whence the shadow on the remaining side? Yet shadow really exists, so the atom is not without directional parts.
- Mutual obstruction (相障): things obstruct each other only because there is a "this face" and a "that face" facing each other such that "the other cannot enter." If the atom has no directional parts, then there is no facing remaining part; how could they obstruct each other? And if they do not obstruct each other, the atoms can occupy the same place, and the aggregate is no more than a single atom's measure (treading again into the fault of Verse 11). If it has directional parts it is not one (it is already many parts); if it has no directional parts there is no shadow or obstruction (contradicting direct perception) — both horns stand. If the opponent rescues with "shadow and obstruction belong to the aggregate, not to the atom," Vasubandhu refutes it: the aggregate is not other than the atom, so shadow and obstruction cannot be fastened to the aggregate alone. So whether arranged as atom or as aggregate, 「俱非一實」 ("neither is a real single thing"). At this point the reality of the atom utterly disintegrates, and Vasubandhu thereupon draws to a close and pivots — directly asking "the real form such as blue and the rest, the object confronting eye and the rest — is its nature one or many?": if many, it treads again into the manifold faults of the atom; if one, it is unreasonable (a single whole patch of blue cannot have a gradual distinction of position; see Verse 14 in detail). The sectarian "partless atom" treated in this verse is the thing-to-be-refuted explicitly named by the treatise-root; [the apparatus] records its position and its refutation, and does not adjudicate its tenet.
5. The ancient masters' readings (古德 · stating their position)
- Dharmapāla, Cheng weishi baosheng lun 《成唯識寶生論》T31n1591 (Tang; tr. Yijing 義淨; a transmitted Chinese witness) — fascicle three expounds this verse, unfolding the dilemma of directional parts, the shadow-and-obstruction illustrations, and the turning-question on the real nature of blue and the rest. 〔Position stated: Dharmapāla has a detailed exposition of the shadow-and-obstruction illustration and the objection from directional parts; this is one of the ancient masters' readings, recorded here only as to its position, not adjudicated.〕
- Vinītadeva, Prakaraṇaviṃśakaṭīkā — extant in Tibetan translation · no transmitted Chinese witness; per convention its name only is recorded, its content not entering as doctrinal ground.
6. Rendering and old-translation comparison (the three Chinese translations)
| Chinese translation | Edition | Comparison |
|---|---|---|
| ★ Tang · Xuanzang T31n1590 | 「極微有方分，理不應成一；無應影障無，聚不異無二」 | A five-character quatrain: the dilemma of directional parts + the shadow-and-obstruction illustration + the severance "the aggregate is not other than the atom"; the auto-commentary refutes the partless [atom] by the two direct-perception facts of the sun's shadow and mutual obstruction; this serves as the base text. |
| Chen · Paramārtha T31n1589 《大乘唯識論》 | Continuous prose unfolding the dilemma of directional parts and the shadow-and-obstruction illustration (an ancient Chinese witness · collated for comparison). | |
| Later Wei · Prajñāruci T31n1588 《唯識論》 | The corresponding verse-passage is given in prose; verbatim follows the CBETA original (an ancient Chinese witness · collated for comparison). |
Verbatim comparison follows the CBETA original text; what has not been verified against the original is not recorded verbatim.
7. Terms (Sanskrit per the ancient transmission · pronunciation aid)
| Chinese | Sanskrit (pronunciation aid) | Sense |
|---|---|---|
| 方分 | digbhāga | a part in a direction (light-facing / away from light; this face / that face); having it, it is not one; lacking it, there is no shadow or obstruction |
| 影 | chāyā | the shadow formed by sunlight; requires a part away from the light in order to form |
| 障（相礙） | āvaraṇa | mutual obstruction; requires a facing part in order to form |
| 青等實色 | — | the real form (blue, yellow, and so on) confronting eye and the rest; at the close [the verse] turns to ask whether it is one or many |
The Sanskrit roman script is a pronunciation aid only; where no ancient-master transliteration is attested, it is left blank, not supplied from modern critical editions.
8. Cross-reading (the Yogācāra sequence)
| Baifa (Hundred Dharmas) | Triṃśikā | Within the Viṃśatikā |
|---|---|---|
| #11 form-dharmas (色法) — the atom and the real form such as blue are all the real form that is clung to; refuting it is refuting form as an external object ⭐ | Triṃśikā v.1 「彼依識所變」 ("that rests on the transformations of consciousness") — blue and the rest too are transformed and appear as if present by consciousness | v.12 Refuting Atoms II — this verse takes up "having parts or not having parts, both are grave faults" and directly refutes directional parts |
| #1 eye-consciousness (眼識) — that which cognizes "the object of eye and the rest, blue and the rest" | Triṃśikā v.8 似色現識 (the consciousness in which form-like appearances arise) — blue and the rest are transformations of consciousness | v.14 Refuting Atoms IV — refutes "blue and the rest being one" by walking, elephants-and-horses, water-creatures |
| #8 storehouse consciousness (阿賴耶識) — that which transforms into form and the rest as semblance-objects | Triṃśikā v.17 「故一切唯識」 ("therefore all is consciousness-only") — the conclusion of §III | v.10 the three objections — this verse seals off the overall ground that "the atom does not hold" |
9. Witness placement
- Chinese verse-text (§Verse): T31n1590 〔p0076a15–16〕 (Xuanzang)
- Chinese auto-commentary (§Auto-commentary): T31n1590 〔p0076a17–b01〕 (Xuanzang; the longest auto-commentary in §III)
- Chinese translations collated for comparison: T31n1589 (Paramārtha, 《大乘唯識論》) · T31n1588 (Prajñāruci, 《唯識論》) — verbatim comparison follows the CBETA original
- Ancient master's commentary: T31n1591 (Dharmapāla, 《成唯識寶生論》fascicle three · a transmitted Chinese witness)