Notes on the Rūpa Section of the Pañcaskandhakavibhāṣā

Notes on the Rupa Section of the Pancaskandhakavibhasa

Jowita Kramer

(University of Munich)

This paper focuses on the section on “matter” (rupa) in Sthiramati’s Pancaskandhakavibhasa and its root text, Vasubandhu’s Pahcaskandhaka, providing an overview of the contents and comparing the Pancaskandhakafyibhasdys positions with parallel explanations found in the Abhidharmakosabhasya, the Abhidharmasamuccava, and the Abhidharmasamuccayabhasya. Moreover, two other commentaries on the Pahcaskandhaka have been consulted, the Pahcaskandhavivarana and the Pahcaskandhabhasya, both available only in their Tibetan translations. The latter two works have been accounted for in the present study whenever their doctrinal standpoints differ from Sthiramati or when they appeal' relevant for other reasons.

Sthiramati opens his commentary on the “heap of matter” (rupaskandha) with the statement that it is impossible to determine the “own existence” (svabhava) of the skandhas due to their lack of a svabhava. But, according to Sthiramati, it is possible to expose the basis [of their designation] (upadana), which in the case of rupaskandha is rupa.

Therefore, in the next passage of the text a detailed definition of the constituents of rupa is given. The Pahcaskandhaka describes rupa as the four basic elements (mahabhuta) and the matter that is dependent (upadaya) on them? In his commentary Sthiramati first provides a number of arguments explaining the term mahabhuta:^

The elements are qualified as maha- because

1) they are “coarse” (audarika) serving as the basis for (subtle) dependent matter,

2) they are present (as solidity etc.) in all aggregations of visible matter etc., and

3) they are of great importance for the composition of matter. Moreover, the four are referred to as bhuta since

1) they come into existence in this or that appearance when the different kinds of dependent matter arise and

2) they have never been non-existent in the beginningless samsara. In contrast to Sthiramati’s commentary, the Pahcaskandhavivarana and the Pahcaskandhabhcisya relate the “greatness” of the mahabhutas to the destruction of the different realms of the world.

The Vivarana explains that the destruction by fire reaches up to the realms corresponding to the first dhyana, the destruction by water up to the second dhyana, and the destruction by wind up to the third.6 The Bhasya assigns the lirst destruction (up to the lirst dhyana) to water and the second (up to the second dhyana) to lire. In addition, it connects the lirst destruction with desire (raga), the second with hatred (dvesa) and the third with delusion (moha).1

All three Pahcaskandhaka commentaries provide arguments aiming to prove that space (akasa) is not to be regarded as an additional basic element. Sthiramati’s main assumption is that space is nothing other than the mere non-existence of impenetrable (sapratigha) rupa, whereas the Vivarana focuses on the idea that, contrary to the mahabhutas, space does not benelit or harm living beings and, being permanent, it is not the result or cause of anything.8 The Bhasya only mentions that in contrast to the mahabhutas space is not subjected to the process of arising and ceasing.9

In connection with the explanation of matter dependent on the mahabhutas, Sthiramati mentions live kinds of how matter derived from the elements (bhautika) is dependent on them: generating (janana), basis [of change] (nis'raya), continuity (pratistha), support (upastambha), and nourishing (brmhana).10 The lirst of these live indicates that derived matter could not arise without the existence of the mahabhutas.

The fact that the mahabhutas are the basis of the matter derived from them means, according to Sthiramati, that dependent matter changes in the same moment as the elements change.12 As long as the elements are produced in a certain continuum, the series of derived matter will not be interrupted—this is the meaning of the third kind of dependence, “continuity”.13 The dependence consisting in “support” indicates that dependent matter does not cease to exist by the power of the mahabhutas “Nourishing” points to the fact that dependent matter can only increase if the mahabhutas it depends on grow.15

The delinitions of the four mahabhutas given by Sthiramati (in accordance with Vasubandhu) in the following section of the Pancaskandhakavibhasa closely resemble the explanations of the Abhidharmasamuccaya and the Abhidharmakosabhasya. Solidity (khakkhatatva) is said to be the nature of earth, fluidity (sneha) the nature of water, heat (usmd) the nature of lire, and lightness and motion (laghusamudiranatva) the nature of wind.16 The respective activities of the four mahabhutas are support (dhrti), cohesion (Samgraha), ripening (pakti), and shifting (vyuhana)P

All three Pahcaskandhaka commentaries provide examples of how the presence of the four elements in certain objects can be inferred.18 In this context Sthiramati explains that the existence of water, fire and air in a solid object like for instance a stone is obvious from its cohesion, its dryness (pakti; Tib. skarn pa), and its ability to be moved. Remarkably, the Pahcaskandhabhasya argues differently with regard to the presence of fire in a stone adducing it from the appearance of sparks when two stones collide.

The fact that water (as “normal” water, not as a basic element) also contains the elements earth, fire and wind is deducible from its capacity to support a piece of wood, from the fact that leaves etc. rot in it and that it can be warm,19 and from the water’s ability to flow. The presence of the other three elements in a burning fire is evident from the stability,20 coherence, and the motion of the flames. And finally, wind consists of all four elements because it is able to support, for instance, leaves,22 does not disperse, and dries clothes.

In addition to the examples of the simultaneous occurrence of all four mahabhiitas the Pancaskandhabhasya quotes a passage from the Garbhavakrantisutra that discusses the results of one of the mahabhutas not being present in an embryo:22 If the water element is missing it dries up or disperses like flour or ashes. If the earth element is not present it flows out like oil. If there is no fire element it becomes rotten and without the wind element it does not increase or develop.

In the section on matter dependent on the basic elements, Sthiramati explains the meaning of pellucid matter (rupaprasada), which is the nature of the five material sense faculties.23 He gives the example of images being reflected in a clear mirror or in a vessel filled with water. In the same way the pellucid matter of the five sense faculties reflects their objects.24 Sthiramati mentions in this context that faith (sraddha) is also considered as prasada, but it is to be distinguished from the indriyas as it does not have matter (rupa) for its nature.25

It is worth investigating the subsequent passage of the Pancaskandhakavibhasa which defines the objects of each sense faculty in detail, insofar as it reveals interesting dissimilarities between the Pancaskandhakavibhasa, the Abhidharmakosabhasya, and the Abhidharmasamuc-caya. First of all, the object of the sense of sight is discussed. The Pancaskandhaka assigns three different categories to the visible (rupci): colour (yania), shape (samsthana}, and representation (yijhapti).26 All three categories are mentioned in the parallel description found in the Abhidharmasamuccaya, whereas the Abhidharmakosabhasya only points out varna and samsthana as parts of the visible and obviously places the bodily vijnapti (kayavijhapti) under samsthana.

When going into details the Pancaskandhakavibhasa, the Abhidharmasamuccaya, and the Abhidharmakosabhasya agree on four kinds of colour and eight kinds of shape.28 But they disagree on the question whether entities like a cloud, smoke, or the sunlight are to be classified as separate categories, or whether they are already included in the categories of colour and of shape. In this context A ABA 6,12 mentions eight additional entities: cloud (abhra), smoke (dhuma), dust (rajas), mist (mahika), shade (chaya), sunlight (atapa), (other) light (clloka), and darkness (andhakara).

In the Abhidharmasamuccaya space (abhyavakasa), vijnapti, and the sky (nabhas) as well as two further shapes, namely fine (rdul phra mo) and rough (rags pa) shape, are added to the eight entities mentioned in the Abhidharmakosabhasya. Sthiramati lists the entities from abhra to nabhas (omitting vijnapti, which he mentions in another section), but he rejects them as separate constituents different from colour and shape. He argues that clouds, smoke etc. are either included in the category samsthana in case they are limited or in the category varna in case they are not.

In contrast, the Pahcaskandhavivarana and the Pancaskandhabhasya list the additional items without pointing out their controversial nature. The Pancaskandhavivarana mentions abhra to nabhas (omitting abhyavakasa and vijnapti), the Pancaskandhabhasya includes the same list as the Pancaskandhakavibhasa (i.e. abhra to nabhas, omitting vijhapti) and adds brief explanations of some of the ten categories.31

While the Pahcaskandhaka mentions vijhapti as a third kind of visible rupa beside colour and shape, the Abhidharmakosabhasya includes the definition of vijhapti in another context, namely in the fourth chapter, which deals with the topic “karma”. The Abhidharmasamuccaya mentions vijhapti as one of the 25 categories forming visible matter, but it does not characterize the matter of vijhapti in more detail. Sthiramati describes vijhapti as a bodily shape that arises from a mind which has this shape as its object and that is called vijhapti because it “makes known” the mind (i.e. the intention) by which it has been aroused.

The examination of vijhapti presented in the Pancaskandhakavibhasa and the Abhidharmakosabhasva shows that there was a dispute over the characterization of the bodily vijhapti (kayavijhapti) within the different traditions. This discussion was related to the general question of whether shape (samsthana) was to be regarded as a real entity (dravya), in the same way as varna, or as a designation for an accumulation of colour atoms arranged in a certain manner.33 The first view (i.e. samsthana being dravyasat) was held by the Vaibhasikas, who considered kayavijhapti to be a kind of shape and therefore a real entity.34 They classified kayavijhapti as shape without there being colour.

The opposite view was taken by the Sautrantikas, who maintained that samsthana was a mere designation and that kayavijhapti was shape not existing as a real entity.36 A similar position is held by Sthiramati in the Pancaskandhakavibhasa. He also explains that shape does not exist as an entity and argues that this is evident for example from the fact that there are no differently shaped atoms in the same way as there are atoms of various colours. Therefore there are for instance no long-shaped atoms in an accumulation of atoms having a long shape.

Sthiramati takes shape to exist as a mere designation and not to constitute vijhapti, which he characterizes as “the uninterrupted arising of the continuum of the body at a different place” caused by an intention (and “making known” this intention to others).38 Remarkably, Sthiramati adds that according to ultimate reality (paramarthatas) varna—in the same way as samsthana—is not the object of the faculty of seeing.

He argues that this is due to the fact that perception (vijhana) does not have any outer objects, because the existence of colour atoms, like that of shape, is not possible from the viewpoint of that level.39 This remark by Sthiramati is one of the very few statements in the Pancaskandhakavibhasa pointing to the doctrine of “representation only” (vijhaptimatrata).

The classifications of sound (sabda), the object of the faculty of hearing, differ in the Pahcaskandhaka (yibhasa), the Abhidharmasamuccaya, and the Abhidharmakosabhasya. In this context it becomes obvious that Vasubandhu and Sthiramati did in some cases neither follow the tradition of the Abhidharmasamuccaya nor that of the Abhidharmakosabhasya. All four texts agree on two kinds of sound:40

1. the sound caused by the basic elements (mahdbhutahetuka) that are appropriated (upatta)

2. the sound caused by the basic elements that are not appropriated (anupatta)

The first kind of sound is identified as the sound of the voice (vac) in the Pancaskandhakavibhasa, the Abhidharmakosabhasya, and the Abhidharmasamuccayabhasya.

The Pancaskandhakavibhasa and the Abhidharmakosabhasya additionally mention the sound of the (clapping) hand (hasta). The sound that is not appropriated is, according to the Pancaskandhakavibhasa and the Abhidharmakosabhasya, the sound of the wind (yayu), of the trees (vanaspati), or of the river (nadi)A The categorization of the remaining types of sound appears to have been controversial. The Abhidharmasamuccaya and the Pahcaskandhaka distinguish a third category of sound: the sound that is both, upatta and anupatta A1 As an example for this kind of sound the Abhidharmasamuccavabhasva and the Pancaskandhakavibhasa mention the sound of a hand and a drum (mpdahga) (i.e. a hand hitting a drum).

Remarkably, this class of sound is rejected in the Abhidharmakosabhasva as a separate category. There it is stated that “others” (apare) say that a sound can be appropriated and not appropriated at the same time, but this is not accepted, as it is not admitted that one atom is based on two tetrads of the basic elements (i.e. the four basic elements of the hand and the four basic elements of the drum).44 This rejection of the sound that is both, appropriated and not appropriated, in the Abhidharmakosabhasya is remarkable insofar as the author of the Abhidharmakosabhasya is generally accepted to be identical to that of the Pahcaskandhaka What is more, two other kinds of sound mentioned in the Abhidharmakosabhasya are absent in the Pahcaskandhaka. In AKBh 6,22 the class of sounds of living beings (sattva) and, final ly, the sounds not belonging to living beings are listed additionally.

The lirst of these two classes refers to the representation of speech (vagvijhapti), the second is described as including all other kinds of sound.46 The Abhidharmakosabhasya adds that all four of the mentioned sounds can be pleasant (manojna) or unpleasant (amanojha), which makes a total of eight different categories of sound.47 In contrast, the Abhidharmasamuccaya lists five additional classes of sound besides the two categories mentioned above.

They include sounds known in the world (lokaprasiddhd), i.e. common talk (laukikabhasa), sounds produced by the siddhas (siddhopamta), fabricated (parikalpita} sounds, and sounds belonging to the common practice of the Aryas (aryavyavaharika) or to the common practice of the non-Aryas (anaryavyavaharika).^ All these sounds can be not only pleasant or unpleasant, as indicated in the explanation of the Abhidharmakosabhasya mentioned above, but also neutral.49 The Pahcaskandhabhasya lists only the three kinds of sounds mentioned in the Pancaskandhakavibhasa, whereas the third commentary on the Pahcaskandhaka, the Pahcaskandhavivarana, additionally provides explanations of the same five sound classes as those appearing in the Abhidharmasamuccaya.56

The sections on smell and taste consist of only three lines in the Pancaskandhakavibhasa. Nevertheless they are noteworthy as they differ from the respective passages in the Abhidharmasamuccaya and the Abhidharmakosabhasya. The latter mentions four types of smell: good (sit-) and bad smell (durgandha) which can both be either constant (sama) or inconstant (visama).

It is added, however, that in the sastra (i.e. in the Prakarana) three kinds of smell are taught: good, bad, and neutral (sama).51 An almost identical statement is made by Vasubandhu in the Pahcaskandhaka, the “neutral” smell being indicated with the phrase “[[[Wikipedia:smells|smells]]] other than that”.32 Sthiramati only explains the smell characterized as sama in his commentary, qualifying it as neither benefiting nor harming the basic elements of the sense of smell.

In addition he refers to another definition, which describes smell as natural (sahajd), like the smell of sandalwood (candand) or saffron (kuhkuma), as arising from a combination (samyogika), like the smell of incense (dhupavarti), and as arising from change (pdrinamika), like the smell of ripe mango fruits (pakvamraphala).^ All six types of smell are listed in the Abhidharmasamuccaya, and in its commentary the last three are illustrated with the same examples of sandalwood, incense, and ripe fruits.55 The same six categories are also mentioned in the Pancaskandhavivarana and the Pancaskandhabhasya.36

In the case of taste, Sthiramati (in accord with Vasubandhu's root text) follows the classification of the Abhidharmakosabhasya, where six types are listed: sweet (madhura), sour (amid), salty (lavana), pungent (katukci), bitter (tiktd), and astringent (kasava).51 He does not mention that taste, according to the Abhidharmasamuccaya, can also be divided into the classes pleasant (manojhd), unpleasant (amanojhd), and neutral or, in analogy to the categories of smell, into sahaja, samyogika, and pdrinamika.33, All these kinds of taste are mentioned also in the Pancaskandhavivarana, whereas the Pancaskandhabhasya lists only the first six types. Both texts provide examples for tastes that are sweet, sour etc.59

When analysing the nature of the tangible, Sthiramati again is closer to the Abhidharmakosabhasya than to the definition of the Abhidharmasamuccaya, parts of which he even refutes explicitly. In AKBh 7,8f. eleven entities are mentioned as being tangible: the four mahabhutas, softness (slaksnatvd), hardness (karkasatva), heaviness (gurutva), lightness (laghutvd), cold (sitd), hunger (Jighatsa), and thirst (pipasa).

The explanation given in the Pahcaskandhaka is more differentiated because, in contrast to the Abhidharmakosabhasya, rupa is divided into the matter of the four mahabhutas and the matter dependent on them. As the tangible is explained in the context of dependent matter, the mention of the four mahabhutas as constituents of this category would contradict the classification (of mahabhuta versus dependent matter) made earlier. Therefore it is said in PSk 3,5 that only a part of the tangible (sprastavyaikadesd) is explained in this context.

The Abhidharmasamuccaya does not mention the mahabhutas under the topic of the tangible and explains that the latter consists of dependent matter including in addition to the seven entities “softness” etc. fifteen other categories like fainting (muircha), strength (bald), and weakness (daurbalya).60 Sthiramati explains that these additional categories are not listed in the Pahcaskandhaka because they are already included in the remaining ones, like for example strength is included in hardness and heaviness and fainting in softness.61 Neither the Pancaskandhavivarana nor the Pancaskandhabhasya mention the additional items listed in the Abhidharmasamuccaya.

Notably, the Pancaskandhakavibhasa and the Pancaskandhabhasya include a list showing the relations between the seven tangible categories, softness etc., and the four basic elements. According to this list softness is mainly a combination of water and fire, hardness of earth and wind, heaviness of earth and water, lightness of fire and wind, and cold of water and wind,62 while hunger is mainly wind and thirst is for the most part fire.63

Moreover, all three Pahcaskandhaka commentaries provide reasons for the fact that hunger and thirst, characterized as the causes for the desire to eat or to drink, are actually mental states and are nonetheless mentioned among the (material) objects of the sense of touch: Hunger and thirst appear in the rupa section as the results of their causes “wind” and "lire”, which undisputedly belong to the rupa category. Sthiramati refers to this “figurative” usage of the terms as “applying the designation of the result (i.e. hunger) to the cause (i.e. wind)”.64

The third and last constituent of matter dependent on the basic elements is, according to the Pahcaskandhaka, the avijnapti, which is explained as “invisible and penetrable matter arisen from vijnapti or meditative absorption” (yijhaptisamadhijam rupam anidarsanam apratigham).

65 A similar statement appears in the Abhidharmakosabhasya, in which avjhapti is described as beneficial (kusala) or unbeneficial (akus'ala) matter, having arisen from vijnapti or samadhi.66 Remarkably, the term avijnapti is not used in the Abhidharmasamuccaya. Instead, the expression samadanika is applied to this kind of karmic matter in the context of the definition of five classes of matter belonging to the dharmayatana.61 However, the Abhidharmasamuccaya does not specify exactly what the matter belonging to the samadanika category is.

According to Sthiramati avijnapti that has arisen from vijnapti belongs to [the sphere of] sensual pleasure (kamapta) and can be divided into four classes:

(1) the restraint (samvara) of the code of precepts (pratimoksa),

(2) the restraint of the Bodhisattva,

(3) the non-restraint (asamvara), which is characterized as engaging completely in practice harming others, and

(4) neither restraint nor non-restraint, which is described as partly engaging in beneficial and harmful [[[activities]]] (ekades-enanugrahopaghatapravrtti).

This classification of avijnapti seems to be an extended adaptation of the analysis of it as found in the Abhidharmakosabhasya. There avijnapti is structured into the three classes samvara, asamvara, and naiva samvaro nasamvarah,69 The class of the bodhisattvasamvara is, as might be expected, missing in the Abhidharmakosabhasya. It is notable in this context that the same structure of three classes is presented in the Abhidharmasamuccaya, however, not explicitly describing avijnapti, but the divisions of karma.70 What is surprising here, is the fact that the Abhidharmasamuccaya does not mention the restraint of the Bodhisattva either.

The first category, samvara, is divided into three subclasses in the Abhidharmakosabhasya and the Abhidharmasamuccaya:1 The first of these is pratimoksasamvara. The other two classes are the restraint of contemplation (dhyanasamvara) and the restraint of the uncontaminated (anasravasamvara). These are the two categories that are described by Sthiramati as arising from meditative absorption (samadhija). The avijnapti that arises from meditative absorption belongs either to the material [[[sphere]]] (rupapta) or is the uncontaminated (anasrava) avijnapti J-If it belongs to the rupadhatu then it originates from contaminated meditative absorption (sasravasamadhija) of the four contemplations (dhyana), of [the stage] before attaining [the first dhyana) (anagamya), and of the states between the [[[first two dhyanas]](dhyanantara). The avijnapti that is uncontaminated arises from uncontaminated meditative absorption (anasrava-samadhija):6

In the Abhidharmakosabhasya the analysis of avijnapti is very comprehensive and the discussion of opposing views with regard to the nature of avijnapti is rather complex and difficult to understand.74 In this context one of Vasubandhu’s main concerns seems to be to oppose the Vaibhasika theory that avijnapti is an existing entity (dravya) and is a component of the category rupa. Both these assumptions were rejected by the Sautrantikas, whose arguments Vasubandhu employs to support his position.

Sthiramati refers only briefly to this discussion75 and seems to accept the view that avijfiapti belongs to the category of matter. However, he does not accept the view of the Vaibhasikas that avijfiapti, in the same way as vijhapti, exists as a real entity (dravyd). At the same time he points out that this non-existence of avijfiapti as a separate entity does not result in the non-existence of the different kinds of restraint etc.

The underlying explanation for this statement is probably to be found in the Vaibhasikas’ objections to the claim that avijfiapti cannot really exist and the Sautrantikas’ defense of their view of avijfiapli not being a real entity, which are presented in detail by Vasubandhu in the Abhidharmakosabhcisva Sthiramati’s argumentation seems to rely at least partly on the concept of the “store mind” (cilavavijnand) and thus goes beyond Vasubandhu’s line of reasoning.

He, for instance, explains the concept of samvara to be an intention to restrain oneself from committing wrong deeds which is produced at the ordination ceremony and which leaves a seed in the alayavijnana and thus provides the source for future intentions of the same kind.78 The functions of asamvara as well as of the dhyana- and anasravasamvara are, according to Sthiramati, also to be explained as the continuation of intentions to engage in certain activities, like doing harm to other beings on the one hand and the restraining from committing misdeeds on the other.79

A question that seems to be controversial to Sthiramati is whether there are any other kinds of matter comparable to avijfiapti. The manner in which Sthiramati discusses this problem shows that the boundary between material and mental factors was disputed among the authors of Abhidharmic texts like the Pancaskandhakavibhasa, the Abhidharmasamuccava, and the Abhidharmakosabhasya. In the SahgTtisutra it is stated that there are three kinds of rupa: one that is visible (sanidarsana) and impenetrable (sapratigha), one that is invisible and impenetrable, and one that is invisible and penetrable.80 The sutra does not explain, however, what exactly is meant by these three classes.

The passage in question is quoted in the Abhidharmakosabhasya^ and Sthiramati seems to refer to it in the Pancaskandhakavibhasa when he explains that all matter is threefold. He identifies visible and impenetrable matter with the object of the sense of sight, invisible and impenetrable matter with the five sense faculties as well as the objects of the other four senses (apart from the sense of sight), and finally the invisible and penetrable matter with the matter of dharmayatana.S2 According to the Abhidharmasamuccaya, the matter of dharmayatana includes five entities: compressed (abhisamksepika) matter, matter of space (abhyavakasikd), matter of commitment (samadanika), imagined (parikalpita) matter, and matter produced by those with meditative power (vaibhutvika)?2

The commentary explains matter that is abhisamksepika as the matter of atoms (paramanu), whereas abhyavakasika is defined as referring to the matter of atoms being separated from other obstructing tangible [[[Wikipedia:matter|matter]]] (tadanyaprativarakasprastavvara-hita).M The term samadanika is explained as an alternative expression for the matter of avijfiapli, and parikalpita is defined as indicating matter of meditative images (pratibimba), as for instance the image of a skeleton (asthisamkalika).85 Vaibhutvika is explained as referring to objects of those who are absorbed in the [eight] liberations (vimoksadhyayigocara).S6 Sthiramati rejects four of hese categories as belonging to rupa and explains why he accepts only the avijnapti as invisible and penetrable matter.

According to his argument, the first two categories, the abhisamksepika and the cibhyavakasika, refer to matter of atoms and thus belong to the category of colour (which is part of the visible matter).87 The objects of the last two categories, the parikalpita and the vaibhiltvika, are nothing other than mental images and are therefore to be regarded as being part of the mind.88 Thus, in the case of invisible and penetrable matter Sthiramati explicitly rejects the teaching of live different entities given in the Abhidharmasamuccaya and follows the tradition of just one entity falling under this category of matter.

This tradition is also evident in the Abhidharmakosabhasya where it seems to be presented as the position of the Vaibhasikas, who state that there is no other invisible and penetrable rupa than avijnapti.89 It is notable that Vasubandhu mentions in this context the view of some Yogacaras who claim that an image that is perceived in contemplation is matter that is invisible and penetrable.90 However, the fivefold division of invisible and penetrable riipa as explained in the Abhidharmasamuccaya and the Pancaskandhakavibhasa is not mentioned in the Abhidharmakosabhasya.

Finally, it should be noted that the explanations regarding avijnapti provided in the Pancaskandhavivarana and the Pancaskandhabhasya resemble Sthiramati’s presentation closely, although the Vivarana is much shorter and does not explain the samvara concept in detail. Both commentaries do not mention the fivefold classification of invisible and penetrable matter and do not refer to the Abhidharmakosabhasyas discussion of the existential status of avijnapti.

The Pancaskandhabhasya only includes some remarks on the avijnapti’s dependence on the four basic elements.91 This topic is dealt with in the Abhidharmakosabhasya in some detail and is indicated by Sthiramati in one sentence, in which he says that avijnapti is “dependent matter since it arises based on the basic elements and complies with them”.92 The Pancaskandhabhasya, moreover, lists (as the only Pancaskandhaka commentary) reasons for giving up the various kinds of samvara as well as the asamvara:93

1. Four reasons for giving up the pratimoksasamvara: giving up the discipline, death, occurrence of the male or female organ, taking up false views (plus a fifth reason for

giving

up temporary discipline: the end of one day and one night)

2. Two reasons for giving up the bodhisattvasamvara: giving up the restraint, taking up false views

3. Four reasons for giving up the asamvara: taking up a restraint, attaining an insight into true reality or seeing the truth, death, occurrence of the male or female organ

4. One reason for giving up “neither restraint nor non-restraint”: giving up an undertaking

5. One reason for giving up the restraint of meditative absorption and the anasravasamvara:

emerging from meditative absorption

This list seems to be related to a parallel explanation in the Abhidharmakosabhasya, in which the causes for losing the different kinds of (non-)restraint (except for the restraint of the Bodhisattva) are discussed in detail.94 The four reasons for giving up the pratimoksa restraint are more or less identical in both texts, however, the Abhidharmakosabhasya lists “cutting off the roots of the beneficial” (kusalamulasamuccheda) instead of the taking up of wrong views. In connection with losing the asamvara the Abhidharmakosabhasya mentions only three reasons, omitting the insight into true reality, and with regard to

“neither restraint nor non-restraint ” six causes are discussed instead of one, the other five including cutting off the force of faith, cutting off an action, cutting off an object, death, and cutting off the roots of the beneficial. As for the last category, the Abhidharmakosabhasya additionally mentions in the context of giving up the restraint of contemplation the possibility of arising in a different level.

Notes

1. Parts of this paper have been previously published in the Nagoya Studies in Indian Culture and Buddhism. Sambhasa 27, 2008, pp. 149-171. They are presented here in a revised and extended form. I would like to thank Jens-Uwe Hartmann and Ralf Kramer for offering very helpful comments and corrections to previous drafts of this paper.

2. PSkV 3b3f. The author of the Pancaskandhabhayya also refers to this topic and notes that the relation between rupaskandha and rupa is comparable to that between heaps of barley or rice and barley and rice themselves (see PSkBh 34a5).

3. The same definition of rupa is found in the Abhidharmasamuccaya(bhasya (AS* 3,12f. and ASBh 3,3f.). In contrast, the Abhidharmakosabhasya describes rupa as consisting of the five sense faculties (indriya), their five objects (artha) and the avijnapti (AKBh 5,20-22). The four mahabhutas are dealt with in a passage that follows the rupaskandha section (AKBh 8,10ff.).

4. mV3b5-4al.

5. On the destructions, see Gethin 1997, 196f.

6. PSkViv 4a7f.

7. PSkBh 34b3-5.

8. PSkV 4a4f. and PSkViv 4b6f. See also AS* 13,6. On Sthiramati’s understanding of akasa, see Kramer 2012, 127f.

9. PSkBh 35alf.

10. PSkV4b2f. The same list appears in ASBh 3,4.

11. P.S'X.V4b4 See also ASBh 3,5f.

12. mV4b4. See also ASBh 3,7.

13. /’SAV 4b4f. See also ASBh 3,8f.

14. PSkV4b5. See also ASBh 3,6.

15. PSAV4b5f. See also ASBh 3,9f.

16. PSkV 5a3-5. Though the terminology found in the Tibetan translation of the Abhidharmasamuccaya is identical to the Tibetan rendering of the PSkV (see AS, 46a4f., PSk llb6f., and PSkVT 198bl-3: sra ba nyid, gsher ba nyid, tsha ba nyid, yang zhing g.yo ba mid), the terms given by Pradhan in his reconstruction of the Sanskrit text differ from those in the PSkV. kathinata, nisvandata, usvata. kampanata (see AS* 3,14-16). The respective terms found in ASBh 8,18 are khara, sneha, usnata, irana.

17. PSAV5blf.

18. PSAV5bl-5, PSkViv 5al-3, and PSkBh 36bl-5.

19. While this is what the Sanskrit text seems to mean (PSkV 5b3: -patradalapakosnata-), the text preserved in the Tibetan translation is as follows: padma'i 'dab ma rgyas pa (translated in Engle 2009, 252, as “the development of a lotus blossom”, and explained on p. 462, n. 58, as “the fact that a plant can develop in water”). See also PSkBh 36b3, where the argument is definitely aiming at the rotting of leaves and warm water.

20. In connection with the presence of the earth element in a burning fire the Pancaskandhabhasya also mentions the ability of flames to support leaves etc. (PSkBh 36b4f.).

21. PSkBh 36b3.

22. PSkBh 36a6-36bl. For the edition and translation of the parallel passage in the Garbhavakrantisutra, see Kritzer (forthcoming). The Tibetan text found in the extant versions of the Garbhavakrantisutra is not identical to the Pancaskandhabhasya’s reading and it is difficult to say if the author of the Pancaskandhabhasya has changed and abbreviated the text or if he was quoting from a different version of the stltra.

23. According to the Pancaskandhaka the matter dependent on the basic elements consists of the five sense faculties, their five objects, and the avijhapti (PSk 2,3-5). The explanation of AS* 3, 16-18 is very similar, though the last category, the avijhapti, is extended to “matter included in the dharmayatana". The answer to the question, what the five indriyas are, given in the Pancaskandhaka is: “pellucid matter having colour as its object”, "... having sound as its object”, etc. (PSk 2,6-10). The definition found in AS* 3,18f. differs somewhat. There it is said that the faculty of seeing is pellucid matter dependent on the four mahabhutas, which is the basis for visual perception (caksurvijhana). See also AKBh 5,25ff., where a similar explanation is given.

24. /'.SWOa.Sf.

25. I’SkV 6a6-b5. For further remarks on this passage in Sthiramati’s commentary and on the corresponding section in the Pancaskandhabhasya, see Schmithausen (forthcoming), § 52.

26. />S7<V7a6.

27. AS* 3,23-26 and AKBh 6,8 and 6,16f. The Abhidharmasamuccaya does not mention the expressions “colour” and “shape” explicitly in this context, but gives a list of different colours and shapes.

28. The four colours are blue (nila), yellow (pita), red (lohita), white (avadata) and the eight shapes include long (dirgha), short (hrasva), square (yrtta), round (parimandala), high (unnata), low (avanata), even (sata), uneven (visata). See PSkV7blf. and 5, AS* 3,24f., and AKBh 6,11.

29. See AS* 3,25f. (and AST 46b2) and also ASBh 3,14f. The Abhidharmasamuccaya states that all the different categories of visible matter can be of three kinds: beautiful (kha dog bzang po), not beautiful (kha dog ngan pa), or neither of these two (see AS* 4,1 and AST 46b3f.).

30. PSX.V7b2f.

31. PSkViv 6al and PSkBh 38b6-39al.

32. PSkV 8a6f. See also PSkBh 39a4-6 for examples of vijhapti “making known” a pure or a hateful mind.

33. See e.g. AKBh 195,7ff.

34. AKBh 192,20ff. and 196,If. For a more detailed description of the discussion of samsthana existing either as a real entity or as a mere designation (prajhaptisat), see Karunadasa 1967, 50-52.

35. AKBh 6,16f. Another example for rupa consisting of mere shape without colour exists in objects seen at a distance (see AKBh 195,121'.).

36. AKBh 195,161'.

37. PSkV 8a*4-7. (The scribe has erroneously omitted a part of the text, which was subsequently added on an additional folio. As both folios are marked as folio 8 in the manuscript, for the sake of clarity I am referring to the additional folio as 8*.)

38. PW8b4-6.

39. mV9a6f.

40. PSkV 9b4, AKBh 6,22, AS* 4,3, and ASBh 3,19f.

41. PSkV 9b5f., AKBh 6,231'., and ASBh 3,19. ASBh 3,20 has only vrksa as the sound that is not appropriated.

42. AS* 4,3 andm-2,13f.

43. ASBh 3,20 and PSkV9b6. See also PSkViv 6a7 and PSkBh 39b3.

44. AKBh 6,24-7,1.

45. See, e.g, Schmithausen 1987, 262, n. 101. Vasubandhu does not explicitly mention the example of the sound of a hand and a drum in the Pancaskandhaka, but he clearly accepts the position of a sound being upatta and anupatta simultaneously (see PSk 2,13f.).

46. AKBh 6,24.

47. AKBh 6,22f.

48. AS1* 4,3f. and ASBh 3,20-22. The siddhopanita and the parikalpita sounds are explained in ASBh 3,21 as sounds communicated by the Aryas (aryair desitah) on the one hand or by the non-Buddhists tirthyair desitah) on the other. The reconstruction of these two and the following two categories offered by Pradhan (see AS* 4,4: siddhopamto va parikalpito va dryair desito va tirthyair desito va) seems to be wrong when compared to the commentary found in the Abhidharmasamuccayabhasya and to the Tibetan translation of the Abhidharmasamuccaya (AST 46b5f.): grub pas bstan pa dangl kun brtags pa dangl 'phags pas tha snyad btagspa dang/ ’phagspa ma yin pas tha snyad btagspa'o. The phrases dryair desitah and tirthyair desitah do not describe additional categories, but are the respective definitions of the siddhopamta and the parikalpita sounds. The correct expression for the last two categories of sound is found in ASBh 3,21: drydnaryavyavaharikau.

49. AS* 4,2.

50. PSkBh 39bl-3 and PSkViv 6a7-6b2.

51. AKBh 7,5f.

52. PSA 3, If.

53. PSkV 9b6. My understanding of the phrase indriyamahabhutanam as “basic elements of the sense of smell]” follows Engle 2009, 259. Notably, the Pahcaskandhabhasya (PSkBh 39b6) explains smell “other than that” (de las gzhan pa) as “neither benefiting nor harming the five sense faculties and the four basic elements” (dbang po Inga dang ’byung ba chen po bzhi la phan par yang mi byed gnodpar yang mi byed pa ste).

54. PSkV lOalf.

55. AS* 4,5f. and ASBh 3,24f.

56. PSkViv 6b3f. and PSkBh 39b5-40al.

57. AKBh 7,4.

58. AS* 4,7-9.

59. PSkViv 6b4-6 and PSkBh 40a2-4.

60. ASr47al-3 (see also the reconstruction in AS* 4,10-12, where in addition usnatva is mentioned).

61. PSkV llb2f.

62. The Pancaskandhakavibhasa mentions only water as the main constituent of cold (PSkV llb4f.). However, the Pahcaskandhabhasya (PSkBh 41a4) and a number of other texts which include a parallel listing (e.g. PSkPra 242bl and AKBhTT tho 48b4) have chu dang rlung (“water and wind”) as the main basic elements of cold. A similar statement is also to be found in ASBh 3,27.

63. PSkV llb4f. and PSkBh 41a3f.

64. PSkV lla6f. See also PSkViv 7alf. and PSkBh 40b4-7.

65. PSk 3,8f.

66. AKBh 8,9. Sthiramati also classifies avijhapti as being kusala or akusala (PSkV 12al). However, the two terms are missing in the Tibetan translation of the Pancaskandhakavibhasa (PSkV 203a7). In contrast to avijhapti, which can never be neutral (avydkrta) (see AKBh 200,25), vijhapti might be kusala, akusala, or avydkrta (see AKBh 201,2 and PSkV 12a2). The reason for this classification of avjhapti as either morally good or bad might be explained by the fact that avijhapti was introduced to justify the karmic results of actions that cannot be perceived directly in opposition to the visible actions of body and speech (i.e. kaya- and vagvijhapti). Therefore an avijhapti not having karmic consequences would be ineffective and purposeless.

67. See AS1* 4,13 and also ASBh 4,4, where the category samadcinika is defined as avijhapti.

68. PSkV 12a2-6. See also A .S'* 58,8f. and ASBh 69,7f. “Neither restraint nor non-restraint” refers to good or bad activities of those who are not committed to a continuous beneficial or unbeneficial way of living like monks etc. on the one hand and those who kill animals as part of their profession (e.g. hunters) on the other. See Sanderson 1994, 39f.

69. AKBh 205,12f.

70. See AS* 57,3.

71. AKBh 205,15 and AS* 57,4f.

72. PSkV 12a6f.

73. PSkV 12blf. In AKBh 201,8-11 Vasubandhu defends the view that avijhapti can only be produced in kama- and rupadhatu, not in arupyadhatu. The arising of avijhapti is impossible in the sphere without matter, as avijhapti is dependent on the mahabhutas, which, of course, do not exist there. Vasubandhu rejects the opponent’s assumption that it should be possible to produce avijhapti in arupyadhatu just as uncontaminated avijhapti is produced by someone existing in the rupadhatu. In contrast to uncontaminated avijhapti, which does not fall under the division of the three dhatus, an avijhapti belonging to arupyadhatu could not be produced dependent on elements which belong to another sphere.

74. See AKBh 9,18-10,5 and 196,411.

75. See PSkV 13a5-14a2.

76. PSkV 13bl.

77. AKBh 196,81'1. and 197,311.

78. PSkV 13b2-4.

79. PSkV 13b4-14a2.

80. See SahSu III.23.

81. AKBh 196,81

82. PSkV 12b3-5.

83. AS* 4,12-14, ASBh 4,3-5, and PSkV 12b5.

84. See also Engle 2009, 264, who translates the explanation of abhyavakasika as “they are intangible in the sense that they do not obstruct anything”. It is difficult to discern the subtle difference between the atomic matter of the category abhisarriksepika and that of abhyavakasika. The first kind of matter possibly refers to the ordinary atoms that constitute the material objects, whereas the second relates to the (dark or light) matter of holes, like the opening of a door or the mouth. See AKBh 18,9-17, where akasadhatu is defined as the material space of the opening of doors etc.

85. PSkV 13a2.

86. ASBh 4,3-5 and PSkV 12b6-13a2. On the eight vimoksas, see Kramer 2005, 1461

87. PSkV 12b6-13a2.

88. PSkV 13a2-4.

89. See AKBh 196,12.

90. AKBh 197,4-6.

91. PSkBh 42b2-4 and 43blf.

92. AKBh 199,161'1 and PSkV 12b2.

93. PSkBh 42b4-7.

94. AKBh 222,17-225,18.

Bibliography

Primary sources

AKBh Abhidharmakosabhasya (Vasubandhu), ed. by P Pradhan, Patna: K. P Jayaswal

Research Institute, 1967.

AKBhTT Abhidharmakosabhasyattka Tattvartha (Sthiramati), Tibetan translation, Derge

4421.

AS* Abhidharmasamuccaya (Asanga), edition and reconstruction by P. Pradhan,

Santiniketan: Visva-Bharati, 1950.

ASt Abhidharmasamuccaya (Asanga), Tibetan translation, Derge 4049.

ASBh Abhidharmasamuccayabhasya (Buddhasimha or Jinaputra), ed. by N. Tatia, Patna:

K. P. Jayaswal Research Institute, 1976.

PSk Pahcaskandhaka (Vasubandhu), ed. by X. Li and E. Steinkellner, Beijing /Vienna:

China Tibetology Publishing House Austrian Academy of Sciences Press, 2008.

PSkT Pahcaskandhaka (Vasubandhu), Tibetan translation, Derge 4059.

PSkBh Pancaskandhabhasya (author unknown, probably not *Prthivfbandhu), Derge 4068.

PSkPra Pancaskandhaprakarana (Candrakirti), Tibetan translation, Derge 3866.

PSkV Pancaskandhakavibhasa (Sthiramati), ed. by J. Kramer, Beijing Vienna: China

Tibetology Publishing House Austrian Academy of Sciences Press (forthcoming).

PSkViv Pancaskandhavivarana (Gunaprabha), Derge 4067.

PSkVT Pancaskandhakavibhasa (Sthiramati), Tibetan translation, Derge 4066.

SariSu SahgTtisutra, ed. by V. Stache-Rosen, Berlin: Akademie-Verlag, 1968.

Secondary sources

Engle, Artemus B. (2009). The Inner Science of Buddhist Practice: Vasubandhu's Summary of the Five Heaps with Commentary by Sthiramati. Ithaca: Snow Lion Publications.

Gethin, Rupert (1997). “Cosmology and Meditation: from the Agganna-Sutta to the Mahayana.” History of Religions 36 (3), pp. 183-217.

Karunadasa, Y. (1967). Buddhist Analysis of Matter Colombo: Department of Cultural Affairs.

Kramer, Jowita (2005). Kategorien der Wirklichkeit im friihen Yogacara. Wiesbaden: Dr. Ludwig Reichert Verlag.

(2012). “Descriptions of ‘Feeling’ (Vedana), ‘Ideation’ (Samjha), and ‘the Unconditioned’ (Asamskrta) in Vasubandhu’s Pahcaskandhaka and Sthiramati’s Pancaskandhakavibhasa" Rocznik Orientalistycznv 65 (1), pp. 120-139.

Kritzer, Robert (forthcoming). Garbhavakrantisutra: The Sutra on Entrv into the Womb. Tokyo: International Institute for Buddhist Studies.

Sanderson, Alexis (1994). “The Sarvastivada and Its Critics: Anatmavada and the Theory of Karma.” In: Buddhism into the Year 2000, Bangkok: Dhammakaya Foundation, pp. 33-48.

Schmithausen, Lambert (1987). Alayavijhana. On the Origin and the Early Development of a Central Concept of Yogacara Philosophy. Tokyo: The International Institute for Buddhist Studies.

(forthcoming). The Genesis of Yogacara-Vijhanavada: Responses and Reflections. Tokyo: