Causality:
The Central Philosophy of Buddhism D avid J. Kalupahana FOREW ORD BY ELIOT DEUTSCH
THE UNIVERSITY PRESS OF HAWAII Honolulu

Contents

Foreword	ix
Preface	xi
Acknowledgments	xvii
I.	Pre-Buddhist Theories of Causation : The Vedic Tradition	1
II.	Pre-Buddhist Theories of Causation : The Non-Vedic Tradition	23
III.	Clarification of Terminology	54
IV.	The Conception of Dharma	67
V.	The Causal Principle and Its Validity	89
VI.	The Causal Explanation of Existence	110
VII.	Later Developments	147
VIII. Causal Correlations : Another Facet of Development	163
IX.	Conclusion	177
Abbreviations	186
Notes	189
Bibliography	228
Index of Chinese Terms	240
General Index	253

Foreword

D. J. KALUPAHANA, chairman of the Department of Philosophy at the University of Hawaii, has carried out, in the tradition of his teacher K. N. Jayatilleke, a masterful articulation, analysis, and interpretation of the doctrines of causation in Buddhist philosophy. Special attention is given to early Buddhist teachings as found in the Pali Nikayas and Chinese Agamas; and by working with both Pah and Chinese sources Kalupahana has broadened considerably the foundations of scholarship in early Buddhist philosophy.

In early Buddhism, Kalupahana maintains, a cause is defined as "the sum total of several factors that gives rise to a consequent' '— the "consequent" being the entire universe as well as a specific thing or event. As with the Greeks (Aristotle in particular), for whom the question of causality was not so much how one thing
(efficiently) causes another, but how one thing can become something different from itself, we have in Buddhist thought a concern to account for 'development,' 'process,' 'change' in the whole of our experience. For Buddhism, causality is at once a problem in metaphysics and epistemology, and its resolution in these areas leads to a variety of ethical considerations.

In recent years Western interest in Buddhism has been directed mainly to Mahayana traditions, principally to the Madhyamika school of Nagaijuna and to Zen. Professor Kalupahana is to be warmly commended for turning our attention back to the philosophical riches of the early schools, where, together with profound spiritual concerns, a good deal of sharp philosophical analysis is to be found. The author offers us as well a comprehensive historical background to Buddhist 'phenomenalism' and unravels many of the complexities in the schools associated with the so-called Theravada tradition and in the later Mahayana developments.

One of the most interesting analyses Kalupahana offers has to do with whether the doctrine of 'dependent origination,' or paticcasamuppada (Skt. pratTtyasamutpada), implies determinism.

His manner of treating this question shows how nicely he refuses to accept any interpretation of Buddhism that does violence to the original texts for the sake of satisfying one's philosophical preconceptions or predilections.

Professor Kalupahana faces directly the many difficulties in the Buddhist doctrines and presents a hard-headed, no-nonsense, sympathetic but not apologetic, analysis. His work should deepen considerably philosophical interest in Buddhism and appreciation for the distinctive genius of the many extraordinary thinkers associated with it.

E l io t D e u t s c h STUDIES ON THE PHILOSOPHY of early Buddhism have so far been confined mostly to the material available in the Pali Nikayas, which represent only one of the early Buddhist traditions. Some Japanese scholars, such as Ui Hakuji and Akanuma Chizen, have examined the teachings embodied in the Chinese Agamas.

Unfortunately, these treatises are not accessible to the present author because he does not know the Japanese language. Akanuma Chizen as well as Masaharu Anesaki have rendered a great service to those interested in comparative studies of the Nikayas and the Agamas by compiling catalogues of the sutras in these two bodies of literature. The present work was undertaken to compare the teachings on the problem of causation in the Pali Nikayas and the Chinese Agamas. The importance of the Chinese Agamas for the study of early Buddhist thought is twofold. First, they supply corroborative evidence for some of the major concepts in the Pali Nikayas, whose authenticity has been questioned by many scholars in recent times. Secondly, they throw much light on some of the obscure concepts in the Nikayas.

Indian thinkers before and during the time of the Buddha put forward a wide variety of views regarding the problem of change and causality. Buddhist theory is a product of criticism, assimilation, and synthesis of these ideas. Most of the earlier theories are examined in the early Buddhist texts, namely, the Pali Nikayas and the Chinese Agamas. In chapter 1 an attempt is made to trace the gradual development of causal theories in the Vedic tradition to establish the historical basis of some of the theories mentioned in the Pali Nikayas and the Chinese Agamas. Here the evidence from the Agamas is mainly corroborative. An examination of philosophical sections of the Vedas, Brahmanas, Aranyakas, and the Upanisads has led to the conclusion that the theories of self-causation (sat-karya) and creation by God (Tsvara-nirmana) were two of the major causal theories in the Vedic tradition. These were referred to in the early Buddhist texts as say am katam (tsu tsao) and issaranimmana (tsun yu tsao), respectively. Buddhist criticism of these theories appearing in the Pali Nikayas and the Chinese Agamas is also examined.

Chapter 2 is devoted to an analysis of pre-Buddhist ideas mainly in the non-Vedic tradition. The causal theories of the Materialists, the AjTvikas, and the Jainas are discussed in detail, especially because of their possible influence on the Buddhist theory of causation. Here the evidence gleaned from the Chinese Agamas is of immense value in understanding some obscure concepts, such as niyatisahgatibhava of the AjTvikas. The close similarity between the theories of moral causation put forward in Jainism and Buddhism has confused some scholars who have written on this subject, so the Jaina standpoint is discussed at length. With the help of the commentaries of Silanka it is possible to determine the relationship between the two schools of thought with some precision. It is pointed out how the philosophical theory of causation formulated in Jainism led to the acceptance of a deterministic theory of moral causation. The influence of some Ajlvika theories on Jainism is not overlooked. Apart from the doctrines of these major schools of thought, we examine several other theories that are mentioned in the early Buddhist texts.

Chapter 3 is an attempt to elucidate the meaning of some of the terms expressing causation in the early Buddhist texts. The different views, classical as well as modern, regarding the use of the terms hetu (yin) and pratyaya (yuan) are discussed. All the evidence from the Pali Nikayas and the Chinese Agamas indicates that during the earliest phase of Buddhism the two terms were used synonymously and that they did not express any distinction comparable to the distinction between 'cause' and 'subsidiary condition.' Doctrinal as well as textual evidence suggests that that distinction originated with the Sarvástivadins.

A discussion of the causal principle involves an examination of the nature of things that are connected by the causal principle. Chapter 4 is therefore devoted to an analysis of the nature of causally conditioned dharma or dhamma. There are many important disquisitions on this subject by modern scholars. Outstanding among them are (i) The Central Conception o f Buddhism and the Meaning o f the Term 1 Dharma! by T. I. Stcherbatsky, which is based primarily on the source material in the Abhidharmakosa, and (ii) Pali Dhamma by Wilhelm and Magdalene Geiger, who examined almost every reference to the term dhamma in Pali literature, canonical as well as commentarial, to bring forth the wide variety of meanings the term connotes. A fresh look at this material was found to be necessary especially in the light of the information supplied by the Chinese Agamas. The conception of dhamma (fa) appearing in the Pali Nikáyas and the Chinese Agamas was found to differ from the conceptions of dharma in some of the major schools of thought, such as the Sarvástivada, the Sautrantika, and the later Theraváda. Some scholars have minimized the difference between the teachings of early Buddhism as embodied in the Pali Nikayas and the Chinese Agamas on the one hand and those of the later schools on the other;1 we endeavor to show that the difference is far too great to be ignored. For example, the acceptance of a logical theory of momentariness (ksanikavada) instead of an empirical view of change and impermanence (anitya) gave rise to a host of problems, and their solution created significant differences not only among the later schools but also between these and early Buddhism. The Sarvástivadins, it is pointed out, accepted an eternal underlying substratum (dravya, svabhava) in things (dharma). It was this fundamental conception of Sarvástivada, a school that gained prominence in India after the third century B.C., that was the target of Nágárjuna's dialectic. Thus, with much justification, the Maháyánists attributed a theory of pudgala-nairatmya (nonsubstantiality of the individual) but not dharma-nairatmya (nonsubstantiality of the elements) to the Sarvástivadins and claimed superiority over them for formulating the latter conception. Since the Sarvástivada was then the most prominent of the Hïnayâna schools, the Mahâyânists, without any justification, extended their criticism to all the Hïnayâna schools.

Moreover, with the emergence of different Buddhist schools and the compilation of the Abhidharma literature and ancillary works, the study of the early sutra literature was relegated to the background. As a result, the Abhidharma came to be looked upon as the primary source for the study of early Buddhism. The Sautrântika school, claiming to base its doctrine on the source material in the sutras (sütrânta), may have emerged in reaction to giving such priority to the Abhidharma. But the Sautrântikas, too, in accepting the theory of momentariness, moved away from the standpoint of early Buddhism.

The view that the Sarvâstivàda school represents the earliest phase of Buddhism seems to have prevailed in the minds of the compilers of the Sandhinirmocana-sûîra (7.30) as well as historians such as Bu-ston.2 Taking the various phases of Buddhism as represented by the schools in India during their own time, these Mahâyâna scholars formulated the conception of tricakra-parivartana,
"the three swingings of the Wheel of the Law." That theory completely ignores the sutra literature, which, as we point out, is if not more exalted at least not much different from the ideas expressed and the critical attitude adopted by Nâgârjuna and his followers.

Stcherbatsky's theory that the earliest form of Buddhism was a Radical Pluralism that eventually led to Monism and finally to Idealism was contested by Schayer, who held the view that Pluralism, Monism, and Idealism were parallel currents of thought in early Buddhism.3 It may be possible to trace the germs of Pluralism, Monism, and Idealism in the Buddhist canon in the same way that one can see different trends of thought in the Upanisads. But the Pali Nikâyas and the Chinese Agamas, which are primarily based on the speculations of one individual, unlike the Upanisads, should lend themselves to a single interpretation. Our examination of the conception of dhamma (fa) in the Nikâyas and the Agamas leads us to conclude that the philosophical standpoint of these early texts represents a form of 'phenomenalism.'
Chapter 5 explains the causal principle in Buddhism. The first part of the chapter examines the nature of the causal nexus, and the second part assesses the validity of the causal principle. The status of the causal principle in Buddhism is discussed in relation to some of the Western schools of thought that are skeptical of causation. The chapter concludes with a discussion of how the causal principle is verified.

Chapter 6 elucidates the various problems that are given causal explanations in Buddhism. The operation of the principle of causation in the spheres of inorganic phenomena, organic life, psychology, social and moral fife, and spiritual life are examined in detail.

The detailed treatment is necessary to eliminate the erroneous belief, created by the writings of some modem critics of Buddhism, that the idea of causation in Buddhism is confined to the twelvefold formula.

A comparison of the theory of causation in the Nikayas and the Agamas with later developments (chapter 7) has convinced us that the view of some scholars that Madhyamika philosophy represents a " Copemican revolution" in Buddhism is unfounded. The Madhyamika philosophers did develop a critical attitude in philosophy as a result of the metaphysical theories propounded by the later Buddhist schools; but the standpoints adopted by the Buddha and the Madhyamika philosophers to avoid such metaphysics are radically different. The concluding chapter, in which we try to interpret the Buddha's silence on metaphysical questions, was written to clarify the difference between these two standpoints. Chapter 8 includes a comparison of the theory of causal correlation (pratyaya) of the Theravada as enunciated in the Patthana with the theories of the Sarvastivada and Yogacara. Correspondences among these theories are noted.

Our final conclusion is that the Buddha rejected contemporary metaphysical speculations, giving instead a scientific explanation of the phenomenal world, and that without getting unduly engrossed in or obsessed by the mystical aspect of the current religious traditions, he utilized that mystical knowledge and experience to achieve freedom (vimutti, chiai i'd) from the trammels of samsaric existence.

Acknowledgments

THE PRESENT W ORK is an outgrowth of the Ph.D. thesis that I submitted to the University of London in 1966. I am greatly indebted to Professors Stanley Weinstein (currently at Yale) and D. L. Friedmann, of the University of London, under whose watchful eyes this work was undertaken and' completed. Thanks are also due Dr. K. P. K. Whitaker and Mr. G. F. Weys, of the same institution, for helping me in various ways, and the latter for giving me my first lessons in classical Chinese. My teacher and later colleague, Professor N. A. Jayawickrema, at the University of Ceylon, has always been a source of inspiration and encouragement and was solely responsible for obtaining leave of absence for me to complete this project. The late Professor K. N. Jayatilleke read through the entire original draft and made many useful suggestions. The untimely death of this distinguished scholar is greatly felt by those interested in Buddhist scholarship. I am grateful to my friend and colleague, Professor Eliot Deutsch, for writing the Foreword. Professor Dennis Twichett of the University of London (now of Cambridge University) and Mr. W. Zwalf of the British Museum have helped me in innumerable ways during the period when this research was carried out. Thanks are due Mrs. Iris M. Wiley of The University Press of Hawaii for the interest she has shown in this work, and to Mrs. Floris E. Sakamoto for her assistance in typing the manuscript. The completion of this research project was made possible by a generous grant from the British Council. My wife Indrani deserves a word of praise for the patience with which she has tolerated the time spent on this work.

Auspicious are the Sun's bay-coloured horses, bright, changing hues, meet for our shouts of triumph. Bearing our prayers, the sky's ridge have they mounted, and in a moment speed round earth and heaven.

This is the godhead, this the might of Surya, he hath withdrawn what spread o'er work unfinished.

When he hath loosed his horses from their station, straight over all Night spreadeth out her garment.3 This conception of a purely physical order appears to have arisen as an inductive inference based on the repeated experience of such phenomena as the daily rising and setting of the sun. As the number of experiences of such phenomena increased, the probability of their repetition approached certainty, and man began to believe in dependable regularities in the external world.

The regularity with which these phenomena function was understood by the Vedic Aryans to be due to the greatness or divinity of the gods.4 Thus, order prevails in the world because of the nature of the gods. After experiencing a uniformity a certain number of times, Vedic Aryans came to expect that it would be repeated on future occasions; as a result, they formed expectations about many phenomena, which are expressed in their hymns.5 One such expectation is found in the hymn to Paijanya.6 The wind blows forth, the lightning falls; the plants shoot up; heaven overflows. Nature is bom when Parjanya quickens the earth with seed.

Give us, O Maruts, the rain of heaven; pour forth the streams of your stallion. Hither with this thunder come, pouring down the waters as the divine spirit our father.

Bellow towards us; thunder; deposit the germ; fly around with thy water-bearing car. Draw well thy water skin unfastened downwards; let the heights and valleys be level.

Thou hast shed rain; now wholly cease; thou hast made the deserts passable again. Thou hast made the plants to grow for the sake of food; and thou hast found a hymn of praise from [thy] creatures.7 The passage above, among a host of others, exemplifies a belief in the uniformity of nature based on a primitive conception of causation. It expresses the idea that rain causes the plants to shoot up; rain has made the heights and valleys level; and rain has made deserts passable again. These three assertions are good examples of the primitive notion of cause. Three features stand out in them: firstly, the assimilation of causation to agency; secondly, the relation of cause and effect as being one of production; and thirdly, the effect as being regarded as relatively passive.8 These features characterize the primitive as well as the common-sense notion of cause that derives from our immediate experience, such as the moving of our limbs.9 Our own efficiency is measured by our volitional actions and consciousness of effort, and it is thus natural for us to interpret external events by ascribing to them actions and volitions that we experience ourselves.

The Vedic Aryans conceived of physical phenomena in the same manner, introjecting their own experiences into external objects. They tried to understand the working of the forces of nature by positing inner wills or agents in them. This comes very close to the activity view of causation,10 except that the Vedic conception is more anthropomorphic than the activity or the common-sense views of causation.11 In the next stage, the regularity observed in the functioning of phenomena was considered to be an unalterable law. This is the conception of rta, or cosmic order, that we come across in the Rgveda.

It is interesting to note that during the time polytheism was in vogue this universal or cosmic order was considered to be independent of the gods. The gods themselves were said to follow the laws of rta.12 Heaven and earth are what they are by reason of rta.13 The whole universe is founded on rta and moves in it. It appears that it was during this stage that Varuna came to be considered the guardian (gopa) of rta ;14 in another context, not only Varuna but all the gods are looked upon as the custodians of rta.15 During the period of transition from polytheism to monotheism, and during the period of monotheism itself, this natural order came to be regarded as the creation of the gods M itra and Varuna. It was believed that M itra and Varuna established rta by means of sacrifice.16 During the monotheistic stage, when Varuna alone attained the position of "the lord of all, including the gods," 17 the natural law became his will, which is unhesitatingly followed by all the gods.18 At this stage, two other conceptions of rta, or law, appear to have come into vogue.19 One is rta as the moral order, and the other is rta as a sacrificial order. The first was a natural development, for we can see how Varuna, the god who established the cosmic law, was also looked upon as the righteous ruler of the world, the dispenser of justice.20 The consciousness of right and wrong issuing out of a belief in the prevalence of a moral order is clearly expressed in the hymns addressed to V aruna.21 The association of rta with the moral order of the universe was developed further when rta came to be considered as the truth in the world and anrta as falsehood. However, Heinrich Liiders, who made an exhaustive study of the conception of Varuna and rta, maintains that in the Vedas rta is never an adjective but always a noun and has only one meaning, 'truth.'22 The conception of rta as a sacrificial order would have arisen only after the originally simple sacrifices of the Vedic Aryans had developed into an elaborate institution under the priestly architects. With the development of monotheism and the gradual emphasis of sacrifices as the cause of the origin of the world,23 rta, in the form of the sacrifice, became the eternal law governing the universe. The smooth, orderly functioning of nature was ensured by the sacrifice, and thus sacrificial acts were looked upon as rta,24 In the Atharvaveda, however, the order of the universe was considered to be more of a magical character. "Order (rta), truth (satya), creative fervor (tapas), sovereignty, asceticism, law and works; past, future, strength and prosperity are in the ucchista." This ucchista, which was considered the foundation or the basis of the universal order, was thought of as a 'force of force.'25 Ucchista,
or the remnants of the sacrificial food, was thus considered to have the magical power of determining everything in the world.

In the Brahmanas, too, the order of the universe was mechanical but magical.26 Rta was identified with Brahman (Brhaspati), the creator of the universe.27 This appears to be a continuation of the conception of order found during the stage of monotheism in the Rgveda. Rta, which denoted merely the regularity of events or the natural law was looked upon by the Brahman priests as no more than the customary mode of divine action as exemplified in the sacrifice; hence its magical character.

This, in brief, is the gradual development of the conception of the uniformity of nature or the natural law. During the earliest period, belief in a natural law was purely a product of observation and inference. It was taken for granted by the later speculative thinkers, whose main preoccupation was to explain how this universe, in which orderliness prevails, originated and developed to its present form. This speculative thought, starting as early as the time of the tenth book (mandala) of the Rgveda, continued unabated during the periods of the Brahmanas, Aranyakas, and the Upanisads, giving rise to a multiplicity of conflicting theories, most of which are referred to in the early Buddhist texts. An examination of these various theories will throw much light on the Buddhist theory of causality and enable a proper appreciation and evaluation of it, for they form the context in which the Buddha preached his doctrine of causality.

Four Pre-Buddhist Causal Theories

The theories mentioned in the Pah Nikayas and the Chinese Agamas may be classified into four main types as follows:

1. Self-causation (sayam katam, attakatam,2B tsu tsao, 29 tsu iso30), a metaphysical theory that was intimately connected with two concepts, the concept of self or soul (atman, P. atta) and the conception of evolution. In the Upanisadic system, it was closely associated with the concept of atman, both in the theories of evolution and in the theories of creation. The basic assumption of this metaphysical postulate was that the cause and the effect are identical in essence.
2. External causation {param katam,31 t 'a tsao, f a tso32),
   which includes several different theories. Silanka, the Jaina commentator, defining the term annakadam {anyakrtam, P. parakatam)
   occurring in the Sutrakrtahga,33 lists the following theories under this category:

• Time (kala). - Creation by God (Tsvarakrtaka).
• Inherent nature (svabhava).
• Action or behavior {karma). - Fate {niyati).

1. Internal as well as external causation {sayam katah ca param katah ca,34- tsu tso t 'a tso3s), which is a combination of the first two theories.
2. Neither internal nor external causation {asayamkaram aparamkaram,36 fe i tsu fe i t'a tso )31 therefore indeterminism (adhiccasamuppanna,38 wuyin ts o 39 yadrccha40). It denies any form of causation.

These four types of theories may be broadly divided into Vedic and non-Vedic. The theories of self-causation and of divine creation included under external causation are found in the mainstream of the Yedic tradition. The rest may be considered non-Vedic because they were developed by schools that were generally opposed to the Vedic tradition although, of course, the germs of these theories are found in the Vedic tradition itself.

Theory Of Self-Causation

The theory of self-causation can be traced to the time of the tenth book (mandala) of the Rgveda. A very important feature of the philosophical hymns of the Rgveda is that they partake of ideas of mechanical as well as creative evolution. In their attempt to trace the origin and gradual development of the universe into its present form, the Vedic thinkers posited various primeval substances such as water and abstract principles such as "year"
(samvatsara) and then explained the universe as a product of the gradual evolution of these original substances. The conception of evolution is one of self-causation or self-origination, where one phenomenon gives rise to or produces another phenomenon by its own inherent power (svadhaya)41 in an orderly sequence.

According to Aghamarsana, who is considered the first philosopher in India,42 warmth (tapas) is the first creative principle, and in it originated law (rta) and truth (satya). These, in turn, produced darkness (tamas), and from darkness was produced water
(apas). Water gave rise to the year {samvatsara) or the time principle, and the year produced in due course the sun and the moon, the heaven and the earth, the firmament and the light, and ordained the days and nights.43 In a similar fashion, Prajapati Paramestin, who is called the Thales of India,44 advanced a theory of natural evolution based on water (salila) as the primeval substance. From water sprang, in due course, everything in the universe, animate as well as inanimate.45 According to these theories, chance has no place in the evolution of the world. The principle of movement or development is inherent in matter itself, and the world evolves from the immanent energy (svadha) of nature; the movement as a whole is self-determined. Herein we also find the germs of the theory of Natural Determinism (svabhavavada), which later developed into a systematic philosophy of nature under the Materialists.

In some of the Vedic hymns and especially in the Brahmanas, this concept of immanent or inherent energy {svadha) in phenomena was given a theological twist. There was a tendency to identify the various principles suggested by the earlier thinkers as the ultimate world-ground, with the personal God, Prajapati,46 considered to be the creator of the universe.47 Thus, theories of evolution came to be associated with theories of creation, giving rise to a form of pantheism. It was often said that Prajapati created the world out of himself, the inherent power of phenomena to evolve thus being considered the power of Prajapati. Prajapati is not merely a creator of beings but part and parcel of them, for he reduces them to order from their chaos by entering them with form (rupa) and name
(naman).48 The Satapatha Brahmana says: "Verily, in the beginning, Prajapati alone was here. He desired, 'May I exist, may I be generated.' He wearied himself and performed fervid devotions: from him thus wearied and heated, the three worlds were createdthe earth, the air and the sky."49 It is interesting to note that this idea of self-causation was used to explain the origin of a sacrificial oblation, the graha (a substance such as soma or ghee that is held in a sacrificial spoon or ladle). A passage in the Satapatha Brahmana runs th u s: " He [the Adhvaryu] offers [the graha] with [the words], 'Self-made thou art (svamkrto'si),' for, this graha being his [ Yajha's]
out-breathing, it is indeed made by itself, born of itself. Hence he says, 'Self-made thou art'—for all powers, divine and earthly—for it is born of itself for all creatures." 50 Moreover, the epithet svayambhu (" self-originating"), applied to the creator God in many places in the Brahmanas,51 suggests that self-causation was widely accepted at this time.

The idea of self-causation appears in a more refined form in the philosophy of Mahldasa Aitareya in the Aitareya Aranyaka.S2 There the problem of causation is presented with the problem of change. M ahldasa appears to have been aware of the problem presented by the 'unceasing mutability of existence,' which is considered one of the earliest as well as one of the most persistent problems in all philosophy.53 He seems to have conceived the idea that within the unity of one thing there is a succession of different states. He believed that all change and diversity in the world have an immutable ground of unity. This is implied in his statement:
"That body into which goes the indestructible [the breath] which we have joined [in meditation], proceeding from the indestructible
[the highest Brahman], that body which the harnessed horses [the senses] draw about, that body where the true of the true follows after—in that body all gods become one." 54 Instead of the conception of change enunciated in some of the Upanisads, change that is merely an illusion of our deceptive senses because it is incompatible with a permanent reality, MahTdasa conceived of change as the transformation of a single bodily reality, transition from the potential to the actual, a concept that is quite similar to the Sankhyan view. This conception of change determined to a great extent the theory of causation in the philosophy of Mahldasa that is set forth in the following passage.

Then comes the creation of the seed. The seed of Prajapati is the gods; the seed of the gods is rain; the seed of rain is herbs; the seed of herbs is the food; the seed of food is the living creatures; the seed of living creatures is the heart; the seed of the heart is the mind; the seed of the mind is the speech; the seed of speech is action; the act done is this man, the abode of Brahman." 55 The use of the term "seed" (retas) is very significant in that it affords a clue to the meaning of causation in the philosophy of MahTdasa. Just as a sprout is produced by a seed, so Prajapati produced the gods; the gods in turn produce rain, and so on. Thus, a chain of causation is established, each link in the chain producing the one following. Speculating on the origin of the universe, Mahldasa says:
Was it water? Was it water? This world was water. This was the root (mula); that the shoot (tula). This the father; those the sons. Whatever there is of the son's, that is the father's; whatever of the father's, that is the son's." 56 This statement further illustrates the connection between two links in the chain of causation. Sayana maintains that there is unity or oneness between the cause and the effect and that there is no complete division or distinction between the two, as between clay and a ja r made of clay.57 It is interesting to note that this is the first reference to the division of the causal process into two compartments, cause and effect. According to M ahldasa's view, the cause
(the root, mula) passes its characteristics on to the effect (the shoot, tula), just as a father bestows some of his characteristics, physical as well as mental, on the son, or as the son inherits the characteristics of the father.

Mahldasa attempted to explain the nature of causality in accordance with his theory of evolution and change. The theory of self-evolution or self-causation formulated by Aghamarsana finds detailed and more systematic treatment in the philosophy of Mahldasa. Yet his analysis is so tempered by speculative metaphysics that he shows greater zeal for First and Final Causes than for a rational explanation of things and their interrelation.

During the time of the Upanisads, greater emphasis was laid not so much on the theory of change, which would have given rise to a theory of causality, but on the theory of permanence. Almost all the thinkers of this period concentrated on demonstrating and proving the permanence of the self (atman), the reality underlying the phenomenal world. Thus, the Svet&svatara Upanisad says:
"That eternal should be known as present in the self (atmasamstha).

Truly there is nothing higher than that to be known. When one recognizes the enjoyer, the object of enjoyment, and the universal Actuator, all has been said. This is the threefold Brahma." 58 The keynote of all the Upanisads, in fact, was the immutable or imperishable nature of Brahman. The Katha Upanisad says: "The wise one (atman) neither is bom nor dies. It has not come from anywhere, has not become anyone. Unborn, constant, eternal, primeval, it is not slain when the body is slain." 59 The Upanisadic thinkers recognized two forms of knowledge, the higher (para) and the lower (apara), and emphasis was laid on the higher form of knowledge, understanding the imperishable Brahman.60 From these few instances, it is clear that the Upanisadic thinkers aimed foremost at understanding the 'Absolute' (Brahman or Atman), which is celebrated as the 'imperishable' (aksara).61 Because that aim was incompatible with investigating the nature of causality, the Upanisadic contribution to the theory of causality was a negligible one. Upanisadic thinkers only systematized some of the theories inherited from the earlier period. These systematizations are found in all theories of evolution, which to some extent make use of the notion of causality. In these atman is considered the chronologically antecedent cause and the manifold universe, the effect.62 Yet these come under the category of metaphysical speculation because they represent investigations into First Causes63 rather than empirical causal events.

The theories of evolution in the Upanisads seem to follow a pattern similar to the theories of self-causation enunciated by Aghamarsana and elaborated by MahTdasa. One of the most important theories of evolution is that of Uddalaka ArunT, set forth in the following passage from the Chandogya Upanisad.
In the beginning, . . . this world was just Being (sat), one only, without a second. To be sure, some people say: "In the beginning this world was just non-Being (asat), one only, without a second; from that non-Being Being was produced." But verily, . . . whence could this be? said he. How from non-Being could Being be produced? On the contrary, . . . in the beginning this world was just Being, one only, without a second. It bethought itself: "Would that I were many! Let me procreate myself!" It emitted heat. That heat bethought itself: "Would that I were many! Let me procreate myself!" It emitted water. . . . That water bethought itself:
"Would that I were many! Let me procreate myself!" It emitted food. Therefore, whenever it rains, there is abundant food. So food for eating is produced just from water.64 Uddalaka then explains how beings are produced from elements that evolve in this manner, and later enumerates this process of evolution in reverse order.

On this point, . . . understand that this [body] is a sprout (suhga) that has sprung up. It is not without a root (mula). What else could its root be than food? Even so, . . . with food as a sprout, look for water as the root. With water
. . . as a sprout, look for heat as the root. With heat… as the sprout, look for Being as the root. All creatures here . . . have Being as their root, have Being as their home, have Being as their support.65 Here the words mula and suhga are used, whereas the philosophy of MahTdasa used the terms mula and tula.

Sankara makes a distinction between the conception of Being
(sat) in U ddalaka's philosophy and the conception of matter
(prakrti) in the Sañkhya philosophy. According to him, the Sañkhya accepts a primordial substance that is insentient, while in U ddalaka's theory Being (sat) is sentient because it is able to make a wish, namely, a wish to procreate, which sets the whole chain of causation in motion.66 The persistent endeavor by Vedic and Upanisadic thinkers to attribute sentience even to material things seems to be the result of an attempt to explain the functioning of phenomena by the analogy of human behavior. It was observed earlier that the primitive notion of cause attributes causation to personal agency. Uddalaka's conception of causation is not much different from this.

Criticism o f the theory o f s e lf causation. Commenting on the word "Being" (sat) occurring in the description of evolution given by Uddalaka, Sankara says that it stands for the entity that is
"mere esse'' (astitama.tr am).61 In the Samyukta, the Buddha rejects the conception of 'Being' (atthita, jo yu - astita) as an extreme view, because it is an unobservable entity, hence a metaphysical postulate.68 He says, "To one who observes with proper understanding the passing away of things of the world, there would not be the belief in Being [existence]." 69 According to the commentator Buddhaghosa, Being (atthita) connotes permanency (sassatam).10 Moreover, the Buddha maintained that the theory of self-causation also leads to the belief in permanency.71 Thus, both the conception of Being (sat) and the theory of self-causation, which are knit together in the philosophy of Uddalaka, lead to one result, the belief in permanency. The pre-Buddhist thinkers maintained the theory of self-causation by assuming an immutable basis such as atm an; hence they considered cause and effect as being identical in essence. This view of cause and effect was accepted by the Sañkhya school and came to be known as the satkaryavada.12 The Buddha rejected this view because it has a metaphysical basis that is not verified by observation. On similar grounds, Nagárjuna criticized it on a later occasion.73 In the Pasadika-suttanta, a list of wrong views concerning the beginning of things (pubbantakapparia, pin shing pin chien) is given.74 One of them refers to the self-causation of the self (atta)
and the world (loka).15 This certainly is a reference to such cosmological speculations as those of Uddalaka and his predecessors.

The Buddhist counterpart of Uddalaka's theory of evolution is found in the Agganna-suttanta of the DTgha Nikaya.76 The keenness of the Indian mind for cosmological speculation was so great that even the Buddhists, who for empirical reasons abstained from discussing the problem of the origin of the world,77 were compelled at least to give a rational explanation of the problem of evolution. But unlike the theories of Uddalaka and his predecessors, where the principle of evolution is one of self-causation, the Buddhist theory applies the general formula of causation to explain the process of evolution.78 The causality of the individual self or soul (atman) is discussed in the Svetasvatara Upanisad. But this has to be supplemented by the discussion of the origin of the self found in the TaittirTya Upanisad: "In the beginning this [world] was nonexistent. Therefrom, verily, Being (sat) was produced. That made itself (svayam akuruta) a Soul (atman). Therefore, it is called the well-done
(,sukrta)." 79 Here the idea of self-causation is clearly exemplified. The causality of the universal, rather than the individual, self is described.

But because the individual self was considered to be a basic part of the universal self,80 what is said of the latter may be true of the former to some extent. Referring to the reincarnating individual self, the Svetasvatara Upanisad says: "Coarse and fine, many in number, the embodied one chooses forms (rupa) according to his qualities." 81 Thus, the nature of the physical form is determined by the actions or qualities of the soul or self (dehT = atman).
This theory is referred to and criticized in the Samyukta,
where the question of who made this body is raised, and bhikkhuni Sela says that it is neither self-made nor wrought by another.

Rejecting the use of a metaphysical principle to explain the causation of the human personality, the Buddhists, after a perusal of observable facts, explained it as being due to a concatenation of causes.82 Applying the theory of self-causation to the sphere of moral responsibility, the Svetasvatara Upanisad says: "Whoever has qualities is the doer of deeds that bring recompense, and of such action surely he experiences the consequences. Undergoing all forms characterized by the three qualities, treading the three paths, the individual self wanders along according to its deeds." 83 Both MahavTra and the Buddha opposed this conception of moral responsibility.

In the Sutrakrtahga, MahavTra is represented as rejecting the view that suffering is self-caused (na tarn sayam kadam dukkham).84 The commentator Silanka says: " 'Caused by oneself' (sayam kadam) means 'caused by one's own effort' (atmana purusakarena krtam)," and he adduces empirical arguments for the rejection of this view by MahavTra. He points out, "If one experiences happiness, etc., caused by one's own effort, then why should there be disparity in, or even absence of, results reaped by servants, traders, farmers and others when they exert equal effort. Some, even those who do not have a profession, such as servants, appear to enjoy great gains. Therefore, nothing can be achieved by one's own effort." 85 This line of argument leaves room for the introduction of another agency to which the effect (phala) can be attributed when one's own exertion does not seem to be the determining factor. Thus, according to the Jaina theory of causation, it is not human exertion alone but some other agency that combines to produce the effect.86 The early Buddhist literature makes numerous references to the theory of self-causation of suffering (and happiness).87 The Samyukta records an interesting dialogue between a man called Acela Kassapa and the Buddha.88 Kassapa raises the question whether suffering is self-caused, to which the Buddha gives a negative answer. The Buddha argues thus: "A person acts and the same person experiences [the result]—this, Kassapa, which you emphatically call 'suffering self-wrought,' amounts to the eternalist theory." 89 This explanation shows that the Buddha was aware that the problem of personal identity was connected with the theory of moral responsibility. For the Buddha the Upanisadic solution was not the least satisfactory. Commenting on the foregoing explanation of the Buddha, C. A. F. Rhys Davids says: "We fare no better in the editorial hands over the way in which the Founder is shown teaching another important application of the causal law. We mean the statement of continuous identity." 90 This criticism may lose its force if we carefully consider the argument in light of the Buddha's attitude toward metaphysical concepts. The Buddha is represented as rejecting the theory of selfcausation of happiness and suffering because he was aware, as pointed out earlier, that acceptance of the theory of self-causation results in the adoption of a metaphysical entity such as a permanent self or soul, and therefore does not solve the problem of personal identity. His empirical attitude prevented him from accepting a permanent and immutable self (atman) serving the functions of both the agent (karta) and the enjoyer (bhokta) of consequences.

This is evident from a conversation the Buddha had with a monk named Moliya Phagguna, who raised the question, Who feeds on consciousness-sustenance? The Buddha pointed out: " It is not a proper question. I am not saying [someone] feeds on it. If I were to say, '[Someone] feeds on it,' then the question, 'Lord, who feeds on it?' would be a fitting one. But I do not say so. And since I do not say so, if some one were to ask me: 'Of what, Lord, is consciousness the sustenance?' this would be a fitting question." 91 This does not mean that the Buddha despised the beliefs of common sense. His attitude is comparable to that of a modern Logical Positivist, who is suspicious of the " unreflecting analysis of those beliefs, which takes the grammatical structure of the sentence as a trustworthy guide to its meaning." 92 Therefore, the Buddha reformulates the question without leaving room for the introduction of any substantive ego, an atman.
Taking up the question of the self-causation of suffering, the Buddha says: "Even those who believe that [happiness and] suffering are self-caused depend on contact (jphassa, cKii) [with the world],
for it is impossible to have any experience without contact." 93 The inference is that the Buddha reduced the substantive ego {atman)
of the Upanisadic thinkers to what may be described in Humean terms as " a bundle or collection of different perceptions." 94 To say something about the self, according to the Buddha, is to say something about sense experiences. In making these arguments, the Buddha rejected the Upanisadic conception of moral responsibility, which in turn was based on the theory of self-causation.

The inherent conflict in the Upanisadic theory of self-causation could not lie dormant forever. Evolution or transformation of an immutable and permanent self (atman) was a paradox. The Upanisadic thinkers, therefore, gradually came to consider change as a mere illusion of our deceptive senses because it could not be reconciled with a permanent and homogeneous bodily reality.95 They were led to a complete denial of plurality.96 Although the search for an essential unity of things was crowned with success, philosophy suffered a severe setback as a result of this transcendentalism. As Deussen puts it: "This unity excluded all plurality, and therefore, all proximity in space, all succession in time, all interdependence as cause and effect, and all opposition as subject and object." 97 Reality was considered to be beyond space, time, change,98 and therefore causality. Change is a mere matter of words, nothing but a name (vacarambhanam vikaro ridmadheyam)."
After this, metaphysical speculation took the upper hand, and any serious attempt to give a rational explanation of the things of experience is lacking in the Upanisads.

Theories Of Divine Creation

We now come to the second category of pre-Buddhist theories of causation: external causation. The theories in that category that belong to the Vedic tradition are those asserting the creation of the world by an omniscient and omnipotent God (issara, tsun y«).100 Many theories of creation are mentioned in the pre-Buddhist literature. They appear to be the product of reasoning as well as of religious experience. Of the two methods, it was by the former that the concept of God and creation, as it appears in the Vedas and the Brahmanas, seems to have been arrived at. The argument from religious experience was mostly adduced during the period of the later Upanisads.

The process of reasoning by which the conception of God was arrived at in the Vedas and the Brahmanas involved two types of arguments, namely, the cosmological and the teleological, or the argument from design.

The cosmological argument is based on the assumption that the infinite regression of time is meaningless. The problem of the infinite regression of time is hinted at in the NasadTya-sukta.101 The conception of the infinity of time (and of space) seems to have been personified in the form of Aditi.102 But the Vedic thinkers could not reconcile themselves to the idea of infinite regression. Prajapati Paramestin appears to have been aware of the problem of the infinite regression of contingent phenomena, and he raised genuine doubts as to whether anybody could say what the beginning of the universe was. But even he had to fall back upon something that he described with attributes that are the opposite of those of existence: "That One, breathless, breathed by its own nature." 103 This is a clear indication of an attempt to avoid the infinite regression of contingent phenomena by resorting to a noncontingent factor.104 While the reluctance to accept the infinite regression of phenomena contributed to the development of the idea of God, it was further supported by the theory of self-causation. According to the theory of self-causation, a phenomenon produces from within itself another phenomenon. Tracing observable phenomena backwards, the Vedic thinkers posited primordial substances such as the
(heavenly) waters (apas). But this is only a material cause that is insentient and inanimate, although the later Upanisadic thinkers attempted to explain such primordial substances as being sentient. Therefore, the necessity of positing an intelligent being as the creator of the universe may have been felt even at an early date.

This theoretic desire to determine the first cause of the world grew keener and keener, and we find several hymns devoted to it, Narayana's "Hymn of Creation" (Purusa-sukta) being one of the best-known.105 Although the so-called cosmological argument led to a belief in an original Being (sat) that possessed characteristics opposite to those of the world of experience, the concept of a personal God as the creator of the universe did not appear. Barua has rightly remarked that it was the conception of Hiranyagarbha and of Visvakarman that showed a considerable advance toward the idea of G od." 106 In one of the hymns of the Rgveda, things of the world are traced back to their causes.107 The sun, which was called the Golden Germ (hiranyagarbha), was looked upon as the. great power of the universe, from which all other powers and existences, divine and earthly, are derived. It represented the 'origin of life.'108 The sun, which denotes fire or the generating principle, was the solar essence. But it was itself contained in the (heavenly)
water. The author of the hymn was not satisfied by explaining the origin and development of the world using water as the first principle, for there conceivably was a higher principle behind it. It was Prajapati, the God of gods, who brought forth water and provided the generating principle and the ordaining power of things. This was the theory posed in reply to the question, "W hat God should we adore with our oblations?" 109 The other argument, the teleological argument, or the argu ment from design, appears to be the basis of the conception of the creator God found in the hymn addressed to Visvakarman. The question raised there is, "What was the tree, what wood in sooth produced, from which they fashioned the earth and heaven?" 110 The critics of the teleological argument have tried to show that the argument does not prove that the God is a creator but only an architect who arranges the material.111 But this criticism does not hold in the case of the Vedic conception, for according to the hymn, the original substance out of which the universe was fashioned derives its being from the creator God. The commentator Sayana says that there is no contradiction in applying the attributes 'created' and 'creator' to the same being because of the ability to assume both these attributes by the power of tapas.212 Thus, the creator is one with the creation, although he is the maker, disposer, and the omniscient one.113 Most of the theories of creation in the Rgveda include mechanical and organic views of creation. But in the hymns addressed to Visvakarman we come across, for the first time, the idea of a personal creator God, which became a favorite topic of speculation during the period of the Brahmanas.

In the sacrificial metaphysics of the Brahmanas, theological speculation centers on Prajapati, who replaces various other concepts found in the philosophic hymns of the Rgveda and the Atharvaveda. In the Taittinya Brahmana, Prajapati is identified with Visvakarman.114 Continuity in the cycle of creation is hinted at when it is said, several places in the Satapatha Brahmana, that Prajapati, after creating beings, became exhausted and was healed by the gods (his offspring) by the power of the sacrifice.115 Although Prajapati was the creator, his ability for continuous procreation was dependent on the sacrifice. In the Brahmanas there are only a few instances where the theory of creation is presented independent of sacrificial metaphysics. Once it is said that Prajapati, who alone was at the beginning (agra) and who thought of procreation, created the three worlds by the power of heat (tapas).116 An attempt is also made to explain how Prajapati created living beings of various species.

Side by side with the conception of Prajapati, we find the conceptions of Brahman (neuter)117 and of Brahma (masculine)118 serving the function of creator God. This is evident in a statement repeated in several texts belonging to different periods. In the Paňcavimsa, or Tandya Maha Brahmana, Brahman (neuter) is said to be the "first born of the divine order." 119 In the Taittiriya Brahmana, the same is said in identical words with regard to Prajapati.120 This suggests that the concepts of Brahman (neuter),
Brahma (masculine), and Prajapati were used without much discrimination during the period of the Brahmanas, which may be considered a formative stage in the conception of a personal G od.121 The rational justification of the existence of G od and of the creation of the world by him continued during the time of the Aitareya Aranyaka. There we find the amalgamation of the two arguments, the cosmological and the teleological, for the existence of God. We have already seen how MahTdasa Aitareya explained the causality of phenomena. Things arise as effects (tula) from causes (mula). Tracing these causes backward, one arrives at the conception of water (apas), the first root of which the universe is the shoot.122 This primordial matter, which was passive and which served as the substratum of change, had to be energized; hence the idea of God brooding over m atter.123 The principle of motion by which passive matter was set in motion was considered to be something other than matter. This was God, the highest truth (satyasya satyam).124 This argument is extensively used by Uddyotakara in his Nydya Varttika to prove the existence of God. He maintains that "Just as an axe, not being intelligent itself, acts [only] after having been directed by an intelligent carpenter, in the same manner do unconscious pradhana,
atoms, and karma act. Therefore, they are also directed by an intelligent cause." 125 Speculative theories based on rational explanations rather than mystical experience seem to be a characteristic of thought during the period of the Vedas and the Brahmanas. Such speculative theories about the existence of God and the creation of the world by him continued during the time of the early and middle Upanisads.

In the Brhaddranyaka Upanisad we find a speculative theory that partakes of most of the ideas expressed in the Yedas, Brahmanas, and the Aranyakas but tries for the first time to explain how the original unitary Being gave rise to the world of manifold variety.126 As is evident, the belief was that the Self (atma), which served the function of a creator God, was solely responsible for the creation of the world of diversity. With the disappearance of the distinction between atman and brahman, on the one hand,127 and the synonymous use of brahman and brahma, which are not strictly distinguished in the early Upanisads, on the other,128 Brahma assumed the role of a personal creator God. It is this stratum of thought that is preserved in the Mundaka Upanisad when it says: " Brahma arose as the first among the gods, the maker of the universe, the protector of the world." 129 It is an echo of the statements made about Brahman and about Prajapati in the Brahmanas referred to earlier.

The intuitional method of verifying the existence of God appears to have been adopted during the time of the later Upanisads.

During this period more and more importance was attached to yogic concentration and the mental powers attained by such methods. Meditation was considered the proper means of beholding God. The Svetasvatara Upanisad says: " By making one's own body the lower friction stick and the syllable 'Om' the upper friction stick, by practicing the friction of meditation (dhyana), one may see the God (deva) who is hidden, as it were." 130 In the same Upanisad, the following question is posed: "Presided over by whom do we live our different conditions in pleasure and pain?" 131 Having rejected some of the theories of causation such as time (kala), nature (svabhava), necessity or fate (niyati)
posited by contemporary thinkers, the Upanisad replied thus:
"Those who have followed after meditation and abstraction saw the self-power of God hidden in his own qualities. He is the one who rules over all these causes, from 'time' to the 'soul'." 132 The repeated occurrence of the terms is and Tsvara in the sense of an omnipotent God is a significant feature of the Svetasvatara Upanisad.133 Also, the word deva occurs in almost every stanza in this Upanisad and indicating that the idea of God as a personal being was the predominant conception. This God (Tsvara) is the creator of all and receives the appellation of visvakarma,134 He is the supreme Lord of Lords, the highest deity of deities.135 The function of the creation of the world attributed to Brahma in the earlier Upanisads is transferred to Tsvara in the Svetasvatara Upanisad. This appears to be the formative stage of the conception of Tsvara, which was to dominate the theological speculations of a later time.136 Criticism o f the theories o f creation. Side by side with this new concept of Tsvara we find the continuation of the earlier concept of Brahma (masculine).137 Therefore, when the Buddhist texts refer at times to Brahma as a personal creator G od138 and, at other times, to Tsvara (issara, tsun yu) playing the same role,139 they are not referring to fanciful accounts of their own imagination140 but are presenting genuine conceptions found in the mainstream of the Vedic tradition.

As we have already seen, the terms brahma and Tsvara were used synonymously in the later Upanisads. This is reflected in the early Buddhist texts, where it is often said: "yo kho so bhavam brahma mahabrahma abhibhu anabhibhuto annadatthudaso vasavattl issaro katta nimmata settho sanjita vasT pita bhutabhavyanam"
("That illustrious Brahma, the Supreme One, the Mighty, the Allseeing, the Ruler, the Lord of all, the Maker, the Creator, the Chief of all, appointing to each his place, the Ancient of days, the Father of all that are and are to be").141 The Chinese Agamas preserve this statement and there, too, the terms ta fan and tsu tsai are used synonymously.142 Moreover, in the Patika-suttanta, the Buddha refers to some teachers who advocated the traditional doctrine of creation of the world by Tsvara, by Brahma,143 and the commentator believes that there the two terms Tsvara and Brahma are used synonymously.144 The intuitional method of verifying the existence of God is referred to and criticized in the Brahmajala-suttanta.14^s There, the Buddha does not adopt the negative approach that treats all forms of religious experience as illusions or hallucinations. He adopts a more sober attitude, comparable to that held by Broad and Stace in our day. According to that view, human beings do, in some religious experiences, come in contact with "an aspect of reality" not encountered in more ordinary types of experience, but that aspect of reality is probably misdescribed by the use of theological language.146 This method of refutation was adopted by the Buddha in criticizing the claims of the sages who maintained that they had witnessed God in their trances. The Buddha's argument is put forward in the form of a parable. He says:
At the dissolution of the world process, some of the beings are bom in the realm of Brahma, and of these, the being who is to be bom in that realm first comes to be of long life, good complexion and is powerful. Beings who follow him are inferior. It so happens that one of the beings who came later, having passed away from that realm, is reborn in this world.

After being reborn here he adopts the life of a religious mendicant and by practicing mental concentration is able to reach such rapture of thought that he can recollect his past births up to some moment [of his life in the Brahma-world] and not beyond. Then with regard to the being who was first bom in the realm of Brahma, he maintains that "he is the great Brahma, the supreme one, the mighty, the all-seeing, the ruler, the lord of all, appointing to each his place, the ancient of days, the father of all that are and are to be, and we must have been created by him." 147 This fanciful account is intended merely to refute the idea of creation. At the same time, it testifies to the existence of people who depended on religious experience to make assertions about the existence of a creator God. The Buddha seems to have been aware of the difficulty into which these religious teachers had fallen. As the parable shows, he rejected their views not because they were illusions or hallucinations but because they were misdescriptions of an aspect of reality that pertains to extrasensory perception.

Further, the view that the origin of the world was due to the creative activity of God was refuted by a counter-theory that appears to reject each of the salient features of pre-Buddhist theory. First, the view that the world process had a conceivable beginning is rejected when it is suggested that the process is one of dissolution (samvatta, huai) and evolution (vivatta, pien), without a beginning. Second, the Upanisadic idea that the first being became as big as a man and woman embracing each other and that the parting of this very body into two resulted in the appearance of man and wife is rejected.148 The Buddhist text maintains instead that just like the first being, another being appeared in this world, having disappeared from the world of Brahma.149 Then it explains how the being who came first misconstrued that he was the creator of other beings who came later. Because the first being hoped for the company of another, he interpreted the appearance of the second being as a response to his wish. As for the other beings who came later, they thought that the being who appeared first was their creator. This story exposes the fallacy of the idea of creation of the world by an almighty God and perhaps also indicts the view that the prior or the preceding is always the cause of the subsequent.

The Buddha's objection to the view that the world of beings, with their happiness and suffering, is created by an omnipotent and omniscient God is based mainly on two grounds. First, it denies the doctrine of the moral responsibility of man, and second, it is detrimental to the religious life.

According to the Mahabodhi Jataka, " If God (issara) were to determine the life of all beings, including their happiness and misery, virtue and vice, then man is carrying out the commandments of God. Therefore, it would be God who would be smeared by their actions." This argument, which makes use of the idea of creation itself to refute the theory of creation, is compared to the bringing down of a mango by striking it with another mango.150 Another argument is adduced to the same effect: "If beings experience pleasure and pain because of theistic determination, then the Niganthas, for example, are created by an evil God because they experience extreme forms of pain; and the Buddha, because he, being freed from defilements, enjoys extreme happiness, would be a creation of a beneficent G od." 151 From these arguments it is evident that the Buddha objects to the idea of creation because it tends to undermine the idea of moral responsibility.

On the other hand, if we were to hold that evils such as murder or theft are due to theistic determination, it would destroy the very foundation of religious fife. The Ahguttara says: "For those who fall back on the idea of creation by God as the essential reason, there is neither desire nor effort, nor the sense of 'ought' and 'ought not.' Thus in the absence of such [disposition and discrimination] in truth and verity, the term 'recluse' cannot be applied [to such a person] because he lives in a state of bewilderment with the faculties unguarded." The doctrine of creation (issaranimmanahetu, yin tsun yu tsao), along with two other views, namely, that everything is due to past action (pubbekatahetu, yin su ming tsao) and that everything is due to chance occurrence (ahetu appaccaya, wu yin wu yuan), are considered to be sectarian tenets (titthayatana, tu ch'u) that lead to a traditional doctrine of inaction (akiriya, wu tso).1S2

Ii. Pre-Buddhist Theories Of Causation : The Non-Vedic Tradition

THE VARIOUS THEORIES of external causation, except the theory of creation by a personal God, are predominantly nonVedic, although the germs of these theories may be found in the Vedic tradition itself. Most of these theories may be classified under the broad category of Naturalism.

Theories of Naturalism In the history of Indian thought, three types of Naturalism have arisen. The first type is synonymous with Materialism, which regards all facts of the universe as sufficiently explained by a theory of m atter.1 M atter is considered the ultimate fact of the universe, and all phenomena, including the phenomenon of consciousness, are reduced by the theory to transformations of material molecules.

The transformation of material molecules takes place according to inherent nature (svabhava). This school of thought is represented in Indian philosophy by the Carvakas, the Lokayata, or the Barhaspatya.

The second type of Naturalism is the one advocated by the AjTvikas, who, while accepting the Materialist conception of the universe, laid emphasis on the theory of complete natural determinism (niyatT). Being influenced by the biological speculations of the time, however, they, unlike the Materialists, believed in transmigration. The third type of Naturalism limits itself to what is natural or normal in its explanation of the universe, instead of appealing to what is supernatural. Its scope is not limited to physical nature but takes in mental phenomena, which are also considered fundamental constituents of the universe. It may be this type of Naturalism Riepe had in mind when he said: "Every school that is materialistic is also naturalistic although it is by no means true that all naturalistic schools are materialistic." 2

Materialism

Let us consider the first type of Naturalism, namely, Materialism. Materialism is generally thought to be a product of the incipient rational temper pervading the pre-Buddhist philosophical atmosphere,3 and especially a revolt against the ritualism of the Vedas and Brahmanas and some of the idealistic metaphysics of the Upanisads. Systematic treatment of Materialism is found quite late in the history of Indian philosophy. The sources by which the teachings of this school can be reconstructed fall into three broad categories. (1) The references to Materialistic teachings in the orthodox as well as the Jaina and Buddhist literature. These accounts are tinged with partiality, since only the aspects of Materialistic teaching that are opposed to the doctrines of the school quoting them are emphasized, to the neglect of other aspects. (2) Accounts in the histories of philosophical systems such as the Saddarsanasamuccaya and Sarvadarsanasamgraha, which, though their authors may belong to a particular faith, present the views of the Materialists as a comprehensive whole. In this category may be included the account of Materialism found in the Santi-parvan of the Mahabharata,4 (3) The Tattvopaplavasimha of Jayarasi Bhatta is a unique work, being the only treatise on Materialism belonging to a Materialistic school.

There is no doubt that the information supplied by the sources of the first category is the earliest. Yet it would be unfair to depend on them for a true picture of the materialistic philosophy. On the contrary, it would be more fitting to evaluate the information gathered from the sources of the first category in the light of the information afforded by those of the second and third, for the latter present a systematic treatment of Materialism.

There are several important discussions by modern scholars of Indian Materialism.5 Some have traced the origins of Mate rialist thinking to the early Upanisadic period.6 Jayatilleke has pointed out that "the Materialists themselves seem to trace their doctrines to the early Upanisads when they quote a statement attributed to Yajnavalkya in the Upanisads in support of their doctrines" .7 Discussing the ontological speculations of Uddalaka in the Chandogya Upanisad, Ruben calls them a form of "hylozoistische Monismus," and traces the germs of Materialism to the thought of U ddalaka.8 On the other hand, we find Sankara making an effort to distinguish Uddalaka's theory from the more materialistic Sankhya and to interpret it as a form of Idealism (see chapter 1). For Uddalaka and his predecessors, as in the case of the Milesians,9 the union of matter and spirit in a primordial substance was an unquestioned assumption. But as time went on, matter and spirit tugged more and more strongly at the bonds that united them until the emergence of completely materialistic as well as naturalistic schools on the one hand and the idealistic schools on the other. Therefore, it is possible to trace the origins of M aterialism as well as of Idealism to the hylozoistic tendencies in the thought of Uddalaka and his predecessors.10 Causation through inherent nature (svabhava) . It was pointed out that a theory of self-causation was at the basis of the philosophy of change accepted by Uddalaka (see chapter 1). The material elements, which were considered sentient, were able to produce out of themselves succeeding elements. Even the Sankhya, which, according to Sankara, considered material elements to be insentient, accepted a theory of self-causation. But being unable to explain how movement can be initiated in insentient matter (prakrti), the Sankhya school posited an external spiritual principle, purusa, that creates the movement in matter. Therefore, in the ultimate analysis, it is the spirit or purusa that sets the chain of causation moving. The Materialists rejected the spiritual principle as accounting for movement and instead attributed change to inherent nature {svabhava). Although no separate work of the Materialists in the preBuddhist period has come down to us, there is no doubt that the school existed independently, fighting against the ritualistic and the idealistic schools of thought current at that time. This is borne out by the Svetasvatara Upanisad, which refers to several contemporary theories of change and causation, two of which are the "theory of elements" (bhutani) and the "theory of inherent nature" (svabhava).11 There is a tendency to indentify Materialism with the theory of elements,12 but not with the theory of inherent nature.

In the later sources of the Materialist school, we find that M aterialism, or the theory of elements, is inextricably connected with the theory of inherent nature.13 Jayatilleke believes that in the Sarvadarsanasamgraha the Materialists have adopted the theory of inherent nature.14 But in the Mahabharata we have an earlier reference to the close connection between Materialism and svabhava.ls Unfortunately, the early Jaina and Buddhist texts make no reference to this aspect of Materialism even though it is mentioned in the Svetasvatara Upartisad. But is this silence on the part of the Buddhists and the Jainas a proof of the nonacceptance of the theory of inherent nature by the Materialists? If Natural Determinism (svabhavavada) was adopted by the Materialists at some point, what was their position before its adoption? Did they propound a theory of chance (yadrccha) ? This is not plausible, because even the Nihilist school of Materialists, as will be pointed out later, accepted a theory of inherent nature. We are inclined to believe that svabhavavada was part and parcel of Materialism, even in preBuddhist times.

Two types o f Materialism. All accounts of Materialism admit the plurality of elements.16 Even the Tattvopaplavasimha, which purports to 'upset all realities' (tattva-upaplava) including the material elements, admits, at the level of conventional truth, the reality of the four elements, earth, water, fire, and air.17 Scholars have generally agreed that the Materialists saw the material or the physical world as the only reality and considered nonmaterial phenomena, such as consciousness, unreal. But this is not so. The Tattvopaplavasimha testifies to the existence of a slightly different philosophy of Materialism.

Comparing the conceptions of reality given in the sources mentioned above, it is possible to classify the Materialists into two schools, (1) those who upheld a theory of evolution (parinati) of physical objects and ascribed reality to them, denying the reality of mental phenomena, and (2) those who upheld a nihilistic theory denying the reality even of physical objects.

Group 1 is certainly the best-known. Both the Saddarsanasamuccaya of Haribhadra and the Sarvadarsanasamgraha of Madhava present this theory. The Saddarsanasamuccaya states it thus: " As a result of the evolution (parinati) of body (deha) by the combination of elements of earth, etc., consciousness arises." The same theory is set out in the Sarvadarsanasamgraha: " Here the elements, earth, etc. are the four realities; from the evolution of bodily form (dehakard) from these [realities], consciousness is produced." 18 Thus, according to this theory, body (deha) and such other physical things are real in that they have evolved directly from real material elements. But these bodies are distinguished from consciousness, which arises in the bodies once they have evolved (parinatebhyah). This implies that the physical bodies are as real as the material particles that constitute them, but consciousness is only a by-product and is therefore unreal. Because Materialists of group 1 accepted perception (and also inference in a limited sense) as a valid form of knowledge, they were able to grant the reality of physical bodies, but they rejected consciousness, etc. as unreal since these are not subject to perception (adrsta).

Belief in the evolution of the physical personality (deha =
rupa) from material elements, and the granting of a greater degree of reality to objects that have evolved in this maimer than to consciousness, may have led these Materialists to accept a personality lasting as long as life. This gave rise to a school of Materialists who believed that the soul is identical with the body (tajjivatacchanravada), a theory referred to in the Buddhist as well as the Jaina texts.19 The teachings of group 2 are represented in the Tattvopaplavasimha. There the constitution of the phenomenal world is described in a slightly different way. It does not speak of evolution but maintains that "earth, water, fire and air are the realities, and as a result of their combination [arise] body, senses, objects and consciousness." 20 The important feature in this statement is that even the body, the senses, and external objects—without distinctionare put into the same category as consciousness. Consciousness is considered by all Materialists as unreal, and therefore the conclusion that this school believes that even physical bodies are unreal is irresistible. Such a theory is quite plausible when we consider the epistemological standpoint of this school of Materialists. As Jayatilleke has pointed out, they denied even perception,21 and thus there was no ground for a belief that physical bodies are real.

Dialectical arguments were adduced by this school of Materialists to refute the conception of causality (hetuphalabhava).
Jayarasi rejected the idea of production (janakatva)22 as well as concomitance (sahotpada).23 That rejection led him to deny the idea of destruction (vinasa).24 Denial of any form of production appears to be the result of the acceptance of the a priori premiss that "What is does not perish, and from nothing comes nothing," 25 which is attributed by Silanka to one of the schools of M aterialism.26 Having rejected destruction (vinasa), Jayarasi had to admit the permanence of all realities.27 This he did without any hesitation when, after criticizing the conception of causality, he concluded that "anterior or posterior activity is not generated by immovable or static matter (avicalitarupa)." 28 This means that the Nihilist school of Materialists upheld a theory of motionless permanence (avicalita-nityatvam). With the acceptance of the principle of motionless permanence, these Materialists were compelled to maintain the unproductivity or barrenness of phenomena. This idea was certainly hinted at by Jayarasi when he said: "The wise do not query about causation or absence of causation in the case of a barren woman's child who is nonexistent." 29 Conception o f svabhava. Of all the doctrines of the Materialists, the one that is most relevant to our study is their conception of Natural Determinism (svabhavavada). Without doubt it was the Materialists who first put forward a systematic theory of inherent nature (svabhava). Both RamatTrtha Svami30 and Nrsimha Asram a,31 commenting on the SamksepasarTrika32 of Sarvajnatma Muni, attributed svabhavavada to the Materialists.

It was noted above that according to the first school of Materialists there is a plurality of elements and the phenomenal world is the product of the evolution of these material elements. The most popular school of Materialists, we are informed, denied the validity of inference.33 If so, how did they explain evolution? W hat was the principle by which the plurality of elements formed the world of experience? Jayatilleke states the problem that the second school of Materialists, according to his classification, faced as a result of their change of epistemological outlook, but he leaves it unsolved. He says: " It is difficult to say whether this school asserted that there was a necessary connection between cause and effect or merely held that concomitance or sequence was only probable and therefore the inference was only probable." 34 Having changed its epistemological outlook, this school of Materialists would be expected to propound a theory of causation. But it must be remem bered that these Materialists changed for practical reasons and as a result of the criticism of the other schools of thought. Though they accepted the validity of inference in a limited sense, they emphasized that the inferable was confined to the sphere of the verifiable. Thus, it was difficult for them to go beyond the school of Materialists that accepted sense perception only as a valid means of knowing and put forward a theory of causation based on the inductive principle.

According to the best-known school of Materialists, "a universal proposition is not established even by the observation of several instances because of the possibility of error, even after a thousand instances have been observed. Though, by the observation of several instances, we come to the conclusion that smoke and fire are concomitant, we cannot know that there is no smoke in the absence of fire, even after repeated observation." 35 This denial of the validity of inference and thus of universal propositions militated against the acceptance of the principle of causation.

Does this mean that the plurality of phenomena perceptible to the senses are destitute of causes? The Materialists of the Sarvadarsanasamgraha raise this question thus: "Nanv adrstânistau jagad vaicitryam âkasmikam syâd iti" (" If what is not perceived is not granted [as existing], is it not that variety in the world is due to chance occurrence?")36 As this school of Materialists was opposed to indeterminism, the answer to that question was in the negative. The use of the word akasmika to denote the idea of chance occurrence is very significant, because Sankara used the very same word to explain yadrcchâvâda. Commenting on the Svetâsvatara Upanisad, he says: yadrcchâ akasmikTpraptih}1 If akasmika is a synonym for yadrccha, then certainly the Materialist theory set forth in the Sarvadarsanasamgraha cannot be considered a "curious admixture of Svabhâvavâda and Yadrcchâvâda" , as Hiriyanna would have us believe,38 because akasmika is rejected as a solution and in its place the belief in inherent nature (svabhâva) is upheld.39 Even the Nihilist school of Materialists, which accepted a doctrine of motionless permanence (avicalita-nityatvam), appears to have believed in a theory of inherent nature (svabhâva).w Defining the word svabhâva occurring in the Svetâsvatara Upanisad (1.2), Sankara says: "Svabhava is the unique power or property restricted to [individual or classes of] objects, like the warmth of fire."41 It is one's uniqueness. Udayana Acarya, in his NyayakusumahjalT, supports and elaborates this view: "W hat pertains to all cannot be inherent nature (svabhavatva) and indeed, the same thing cannot be the nature of more than one; otherwise there is contradiction."42 Vardhamana, commenting on this passage, says: " Svabhava is said to be the property restricted to one [class of]
object[s]. If that pertains to everything, then there would not arise the state of inherent nature {svabhavatva) or uniqueness (asadharanatva)."43 Vardhamana's explanation is significant in equating inherent nature {svabhava) with uniqueness {asadharanatva). This means that svabhavavada involves the idea of necessary connection or invariability {niyamatva).
Udayana Acarya defines invariability as "the dependence of the effect on the cause" and goes on to argue that if the Svabhavavadins are to accept such a theory of invariability, then this svabhavavada may be acceptable.44 Commentator Varadaraja also maintains that "This itself, [invariability] is the dependence of the effect on the cause [in such a way] that it happens only when that exists. If an assisting cause devoid of invariability is not meant [by svabhavavada], then svabhavavada would be accomplished. . . . If there be a svabhavavada [according to which] smoke exists when there is fire, it would be acceptable to us."4S
This interpretation of svabhavavada makes it a theory of causation that maintains the invariable concomitance between two things such as fire and smoke. It would therefore be a recognition of the validity of a universal proposition that was categorically denied by the Materialists. As Varadaraja himself points out, the svabhavavada accepted by the Materialists is different from this.46 The example quoted by Udayana Acarya to illustrate the M aterialist theory of svabhava seems to refute the idea of interdependence.47 Varadaraja also considers that the belief in permanence is intended to affirm nondependence {anapeksatva).4S Belief in the permanence of material elements was a cornerstone of the Materialist creed. Therefore, the svabhavavada propounded by the M aterialists was clearly opposed to interdependence.49 For the Svabhavavadin, a phenomenon does not depend on another phenomenon or group of phenomena for its existence. It depends on its inherent nature {svabhava). Inherent nature was the only cause {karana). Therefore, Nrsimha Asrama, commenting on the SamksepasarTrika, says: "The Carvakas maintain that inherent nature (svabhava) is the cause because of the inadmissibility of positing a theory of cause and effect apart from inherent nature." 50 This is because the Materialists were reluctant to draw any inferences beyond what is perceived. They refused to depend on past experiences to draw inferences for the present or the future. Denying induction, they were forced to abandon causality, and they maintained that all things—for example, the sharpness of the thorns, the variegated instincts of the birds and the beasts—are born of inherent nature. To maintain this it was not necessary to assume the validity of something unseen. It was an inference that did not go beyond- the verifiable, or, more correctly, the perceptible.

The Materialists' rejection of interdependence and any form of causation except inherent nature {svabhava) earned them the appellation of noncausationists {ahetuvada) (see below). But it must be emphasized that they were not indeterminists, for they accepted the determinism of nature {svabhava).
The interpretation of svabhava as the unique power or property of an object or a class of objects implies the classifiability of the things of the world according to their resemblance to one another. This leaves diversity or plurality as the ultimate characteristic of the universe. The prefix sva in the term svabhava means
" one's own," implying contrast with " other," and is therefore definitely opposed to monism.51 In fact, the Tattvopaplavasimha emphasizes the diversity in the world: "Because things are determined, each according to its own nature, by n ature,. . . they partake of individuality or diversity." 52 If svabhava is to be interpreted as inherent nature or selfnature, why did the Buddhists and Jainas include it under the category of external causation (param katam, t'a tso), in opposition to self-causation {sayam katam, tsu tso)'l The idea of self-causation, as we saw earlier (see chapter 1), was wedded to the conception of atman and considered to be the reality in man as well as in things. But the Svabhavavadin, who recognized no such entity as atman as a reality, formulated this theory of inherent nature {svabhava) to explain the force at work in material phenomena. It was a purely physical law. The individual was only an automaton functioning according to the dictates of the stuff out of which his physical personality was composed. Nature was a power over which human beings had no control ;S3 in this sense it was external to them. Hence its inclusion under the category of external causation by those who recognized the validity of human exertion as a causal factor.

This being the case, it is difficult to subscribe to the view put forward by Basham that according to the svabhavavada the individual was "rigidly self-determined by his own somatic and psychic nature." Such a wider application of svabhavavada to include psychic phenomena can not be found, especially in the teachings of the Materialists, who alone, apart from the AjTvikas, put forward a theory of inherent nature (svabhava) before the rise of Buddhism.

Basham himself maintains that the AjTvikas "viewed the individual as determined by forces exterior to himself." 54 If even the AjTvikas, who accepted such phenomena as rebirth, did not consider the psychic personality as being self-determined, it is difficult to see how the Materialists could do so. In fact, Basham's interpretation derives from a consideration of svabhavavada as a philosophy distinct from Materialism.

A Jtv Ik A D E T E R M In Is M

Another conception coming under the category of external causation that has very close connections with svabhavavada is niyativada, complete determinism or fatalism. Much has been said about the conception of niyati, which was put forward by the AjTvika school of thought.55 Very definite ideas have been expressed on many problems connected with AjTvikism. We feel that many of these views should be reconsidered and modified. Basham, explaining the close connection between svabhavavada and niyativada,
says that some heretics exalted svabhava to the status of niyati in the regular AjTvika system. To illustrate this connection he quotes from the Prasnavyakarana of Jnanavimala and the TarkarahasyadTpika of G unaratna.56 One of the earliest exponents of niyativada, or fatalism, was Makkhali Gosala. His teachings are recorded in the Samahhaphala-suttanta thus:
There is neither cause nor basis for the impurity of living beings; they become impure without cause or basis. There is neither cause nor basis for the purity of living beings; they become pure without cause or basis. There is no deed per formed either by oneself or by others, no human exertion or action, no strength, no courage, no human endurance or human prowess. All beings, all that have breath, all that are born, all that have life are without power, strength, energy; have evolved according to destiny (niyati), species (sahgati),
and nature (bhava); and experience pleasure and pain in the six types of existence."57 The crucial phrase in this passage is niyati-sahgati-bhavaparinata. Basham translates it as "developed by Destiny {niyati),
chance {sahgati) and nature (bhava) N Jayatilleke equates sahgati with yadrccha (chance).s8 Both scholars seem to have been guided by a rarer meaning of sahgati given by lexicographers. MacDonnell gives the following meanings: "meeting with, resorting to a place, frequenting, association, intercourse; alliance (rare); accidental occurrence (rare); fitness, appropriateness; connection, relation; by chance, haply. Monier-Williams gives a similar list of meanings.59 Since chance (yadrccha) is opposed to both forms of determinism, namely, destiny (niyati) and inherent nature (svabhava),60 Jayatilleke tries to reconcile the contradiction by maintaining that Makkhali Gosala is a " syncretic thinker" and that the central concepts of different schools are welded together in his teachings.61 Basham and Jayatilleke seem to have overlooked the traditional explanation of the word sahgati. An examination of the comments of Buddhaghosa and Silanka in the light of the Ajlvika cosmology shows that the term sahgati, in the description of Makkhali Gosala's teaching, does not stand for chance occurrence (yadrccha).
Silanka explains the term samgaiyam62 as samgaiyam'ti samyak svaparindmena gatih63, (" development or progress according to proper self-evolution"). The use of the word samyak is extremely important in that it points to the absence of any incongruity or inconsistency. To understand the full significance of the statement above, it should be examined in the light of the rest of Makkhali Gosala's teaching. We have already seen that the Svabhavavadins advocated plurality and the classification of this plurality according to the resemblance the elements bear to one another. Makkhali Gosala accepted svabhdvavada as well as the classifiability of things, and in accordance with them presented the theory of the six types of existence (cha abhijati, liu sheng).64 Buddhaghosa defines sahgati as channam abhijatmam tattha tattha gamanam ("movement or progress according to any one of the six types of existence" .65 The explanations of sahgati by Buddhaghosa and Silanka seem to be very similar, the only difference being that the former gives a more specific description of the way things or beings are evolved, i.e., according to the six types of existence, while the latter explains it in very general terms as self-evolution (svaparinama).66 Moreover, Buddhagosa does not consider the words satta, pana, bhuta and jiva, occurring in the statement of M akkhali's teaching, as synonyms but as references to different types of existence:
satta = camels, buffaloes, donkeys, etc.

pana = beings with one sense, with two senses, etc.

bhuta = beings born from eggs, bom in the womb, etc.

jiva = rice, barley, wheat, and such other plants.67 The description of niyativada in the Chinese version of the Samahhaphala-suttanta, which is more lucid and less obscure than the Pali version, confirms Buddhaghosa's analysis of beings into different species. The Chinese equivalent of the phrase sabbe satta sabbe pana sabbe bhuta sabbe jiva reads thus: i ch'ieh chung sheng yu ming chih lui,68 and may be literally rendered as "all beings, species of living things."
A comparison of the Pah and Chinese versions of the Devadaha Sutta of the Majjhima would throw much light on the exact meaning of sahgati. There five pre-Buddhist theories are mentioned:69 Pali version Chinese version 1. pubbekatahetu,----------------------- ► 1. yin pen tso 2. issaranimmanahetu, —* 2. yin ho hui 3. sahgatibhavahetu, -— 3. yin wei ming 4. abhijatihetu, , 4. yin chien 5. ditthadhamma-upakkamahetu,''* 5. yin tsun yu P(l) and C(l) represent the Jaina theory of karma (see below). P(2) and C(5) represent the theory of creation (see chapter 1). P(5) and C(4) refer to a theory put forward by those who upheld the validity of human exertion. The phrase ho hui in C(2), meaning
"coming together" or "harmony," were used to render the Pali term sannipata, which occurs in the passages describing the process of rebirth, or more properly, the conception of a being.70 Therefore, in the present instance, ho hui may be taken as representing the term sahgati. Thus, C(2) is the equivalent of P(3). Then we are left with P(4) and C(3). C(3) may be translated as "by reason of destiny" (= niyatiT), which along with C(2) would constitute the ÁjTvika theory of niyatisahgatibhavahetu. But P(3) and P(4) appear to overlap each other because, according to Buddhaghosa's analysis, sahgati can be explained on the basis of the theory of the six types of existence (abhijati). Thus, it appears that the Chinese version of the Devadaha-sutta presents the ÁjTvika conception of niyatisahgatibhava, having split it into two parts— wei ming referring to niyati and ho hui representing sahgati (bhava)—
while in the Pali version we find a repetition of the conception of species (sahgati).
That even Gautama Sanghadeva, in his translation of the Devadaha-sutta, has understood the term sahgati to mean "coming together" or "harmony," and not "chance" (yadrccha) is evident from his rendering of the term as ho hui. The problem would then be how to interpret this "coming together" or "harmony." Another Chinese rendering of the phrase niyatisahgatibhavaparinata becomes helpful in solving this problem. It is the phrase ting fén hsiang hsu chuan pien, where ting fén represents niyati, hsiang hsu represents sahgati (bhava), and chuan pien, parinatá.11 The phrase hsiang hsu is generally used in Buddhist Chinese to mean "stream" or "continuity" (santati, santana).72 Therefore, the coming together or harmony represented by the characters ho hui may be understood as the "harmony of the characteristics that constitute one stream." According to the ÁjTvika theory of existence, a thing has to fall into one of the six categories of existence. The particular characteristics possessed by a thing determine the nature of the species into which it falls. It is the "concurrence" of these characteristics that is denoted by the term sahgati. The concurrence is not haphazard. It is strictly determined by destiny (niyati). Once the nature of the species (sahgati) is determined by Destiny (niyati), that species begins to evolve (parinama) according to its nature (bhava ~
svabhava). This may be the proper self-evolution (samyak svaparinama) that Šilaňka had in mind.

Considering the three factors separately, as Buddhaghosa does,73 we find that they are presented in a particular order. Destiny (niyati) is placed at the beginning because of its universality and all-comprehensiveness. It is the cause that accounts for the manifold diversity of the universe. Then comes a more specific factor, species (sahgati), and lastly, nature (bhdva), that is, the nature of that particular species. This interpretation of the concept of sahgati has the support of Belvalkar and Ranade, whose translation of the passage in the Samahhaphala-suttanta, although done about four decades ago, has gone unnoticed. They have paraphrased the last part of the passage thus: "They [the beings] are bent this way and that by their fate, by the necessary conditions of the class to which they belong, by their individual nature, and it is according to their position in one or the other of the six classes that they experience ease and pain." 74 This description of the evolution of the different types of existence is reminiscent of the biological speculations of the earlier thinkers, such as MahTdasa Aitareya.7S In fact, Basham, in his analysis of the AjTvika doctrines, also recognized the impact of the earlier biological speculations on the teachings of the AjTvikas.76 Thus, according to Makkhali Gosala, the evolution of things is determined by three factors, destiny (niyati), species (sahgati) and inherent nature (svabhava). This explanation would make it unnecessary to consider Makkhali Gosala a syncretic thinker or to give a special explanation why the idea of evolution came to be associated with the teachings of Makkhali Gosala.77 Evolution was the very basis of his biological speculations, as it was with some of the thinkers of the Vedic tradition (see chapter 1). Even a group of Materialists accepted a theory of evolution.

If sahgati is to be understood in the above manner, then the Sutrakrtahga passage may be translated as follows: "Samgaiyam means development or progress according to proper self-evolution.

Whatever experience of pleasure and pain, no matter whose, in whatever time or place—that is according to one's species. Destiny is one's natural lot. They say that since pleasure and pain, etc., are not produced by human exertion and so on, evolution is caused by destiny and species." 78 In fact, Silanka identifies sahgati with niyati?9 Thus it is possible to eliminate the idea of chance (yadrccha) from the teachings of Makkhali Gosala,80 and the evidence above would show how, in the words of Basham, " some heretics exalted svabhava to the status of niyati in the regular AjTvika system." 81 Summing up the doctrine, Buddhaghosa says: "Whatever should happen will happen in that same way. Whatever should not happen will not happen." 82 Thus, it is complete determinism, but not indeterminism, that is the basis of Ajlvika fatalism. If the phrase niyatisahgatibhdvaparinata is interpreted in this manner, it is clear that chance (yadrccha) has no place in the teachings of M akkhali Gosala and, therefore, of the Ajivikas. B. C. Law has rightly observed that "Gosala maintains that everything happens according to the unalterable laws of nature, that is to say, he banishes chance from the whole of experience. He seeks to explain things as a biologist in the light of these principles: (1) Fate,
(2) Species, and (3) N ature." 83 This analysis leads to a very important conclusion. Makkhali Gosala's was not an attempt to reconcile the central teachings of different schools of thought. With due recognition for his ingenuity, it may be held that he was presenting a set of beliefs, logically and mutually consistent, leading from a theory of complete Natural Determinism to a doctrine of Fatalism.

Jayatilleke has rightly observed the influence of the earlier speculations regarding the problem of time (kala) on the determinist thesis of the Ajivikas.84 In the Atharvaveda time (kala) is conceived as an hypostatized entity that has everything in the world under its control. Time had produced what was in the past and would produce what would be in the future.85 This conception of time (kala) as the cause of the things in the world was mentioned in the list of theories given in the Svetasvatara Upanisad.86 The influence of this conception of time is to be found in the Ajlvika theory of salvation.

In keeping with his Determinism, Makkhali Gosala propounded a theory of transmigration that, in the words of Basham,
" seems to have been thought of on the analogy of the development and ripening of a plant." 87 As may be expected, this kind of rebirth has its appointed end. Thus, Makkhali Gosala maintained that "samsara is measured as with a bushel, with its joy and sorrow, and its appointed end. It can neither be lessened nor increased, nor is there any excess or deficiency of it. Just as a ball of thread will, when thrown, unwind to its full length, so fool and wise alike will take their course and make an end of pain." 88 This theory came to be known as samsarasuddhi, or "purification through-wandering in the cycle of existence.89 It is interesting to note that this theory is said to have been propounded by the noncausationists {ahetuvadT), because they denied any form of causation other than species and nature.90 Let us examine the moral and ethical implications of the conception of Natural Determinism, for it is this aspect that comes under the persistent criticism of the Jainas and the Buddhists.

The Santi-parvan of the Mahabharata presents us with a model account of svabhavavada. Though a later account, it is the most comprehensive. It combines the main features of Lokayata M aterialism and Ajlvika Determinism.91 There, Blusma quotes an old story of a discussion between Prahlada and Indra to dispel doubt as to whether man is the doer of actions producing consequences. Prahlada upholds svabhavavada, for he maintains that everything comes into being and then ceases in consequence of its own nature (svabhava). He draws several conclusions from this main thesis. Firstly, there is no personal exertion as such because nothing is achieved by it. Secondly, in the absence of personal exertion, no personal agent exists. Thirdly, there is no effect of good and bad deeds, no moral responsibility. Lastly, freedom and emancipation come through inherent nature {svabhava).
The denial of human exertion is a necessary corollary of svabhavavada. As we have already pointed out, nature {svabhava)
is a force external to man in the sense that he is unable to control or change the course of nature. He has no power over his own physical personality because his physical frame is subject to the physical laws that govern nature. Jayatilleke has pointed out that the arguments adduced by the Niyativadins against causal determination, are said to hold against the connection between human exertion and its fruits.92 Jayatilleke quotes only Silanka (9th century), although the argument is stated in the same manner in the Mahabharata. There the argument is attributed to the Svabhavavadin, thus showing the close connection between svabhavavada and niyativada. It is stated thus: "Even among persons doing their utmost, the suspension of what is not desired and the attainment of what is desired are not seen. W hat comes then of personal exertion? In the case of some, we see that without any exertion on their part, what is not desired is suspended and what is desired is achieved. This then must be the result of nature {svabhava)".93 Whatever the philosophical implications of this argument,94 it has been used by the Svabhavavadins to reject the validity of human exertion.

The denial of the validity of human exertion is, according to the Samahnaphala-suttanta, one of the main theses of the fatalist Makkhali Gosala.95 Thus, the Materialists, who upheld a theory of nature (svabhava), and the AjTvikas, who accepted Determinism
(niyati), agree in repudiating human exertion as having any influence on the course of nature.

If we accept the Mahabhdrata description of svabhavavada as accurate then we are compelled to admit that the absence of personal exertion implies the absence of a personal agent. This brings together the teachings of Makkhali Gosala and Pakudha Kaccayana because, while the former believed in the absence of human exertion, the latter reiterated the absence of a personal agent. Pakudha Kaccayana maintained that "No man slays or causes to slay, hears or causes to hear, knows or causes to know," because " even if a man cleaves another's head with a sharp sword, he does not take life, for the sword cut passes through or between the elements." 96 In the same tone the Svabhavavadin of the Mahabhdrata says: "When one slays another, one slays only the other's body." 97 The Chinese version of the Sdmahhaphala-suttanta attributes this doctrine to Purana Kassapa,98 who according to the Pali version was a Materialist. Thus, Pakudha Kaccayana, who was an AjTvika, seems to have shared some of the dogmas accepted by the Materialists. He is represented as putting forward a theory of motionless permanence (' avicalita-nityatvam) ." We have shown that the Nihilist Materialists also propounded a theory of motionless permanence while accepting svabhavavada as a central tenet.

Therefore, it is not improbable that Pakudha Kaccayana accepted a theory of Natural Determinism {svabhavavada).
Further, according to the Svabhavavadin of the Mahabhdrata,
the denial of human exertion and the repudiation of a personal agent leads to the denial of moral responsibility. He maintains that " What we have now become is neither due to any act of ours nor of others. Everything is due to inherent nature {svabhava).'"100 The denial of moral responsibility is explicitly stated as part of the teachings of Makkhali Gosala, Ajita Kesakambali and Purana Kassapa, and is implied in the teachings of Pakudha.101 This means that the Materialists, as well as the Ajlvika Determinists who accepted svabhavavada, held the view that there is no moral responsibility, no effect of good and bad deeds.

The teachings attributed to Makkhali Gosala in the Samahhaphala-suttanta are in another place102 called ahetuvada-, the teachings of Purana Kassapa, akiriyavada, and the teachings of Ajita Kesakambali, natthikavada. The three terms ahetuvada, akiriyavada, and natthikavada are used in another context as synonyms.103 This is evident from the Chinese renderings of at least two of the terms. The term akiriyavada is rendered into Chinese as shuo wu tso, and the phrase natthikavada as shuo wu yeh,104- the only difference being that the tso expresses a more active meaning than does yeh. The rendering of natthikavada as shuo wu yeh, "he who maintains that there is no [effect of] action,"
shows that the definition of the term in the PTS Dictionary105 is inaccurate.

Basham has observed that "In certain passages of the Pali Canon the description of doctrines among the six teachers is significantly altered, in a way which strongly suggests that the credos ascribed in the Samahhaphala-suttanta to Makkhali, Purana, and Pakudha were aspects of a single body of teaching" .106 The examination of svabhavavada described in the Mahabharata confirms this and points to the close connection between Materialism and Ajlvika Determinism. Therefore, it is difficult to question the authenticity of the Chinese version of the Samahhaphalasuttanta, as Basham himself does,107 although it does not agree with the Pali version in the description of the teachings of these six heretical teachers.

Thus we are led to the conclusion that svabhavavada, as described in the Mahabharata, is a synthesis of Materialism and Ajlvika Determinism. It incorporates the teachings of four of the six heretical teachers, Ajita Kesakambali, Pakudha Kaccayana, Purana Kassapa and Makkhali Gosala. In spite of their differing emphases all of them were Naturalists, accepting Natural Determinism {svabhavavada) as the basis of their teachings.

Buddhist criticism. The Devadaha-sutta of the Majjhima Nikaya, though mainly devoted to a refutation of the Jaina theory of karma, is also an attempt to demonstrate that belief in external determination undermines belief in moral responsibility. The Buddha argues that if happiness and suffering are caused by destiny (niyati, wei ming) or one's lot or species (sahgati, ho hui), then the Niganthas, because they undergo extreme forms of suffering, have an evil destiny or are of evil species. The Buddha, who is freed from all defiling tendencies and who therefore enjoys extreme forms of happiness, has a good destiny or is of good species.108 This criticism represents an attempt to show the fallacy of positing an external agency, even in cases where individual responsibility is clearly manifest, because the Niganthas' severe pain is inflicted upon themselves by themselves, while the Buddha's happiness is a direct result of his untiring effort.

Another criticism of this theory of external causation is found in the Mahabodhi Jataka: "If man's behavior depends on one's species (sahgatya) and nature (bhavaya), then his actions, which should or should not have been committed, are committed without any intention [on his part]. If so, who is it that would be smeared by the evil consequences when these actions are unintentional." 109 This argument was adduced to refute the belief in a creator God
(issara), too.

Therefore, whenever the Buddhists criticize the doctrines of the heretical teachers, they seem to refer only to those aspects with which they disagreed. The conception of inherent nature
{svabhava) that was common to the teachings of four of the six heretical teachers is not referred to, except that of Makkhali Gosala. And reference to his conception of svabhava may have been made because it was exalted to the status of destiny {niyati). The word svabhava is never even mentioned in the early Buddhist texts, and when it occurs in the later commentaries it connotes
"truth." 110 Criticism of the philosophical theory of nature (svabhavavada) is conspicuous by its absence, although the theory definitely existed during the time of the §vetasvatara Upanisad This does not mean that the Buddhists accepted in toto the theory of nature
{svabhavavada) of the earlier teachers. The Buddhists would have certainly objected to the attempt to limit svabhava to physical nature alone, ignoring the mental or psychological and moral aspects of nature that Buddhists consider the fundamental constituents.111 But the silence of the Buddhist texts on svabhavavada may be accounted for if we admit the influence of this theory on the Buddhist theory of natural causal Determinism (see chapter 4).

The third type of Naturalism referred to at the beginning of this chapter does not fall into the category of either Materialism or complete Determinism. Naturalism in this sense is expressed in the Buddhist texts by the term dhammata. ("nature of things").

While the word is used to explain the behavior of physical phenomena,112 it is not confined to that. Even psychological attitudes are given naturalistic explanations and are illustrated by examples from physical nature. The Kosambiya-sutta says: " It is the nature {dhammata) of a person endowed with right understanding that whatever kind of offense he falls into, . . . he confesses it, discloses it, declares it quickly to the teacher or to intelligent coreligionists; having confessed, disclosed and declared it, he comes to restraint in the future, just as an innocent little baby lying on its back draws back its hand or foot if it has touched a live ember. . . ," 113 It is also "the nature {dhammata) of a person endowed with right understanding that if he is zealous concerning those many duties to be done for coreligionists, then he becomes of strong aspiration for training in higher conduct {adhisila), higher thought (<adhicitta), and higher intuitive wisdom {adhipahha). Just as a cow with a young calf, while she is pulling the grass, keeps an eye on the c a lf 14 Even extrasensory perceptions and emancipation are not considered supernatural occurrences in Buddhism. They are natural causal occurrences.

It is in the nature of things {dhammata, fa tsu jari), that a person in a state of concentration knows and sees what really is. A person who knows and sees what really is does not need to make an effort of will to feel disinterested and renounce
[the things of this world]. It is in the nature of things that a person who knows and sees [the world] as it really is, feels disinterested and renounces. One who has felt disinterested and has renounced does not need to make an effort of will to realize the knowledge and insight of emancipation {vimutti, chiai fo). It is in the nature of things that one who has felt disinterested and has renounced realizes the knowledge and insight of emancipation.115 According to this statement, mental concentration, which is a product of training and effort, is a causal factor {upanisa, yin)
in the production of knowledge leading to emancipation. It is categorically stated that in the absence of right mental concentration, the cause for the production of knowledge and insight is absent.116 This does not mean that these processes are absolutely determined by destiny (niyati, wei ming) or any such thing, as the AjTvikas believed, but that they are natural causal occurrences. The Buddhists, unlike the Materialists, did not confine their Naturalism to physical nature. Thus, in the later commentaries, which attempt to systematize the teachings found in the early sutras, five kinds of causal patterns are enumerated. They are in the realm of (a) the physical (inorganic) world (utuniyama), (b) the physical (organic) world (bTjaniyama), (c) the sphere of mental fife
(cittaniydma), (d) the moral sphere (kammaniyama), and (e) the higher spiritual life (dhammaniyama) (see chapter 6). Thus the teachings of the Buddha may, in certain respects, fall into the third category of Naturalism.

Criticism Of External Causation

The above-mentioned theories of external causation {param katam, t'a tso) were criticized by the Buddhists for two main reasons. First, because they implied a denial of the validity of human exertion and posited a principle external to man as the cause of his pleasure and pain. Second, because they led to a belief in annihilation. The second criticism is stated in the early Buddhist texts thus: "The theory of external causation of suffering, . . . according to which one acts and another experiences, amounts to a theory of annihilation (uccheda, tuari)."111 If we understand annihilation {uccheda, tuan) in the usual sense of destruction of life and the absence of rebirth,118 the statement that external causation leads to a theory of annihilation presents a problem. How are we to include the Ajlvika theory of determinism {niyati, wei ming) under this category of annihilation? The AjTvikas, unlike the Materialists, did not hold that a being is cut off and completely destroyed at death; they believed in some kind of survival.119 This problem may be easily resolved if we distinguish several different uses of the term uccheda {tuan). In the context above, "annihilation" may be interpreted as the annihilation of action (kamma), that is, elimination of the connection between an action and its consequences.120 Annihilation in this sense can be taken as denial of moral responsibility and may be synonymous with natthikavada. Thus, ucchedavada and natthikavada mainly represent the theories denying moral responsibility, though they are used more specifically to denote nihilistic systems.

Criticism of the theory of external causation brings us once again to the problem of personal identity. It was mentioned that the theory of self-causation led to a belief in permanence, a belief in a permanent entity underlying the empirical reality (see chapter 1). According to it, the person who acts (the cause) and the person who experiences (the effect) are one and the same, the identity of the two individuals (or the cause and the effect) being maintained on the basis of a permanent substance. The theory of external causation leads to an opposite result, namely, a belief in annihilation (uccheda, tuari). This means that the person who acts and the person who experiences the result are two different persons; the cause and the effect are different. All the statements in the early Buddhist texts to the effect that a person acts and reaps the consequences121 are made to refute the theory that denies the identity of the person who acts and the person who experiences the results.

But these statements should not be taken as referring to an ultimate reality; they are only empirical statements and the identity is maintained not by positing an extraempirical entity such as a soul
(atman), but by a theory of causality (see chapter 6).122 Thus, the Buddha criticized two main theories: (1) the doctrine that posited a permanent soul (atman) on the basis of which personal identity, moral responsibility, and survival were explained, which came to be known as atthikavada (shuoyu lun che), and (2) the doctrine that did not posit such a soul, denied personal identity, moral responsibility, and survival, which came to be known as natthikavada (shuo wu lun che). Both these theories were rejected by the Buddha on empirical grounds. We have already stated the argument for the rejection of the first (see chapter 1). In the same manner, the Buddha appealed to experience in his refutation of the second. He maintained that "to one who sees, with proper understanding, the arising of the things in the world, the belief in nonexistence would not occur." 123

The Jaina Theory Of Causation

The third of the four main types of pre-Buddhist causal theories is that which upholds internal as well as external causation
(sayam katañ ca param katañ ca, tsü tso t'a tso). This is a relativist theory that attempts to combine the first two types of theories discussed above, self-causation and external causation. There appear to be two such theories in Indian thought. One is propounded by the Jainas, who were recognized as relativists.124 The other, as pointed out by Jayatilleke, is mentioned in the Svetásvatara Upanisad.125 We have seen how the Upanisadic thinkers conceived of Being (sat) or reality as permanent, immutable, and eternal, and how they rejected impermanence, mutability, and change as illusions (chapter 1). The Jaina thinkers, in the desire to account for the various forms of experience such as change, continuity, impermanence, and duration, and supported by their epistemological standpoint that absolute judgments are not possible at the mundane level,126 maintained that Being (sat) is multiform in that it exhibits the characteristics of productiveness (utpada), destructiveness (yyaya), and permanence or durability (dhrauvya).127 This came to be known as parinamanityavada,126 a theory that comes very close to that of MahTdasa Aitareya (see chapter 1).

It was pointed out that the rejection of change and mutability as illusory impeded fruitful speculation on the problem of causality during the period of the Upanisads. When the Jaina thinkers accorded greater reality to experiences such as change and mutability, they initiated serious discussion of the problem of causality.

It has been held that the first attempt at systematic analysis of the causal problem was made by the ÁjTvikas.129 But because of their theory of strict Determinism and Fatalism they were reluctant to accept any cause other than destiny. On the other hand, the Jainas, with their relativist epistemological outlook, made a genuine effort to determine the nature of causality.

The locus classicus of their theory of causality is the Sütrakrtáñga and its commentary by Siláñka. In the former, MahavTra criticizes several theories concerning the causality of suffering. It is interesting to note that some of the same theories came under the persistent criticism of the Buddha. These theories thus had a historical basis and were not mere imaginations of the Buddha or MahavTra. MahavTra says: " Suffering is not caused by oneself (sayam kadam); how could it be caused by another (annakadam)? Happiness and suffering, final beatitude and temporal
[pleasure and pain] are not caused by themselves or by others; they are due to one's own lot or species (.samgaiyam). This is what they [the Fatalists] say." 130 The phrase "caused by oneself" {sayam kadam) is explained by Silanka as "caused by one's own exertion" (atmana purus akarena krtam). The Jainas rejected it because they found that even when there is equal human effort, sometimes the results differ or there are no results. Explaining the phrase "caused by another" (annakadam), Silanka lists several existing theories of causality: destiny (niyati), time (kala), God (Tsvara), nature (,svabhava), and action (ikarma) .131 Destiny is identified with species (sahgati) and is taken up for criticism later.132 The other conceptions are considered and argued against in turn.

MahavTra then says: "Those who proclaim these views are fools who fancy themselves learned; they have no knowledge and do not understand that things are partly determined and partly undetermined (niyayaniyayam samtam)." 133 Silanka's commentary on this statement is very important. Below it is reproduced in full.

Because they unconditionally (ekantena) resort to the theory of creation by destiny [;niyatikrtam, "predestination"], "when things are partly determined and partly undetermined," that is to say, happiness, etc. are partly determined, brought about necessarily (avasyambhavyudayaprapitam) and partly undetermined, brought about by one's own exertion (atmapurujakara), God (Tsvara), etc., therefore, they are ignorant; they have no knowledge of the cause of happiness, suffering, etc., and are devoid of knowledge. But in the teachings of the Jainas (arhatdndm), some part of happiness, suffering, etc., is due to destiny [or predetermination, niyatita eva]—it is said to be caused by destiny [or predetermination] because of the necessary manifestation of [past] karma as a cause at some moment or other. Similarly, some part is undetermined (aniyatikrtam), that is, caused by human exertion (purusakara),
time (kala), God (Tsvara), nature (svabhava), karma, etc. Herein, the effectiveness, in some way or other, of human exertion in the case of (the production of) happiness and suffering, etc., is maintained. Since an action yields results, and the action depends on human exertion, it has been said:
"One should not give up one's effort thinking (that everything) is due to destiny (daivam). Without effort who would be able to obtain oil from sesame seeds?" But the inequality of results obtained when there is equal human exertion, although said to be a fallacy, is not a fallacy. Because, in such a case, the difference in human exertion is the cause of the inequality of the results. If for some person equal exertion produces no result, then it is the work of something unseen
([adrstakrtah). That too we consider a causal factor. Thus, time (kala) also is a cause because the bakula [Mimusops elengi], the campaka [Miceliya campaka], the asoka [Jonesia asoka roxb], the punnaga [Rottleria tinctoria], the naga [Mesua roxburghii], the sahakara [a kind of mango] and such other trees are seen to bloom and bear fruit during the proper season, but not always. The statement that variety in the world is not possible because of the oneness of time (kalasyaikarupatvat) does not hold good in the case of our theory. For we do not accept the agency of time (kala) alone, but accept karma too; therefore, cosmic variety is not a fallacy.134 In a similar way, Silanka assesses the other causal theories.

A careful examination of M ahavlra's statement in the light of Silanka's commentary reveals two main features of the Jaina theory of causation. First, because of their epistemological standpoint, the Jainas refused to posit unconditionally (ekantena) one single cause. They examined each one of the causes posited by various thinkers in their explanation of the causality of suffering, etc.—
causes such as human exertion (purusakara), destiny (iniyati), time
(kala), God (Tsvara), nature (svabhava), action (karma)—and showed that these do not, by themselves, constitute causes. But in the end they agreed that these are causal factors depending on the point of view from which they were considered. Taking up the problem of human exertion, Silanka maintains that if it fails as a cause, that is because there are other causes that are not directly seen (adrsta).
Thus, Jaina theory partakes of relativity as well as plurality.

But this plurality of causes can be divided into two broad categories:
(1) internal causes such as human exertion (purusakara), and (2) external causes such as time, God, nature, and karma. The first group can be included under the category of self-causation
(sayam katam, tsu tso), and the second under the category of external causation (param katam fa tso). Although these are considered to be untenable as causes when taken individually, when considered in a group or from different standpoints their causal status can be defended. Thus, the theory referred to in the early Buddhist texts as "internal as well as external causation"
(sayam katan ca param katah ca, tsu tso t'a tso) can be attributed to the Jainas. This may be why MahavTra is represented as criticizing each of the two theories of self-causation {sayam kadam) and external causation (annakadam), but not a combination of both.

A different grouping of causal factors appears in Silanka's classification of those that are destined to occur (niyata) and those that are not (aniyata). The only example Silanka gives of the former is karma "because it can necessarily manifest itself [as a cause] when the opportunity is available" {kasmimscid avasare
' vasyambhavyudayasadbhavdt). Karma is also included with undetermined {aniyata) causes, along with human exertion {purusakdra), time (kola), etc. Therefore, a distinction has to be made between these two forms of karma.
We are inclined to believe that the karma that is determined
{niyata) to occur refers to past karma, while that which is undetermined {aniyata) refers to present and future karma. If so, the connection between past karma and its characteristic of being determined {niyata) seems to be very significant. It shows that just as the Ajlvikas raised species {sahgati) or nature {svabhava)
to the status of niyati, the Jainas raised karma to the status of niyati. It may be mentioned that the Jainas did not have to face the problem of personal identity because they believed in the existence of individual souls, which were considered to be substances and therefore eternal.135 By maintaining that karma is a necessary cause, the Jainas viewed the present as being strictly determined by past karma. If one's personality is so strictly determined, it appears that there is no freedom of will. But this conclusion is avoided by the way in which the undetermined
{aniyata) cause is explained. It was pointed out that what is undetermined is what is caused by one's human exertion, God, etc. The Jainas recognized human exertion as a causal factor under certain circumstances. This means that although one's present state is determined by one's past karma, one can change the future because one's human exertion is an effective cause. Therefore, man is endowed with freedom of will. This belief is referred to in the early Buddhist texts as pubbekatahetu or yin pen tso or yin su ming tsao.136 A representative passage in the early Buddhist texts runs thus:
Whatever the individual experiences, whether painful or pleasant or neutral, all is due to what was done previously. Thus by burning up, by making an end of, past deeds, by the nondoing of new deeds, there is no overflowing into the future. From there being no overflowing into the future comes the destruction of deeds; from the destruction of deeds comes the destruction of anguish; from the destruction of anguish comes the destruction of feelings; from the destruction of feelings all anguish will be worn away. Thus say . . . the Niganthas.137 From this passage it is evident that a knowledge of causes and conditions is behind the Jaina theory of existence. The first part of the theory is tempered by a belief in strict determinism
(niyata): everything a human being experiences is completely determined by his past karma and there is no escape from it. The second part partakes of the idea of conditionality such that when A happens B happens. Herein, causal factors such as human exertion, God, and nature are recognized. This seems to have been taken by the Jainas as indeterminism (aniyata). Thus, MahavTra's belief that things are "partly determined and partly undetermined" (niyayaniyayam) is reflected in the Buddhist statement of the Jaina theory of karma.

The account of the Jaina theory of karma given by Barua is very confusing. He starts by attributing a certain theory to the Jainas and ends by accepting an altogether different proposition. He says:
The Buddha understood that MahavTra, in opposition to current beliefs that our happiness and misery are caused by others—determined wholly and solely by external factors and conditions—formulated a new theory, namely, that they are caused by the individual agent of our free will. That our weal and ill are conditioned solely by or dependent upon external causes is one extreme, and by opposing to this a new individualistic theory, MahavTra ran to the other extreme, neither of which can a man with true insight reasonably accept.138 Barua makes such a statement because he considers the doctrine of self-causation (sayam katam) mentioned in the Pali Nikayas a reference to the Jaina theory of karma.139 He fails to see that this same theory was rejected by MahavTra. Moreover, after comparing descriptions of the theory of karma in the early Buddhist and the Jaina sources, he maintains that it is "hardly possible for us to detect any differences between their opinions." 140 But being unable to ignore the many passages in the Pali canon where the Buddha is represented as criticizing the Jaina theory of karma, Barua is forced to examine whether there is any difference between the two schools of thought. Eventually he comes to a conclusion that contradicts his earlier statement regarding the Jaina doctrine. He says: "In accordance with MahavTra's view I am not, as a thinking subject, wholly and solely the maker of my moral being, but I am partly a creature of circumstances." And in support of this view he quotes MahavTra's statement that "things depend partly on fate and partly on human exertion." 141 We have pointed out that this relativistic theory is mentioned in the Pali Nikayas and the Chinese Agamas, where it is presented as a combination of self-causation and external causation (see chapter 1). The theory of self-causation is therefore not a Jaina theory, as Barua seems to think, but a theory formulated by the thinkers of the Vedic and Upanisadic traditions.

Buddhist Criticism Of The Jaina Theory

In the Pah Nikayas and the Chinese Agamas, the theory of complete Determinism in the sphere of moral responsibility—that everything we experience is due to past karma (pubbekatahetu*
yin pen iso)—is clearly attributed to the Jainas, but not the relativistic theory of internal and external causation {sayam katan ca param katan ca, tsu tso fa tso). The latter is not attributed to any specific school. What is the connection between these two theories?

On the authority of the Jaina commentator, Silanka, it was pointed out that for the Jainas self-causation {sayam kadarri) stood for causation by one's own exertion, while external causation
{annakadam) included the work of time, God, nature, etc. We agreed with Barua that "MahavTra appears to be in sharp antagonism with Gosala." 142 This is because MahavTra disagreed with Gosala's denial of moral responsibility and free will. But MahavTra did not reject the theory of Determinism {niyatf) advocated by Gosala. This is evident from MahavTra's dictum that "things are partly determined and partly undetermined" (niyayaniyayam).

This, then, was the main problem faced by MahavTra. While accepting a theory of Determinism (niyati), MahavTra wanted to uphold moral responsibility and free will. Acceptance of the doctrine of karma accomplished the latter. By asserting that man's present life is completely determined by his past karma, he was able to make room for the theory of Determinism (niyati). Thus, the theory of moral responsibility, according to which everything a person experiences is due to past behavior (pubbekatahetu), is only a corollary of the philosophical theory that combines selfcausation and external causation.

The theory of moral responsibility put forward by the Jainas cannot be considered a strictly individualistic theory. That would defeat the very purpose for which it was formulated by MahavTra. In fact, as mentioned earlier, he rejected the individualistic theory of moral responsibility. MahavTra would maintain that the individual is responsible for his acts, but once he has committed an act
(karma), it completely determines his future and thus becomes something external to him, for he cannot control it. It is interesting to note that the Buddhists group this theory with two other theories, theistic determinism (issaranimmana, tsun yu tsao) and indeterminism (ahetu appaccaya, wu yin wu yuan), and call them all sectarian tenets (see chapter 1). The theory of creation positing an external personal agent was a widely prevalent view.

Opposed to this personal agent was the impersonal principle, namely, karma. In addition to these two major theories of moral causation, a third existed that denied any form of causation. It was specifically stated that the theory of pubbekatahetu, along with the other two, lead to a denial of moral responsibility (akiriya). Why did the Buddhists consider the Jaina theory of karma, which purports to explain moral responsibility, as leading to a denial of moral responsibility? Perhaps because the Jaina theory partakes of determinism. Moreover, if the identification of pubbekatahetu with the theory of karma referred to by Silanka is correct,143 Silaiika was justified in including it under external causation, although it was intended to occupy an intermediate position between self-causation and external causation.

Or, perhaps in the belief that man is almost powerless to control already committed actions, as the Buddhist texts would have us believe, the Jainas attempted to expiate past actions by the practice of severe austerities and to prevent further accumulation of karma in the future by nonaction.

The Buddha's criticism is levelled against the first part of the theory, which maintains that every experience that a human being has in this existence or moment is completely determined by his past behavior (pubbekatahetu, yin pen iso). Looking upon karma as an external force, as the Jainas themselves did, the Buddha says: "If one's experiences of pleasure and pain are due to what was done in the past, that person is paying his debt, to wit, his former sins. Who, when freed from debt, would be smeared by his sins?" 144 The argument that was used to refute the idea of creation and of "species" is thus also employed to refute the Jaina theory of karma. It is maintained that " if the experiences of a human being are determined by past karma, the Niganthas, who in this life undergo extreme forms of suffering, would have been of evil behavior in their past lives" , and "the Tathagata who experiences extreme forms of happiness was of good behavior in his past life" .145 Apart from these few instances, the Buddha's criticism is directed, not so much at the theory itself, but at the epistemological basis of the theory. The Buddha asks the Jainas whether they knew that they existed in the past and whether they knew that they committed such and such an act. The Jainas replied in the negative and claimed that they depended on the testimony of their teacher, Nigantha N athaputta.146 In the Svetasvatara Upanisad we come across another relativistic theory of causation comparable to that of the Jainas. While the Jainas posited karma as an external cause and upheld the validity of human exertion as an internal cause, the Theists mentioned in the Svetasvatara Upanisad believed that God is the external cause and that the person assumes various forms according to his own deeds.147 As Jayatilleke has pointed out, Dhammapala, commenting on the Udana,14B referred to this theory thus: " It is the belief of some that God, etc. [nature, destiny, etc.], in creating the soul and the world do not create them entirely of their own accord but take into consideration the good and evil of each being
[considered as] a cooperative cause, and thus the soul and the world are [both] self-caused and caused by another" .149 Theory of Noncausation The last of the four main types of causal theories is the theory of "chance occurrence" (yadrccha), or the theory of " fortuitous origination" (adhiccasamuppada, wu yin wu yuan), which is a denial of any form of causation. In the early Buddhist texts the term adhiccasamuppada replaced the earlier term yadrccha. Jayatilleke has identified this theory with the niyativada of the AjTvikas, firstly because he believes that the word sahgati occurring in the description of niyativada connotes the idea of "chance," and secondly, because niyativada was considered to be a form of ahetuvada.150 We cannot accept the first argument since we understand sahgati as "species," not "chance." With regard to the second argument, although niyativada came to be described as ahetuvada, it was only in the sense that there was no cause other than niyati, not in the sense of chance occurrence. It should be noted that the ahetuvada of Makkhali Gosala falls into the category of external causation {param katam, fa tso), whereas fortuitous origination implies a denial of both internal and external causation (sayam katah caparam katah ca,fei tsu fei fa tso).1S1 The theory of "fortuitous origination" (adhiccasamuppada),
although it does not imply any form of determinism such as niyativada, may be designated a theory of noncausation (ahetuvada). This is what the Buddhist commentator Dhammapala meant when he said: "Adhiccasamuppanna means 'arisen by chance'; it is called the theory of 'fortuitous origination' because events arise without any cause. Therefore, even ahetukavada is to be included in it." 152

Iii. Clarification Of Terminology

ONE OF THE TERMS used most frequently in the early Buddhist texts to denote both causation and causality is paticcasamuppada (Sk. pratTtyasamutpada), which is a combination of the two terms paticca, 'having come on account of' *{prati + y / f + {t)ya * and samuppada, 'arising'. Buddhaghosa defines it thus: "causation or dependent arising *{paticcasamuppada)* is the mode of causes
(paccayákara). The mode of causes is that according to which coordinate phenomena are produced mutually. Therefore, it is called causation." 1 Again, he explains the term 'arisen' (samuppanna), as:
"When arising, it arises together, that is, coordinately, not singly or without a cause," 2 and "that which has arisen depending on causes." 3 These definitions emphasize the existence of a group of causes and their occurrence together.

Another term used in the early texts to denote causation or causality is idappaccayata (Bsk. idampratyayata4), which means
'conditionality' or 'relativity.' It is an abstract noun derived from the combination of the terms ida or idam (neuter of ayam), meaning
'this,' and paccaya {prati + yfY) 'foundation', 'cause,' or 'basis.'
The PTS Pali-English Dictionary explains it as "having its foundation on this, that is to say, causally connected, by way of a cause." Buddhaghosa suggests the following meaning: " From the standpoint of the condition Qietu) or group of conditions (hetusamuha)
that give rise to such states as decay and death, as stated, there is said to be conditionality." 3 Edgerton states that the two words idampratyayata and pratTtyasamutpada constitute one compound and that the editors of the Vinaya Pitaka have wrongly separated them.6 The different uses of the term idappaccayata in the Pali texts do not justify that opinion. In the Pali texts we come across two main uses of the term idappaccayata. Sometimes it is used to denote a characteristic of paticcasamuppada, as, for example, in the statement, " Causation is said to have [the characteristics of] objectivity, necessity, invariability and conditionality." 7 Here the two words are clearly used separately, the term idappacayata defining one of the characteristics of causation {paticcasamuppada). Candraldrti seems to take idampratyayata in the sense of relativity, as is implied in the statement, "when this exists, that exists" (asmin sati idam bhavati). This is distinct from active causation (pratTtyasamutpada), which is referred to by the statement, "when this arises, that arises" {asyotpadad idam utpadyata).8 But sometimes the term idappaccayata is used as a synonym of paticcasamuppada and appears along with it.9 In such cases, the term paticcasamuppada is almost always preceded by the term idappaccayata. Of the four characteristics of causation mentioned above, the importance of idappaccayata (see below) may be taken as the raison d'etre for using it as a synonym of paticcasamuppada. It is probably this usage of the term that misled Edgerton to declare that the two terms constitute one compound.

Moreover, a comparison of some of the Pah passages in which idappaccayata occurs with the corresponding passages in the Chinese Agamas shows that the Agamas, except in one instance, do not have a special translation for this term. In the example quoted in the paragraph above, (from S 2.26, the Chinese version in TD 2.84b; Tsa 12.14), the term tathata is rendered as ju fa erh; avitathata as fa pu li ju; anannathata as fa pu i ju; and sui shun yuan ch'i may be taken as a rendering of the term idappaccayata. The absence of a special translation for idappaccayata or idampratyayata is more evident in the Chinese translations of the later Buddhist texts such as the Bodhisattva-bhumi.10 That may be because when the Buddhist texts were translated into Chinese, the usage of idappaccayata as a synonym for paticcasamuppada had gained currency, with the result that the characters used to translate paticcasamuppada were also used to render idampratyayata.
The definition of paticcasamuppada given by Buddhaghosa, as well as some of the discussions of the theory of causation in the early texts11 imply the recognition of a group or number of causes.

A 'cause' implies a 'harmony of causes' that constitute one cause having the capacity to produce an effect. Thus, Buddhaghosa maintained that if there were a deficiency in any of the several causes that constitute a single cause, there would be no effect.12 The group of causes (hetusamuha) producing an effect would not be able to do so if they were mutually independent or if some of them were lacking. Therefore, through mutual dependence, equally (samam)
and together (saha), they produce the effect or the resultant states.13 In the Chinese Agamas we come across several different translations of paticcasamuppada. Two of the most popular renderings are yin yuan fa and yuan ch'i fa .14 In the first, the character yuan,
which is generally used to express the term pratyaya, is used as a verbal form, like ch'i in the second rendering, to express the idea of 'arising.' In some instances it is possible to find yin and yuan used in compound form but denoting hetu and pratyaya, respectively. For example, the phrase hetum paticca sambhutam has been rendered into Chinese as yin yuan hui erh sheng and yin yuan ho ho sheng.15. Although the Pah version refers to the cause in the singular as hetum (a collective noun), the Chinese versions definitely imply a harmony of several causes. Even though in classical Chinese yin means direct cause as opposed to yuan, which refers to an indirect cause, this distinction cannot be made in the early Buddhist texts, because the technical meaning yin and yuan acquired in Buddhist Chinese is based on the meaning of the words they represent, namely, hetu and pratyaya. Even if we translated the two Chinese passages as "arising on account of the harmony of hetu and pratyaya," the usage of the two terms hetu and pratyaya in the early Buddhist texts would not allow distinction between them as 'cause' and 'supporting condition,' respectively.

The idea of a group of causes has misled some scholars to think that Buddhism recognizes a difference between hetu and pratyaya, two of the words most frequently used to denote the idea o f 'cause.' Monier-Williams, defines pratyaya, as: " [with the Buddhists] a co-operating cause, the concurrent occasion of an event as distinguished from its approximate cause." Soothill and Hodous as well as Jeschke and Saratchandra Das, have followed MonierWilliams' interpretation.16 Soothill and Hodous went further when they said (in the same place): " It is circumstantial, conditioning or secondary cause, in contrast with yin (hetu), the direct or fundamental cause. Hetu is the seed, pratyaya the soil, rain, sunshine, etc." De la Vallée Poussin also believes that "A distinction is to be made between the cause (hetu) and the conditions (pratyaya)."17 In an article on the Buddhist concept of identity, he seems to elaborate on this point: "Yet like all the Buddhists, the VStsTputriyas admit—basing their faith on scripture and experience—that a flame is always being renewed, and that it never remains for one moment identical with itself. The flame of the lamp on the third watch of the night is the continuation of the flame in the first watch, these two flames form a series (santati): the first is the cause (hetu) of the second, for they have both the same nature, the wick and oil are not causes but only co-efficients {pratyaya)." 18 Do the early Buddhist texts recognize a distinction between cause (hetu, yin) and condition (pratyaya, yuan)''! Words expressing the idea of cause in the Pali Nikâyas are numerous. Buddhaghosa has given a list, including paccaya, hetu, kârana, nidàna, sambhava,
and pabhava, and maintains that although the words are different, they express the same meaning.19 To the above list may be added upanisâ.20 A similar list of synonyms is given in the Sphutârthâbhidharmakosa-vyâkhyâ of Yasomitra.21 Of these, paccaya and hetu occur very frequently in the Pali Nikâyas, and the former appears to have been most favored. In the Chinese Agamas the characters yin and yuan, representing hetu and paccaya, are widely used.

It is almost certain that the hetu (yin) and paccaya (yuan) were used synonymously in the Pah Nikâyas and the Chinese Agamas, although some of the early interpreters have mistakenly rendered them as 'cause' and 'effect,' respectively.22 The locution "What is the cause, what is the condition [reason]?" occurs very frequently in the Nikâyas and the Agamas and also in later texts such as the Saddharmapundanka, Suvarnaprabhâsottama-sütra, and Lalitavistara.23 Another example illustrating the synonymous use of the two terms is found in the Nikâyas and the Agamas: "There are two causes, two conditions for the arising of right view (sammâ ditthi, chêng chien). Which two? Testimony of another (parato ghosa, ts'ung f a wen) and proper reflection (yoniso manasikara,
cheng szu we/)." 24 This passage was quoted by Yasomitra when he wanted to illustrate the synonymous use of the two terms in the early Buddhist texts.25 Therefore, the view expressed by some of the modern critics of Buddhism that "hetu is the seed, pratyaya the soil, rain, sunshine, etc." (see above) does not seem to be applicable in the literature cited above. Such definitions would wrongly imply that in the early Buddhist texts hetu (yin) denotes 'cause' and paccaya (yuan) stands for 'condition,' thus making only a commonsense distinction.26 Let us consider the example quoted by Soothill and Hodous to illustrate the distinction between 'cause' and 'condition.' This very example is found in the Samyukta, both in the Pali and the Chinese versions.27 There, BhikkhunI Sela rejects two of the existing theories of causation, namely, self-causation (attakatam, tsu tsao)
and external causation (parakatam, fa suo tso), saying "This personality is not caused by oneself, nor is it caused by another. It comes into being on account of a cause (hetum paticca)29 and disappears when that cause is destroyed." 29 Then she cites a simile:
"Just as a seed that, when sown in a field, will grow if it is supplied with the essence of the earth and moisture, so that [five] aggregates, the [eighteen] elements and the six senses come into being on account of a cause and disappear when that cause is destroyed." 30 In the Pali version, hetum, the word denoting 'cause,' is in the singular, but in the simile three causes or conditions are enumerated: (1) the seed, (2) the fertility of the soil, and (3) moisture.

Does it mean that the seed is the cause and the others are secondary conditions? If so, then the seed should be able to produce the sprout without the other factors, because according to the Buddhist theory of causation a cause must be able to produce or give rise to an effect invariably (see chapter 4). But in the above simile the seed is not capable of producing the sprout without the support of the other two factors. Therefore, in this case the seed itself cannot be considered the cause. Moreover, the seed has to satisfy several other conditions: it should be unbroken, not rotten, not destroyed by the wind or sun, and must be fresh and well planted. Otherwise, it will not sprout forth or grow up and attain maturity (S 3.54; TD 2.8c-9a; Tsa 2.7). A cause in this context can be described as the sum total of the various conditions.31 This point can be illustrated further by another example from the Majjhima Nikaya. There we find an exact linguistic equivalent of the phrase hetum paticca in paccayam paticca,32 which clearly shows that hetu and paccaya were interchangeable and were used synonymously. It occurs in a passage describing the causation of perception. Although the text does not refer to all the conditions that have to be satisfied to make possible an act of perception, it refers to two of the conditions, the sense organ and the corresponding sense object. But as in the example quoted earlier, although there is more than one condition, they are referred to in the singular as paccayam.
These few examples illustrate two main features of the early Buddhist theory of causation that have been misunderstood by some scholars. First, early Buddhist theory transcends the commonsense notion of causation. While recognizing several factors that are necessary to produce an effect, it does not select one from a set of jointly sufficient conditions and present it as the cause of the effect. In speaking of causation, it recognizes a system whose parts are mutually dependent. This dependence has been designated the 'dependent origination' (paticcasamuppada), which conforms with the definition given by Buddhaghosa.33 Thus, although there are several factors, all of them constitute one system or event and therefore are referred to in the singular. Only if a cause includes all the necessary factors will it give rise to the effect. In taking this position, early Buddhism did not make any distinction between cause {hetu, yin) and condition (pratyaya, yuan),34 even though current convention did recognize such a distinction.

The definition of a cause as the sum total of several factors that gives rise to a consequent led to further developments in the Buddhist theory of causality later. During the period of the Abhidhamma, the Buddhists started investigating the nature of the several factors that constitute a cause. They found that each of the several factors stands in a different relationship to the effect. These different types of relation were analyzed in the Patthdna of the Theravadins35 and philosophical treatises of the other schools of Buddhism.

When the analysis o f 'jointly sufficient conditions' was undertaken during the period of the Abhidhamma, the meaning of hetu
{yin) was restricted to 'root' or 'primary,' but pratyaya {yuan),
which stood for 'cause' in general, came to be prefixed by various terms such as hetu and alambana. In this manner arose the compound hetu-pratyaya {yin yuan), which means 'primary or root cause.' In this case hetu is only an adjective qualifying the word pratyaya, and hetu alone does not seem to have been used to mean
'primary cause.' With this specialization of the meaning of hetu, its former function of denoting 'cause' was taken over by pratyaya. But the Yogacarins also retained the earlier meaning of hetu as 'cause' in general (see below). According to the Abhidhammika definition, paccaya is "that depending on which the [fruit or effect] derived comes," 36 with hetu meaning 'root'.37 A similar definition of pratyaya is given by Nagarjuna in the Mulamadhyamaka-karika: "Those which, through dependence, give rise [to effects] are called the pratyayas." 38 Viewing the simile quoted by Soothill and Hodous in light of this new analysis, we maintain that the seed would be hetu-pratyaya
{yin yuan), meaning 'root cause,' not hetu {yin). The essence of the earth, moisture, and temperature may be put into the category of nutritive cause (ahara-paccaya).39 The earth (ti), as given in the Chinese texts, may be considered a supporting cause {nissayapaccaya).40 Paccaya or pratyaya or yuan, therefore, stands for
'cause,' whether as a root or in some other capacity. As pointed out earlier, the use of hetu {yin) to denote cause in general was not completely abandoned, for we find references to ten hetus (sometimes called kdrana or neng tso), which again are qualified by various prefixes, for example, "supporting cause" {sahakari-hetu,
t'ung shih yin and sahakari-kdrana, t'ung shih neng tso41). It is important to note that these different types of causes do not, by themselves, invariably give rise to effects. Although for purposes of examination various causes are distinguished, they do not make any difference in the production of the effects.42 It was the Sarvastivada school that appears to have distinguished between hetu {yin) and pratyaya {yuan), which misled the scholars mentioned above into believing that all Buddhists did so.

The Sarvastivadins formulated a theory of six hetus:

1. kdrana-hetu, neng tso yin.
2. sahabhu-hetu, chu yu yin.
3. sabhaga-hetu, hsiang ying yin.
4. samprayuktaka-hetu, t'ung luiyin. 5. sarvatraga-hetu, pien hsing yin.
5. vipaka-hetu, i shu yin.43 They also formulated a theory of four pratyayas:
6. hetu-pratyaya, yin yuan.
7. samanantara-pratyaya, teng wu chien yuan.
8. alambana-pratyaya, suo yuan yuan.
9. adhipati-pratyaya, tseng shang yuan.44 That the Sarvastivadins were the first to formulate a theory of causality with two aspects, hetu (yin) and pratyaya (yuan), is sufficient evidence that they were the first to make a distinction between hetu and pratyaya.45 Moreover, this distinction does not carry the imprint of authority from the sutras, as is evident in the inability of the Sarvastivadins to quote a statement therefrom to support it. They merely say that the sutras (that deal with the six hetus) are lost, while at the same time they quote statements from the sutras to justify the synonymous use of the two terms.46 Apart from the formulation of a theory of causality with two facets, there is a statement in the Abhidharmakosa-bhasya that indicates recognition of a distinction between hetu and pratyaya. It is a quotation from an unnamed sutra and is preserved in the Sanskrit version as well as in two Chinese translations, one by Hsiian Tsang and the other by Paramartha. The Sanskrit version reads, Tatha caksur bhikso hetu rupani pratyayas caksurvijnanasyotpadaya, and may be translated, In this manner, monks, the visual organ is the cause (hetu) and form the condition (pratyaya) for the arising of visual consciousness.47 The specific use of hetu to describe the visual organ and pratyaya to refer to the external object is very significant.48 Analyzing this causal process of perception in light of the common-sense notion of causation, it can be maintained that the visual organ is the
   'cause' and the external object the 'condition' or 'contributory cause.' In an act of perception, common sense would suggest that the visual organ is more important as a cause than an object of perception.49 Therefore, it would be possible to say that the Sarvastivadin distinction between hetu (yin) and pratyaya (yuan) corresponds to the distinction between cause and condition in the common-sense notion of causation.

It is of interest to investigate the source of the quotation above from the Abhidharmakosa. In his edition of the Chinese version of the Abhidharmakosa, Saeki Kyokuga identifies this passage with a statement in the Samyukta Agama (Kando ed., 9.6). This seems to be a reference to the wood-block edition of the Chinese Tripitaka and is thus not easily accessible. The question is resolved by Nishi Giyu. In his Japanese translation of the Abhidharmakosa appearing in the Kokuyaku Issaikyd, he refers to the Taisho edition of the Agamas, where the Samyukta statement occurs as yen yin yuan se yen shih sheng.50 The Pali version of this statement is found elsewhere as cakkhun ca paticca rupe ca uppajjati cakkhuvinhanam. It occurs in several places in the Pali Nikayas, the Chinese Agamas, and even in the Buddhist Sanskrit texts of a later date.51 It may be translated, Visual perception [or consciousness] depends upon the visual organ and the visible object in order to arise.

This is a sterotyped description of the causation of perception found in the Buddhist texts.

It is quite evident that there is a difference between the statement in the Abhidharmakosa-bhasya and its Chinese versions on the one hand and the statement in the Pali Nikayas and the Chinese Agamas on the other. The Abhidharmakosa-bhasya version presents the visual organ (caksu, yen) as the hetu (yin) and the external object (rupa, se) as the pratyaya (yuan). But no such distinction is implied in the Pali Nikaya and the Chinese Agama versions. In fact, even the later Pah scholiasts seemed to consider the sense organ and the sense object on a par as far as their causal capacity is concerned.52 In the Nikayas and the Agamas no statement corresponds to that in the Abhidharmakosa-bhasya. This leads us to conclude that if the Sarvastivadins were actually quoting from the sutras (included in the Nikayas and the Agamas), they changed the statement found in the sutras to suit their own theory of causation.

If the Sarvastivadins recognized a distinction between cause and condition, their theory of causation would fall in line with the common-sense notion of causation. The distinction between cause and condition is said to be the result of the common-sense notion o f a 'thing.' It is believed that "commonsense distinguishes between a thing and its states." 53 The thing or phenomenon is regarded as something substantial—a substance persisting through a period of time—but it has states that change. This common-sense notion of substance was, no doubt, a basic part of the Sarvastivada teachings.

For the first time in the history of Buddhist thought, the Sarvastivadins accepted a bifurcation of elements as having substance and characteristics (see chapter 4). This came to be called dravyavada and is believed to be the result of Vaisesika influence.54 The acceptance of such a bifurcation leads to the recognition of a distinction between cause and condition. For example, if clay is considered the substance, and the form it assumes (jar, etc.) its characteristics or states, then clay itself would be taken as the cause, while the potter, and the potter's wheel, etc., which give clay its shape, would be only subsidiary conditions.55 Thus, all evidence, textual as well as doctrinal, proves that the Sarvastivadins were the first to make a distinction between a cause (hetu, yin) and condition
(pratyaya, yuan).
Therefore, when Yasomitra wrote the Sphutarthabhidharmakosavyakhya, the Buddhist scholars differed as to the nature of hetu (yin) and pratyaya (yuan). In one place Ya§omitra refers to the earlier view in which the terms were used synonymously.56 A
quotation from the sutras is given to prove this point. Later, commenting on the statement caksur bhikso hetur iti (see above), he refers to several dissenting views thus: "Hetu is the proximate cause, the remote one is the pratyaya', others say hetu is what generates [or produces], whereas pratyaya is only the supporting condition; still others maintain that the two are synonymous." 57 As far as we know, the only Theravadin text that upholds this distinction between hetu and paccaya is the Nettippakarana,
included in the Khuddaka Nikaya. Discussing requisites or conditions (parikkhara-hara), it says, "two things give rise to or produce [a phenomenon], cause and condition." 58 Explaining the characteristics of a cause and a condition, the same treatise points out that the cause has the characteristic of being unique, and the condition the characteristic of being common.59 The example of the sprout is given to illustrate this distinction: the seed is the unique
'cause' for the arising of the sprout, while the earth and water, being common, are only 'conditions.'60 The discussion concludes with, "intrinsic nature is the cause, extrinsic nature the condition; cause is internal, condition external; the cause generates [or produces], the condition supports; that which is unique is the cause, that which is common is the condition." 61 We agree with Nanamoli's comment, on the analysis of the category of requisites (parikkharahara) in the Nettippakarana, that the distinction between hetu and paccaya seems peculiar to this work and that in the suttas no such difference is discernible.62 Since a definition implying a distinction between hetu and paccaya cannot be found in the other canonical texts of the Theravádins, the compilers of the Nettippakarana may have been influenced by the ideas expressed on the subject at the time the Sphutárthábhidharmakosavyakhya was compiled.

While the Sarvástivadin theory of causation consisted of the two facets, hetu (yin) and pratyaya (yuan), the Vijñánavádins and the Theravádins emphasized the theory of pratyayas. The Vijñánavádins even extended the theory of pratyayas by enumerating twenty subdivisions of the hetu-pratyaya (y in-yüan).63 Ten of them are referred to in the Madhyantañbhagabhásya of Vasubandhu64 and in the Vijñaptimátratásiddhi (Dharmapála's version65). In the former, they are called karana (néng tso), and in the latter they are termed hetu (yin). This shows that even when the jointly sufficient conditions were being analyzed, the words hetu (yin), karana (néng tso) and pratyaya (yuan) were used synonymously. The subdivisions of hetu-pratyaya (yin-yüan) are called karana (néng tso) in one text and hetu (yin) in another. The division of hetu-pratyaya into twenty karanas in the Abhidharmasamuccaya appears to be very significant. With the three other pratyayas, they make a total of twenty-three, which corresponds closely to the list of twenty-four enumerated in the Patthana of the Theravádins (see chapter 8).

After the Sarvástivádins put forward a theory of causation with two facets, the Vijñánavádins seem to have attempted to resolve the problems created by this dichotomy by fusing the two theories together. The Abhidharmasamuccaya describes several

samprayuktaka-hetu

samanantara-pratyaya alambana-pratyaya Fig. 1: Forms of hetu and pratyaya in the Sarvastivada
(Abhidharmakosa) and Vijfianavada (Abhidharmasamuccaya) Schools ahipati-pratyaya ways in which a primary cause {hetu-pratyaya, yin-yuan) can be recognized.66 They are by way of1. Self-nature (svabhdva, tzuhsing).

2. Diversity (prabheda, ch'a pieh). 3. Assistance (sahaya, chupan).
3. Coexistence (sampratipatti, teng hsing). 5. Increase (vrddhi, /)• 6. Opposition (paripantha, chang ai). 7. Grasping (parigraha, she shou).
   In his commentary on the Abhidharmasamuccaya, Haribhadra maintains that the first two characteristics, self-nature and diversity, together establish the karana-hetu (neng tso yin), and the remaining five characteristics, in order, elucidate the sahabhu- (dzii yw), sabhaga- {hsiang ying), samprayuktaka- (t 'unglui), sarvatraga- (pien hsing) and vipaka- (i shu) hetus.61 This clearly indicates an attempt by the Vijnanavadins to reconcile the two theories put forward by the Sarvastivadins. While the Abhidharmakosa includes all five
   /zei«.y, except karana-hetu, under hetu-pratyaya,68 the Abhidharmasamuccaya goes one step further to include even the karana-hetu under this category. Moreover, Haribhadra concludes by saying that all can be included under the category of karana-hetu,69 thus making karana-hetu and hetu-pratyaya identical. The difference of standpoints of the two schools is apparent in Figure 1.

The purpose of the classification by the Vijnanavadins was to define hetu-pratyaya so that whatever causes (hetu, yin) appeared in the Sarvastivada classification could be brought under hetupratyaya. Even though the terms hetu and karana were retained by the Vijnanavadins, they do not differ in meaning; they are interchangeable and are used synonymously. Therefore, we may conclude that early Buddhism as embodied in the Pah Nikayas and the Chinese Agamas, as well as the later Theravada and Vijnanavada schools, did not recognize a difference between 'cause' and
'condition' and that the words hetu and pratyaya did not denote any such distinction.

Iv. The Conception Of Dharma

IN OUR DAILY EXPERIENCE "we are accustomed to distinguish between occurrences that we regard as being regularly connected and occurrences that we consider to be accidentally or casually conjoined." 1 There is no doubt that primitive man discovered some minor uniformities, and that he made use of this knowledge as a guide to his daily activities. But where such uniformities could not be discovered he resorted to rituals and magical practices.2 His ritual practices may therefore be explained as unconscious attempts at overcoming or avoiding what he considered to be accidental occurrences. The two types of events enumerated above, namely, those that regularly occur and those that occur accidentally, have been called uniformities and multiformities, respectively.3 Scientific knowledge is said to consist in resolving these multiformities into a uniformity of a higher generality and greater abstraction, or to explain the causation of what have been described as accidental occurrences, so that belief in events that sometimes happen may be replaced by belief in events that always happen.

The Buddha made a similar discovery when, with the insight he gained as he sat under the Bodhi tree on the banks of the river Neranjara, he was able to penetrate into the nature of dhamma.4 The truth he discovered is summarized in a discourse he delivered to his disciples later. There he speaks of two aspects of his discovery,
(1) 'causation' (paticcasamuppada, yin yuan fa), and (2) 'causally produced dhammä' (paticcasamuppanne ca dhamme, yuan sheng fay The So-ch'u-ching makes this distinction very clear in using the two phrases yin yuan ch'i and yin yuan ch'i so sheng fa to denote paticcasamuppada and paticcasamuppannadhamma, respectively.6 The former is further explained in terms of the causal formula:
"When this is present, that comes to be; from the arising of this, that arises. When this is absent, that does not come to be; on the cessation of this, that ceases." This causal pattern, according to which things are conditioned, is explained in terms of the twelve factors of the special application of the causal formula (see chapter 5). Thus, it is a distinction between the causal relation and the causally related. The problem of causation, therefore, involves two aspects, the rule or pattern according to which things change, and the things themselves that are subject to change.7 Analysis of the nature of causally produced dhammä throws much light on the problem of causation. The conception of dhamma (fa) is fundamental to Buddhist philosophy. Conze has observed that " In its essentials the Dharma-theory is common to all schools, and provides the framework within which Buddhist wisdom operates." 8 The term dhamma is used in a wide variety of meanings.9 The implications of the term have been examined by the commentators, and the various uses are given in two slightly different lists.10 Wilhelm and Magdalene Geiger have amalgamated these two lists, giving five different uses as follows:

1. guna, Eigenschaft, Fähigkeit, Tugend.
2. desanä, Lehre, Predigt. 3. hetu, U rsach e. 4. pariyatti, heiliger, kanonischen Text.
3. nissatta (= nijjiva), Unbelebts, Ding, Sache.11 We are primarily concerned with the third and the fifth uses, which, for the sake of convenience, may be considered similar. Dhamma in this sense has undergone multifarious changes in the different schools of Buddhist thought. Our main attempt in this chapter will be to examine the conception of dhamma in the Pali Nikäyas and the Chinese Ägamas, referring to other developments when necessary.12 A wide variety of translations of dhamma have been suggested by scholars. Stcherbatsky renders it as 'elements.' C. A. F. Rhys Davids seems to prefer 'phenomena.' Wilhelm and Magdalene Geiger have translated as 'Ding(e),' or more properly, 'die empirischen Dinge.'13 We propose to leave the term untranslated until the conception in the Pali Nikayas and the Chinese Agamas is fully examined and assessed in light of later developments. Then we can determine which of the translations would best express the meaning of the term in these early texts.

The most important characteristics of dhamma are said to be impermanence (anicca, wu ch'ang), unsatisfactoriness (dukkha, k 'u)
and nonsubstantiality (anatta, wu o).14 Various other characteristics have been given, but they all are representations or even further elaborations of the three major characteristics. There is, for example, the triad consisting of impermanence, unsatisfactoriness, and change or transformation (viparinamadhamma, pien i fa ).1$ The last characteristic replaces nonsubstantiality {anatta),
because change represents the opposite of substantiality or substance {atta, o), reckoned as the immutable substratum of empirical reality. Of the three characteristics mentioned above, the most important is impermanence {anicca, wu ch'ang)', the other two may be regarded as corollaries.16 Discussing the Buddhist theory of the impermanence of dhamma, Stcherbatsky makes the following observation:
The elements of existence are momentary appearances, momentary flashings into the phenomenal world out of an unknown source. Just as they are disconnected, so to say, in breadth, not being linked together by any pervading substance, just so they are disconnected in depth or in duration, since they last only one single moment {ksana). They disappear as soon as they appear, in order to be followed in the next moment by another momentary existence. Thus a moment becomes a synonym of an element {dharma), two moments are two different elements. An element becomes something like a point in time-space. The Sarvastivadin school makes an attempt mathematically to determine the duration of a moment. It, nevertheless, admittedly represents the smallest particle of time imaginable. Such computations of the size of the atom and the duration of the moment are evidently mere attempts to seize the infinitesimal. The idea that two moments make two different elements remains. Consequently, the elements do not change, but disappear, the world becomes a cinema. Disappearance is the very essence of existence; what does not disappear does not exist. A cause for the Buddhist was not a real cause but a preceding moment, which likewise arose out of nothing in order to disappear into nothing.17 Stcherbatsky attributes this conception of dharma to the "first period" of Buddhist philosophy.18 This attribution is not wellfounded because the Pali Nikayas and the Chinese Agamas, which represent the "first period," contain no such conception of dharma.
Evidence suggests, rather, that the conception of dharma described by Stcherbatsky belongs to the period represented by the Abhidharma.
The Abhidharma Pitaka, though regarded by the scholastics as having originated with the Buddha himself,19 is no doubt a product of a later age. It has been observed that the teachings embodied in the Abhidharma Pitaka represent merely an explicit form of the Dharma implicit in the Sutra Pitaka.20 This tendency to minimize the difference between the teachings of the Sutra and Abhidharma Pitakas is the cause of much misunderstanding regarding the relative positions of the different schools of thought in Buddhism. Evidence strongly suggests that early Buddhism, as embodied in the Pali Nikayas and the Chinese Agamas, represents a form of phenomenalism.21 It is true that in speaking of the external world, or more properly of 'form' {rupa), the Buddha makes reference to primary existents (mahabhuta) and 'secondary form' (upaddya rupa).22 But this division is not equivalent to the division of matter into primary and secondary qualities found in many of the realist schools. The so-called primary existents represented by earth {pathavi), etc. are nothing but sense data. Hence, earth, for example, is defined as grossness (kakkhalata).23 On the other hand, there are specific statements that knowledge of the external world is based on experience (vedana), and that this experience is dependent on contact with sense data (phassa).24 Hence, any theory about the nature of the external world has to be based on sense data {phassa), and speculation that goes beyond sense data would be metaphysical and futile. Such theories are based on hypothetical ideas about what reality ought to be rather than on verifiable data. In other words, they are beyond the sphere of experience (avisaya, fe i ching chieh).25 But with the Abhidharma we notice a gradual change in this philosophical outlook. The origin of the Abhidharma has been traced to an attempt to preserve the fundamental teachings of the Buddha after his demise.26 The method adopted to achieve this end was to pick out the central teachings and analyze and classify them. Once the central tenets were determined, the next step was to classify and group them into various categories, sometimes in numerical order.27 This process of analysis and classification continued until the formulation or compilation of texts such as Dhammasahgani and Vibhahga of the Theravadins and Jhanaprasthana and other texts of the Sarvastivadins. There we find the analysis and classification of dharma into material (rupa) and mental (citta, caitasika) groups. Such analyses and classifications had to be complemented by a system of definition, and in defining these categories the Abhidharmikas seem to have followed their own ideas rather than those found in the early texts. For example, rupa or form came to be regarded as nonmental (acetasika, cittavippayutta).28 Such definitions led to a clear demarcation between mental and physical events comparable to the division of reality into mind and matter. Thus the philosophy of the Abhidharma assumed the form of a na'ive realism or pluralism, which was very different from the philosophical outlook of early Buddhism.

The Abhidharma tradition in India then became exposed to various external, non-Buddhist influences. Philosophical speculation continued in the wake of the emergence of such pluralistic and realistic schools as the Vai§esika, and the Abhidharmikas also succumbed to speculation, engaging in an endless analysis of dharmas into their minutest forms. This process of analysis reached its logical conclusion when the Abhidharmikas accepted the view that a dharma is a point in space-time. Thus, the Buddhist schools in India came to accept the theory of atoms (paramanuvada) and a theory of moments (ksanavada). As Stcherbatsky himself points out, " such computations of the size of the atom and the duration of the moment are evidently mere attempts to seize the infinitesimal." 29 These are the theories that dominated the philosophical atmosphere during the period of the Abhidharma, although Stcherbatsky has attempted to throw them back to the time of early Buddhism. They are conspicuously absent, even in the Theraväda tradition before Buddhaghosa (see chapter 7). Setting aside the theory of atoms for the time being, let us consider the theory of moments, which is closely connected with the theory of impermanence (anicca) and which has created innumerable problems in understanding Buddhist conception of causality.

The difficulties faced by the Buddhists who accepted a theory of moments (ksanaväda) is illustrated not only by the criticism of non-Buddhist thinkers such as Sankara but also by the objections raised by Buddhists themselves. Sankara pointed out that "Those who maintain that everything has a momentary existence only admit that when the thing existing in the second moment enters into being, the thing existing in the first moment ceases to be. On this admission, it is impossible to establish between the two things the relation of cause and effect, since the former momentary existence ceases or has ceased to be, and so has entered into the state of non-existence, cannot be the cause of the later momentary existence." 30 In the Tattyasahgraha, Säntaraksita cites the view of Bhadanta Yogasena that causal efficiency cannot be maintained in the case of momentary existences.31 Yamakami Sögen wrongly accuses Sankara of "complete ignorance of the Buddhist doctrine of Universal impermanence." 32 On the contrary, Sankara has convincingly shown the logical implications of the theory of momentariness. His criticism does not affect the theory of impermanence (anicca, wu ch 'ang) as expounded in the Pali Nikäyas and the Chinese Ägamas, or even the preBuddhaghosa Theraväda. The theory of moments, which is a product of psychological or even logical analysis of the theory of impermanence, presents a problem, as Sankara and Yogasena have pointed out, with regard to the conception of causality. If a thing exists for one moment only, a moment being reckoned as the smallest particle of time imaginable, how can the causal efficiency of that moment be made intelligible?

The later Buddhists adopted two methods for solving the problem of causal continuity created by the acceptance of a theory of moments. The first was to recognize an unchanging substratum underlying the momentary flashes of the apparent phases of dhamma. The second was to formulate a theory of immediate contiguity (samanantara) and grant causal efficiency (arthakriyakaritva or paccayata) to the immediately preceding dhamma. As will be seen (chapter 7), the Sarvastivadins adopted both these methods, while the Sautrantikas and the later Theravadins criticized the first and adopted the second.

In his anxiety to defend the Buddhist theory against Sankara's criticism, Yamakami Sogen, says: "The substratum of everything is eternal and permanent. W hat changes every moment is merely the phase of the thing, so that it is erroneous to affirm that, according to Buddhism, the thing of the first moment ceases to exist when the second moment arrives." 33 This distinction between the substratum of a thing and its changing phases, a distinction similar to the common-sense notion of a thing as having primary characteristics and causal characteristics,34 has been the keynote of Sarvastivada teachings.

In his discussion of the Sarvastivada teachings, Stcherbatsky uses the term "element" to translate the word dharma. " Element" has been defined as "that which cannot be reduced to simpler terms under the conditions of investigation." 35 Thus, in science, the term refers to the different kinds of atoms, the sort of material of which the world is composed. The use of the term "element" to render the word dharma may be in keeping with the teachings of the Sarvastivadins. Dharmas are, according to them, the simplest elements to which an empirical object can be reduced. Stcherbatsky points out that these elements were considered to have four salient features: (1) nonsubstantiality (all dharmas are anatman)—this refers to all seventy-five elements, whether eternal or impermanent, (2) nonduration (all samskrta-dharmas are anitya)-X\n.s refers only to the seventy-two impermanent elements of phenomenal existence,
(3) unrest (all sasrava-dharmas are duhkha), and their unrest has its end in (4) final deliverance (their nirvana alone is santa).
After enumerating the four salient features of a dharma,
Stcherbatsky explains the first of these in detail. He maintains that "the term anatman is usually translated as 'non-soul,' but in reality atman is here synonymous with a personality, an ego, a self, an individual, a living being, a conscious agent, etc. The underlying idea is that, whatsoever be designated by all these names is not a real and ultimate fact, it is a mere name for a multitude of interconnected facts, which Buddhist philosophy is attempting to analyze by reducing them to real elements (dharma)." Thus, according to the Sarvastivada teachings, as interpreted by Stcherbatsky, a being
(pudgala) is nothing but a congeries of elements; it is in itself no 'ultimate reality' (not a dharma)?6 This implies that although the individual is not a real individual, the elements (dharmas) that constitute him are ultimate realities. But these ultimate realities are separate., disconnected elements that, according to the second of the four salient features, are momentary ( ksanika). To use Stcherbatsky's own expression, "they are momentary appearances, momentary flashings into the phenomenal world out of an unknown source." 37 Thus came about the dichotomy of an element as having a substance (svabhava) that is unknowable and having a phenomenal appearance (laksana) that is causally conditioned.

While the underlying substratum came to be denoted by the words svabhava and dravya, which are synonymous, the causal characteristics came to be differently conceived by the various Sarvastivada teachers.

The Abhid.harma.kosa, as well as the other texts dealing with the doctrines of Sarvastivada, refers to four main theories put forward by the school on the nature of the causal characteristics
(laksana)?* The first is the view advocated by Dharmatrata, who taught a theory of change of state (bhdvdnyathatva, lui i). He maintained that a thing existing during the three periods of time changes its state (bhava) but not the underlying substance (dravya).
This process is compared to the different shapes assumed by gold, which itself does not change. Thus, the three epochs of time—past, present, and future—are differentiated by the nonidentity of the states (bhava). Ghosaka, another teacher of this school, contributed the second theory, the view that only the characteristics of a thing are subject to change (laksananyathatva, hsiang i). According to him, an entity always takes three courses. When a thing is present, it has the seeds of the past and the future. Likewise with things of the past and the future. This is illustrated by the example of a man who is attached to one woman but who is not at the same time detached from other women. Third is the view adopted by Vasumitra, who believed that a thing, when passing through the three periods of time, does not change its nature but changes its condition (avasthanyathatva, wei i). The condition is determined by causal efficiency (karitra). "When efficiency is present, the thing is said to be present; when efficiency has been given up, it is said to be past; and when it is going to have efficiency, it is future." 39 And fourth, Buddhadeva upheld a theory of change of relations
(anyonyathatva, tai i). A thing is said to change because of the change of its relation with the past and the present. Thus, a woman can both be a daughter and a mother. A refutation of these views is to be found in the TattvaSamgraha-pahjika of KamalasTla.40 Thus, the Sarvastivadins recognized "two hemispheres" in the world of empirical reality.41 One is the world of experience and knowledge; it has no ultimate reality since everything there consists of fleeting momentary appearances. The second is the world of reality, the reality behind the momentary appearances.

It is true that the Sarvastivadins denied the substantiality of the individual (pudgala). But compelled by the need to explain the problem of continuity resulting from the acceptance of discrete momentary dharmas, they came to believe in an underlying substratum (svabhava, dravya) considered to be eternal (sarvada asti, heng yu ).42 We agree with Ninian Smart that the "difficulties of this kind were one motive for the Realist school to insist strongly upon the existence of everything past, present and future: so that events could enter into relations with one another."43 By his interpretation of the term dharmata, Stcherbatsky seems to authenticate the Sarvastivada theory of eternal elements. A passage in the Madhyamikavrtti that had already appeared in the Pali Nikayas and the Chinese Agamas declares, "Whether the Tathagatas were to arise or whether the Tathagatas were not to arise, this nature of dharmas remains." (utpadad va tathagatanam anutpadad va tathagatanam sthitaivaisa dharmaridm dharmata).44 Stcherbatsky seems to have taken dharmata in this context to mean the "essence of dharmas," i.e., their svabhava as opposed to their causal characteristics (laksana).45 Therefore, he translates the word dharmata as " ultimate realities".46 And as the statement in the Madhyamikavrtti goes, these realities are eternal irrespective of whether the Tathagatas were to arise or not. As is pointed out below (chapter 5), dharmata (P. dhammata) refers to the causal connection between two dharmas rather than an underlying substratum of dharmas. If dharmata stands for the causal connection, it cannot mean an ultimate reality (dharmasvabhava) as the Sarvastivadins understood it, because Nagarjuna and his followers rejected the conception of svabhava, using the argument that svabhava is opposed to causality (see below).

This theory that all dharmas in their ultimate nature (svabhava)
exist during the three periods of time, past (atTta, ch'ii), present {vartamana, chin), and future (anagata, lai), gave the Sarvastivadins their appellation.47 Asserting that the theory of the Sarvastivadins represents the earliest phase of Buddhist thought, Stcherbatsky quotes a passage from the Samyukta Agama. The Buddha is asked what is meant by "everything exists" (sabbam atthi or sarvam asti or i ch'ieh yu), and he replies, " 'everything exists' means the twelve
'gateways of perception' (ayatana) exist."48 Stcherbatsky comments: "Now the twelve ayatanas are merely one of the many classifications of the elements of existence of matter and mind. The Sarvastivadin school admitted seventy five such elements. These elements were called dharmas."49 We believe that Stcherbatsky misunderstood the implications of the above text. The statement "everything exists means the twelve ayatanas exist," did not mean that the twelve ayatanas, past, present, and future, exist, as the Sarvastivadins would have interpreted. There is no reference to the past and the future, or, for that matter, to any conception of time. Early Buddhism does not deny present sense experiences, or therefore, their causes, namely, the sense organs and sense objects. This idea is clearly expressed in the sutra that immediately follows the one Stcherbatsky quoted: "The visual organ and the visible object produce visual consciousness and contact. As a result of visual contact (yen ch'u) there arise feelings that are either pleasant, unpleasant, or neutral. This is what is meant by 'everything exists.' " 50 This passage clearly shows that the reference is to present sense experiences and their causes.

Moreover, the Sarvastivada theory of 'everything exists'
{sarvam asti) not only implies the real existence of the twelve ayatanas, but also the existence of an eternal substratum (dravya)
or ultimate nature {svabhava). This idea is denied by the very sutra quoted by Stcherbatsky because it emphasizes that positing anything beyond the twelve ayatanas would be beyond the sphere of experience {fei ching chieh = avisaya). Thus, even the so-called 'elements' {dharma) that, according to Stcherbatsky, were con sidered ultimate realities by the Sarvastivadins are looked upon as being nonsubstantial (wu o = anatta). They have no substance (svabhava) that survives during the three periods.51 The theory that everything past, present, and future exists is contradicted by yet another sutra in the Nikayas and the Agamas.52 According to the analysis in this sutra, the Sarvastivada theory that dharmas past, present, and future exist oversteps the limits of linguistic convention. The version in the Pali Nikayas reads thus:
"There are these three linguistic conventions or usages of words or terms that are distinct—have been distinct in the past, are distinct at present, and will be distinct in the future —which [conventions] are not ignored by the recluses and brahmins who are wise. Which three? Whatever form (rupa) there has been, which has ceased to be, which is past and has changed is called, reckoned, and termed 'has been' (ahosi);
it is not reckoned as 'it exists' {atthi) nor as 'it will be' (bhavissatf). [The same is repeated with regard to the other four aggregates.] . . . Whatever form is not arisen, not come to be, is called, reckoned, or termed 'it will be' (bhavissati) and is not reckoned 'it exists' or 'it has been'. . . . Whatever form has been and has manifested itself is called, reckoned, or termed 'it exists' and is not reckoned 'it has been' or 'it will be.'"
This statement should have served as a warning for the Sarvastivadins to avoid the mistake of maintaining that dharmas in their ultimate reality exist during the past, present, and future. The Sarvastivada theory may therefore be taken as a new development in the history of Buddhist thought resulting from the acceptance of the theory of momentariness. If so, it would be unfair to attribute the misconception (viparyasa) of "perceiving a self in things without self, thinking of nonexistent things as existent" to all the so-called Hlnayana schools, although it may be attributed to the Sarvastivada school.53 Discussing the nonsubstantiality of the individual (pudgala),
Stcherbatsky writes: "The underlying idea is that, whatever be designated by all these names, it is not a real and ultimate fact, it is a mere name [sammuti] for a multitude of interconnected facts." 54 This view is, of course, based on the statement in the Pali Nikayas and the Chinese Agamas that illustrates the doctrine of nonsubstantiality: "Just as the word 'chariot' exists on the basis of the aggregation of parts, even so the concept of 'being' exists when the five aggregates are available." ss Examining the simile in that statement, it would seem that the chariot is unreal (nonsubstantial)
because it is merely a name for a multitude of different parts (wheels, axle) assembled in a certain way. But are the individual and separate parts real or substantial? On closer scrutiny one finds that they too are unreal or nonsubstantial in that they are causally produced, just like the whole, the chariot.

Therefore, the view was accepted that just as the individual is unreal, so the component parts, the aggregates, are unreal in that they have no substance (atman = svabhava), being subjected to becoming (bhuta), composition, (sahkhata), and causal production (paticcasamuppanna). Thus, anatman becomes a synonym of nihsvabhava. That the aggregates (skandha), taken not only in combination but also separately, are nonsubstantial is emphatically stated in the sutras.56 The Chinese Agamas seem to go further in maintaining that even the aggregates taken separately are nonsubstantial (wu o = anatta) and unreal (k'ung = suhha).51 It was this line of argument that was adopted by Nagarjuna to refute the Sarvastivada conception of reality. He devoted one whole chapter of the Mula-madhyamaka-karika to refuting the doctrine of the substantiality of the dharm as.58 Starting from the fundamental proposition in Buddhist philosophy that there is nothing in this empirical world that is not causally produced,59 Nagarjuna raises the question, "How could a contingent svabhava be possible?" 60 Candraldrti replies, "The concept of contingence
(krtakatva) and substance {svabhava) cannot be combined in one meaningful unity {asamgatartham), for they are contradictory
(parasparaviruddhatvad). According to realistic logic, the term svabhava has more or less the meaning of 'thing-in-itself' (jvo bhavo). Even with regard to this definition, nobody in the world would designate contingent reality as svabhava. So, for example, heat [as a property] of water [is a contingent reality and for that reason is not its svabhava]."
Then Candraklrti anticipates the reply of the opponent, which he states thus: "Svabhava is noncontingent (akrtaka), as for example, with heat as a property of fire, for in this example the inherence
{samparka) of the thing (padartha) and its attribute (antara) is not causally produced (ajanita). Therefore one can speak of svabhava."61 To this the Madhyamika replies:
It must be stressed, of course, that this acceptance of a noncontingent svabhava is only true from the point of view of common-sense experience (lokavyahara). On the contrary, we maintain that heat as a property of fire is not a svabhava, for fire is itself contingent (krtaka). It originates in correlation (sapeksata) with certain causes and conditions through the cooperation of various factors: the lens, fuel, the sun, or owing to the friction of pieces of wood. There is no heat independent of fire. So heat is also produced in correlation with causes and conditions and is therefore a contingent {krtaka). And being contingent, it cannot be a svabhava, just as heat of water cannot be a svabhava.62 Here, too, the Madhyamikas were attempting to show that the dharmas are devoid of substance {svabhava) because they are causally produced or are contingent. This is the very argument adduced in the Pah Nikayas and the Chinese Agamas to show the nonsubstantiality of the dharmas.

Moreover, the Madhyamikas quoted a statement of the Buddha to justify their rejection of the Sarvastivada conception of the eternality of the dharmas. Nagarjuna says: " Recognizing the problems of 'Being' and 'non-Being,' the Buddha has, in the Katyayana-sutra, rejected the two concepts 'it is' and 'it is not.' " 63 Candralarti points out that this sutra is studied in all schools of Buddhist thought.64 Also in the Nikayas and the Agamas the two concepts 'everything exists' and 'everything does not exist' are rejected because they are said to lead to a belief in permanence
{sassata-ditthi, cKang chien) and to a belief in annihilation {ucchedaditthi, tuan chien).65 For the Buddha, these were metaphysical problems. The Sarvastivada theory that dharmas in their own nature or substance {svabhava) exist during the three periods of time may be considered a result of metaphysical speculations on the problems of time and continuity.66 There is no doubt that the Sarvastivada theory leads to belief in permanence, although Stcherbatsky attempts to show that it does not.67 This becomes clear from the fact that Nagarjuna viewed the theory of svabhava in the same way as the Buddha viewed the Upanisadic conception of 'Being' (sat).68 Therefore, in his attempt to counteract the Sarvâstivâda conception of svabhâva, Nâgâijuna found it appropriate to quote the statement of the Buddha refuting the Upanisadic conception of 'Being' (sat).
Even a later Theravâda text such as the Kathâvatthu is unequivocal in its criticism of the Sarvâstivâda conception of the
'substantiality of dharmas' (dharma-svabhâva).69 Ignoring this explicit criticism in the Kathâvatthu, Murti says : "It is a mistake to think that the Mahâyâna schools reversed the denial of the soul and reaffirmed its reality. If anything, they were more thorough in carrying out the Nairâtmya doctrine. They denied not only the substance [of the individual, pudgala-nairâtmya] but also extended the denial to the elements (dharma-nairâtmya), which the Hmayâna schools had uncritically accepted as real." 70 The theory of the nonsubstantiality of the dharmas, as pointed out above, was not new to the Pali Nikâyas and the Chinese Âgamas. In the Pali Nikâyas we find specific references to the doctrine of the nonsubstantiality of all dharmas (dharma-nairâtmya) in the locution sabbe dhammâ anattâ.11 Not being able to accept that early Buddhism made such a denial, L. de la Vallée Poussin, quoting a passage from the Ahguttara Nikâya where this locution occurs, changed it to sabbe sahkhârâ anattâ, indirectly implying that it is either a misreading or a later interpolation.72 This, no doubt, is another example of the way in which a modern scholar may be tempted to interpolate early texts to suit his own interpretation of the ideas embodied in these texts. Fortunately, the Chinese Àgamas seem to support the statement in the Pali Nikâyas, for we find the exact equivalent of this statement (z chieh fa wu o) in several places.73 The conception of dharma in early Buddhism, as depicted in the Pali Nikâyas, the Chinese Àgamas, and the Theravâda Abhidhamma as represented by the Kathâvatthu, would therefore be much different from the Sarvâstivâda conception. Hence we maintain that it was the Sarvâstivâdins who propounded a theory of the substantiality of dharmas and that there is no justification for extending that criticism to the other Hïnayâna schools.

The second method of reconciling the doctrine of causal continuity with the theory of momentariness was adopted by the Sarvâstivâdins and by the Sautrântikas and the Theravâdins with slight variations. The Sarvâstivâdins accepted four moments of a conditioned thing (samskrta), one of which is the static (sthiti).74 This static moment was further defined according to its causal efficiency. "The production of a result (phalaksepa) by a dharma endowed with potency gained as a result of coming into [present] existence and the harmony of [external and internal] conditions, is said to be the causal efficiency (kantra).,'ls To maintain causal continuity among such momentary dharmas, they formulated the theory of immediately contiguous cause (samanantarapratyaya, teng wu chien yuan).

The Sautrantikas, who did not recognize the static moment,76 believed that a dharma disappears immediately as it appears.77 For them, what is perceived as a static moment or moment of duration is a series of successive moments with a continuous flow.78 They too employed the theory of immediately contiguous cause to explain the connection between two successive moments. Yet the conception of momentariness presented other problems. If existence is a series of successive moments, how can birth, decay, and destruction be explained? The Sautrantikas attempted to solve this by maintaining that birth is the beginning of a series; decay represents the fact that in a given series each successive moment is slightly different from the preceding one; and lastly, destruction is the end of the series.79 The causality of each individual moment in a series is then reduced to invariable antecedence.80 But still they had to explain the origin or beginning of a series. It was to explain this problem that they presented the theory of causation that came to be known as abhiitva bhava utpada, or pen wu chin yu sheng, according to which the first member of the series being nonexistent
(iabhutva, pen wu) comes into existence {utpada, sheng).61 This theory of causation will be discussed in detail in chapter 7.

The Sautrantika theory, which denies the static moment
(sthiti-ksana) of a dharma, was given authority and sanctity when it was attributed to the Buddha himself. Kamalasila, in his commentary on the Tattvasahgraha, records the Buddha as saying:
"All forces are instantaneous. But how can a thing that has no duration nevertheless have the time to produce something?" The following answer is given: "That is because what we call 'existence' is nothing but efficiency (kriya), and this very efficiency is called a creative cause." 82 This means that a dharma represents an eternally changing process.

Yet the problems raised against the theory of moments remained unsolved. The Sarvastivada solution to the problem, which was based on acceptance of a substance (svabhava), although contrary to the Buddha's doctrine of nonsubstantiality (anatma), was nevertheless more straightforward. On the other hand, the Sautrantikas, while denying substance, merely assumed the causal efficiency of the momentary existence. But this very assumption was being questioned.

For some time the Theravada tradition, as represented by the Pali Abidhamma Pitaka, did not have to face these problems. The reason was that the metaphysical theory of moments did not appear in the Theravada tradition until the time of the great commentator Buddhaghosa (see chapter 7). Therefore, when the doctrine of momentariness appeared in the Theravada tradition, the solutions presented by the Sarvastivadins and the Sautrantikas came along with it. The Sarvastivada doctrines appeared in disguise in the works of Buddhaghosa, Dhammapala, and the writers of the Polonnaruva period. The Sautrantika ideas came to be advocated by Ananda of MulatTka fame.

The foregoing account shows how the Sarvastivadins and the Sautrantikas, in their attempt to present a logical analysis of the doctrine of impermanence (anicca), came to accept a theory of moments, which in turn led to several theories not consistent with early Buddhism. While the Sarvastivadins accepted a belief in an underlying substratum in dharmas, thus going against the nonsubstantialist (anatma) standpoint of early Buddhism, the Sautrantikas were led to adopt a theory of causation that was very similar to the one rejected by the Buddha (see chapter 7). These differences suggest that the teachings of the Abhidharmakosa of Vasubandhu are not representative of the earliest phase of Buddhism.

What, then, is the theory of impermanence found in the early Buddhist texts? Hardly any evidence can be gathered from the Pali Nikayas and the Chinese Agamas to support the view that things were considered to be momentary (ksanika, ch'a na). We do not come across any statement such as, "All forces are momentary." 83 The theory of momentariness is not only foreign to early Buddhism but is contradicted by some statements in the Nikayas and the Agamas. For example, two suttas in the Samyukta called Assutava describe how a man should give up attachment to the physical body made up of the four primary existents because the body grows and decays, comes into being and perishes. Comparing the vacillation of the mind with the change taking place in the physical body, it continues: "This physical body made up of the four primary existents exists for one, two, three, four, five, ten, twenty, thirty, forty, fifty, a hundred or more years. That which is called the mind, thought, or consciousness arises as one thing and ceases as another whether by night or by day." 84 That description of mind and body is not inspired by a theory of momentariness. In fact, it seems to refute the idea of momentariness when it says that the physical body is comparatively more stable than the mind. Physical bodies are experienced as enduring for some time, although they are subject to change and decay, which change is not perceived as occurring every moment. N or is there any suggestion that the mind is subject to momentary changes. The suttas merely emphasize the relative speeds at which the body and mind change.

This is an empiricist account of change. The statement, "All conditioned things are impermanent," it has been observed, "is not given as a result of metaphysical inquiry or of any mystical intuition, but' a straightforward judgment to be arrived at by investigation and analysis. It is founded on unbiassed thought and has a purely empirical basis." 85 Buddhaghosa's commentary on these suttas betrays an overwhelming influence of the conception of momentariness. His attempt, to again use Stcherbatsky's words, "is to seize the infinitesimal." Buddhagosa maintains: "Just as the flame of a burning lamp, without leaving the area of the wick, breaks up then and there and when it burns or flickers in succession throughout the night it is called a lamp, even so, taking the succession [of states]
this body is presented as enduring for a long time." 86 Here he is trying to explain the perceived duration of the body by resorting to a theory of moments. It is reminiscent of the Sautrantika solution to the problem of duration. In fact, the Sphutarthabhidharmakosavyakhya uses the simile of the flame to explain the momentariness of existents (bhutani). The explanation runs th u s: "The movement of the stream of elemental properties, whose nature is such that it appears in successive places, is like the flame because of its momentariness. The comparison with a flame is given because the momentary character of a flame is an established fact." 87 With regard to mental phenomena, Buddhaghosa says: "There is no single thought that can endure for one night, or even for one day, for during a moment of the snapping of the fingers, there arises several hundred thousand myriads of thoughts." 88 Does this explanation of Buddhaghosa represent the original position with regard to the theory of impermanence? In view of the empiricist attitude of early Buddhism, the answer would be no. According to early Buddhism, things are impermanent, not because they are momentary, but because they are characterized by birth (uppada, ts'ung ch'i), decay or transformation (thitassa ahhathatta, ch'ien p 'ien), and destruction (vaya, mieh chin).89 Whatever is born is impermanent,90 since that which is born is sure to perish. W hat is conditioned or compound (sahkhata) is also impermanent,91 and so is that which is subject to decay.92 In short, impermanence is a synonym for arising and passing away, or birth and destruction.93 This pattern of things—that all conditioned things are impermanent (anicca, wu chang), unsatisfactory (dukkha, k 'u), and that all dharmas are nonsubstantial (anatta, wu o)—is eternal.94 The term dhamma, when applied to empirical things, is always used in the sense o f'causally conditioned dhammas' (paticcasamuppanna-dhamma). The realization of this fact may have prompted Nagarjuna to declare: "There is nothing in this world that is not causally produced." 95 In the famous dictum that is held to summarize the Buddha's teachings, it is said: "The Great Recluse says that the Tathagata has spoken of causally produced (hetuppabhava) dhammas, their cause [or causes] and their cessation." 96 The Nikayas and the Agamas abound in statements such as:
"sabbe sahkhara anicca, sabbe sahkhara dukkha, sabbe dhamma anatta."91 The last statement seems significant. Here the occurrence of the two terms, sahkhara and dhamma, is intriguing. Do they refer to two different things? Or, are they synonymous?

The term sahkhara, when it refers to a psychological fact, certainly means 'disposition.' But there are occasions when it is used in a very broad sense to refer to everything in this world. One prominent example is from the Maha-Sudassana-suttanta where, referring to the glories of the famous king of the past, MahaSudassana, his cities, treasures, palaces, elephants, horses, carriages, women, etc., the Buddha says: "Behold, Ananda, how all these things {sahkhara) are now dead and gone, have passed and vanished away. Thus impermanent, Ananda, are the sahkharas', thus untrustworthy, Ananda, are the sahkharas. And this, Ananda, is enough to be weary of, to be disgusted of, to be completely free of, such sahkharas."98 The past participial form sahkhata (wei) is generally used to refer to anything that is 'compound,' 'organized,' or deliberately 'put together,' hence synonymous with kata (tso) 'made,' 'done.'
Thus, sahkhata describes anything in which m an's dispositional tendencies (sahkhara) has played a major role. In such cases, the things that are carefully selected for compounding or putting together, that is, the components, may also be called sahkharas. In the Nikayas and the Agamas, the human personality consisting of the five aggregates is called 'a bundle of components' (sahkharapuhja) ." Therefore; it is possible that sahkhara in the context above is intended to denote all the aggregates, including 'dispositions.'
In contrast to these sahkharas, there are things that are not so deliberately compounded, but that are 'natural' or 'causally conditioned' (paticcasamuppanna). This means thatpaticcasamuppanna dhammas are to be distinguished from sahkharas (or sahkhata dhammas). Thus all sahkharas are dhammas, but not all dhammas are sahkharas, because some dhammas are natural occurrences. If sahkhara is understood in this broad sense, then the above statement may be translated as: "All components are impermanent; all components are unsatisfactory; all dhammas are nonsubstantial."
If dhamma is all comprehensive (and includes even sahkharas),
the statement sabbe dhamma anatta will imply that all things, including the sahkharas, are nonsubstantial. Therefore, it is not possible to maintain that, according to early Buddhism, the 'aggregates' (khandha) are 'substantial.' In spite of the statement "sabbe dhamma anatta'' occurring in the Nikayas and the Agamas, where dhamma include all sahkharas or 'components' such as the aggregates {khandha), the editors of the PTS Pali-English Dictionary,
commenting on the nature of the khandhas, say: " Being the 'substantial' factors of existence, birth and death depend on the khandhas." 100 In a similar tone, Murti says: "As a matter of dialectical necessity then did Buddha formulate, (or) at least suggest, a theory of elements. The Mahayana systems clearly recognize this dialectical necessity when they speak of the pudgalanairatmya—the denial of substance (of the individual)—as intended to pave the way for Absolutism. Sunyata is the unreality of the elements as well (dharma-nairatmya)." 101 These comments imply that according to early Buddhism the individual (pudgala) is unreal but the components (khandha or skandha) are real or have substance. But that view is contradicted by a large number of statements in the Nikayas and the Agamas that emphasize the unreality or nonsubstantiality of the khandhas as well. The most explicit denial of the reality or substantiality of 'components' is quoted by the Madhyamikas themselves: "All form is comparable to foam; all feelings to bubbles; all sensations are mirage-like; dispositions are like the plantain tru n k ; consciousness is but an illusion; so did the Buddha illustrate [the nature of the aggregates]." 102 Moreover, the aggregates {khandha) are considered to be causally produced (hetum paticca sambhuta).103 The characteristics of the dhammas are said to be found also in the causes. They are said to be impermanent (anicca), conditioned or compounded (sahkhata), and causally produced (paticcasamuppanna) and are therefore not substantial.104 Speaking of these three characteristics, C. A. F. Rhys Davids has rightly rem arked: "Hereby we see how interwoven are these three concepts. . . . And they are held in such a way as to elude the metaphysical problems of (a) realism and idealism, and (b) of mechanism and atomism." 105 Thus we see that the causes and the caused have been described in similar terms. Therefore, it is possible to conclude that early Buddhism, as depicted in the Nikayas and the Agamas, does not posit the substantiality of dhammas as the Sarvastivadins did. It was the awareness of this fact that led CandrakTrti to make the following categorical statement: "Indeed, the Tathagatas never posit the real existence (astitva) of a soul or the aggregates." 106 That statement directly contradicts the view expressed by Murti that the Buddha formulated a theory of elements as a dialectical necessity.

Another rendering of the term dhamma that may mislead a student of early Buddhism is 'phenomena.' There are two different theories of phenomenalism: (1) that all knowledge is limited to phenomena (i.e., things and events in time and space) and that we cannot penetrate into reality itself, and (2) that all we know is phenomena, that is, reality present to consciousness, either directly or reflectively, and that phenomena are all that is to know, there being no 'thing-in-itself' or object out of relation to consciousness.107 'Phenomenon' in the first sense may be a better rendering of the Sarvastivada conception of dharma-laksana. That is because it is opposed to dharma-svabhava, which may be considered the equivalent of 'thing-in-itself,' thus coming under the category of the unknowable. 'Phenomenon' in the second sense may come very close to the conception found in the Nikayas and the Agamas, where the emphasis is on sense data as the content of our empirical knowledge, with denial of any real substratum behind phenomena.108 Because early Buddhism did not recognize an underlying substratum in dhammas and because all dhammas were considered to be empirical, the rendering of the term as "die empirischen Dinge" by Wilhelm and Magdalene Geiger (see above) seems satisfactory.

Stanislaus Schayer has put forward a theory of dharma different from those discussed above: "The extension of the term dharma to all elements of the mundane and supramundane existence is an innovation of the later scholiasts and . . . the antithesis of rupa and dharma is a survival of pre-canonical Buddhism which actually divided the world into two opposite categories of rupa and dharma."109 He bases his conclusion on (1) the theory of "two bodies" (dvikaya), namely, the rupa-kaya, representing the physical personality of the Buddha, and dharma-kaya, the spiritual personality and (2) the use of the term dharma in the phrases dharm'-
ayatana and dharmadhatu, where it represents mental objects or concepts. If his theory is correct, then the use of the term dhamma in the famous passage that begins "ye dhamma hetuppabhava . . ." (see note 101, above) and the statement summarizing the Buddha's teaching, ("paticcasamuppadan ca vo bhikkhave desissamipaticcasamuppanne ca dhamme ["I shall preach to you, O monks, on causality and causally conditioned phenomena"]) may have to be taken as referring to 'mental facts' only. But this is not so, for we find even 'form' {rupa), especially namarupa (ming se) denoting not only the psychic but also the physical personality, included under the category of paticcasamuppanna-dhamma.110 There dhamma is used in a very broad sense to include physical as well as mental facts. Therefore, the relationship between dhamma as signifying 'concepts' and paticcasamuppanna-dhamma as implying everything empirical can

be diagrammatically represented thus:
In light of the above analysis of the conception of dhamma,
it is difficult to agree with Stcherbatsky, Murti, Schayer, and others who interpret early Buddhism as a form of radical pluralism. Moreover, according to early Buddhism, pluralism {nanatta) and monism (ekatta) are metaphysical views,111 which, as the commentator has rightly remarked, are similar to or associated with annihilationism and etemalism.112

V. The Causal Principle And Its Validity

IN CHAPTER 4 we investigated the nature of causally produced dhamma, which is one aspect of the Buddha's discovery. This chapter examines the second aspect of the Buddha's discovery, the pattern according to which change takes place in things {dhamma). The change in things is not haphazard or accidental. It takes place according to a certain pattern, and this pattern of things, this orderliness in things, is said to be constant. It is a cosmic truth eternally valid and independent of the advent of the Tathagatas.1 This pattern has been variously described as 'conditionality' {idappaccayata, sui shun yuan ch'i) and as 'causality' {paticcasamuppada,
yin yuan fa ) (see chapter 3). Thus, according to the Buddha's philosophy, there are no accidental occurrences; everything in the world is causally conditioned or produced {paticcasamuppannam).
The realization that every occurrence is a causal occurrence is said to clear the mind of all doubts, a characteristic of the state of perfect knowledge and enlightenment.2 This truth the Tathagata discovers and comprehends; having discovered and comprehended it, he points it out, he teaches it, lays it down, establishes, reveals, analyzes, clarifies it and says "look." 3 The significance of the discovery is such that, according to the Buddhist texts, he who perceives the causal law sees the truth, and he who sees the truth perceives the Buddha.4 This theory of causation has been called the 'middle path'
(majjhima patipada, majjhimena, chung tao),5 because it steers clear of the two extremes (anta, pieri) with regard to causation; selfcausation, which leads to a belief in eternalism, and external causation, which leads to a belief in annihilationism (see chapters 1 and 2).

The general Buddhist formula of causality is often stated in the following manner:
Pali version.

"Imasmim sati idam hoti, imassa uppada idam uppajjati." " Imasmim asati idam na hoti, imassa nirodha idam nirujjhati." 6 Buddhist Sanskrit version.

" Imasya sato, idam bhavati, imasya asato idam na bhavati." 7
"Imasyotpadad idam utpadyate, imasya nirodhad idam nirudhyati." 8 Chinese version I
"Tz'u yu ku pi yu, tz'u ch'i ku pi ch'i."
"Tz'u wu yu ku pi wu yu, tz'u mieh ku pi mieh." 9 Chinese version II
"Jo yu tz'u tse yu pi, jo wu tz'u tse wu pi." 10 These may generally be rendered into English as follows:
When this is present, that comes to be; from the arising o f this, that arises.

When this is absent, that does not come to be; on the cessation o f this, that ceases.

There are two main points in the versions quoted above that should be clarified at the outset. First, in the Pali version of the formula, the same demonstrative adjective "this" {idam) is used, not the pair "this" and "that" {idam, asau), as in some of the Buddhist Sanskrit versions, although we have used the two adjectives in our English translation. Commenting on this question C. A. F. Rhys Davids said: "This should not lead the reader to see in the formula a set of merely identical propositions. Pali diction does not distinguish between the two terms in our way; but the context invariably shows that there are two terms and not one." 11 Moreover, in any statement of causation it is held that the referents, in this case, those denoted by the demonstrative pronouns, "must differ from one another in at least one respect." 12 The Buddhist Sanskrit version of the Pratityasamutpada-sutra, discovered in fragmentary form, uses the two words "this" (idam) and " that" (asau).13 In the English translation above we have followed the Chinese translators, who almost always distinguish the two terms by the use of the two characters tz'u ( " this" , " these" ) and pi
( " that" , " those" ).

Second, it has been argued that the form of the causal principle—"C, therefore E," or " E because C "—is inadequate to pour causation into because it has the form of an explanatory statement.

It suggests only a reason, and does not express the idea of conditionality.14 This is true with regard to the Chinese version I, where the character ku denotes only a reason. But the locative absolute construction in Pali and Sanskrit (as in the statements imasmim sati idam hoti or asmin satidam bhavati), and the use of the characters jo and tse in the Chinese version II, seem to express the idea of conditionality in a more definite form. Those versions may therefore serve as a corrective to the Chinese version I.

Causality or causation (paticcasamuppada, yin yuan fa), as described in the Samyukta, is synonymous with the causal nexus, for example, as between 'ignorance' (avijja, wu ming) and 'dispositions' (sahkhara, hsing). This causal nexus is said to have four main characteristics, (1) 'objectivity' (tathata, ju fa erh), (2) 'necessity' (avitathata, fa pu li ju), (3) 'invariability' (anahhathata,
fa pu i ju), and (4) 'conditionality' (idappaccayata, sui shun yuan c/z'/).ls A. B. Keith, who made a persistent attempt to restrict the Buddhist theory of causation to the so-called chain of causation, seems to have overlooked the importance of this passage when he said: " The lover of causation would have insisted on each link; for the practical Buddhist all that was necessary was to show that evil was caused and the minor details could be left vague." 16 Let us examine these four characteristics of the causal nexus in detail.

The first characteristic, 'objectivity' (tathata, ju fa erh), describes the status of causation in Buddhism. We have already shown that some of the Upanisadic thinkers considered change, and consequently causation, a mental construct, a purely subjective phenomenon (see chapter 1). For them causation had no ontological status; it was a purely epistemological category belonging solely to the description of human experience. If causation were only a mental construct, then it would be a concoction or fabrication of man, a hypothesis without any real basis. Hence a very pertinent question is raised in the Samyukta Agama11 as to who constructed or fabricated this theory of causality, the Buddha or some other person. Buddha's reply to this question was: "It is neither made by me nor by another. Whether the Tathagatas were to arise in this world or not, this pattern of things {fa chieh18 - dhammadhatu) is eternally existent. Concerning this [pattern of things] the Tathagata has insight, is fully enlightened." The Buddha's reply certainly emphasizes the objective validity of the causal law.

The 'objectivity' of causation is further illustrated in the Buddha's comparison of its discovery to the discovery of a buried city.19 The metaphor is stated thus:
Suppose a man faring through the forest, through the great woods, sees an ancient path, an ancient road traversed by men of former days. And he goes along it and sees an ancient city, a former prince's domain, where men of former days lived, a city adorned by gardens, groves, pools, foundations of walls, a beautiful spot. . . . Just so did I behold an ancient path, an ancient way traversed by former Buddhas. . . . Following that path, I came to understand fully decay and death, their arising, their cessation and the path leading to their cessation.20 Of the four Noble Truths discovered by the Buddha, the second and the third refer to the theory of causation. These references would be sufficient to show that according to early Buddhism, as embodied in the Nikayas and the Agamas, causation is not a category of relations among ideas but a category of connection and determination corresponding to a feature of the actual world, both subjective and objective, so it has an ontological status. It is a component of experience because it is an objective form of interdependence in the realm of nature.

Unfortunately, Buddhaghosa's explanation seems to have missed the point completely. According to Buddhaghosa: "As those conditions alone, neither more nor less, bring about this or that event, there is said to be 'objectivity.'" 21 Following this, Nathmal Tatia has translated the word tathata as "regularity of sequence" and considers it to be the positive characteristic of the causal law, while avitathata, rendered as " absence of irregularity,"
is considered to be the negative characteristic.22 According to our understanding, this does not bring out the real implications of the term tathata. Tathata (from tatha > tatha) in the early Buddhist texts means 'correspondence,'23 and as a characteristic of causation, therefore, it is used to mean that causation is not merely an idea or thought-construction without any objective validity, but an idea that corresponds to what is found in nature. This is very clearly expressed in the Chinese rendering of the term as ju fa erh. This interpretation of the conception of tathata finds support in M ahayana Buddhism, wherein the term is used to mean 'true essence,' 'actuality,' 'truth,' or 'ultimate reality,' and is employed as a synonym of satya,24 The use of tathata as a characteristic of causation seems to be very significant, especially at a time when causation was either considered a thought-construction or was completely denied. In this context, therefore, the word can be interpreted as "what corresponds to reality."
The second characteristic of the causal nexus, avitathata (fa pu li ju), has been rendered as 'necessity' which conforms to the explanation given by Buddhaghosa: " Since there is no failure, even for a moment, to produce the events that arise when the conditions come together, there is said to be 'necessity.' " 25 Whether the concept of necessity should be included in an adequate formulation of the causal principle has been the subject of much discussion in recent years. The traditional anthropomorphic meanings attached to the word 'necessity' have been rejected, and the empiricist view that it denotes a lack of exception or the existence of regularity has been accepted. The word avitathata, even according to Buddhaghosa's definition, means necessity in the sense of lack of exception. It has been observed that " If the notion of necessity is stripped of its anthropomorphic and fatalistic associations, it is reduced to . . . constancy and uniqueness," 26 a view that is also implied in the early Buddhist conception of avitathata.
The third characteristic of causation or the causal nexus is
'invariability' (anahhathata,fapu i ju), which Buddhaghosa defines in the following manner: " Since no effect different from [the effect]
arises with [the help of] other events or conditions, there is said to be 'invariability.'" 2'' This definition should not be understood as implying "same cause, same effect" or "every event has a cause, and this cause is always the same." A theory of causation maintaining that if the same cause is repeated, the same effect will result, is said to have the shortcoming of emphasizing the sameness of causes and effects.28 It has no scope at all, " since the same cause never occurs exactly." 29 But 'invariability' in the early Buddhist texts does not refer to the nature of causes and effects. On the contrary, it refers to the nature of the relation existing between causes and effects. It only states that there is a constant relation between causes of certain kinds and effects of certain kinds. It emphasizes the constancy of the relation rather than the sameness of causes and effects.

The fourth characteristic of causation is 'conditionality'
(idappaccayata, sui shun yuan ch ' z). That it was used as a synonym of 'causation' (paticcasamuppada) in the early Buddhist texts (see chapter 3) points to its great importance. Buddhaghosa's definition runs thus: " From the condition or group of conditions that gives rise to such states as decay and death there is said to be 'conditionality.'" 30 Buddhaghosa's explanation seems to imply that a thing comes into existence only if the necessary condition or group of conditions is available. Conditionality as a characteristic of causation is still more important in that it prevents causality from being considered a form of strict determinism. It places causality midway between fatalism (niyativada) and accidentalism (yadrc-chavada) (see chapter 2). Fatalism, or strict determinism, and accidentalism are said to be the two extreme forms of lawlessness.31 If so, conditionality may be called the 'middle path' because it avoids the two extremes, namely, the unconditional necessity asserted by fatalism and the unconditional arbitrariness assumed by accidentalism.

It has been observed that "the statements of causal laws, and in general, scientific laws, do not assert that something will inevitably happen under all circumstances, regardless of the past or the present conditions. Quite on the contrary, statements of causal laws assert that if and only if certain conditions are met with certain results would follow." 32 While this is implied in Buddhaghosa's definition of conditionality, it is also clearly expressed in the use of the locative absolute phrase in the general formula of causation— " When this exists, that exists"—where the word when represents the conditional particle " if."
Keith confused the conception of conditionality in early Buddhism with determinism when he made the following accusation against Buddhism:
Moreover, man has the power to act; strange as it may seem when one ground of the denial of the self is remembered, and the apparent determinism of the Chain of Causation, the Buddha has no doubt whatever that the determinism of Makkhali Gosâla is the most detestable of all heresies. The position is the more remarkable because one of the arguments in the Canon and later against the existence of the self is that such a thing must be autonomous, while all in the world is conditional and causally determined. But the issue is solved by the simple process of ignoring it and Buddhism rejoices in being freed from any error of determinism to menace moral responsibility.33 Oltramare seems to agree with Keith, for he puts forward a similar theory that " Le Bouddhisme a poussé jusqu'aux dernières limites son explication phénoméniste et déterministe des choses." 34 These views are based entirely on a wrong understanding of the conception of causation in early Buddhism and its explanation of the problem of moral responsibility. This question will be examined in detail in chapter 6.

'Necessity,' (avitathata, fa pu li ju ) in the sense of lack of exception and invariability (ananhathata, fa pu i ju), is also expressed in the positive aspect of the general formula of causation :
" When this exists, that exists; on the arising of this, that arises." On the other hand, conditionality is emphasized by the negative aspect, which is expressed by the second part of the causal formula: " When this does not exist, that does not exist; on the cessation of this, that ceases."
In addition to these four characteristics of causation, which are specifically mentioned in the early Buddhist texts, there are a few other im portant characteristics that are not directly stated but are clearly implied by the causal principle. We have seen how early Buddhism criticized the idea of self-causation, which implied the prior existence of the effect (satkâryavâda), and the idea of external causation, which accounted for a previously nonexistent effect (asatkaryavada) (see chapters 1,2). The rejection of these two views may suggest that the Buddhist theory of causation expresses merely the constant conjunction of two things.35 The first part of the general formula of causation, " When this exists, that exists," certainly expresses the idea of constant conjunction or association.

While criticizing self-production and production of a nonexistent effect, however, early Buddhism was not prepared to reduce causation to constant conjunction, especially to constant association of successives, i.e., regular succession, because such a form of causation " does not represent a category of determination through change, but only as an antecedent." 36 Empiricists such as David Hume have reduced causation to mere succession or constant conjunction of impressions—supposedly based on experience. But such a reduction of causation to mere succession is meaningful if experience is analyzed only in terms of momentary impressions, moments being reckoned as the smallest and indivisible units of time. A similar definition of experience is encountered in a later school of Buddhism, the Sautrantika (see chapter 4), which as a result of accepting the theory of moments ( ksana), reduced causation to a mere succession of momentary appearances. In fact, this school's theory led to a denial of causation, as did Hume's. It is true that early Buddhism depended on experience (i.e., 'contact,' phassa, ch'u or 'sensation,'
vedana, shou) to verify the nature of reality. But such experience was not considered momentary (see chapter 4). Therefore, the causal connection itself becomes an object of experience. Thus, the statement, " When this exists, that exists," is immediately followed by, " On the arising of this, that arises" (imassa uppada idam uppajjati), thereby combining the principle of lawfulness or constant conjunction with that of productivity. The use of the word
"arising" (uppada, ch'i) with the term "exists" (bhavati,yu) is therefore not mere repetition or only the statement of a concrete formula as opposed to the abstract formula given first.37 It represents, rather, a deliberate effort to include the idea of productivity in the statement of causation. This is further exemplified by the use of
" dependent arising" or "dependent origination" (paticcasamuppada) to express the idea of causation.38 Even the Buddhists of a later date who had accepted the theory of momentariness and emphasized the constant conjunction of things attempted to accommodate the idea of production when they defined a momentary thing as having the capacity to produce the effect (karitra or arthakriydkdritva) (see chapter 4).

Another interpretation of the statement, " When this exists, that exists," is given by Nagarjuna in RatnavalT. He finds in the statement only the idea of relativity, not active causation. The example he gives is that the idea of shortness exists only in relation to the idea of length.39 The determination of a thing or object is possible only in relation to other things or objects, especially by way o f contrast. Nagárjuna maintains that the relationship between the ideas of 'short' and 'long' does not owe to intrinsic nature
(svabhava).A0 This rare interpretation of the causal principle is not completely foreign to the Pali Nikayas and the Chinese Agarnas, for in one place in the Samyukta the idea of relativity is clearly expressed: " That which is the element of light… is seen to exist on account of [in relation to] darkness; that which is the element of good is seen to exist on account of bad; that which is the element of space is seen to exist on account of form (rwpa)." 41 Comparing the theories of causation advanced by the practical agent and the scientific investigator, Stebbing says: "The practical agent, however, is content with a relation that is determinate only in the direction from cause to effect: whenever X occurs, E occurs. Such a relation may be many-one; given the cause, the effect is determined, but not conversely. But a scientific investigator wants to find a relation that is equally determinate in either direction, that is, he seeks a one-one relation: whenever X occurs, E occurs, and E does not occur unless X has occurred." 42 The general statement of causation, "Whenever this exists, that exists or comes to be,"
when coupled with the negative aspect, " Whenever this does not exist, that does not exist or come to be," seems to establish a oneone relation which, according to Stebbing, is a scientific theory of causation.

Apart from the one-one relation discussed above, we come across the "practical commonsense view," 43 which implies the existence of a plurality of causes. It has been noted that: " When a plurality of causes is asserted for an effect, the effect is not analysed carefully. Instances which have significant differences are taken to illustrate the same effect. These differences escape the untrained eye, although they are noticed by the expert."44 In the Dvayatdnupassana-sutta of the Sutta-nipata the problem is raised of how suffering (dukkha) originates and how it can be ended.45 The Buddha replies that it is due to the substratum of rebirth (upadhi). But the Buddha seems to sense the interlocutor's wish to know of other causes, for he says that according to another standpoint
(amena pariyayená) ignorance (avijja) is the cause of suffering.

Then he proceeds to enumerate ten different causes, which are shown in Figure 2. This is an instance of a many-one relation that is determinate in one direction only, from cause to effect.

Thus, the causal principle as stated in the Pali Nikáyas and the Chinese Ágamas seems to include all the features of a scientific theory of causation—objectivity, uniqueness, necessity, conditionality, constant conjunction, productivity, relativity—as well as one-one correlation. But the existence of such "practical commonsense views" side by side with a philosophically advanced theory may confuse the student of Buddhist thought. Such confusion can be avoided if we distinguish the different types of people to whom the Buddha's teachings were addressed. They were meant not only for those who were philosophically mature and spiritually advanced, but also for untrained (sekha), ordinary people (puthujjana). The Buddha was reluctant to confuse the minds of the latter speaking of highly philosophical theories. His was a gradual path of instruction; hence, during the initial stages of instruction, the Buddha spoke to an ordinary man in terms intelligible to him.

The question of the truth about causation and the validity of the causal law was discussed briefly when the characteristic of
'objectivity' (tathatajufa erh) was examined. A detailed examination of this problem seems to be in order because of the doubts raised on this issue at every stage in the history of philosophy, both Eastern and Western. In the history of Indian philosophy before the Buddha, the first to deemphasize the principle of causation was the idealist school of the Upanisads, which, as was pointed out in chapter 1, denied change and therefore causation. This was because of the Upanisadic view that reality (atman) is permanent and eternal. Thus, the denial of an entity such as atman and emphasis on change as a matter of fact opened the way to fruitful speculation regarding causality. Such speculation actually gave rise to the very significant theory of the Materialists, that of inherent nature
(svabhava). Unfortunately, because of their extreme aversion to the idealistic metaphysics of the Upani§ads, the Materialist thinkers not only rejected belief in a soul {atman), but even denied the truth and validity of mental phenomena. Their suspicions about the different sources of knowledge such as perception and inference led them to the metaphysical theory of inherent nature {svabhava) as an explanation of the pattern of change in physical phenomena.

Following a 'middle path,' the Buddha emphasized the objective validity of the causal propositions and the possibility of their verification through perception, both normal and paranormal, and through inductive reasoning. An attempt to find out the nature of the causal law in Buddhism, therefore, involves an examination of the epistemological standpoint adopted by the Buddha. The most thorough analysis of the early Buddhist theory of knowledge based on the Pali Nikayas has been done by K. N. Jayatilleke.46 W ithout going over trodden ground we shall confine ourselves only to a few problems connected with the verification of the theory of causation.

Jayatilleke maintains that inductive inferences in Buddhism are based on a theory of causality.47 But according to some modem epistemological theories, the theory of causation itself is based on inductive inference, thus leading to a circularity in the argument.48 In light of the Buddhist theory that will be explained in the following pages, we propose to distinguish between 'causation' and 'causal uniformity' or 'causality' and maintain that only causal uniformity or causality is based on inductive inference and that causation itself is given in experience. In the Western world, the view that 'causation' is based on inductive inference seems to be a consequence of the Humean analysis of experience. A detailed investigation into this problem is essential to understanding the status of causation and causal uniformity or causality in Buddhism.

Empiricism in modem Western philosophy is said to have started with John Locke and through George Berkeley reached its culmination in David Hume. Locke, appealing to experience, attempted to eliminate the Cartesian dualism as well as a belief in
'substance.' But Berkeley came dangerously close to positing a
'mental substance.' Hume, in his endeavor to reject the belief in a
'mental substance,' fell back on the introspective method. Thus, Hume came to adopt the Cartesian method of investigation in his desire to eliminate the belief in a 'self' (or substance), which he considered pernicious. To illustrate this method of Hume, let us quote from his Treatise.
For my own part, when I enter most intimately into what I
call myself, I always stumble upon some particular perception or other, of heat or cold, light or shade, love or hatred, pain or pleasure. I never catch myself at any time without a perception, and never can observe anything but the perception. When my perceptions are remov'd for any time, as by sound sleep; so long am I insensible of myself, and may truly be said not to exist. And were all my perceptions remov'd by death, and cou'd neither think, nor feel, nor see, nor love, nor hate after the dissolution of my body, I shou'd be entirely annihilated, nor do I conceive what is farther requisite to make me a perfect non-entity.49 The method adopted by the early Buddhists to reject belief in a 'self' (atman) was very similar to that adopted by Hume. But the Buddha, and the early Buddhists, did not arrive at the conclusion to which Hume arrived, indicated in the second part of the paragraph quoted above. We pointed out above that in Buddhism the rejection of the 'self' did not lead to annihilation. In fact, the causal process (which was referred to as the 'middle' between the two extremes of etemalism and annihilationism) was considered sufficient to explain the continuity of a thing without positing a
'self' or a 'substance.' But Hume was unable to accept that sort of idea because he considered causation nothing but a succession of discrete momentary impressions. Let us look at his argument:
I may venture to affirm of the rest of mankind, that they are nothing but a bundle or collection of different perceptions, which succeed each other with an inconceivable rapidity and are in a perpetual flux and movement. Our eyes cannot turn in their sockets without varying our perceptions. Our thought is still more variable than our sight; and all our other senses and faculties contribute to this change; nor is there any single power of the soul, which remains unalterably the same, perhaps for one moment. The mind is a kind of theatre, where several perceptions successively make their appearance; pass, repass, glide away, and mingle in an infinite variety of postures and situations.50 Though he declared that "Our thought is still more variable than our sight," Hume does not seem to have distinguished between the different patterns of change available to experience. For example, change taking place in a flame appears to experience as being more rapid than change taking place in a piece of stone. Similarly, the pattern of change available or observable in thought is different from that of the external world. His analysis of space is found in the following passage from the Treatise:
The table before me alone is sufficient by its view to give me the idea of extension. This idea is then borrow'd from, and represents some impression, which this moment appears to the senses. But my senses convey to me only the impressions of colour'd points disposed in a certain manner. If the eye is sensible of any thing farther, I desire it may be pointed out to me. But if it be impossible to shew any thing farther, we may conclude with certainty, that the idea of extension is nothing but a copy of these colour'd points and the manner of their appearance.51 Hume reached this conclusion regarding the nature of extension, according to which an object is analyzed in terms of points
(similar to the atoms, paramánu, posited by Indian thinkers) after his speculation regarding the nature of time, which was a necessary condition of his entire philosophy. His conception of time is summarized in the following passage:
All this reasoning takes place with regard to time; along with an additional argument which it may be proper to take notice of. 'Tis a property inseparable from time, and which in a manner constitutes its essence, that each of its parts succeeds another, and that none of them, however contiguous, can ever be co-existent. For the same reason, that the year 1737 cannot concur with the present year 1738, every moment must be distinct from, and posterior or antecedent to another. 'Tis certain then, that time, as it exists, must be compos'd of indivisible moments. For if in time we cou'd never arrive at an end of division, and if each moment, as it succeeds another, were not perfectly single and indivisible, there would be an infinite number of co-existent moments, or parts of time; which I believe will be allow'd to be an arrant contradiction.52 Now it is possible to see how Hume's arguments are ordered.

Speculation on the problem of time, which he considered a necessary condition for the analysis of experience, led him to the view that time consists of indivisible moments, never coexisting but succeeding one another. If experience is analyzed in terms of time, time itself being considered momentary and discrete, the experience of external objects also has to be explained in this manner. Hence, experience of the objects of the outer world came to be analyzed in terms of points, discrete and momentary.53 Once the experience of the outer world is analyzed in this manner, it becomes difficult to account for the causal efficiency of such discrete and momentary entities or even any kind of relation among them. The connection between them would merely be one of succession.

The conception of time and space arrived at by Hume therefore seems to be almost identical with the conception of moments
( ksana) and atoms (paramánu) held by some of the later Indian philosophers, especially the Vaisesikas (H indu) and the Sautrantikas (Buddhist). We have already pointed out that the Sautrantikas, as a result of their acceptance of the theory of momentary experiences ( ksanikavada), failed to account for causal continuity.

For them causation was merely a succession of immediately contiguous (samanantara) point-instants. This, no doubt, was also the theory of causation accepted by Hume.

The theory of moments (ksana) may be a consistent logical development of the theory of impermanence, a theory intended to eliminate belief in 'substance.' It could also be considered a product of the. introspective analysis of experience that was part of the Buddhist Abhidharma tradition. Judging from the sections quoted from the Treatise, it appears that Hume, too, emphasized momentary, discrete, and indivisible impressions as a result of his speculation on the problems of time and space as well as his introspective analysis of experience. After Hume, the idea that experience consists of momentary and discrete impressions gained such popularity among philosophers that causation came to be looked upon as a mere inference, not part of what is given in experience. In fact, B. A. W. Russell went so far as to maintain that " The law of causality . . . like much that passes muster among philosophers, is a relic of a by-gone age, surviving, like the monarchy, only because it is erroneously supposed to do no harm ." 54 Russell's view owes, no doubt, to the circularity of the argument when causation itself is considered to be an inductive inference (not part of experience)
and inductive inferences themselves are supposed to be based on the theory of causality.

But the position is not the same in early Buddhism. We have already pointed out that in early Buddhism experience was not analyzed in terms of moments (see chapter 4). According to the Assutava-sutta, early Buddhism recognized several patterns of change. One was the experience of change taking place in one's own psychic process, and another was the experience of change in the outer world. In the case of the former, the experience is that of rapid change. In the case of the latter, the experience is of a rather stable objective reality. With regard to the perceived physical body, it was said: " This physical body, made up of the four great existents, is seen to exist for one, two, three, four, five, ten, twenty, thirty, forty, fifty, hundred or more years." With regard to the thought process, it was said: " That which is called mind, thought, or consciousness arises as one thing and ceases as another whether by day or by night." Hume seems to have made the same distinction when he said, " Our thought is more variable than our sight," although he does not remain faithful to that idea when he goes on to analyze experience. According to early Buddhism, physical bodies are experienced as enduring for some time, and although they are subject to change and decay, this change is not experienced or perceived as occurring every moment. This, as we have pointed out (chapter 4), is an empiricist account of change. But our perception of our own thought process is much different. It is experienced as being in perpetual flux, palpitating like a fish out of water. It is therefore extremely difficult to control (durakkham) and guard (dunnivarayam).55 The purpose of meditational practices is to gain mastery over our thought process.

If the objective world is explained in this manner, then it is possible to maintain that we have experience not only of individual objects but also of the causal connections between them. The following argument could be raised against that view: Although one is able to perceive the causal connection between two events that succeed one another without a pause or temporal gap (e.g., the connection between touching a live electric wire and getting a shock), one cannot directly perceive the relationship between two events that are separated in time and space, although action at a distance is recognized as a fact. In the latter case, one has to depend on one's memory, which may not be reliable. This, of course, would not be considered a problem by the Buddhist, who accepts the validity of extrasensory perception.

According to the early Buddhist texts, the Buddha claimed to have attained the sixfold higher knowledge. Its forms are as follows:

1. Psychokinesis (iddhividha, ju i tsu).
2. Clairaudience (dibbasota, fie n erh). 3. Telepathy (cetopariyahana, t 'a hsin chih). 4. Retrocognition (pubbenivasanussatihana, su ming chih).
3. Knowledge of the decease and survival of beings (cuVupapatahana, sheng szu chih), also known as clairvoyance
   (dibbacakk.hu, fien yen ).
4. Knowledge of the destruction of defiling impulses ( asavakkhayahana, lou chin chih).56 The first is not relevant to our discussion here because it represents a form of psychic power rather than an actual form of knowledge. The second is very important in that it suggests the ability to perceive sounds well outside the range of normal hearing.

It is said: "With clairaudience, clear and transcending human hearing, one hears two kinds of sounds, human and divine, far and near." S7 This extension of auditory perception, both in extent and depth, would enable a person to perceive directly certain correlated phenomena that are only inferred by others. The Majjhima records an instance where a conversation between two people, the brahman Bhâradvâja and the ascetic Mâgandhiya, was heard at a distance by the Buddha.s8 Perceptions like these would undoubtedly give the Buddhists greater certainty about certain causal correlations.

The faculty of telepathy (cetopariyahana, t'a hsin chih) enables one to know the general state as well as the functioning of another's m ind.S9 One who has developed this faculty is said to be able to comprehend the minds of others in the following manner: "He knows it as a passionate mind, a dispassionate mind, a mind full of hatred or free from hatred, ignorant or devoid of ignorance, attentive or distracted, exalted or unexalted, inferior or superior, composed or not composed, emancipated or not emancipated." It is as if " one were to observe one's face in a mirror or a pan of water and notice whether there is a mole or not." 60 In the same way one is able to perceive the workings of another's mind.

The Ahguttara Nikâya refers to four ways of knowing another's m ind.61 They are (1) by observing external signs (nimitta),62 (2) by obtaining information from others, (3) by listening to the vibrations of thought of another as he thinks and reflects, and (4) by comprehending the mind of another and observing how the mental dispositions are ordered so that he is able to predict that such and such thoughts are likely to arise. The first two ways fall within normal perception, and the last two are forms of paranormal perception. The first two ways can be called 'mediumistic;' the last two are direct perceptions. This direct perception of thought processes enabled the Buddha and his disciples who had developed such faculties to feel more certain about the functioning of mental phenomena. In fact, their certainty was so great that they were able to say, after observing a good number of cases, that "such and such thoughts would follow such and such thoughts invariably." 63 The difficulty of knowing another's mind was, therefore, the raison d'être of scepticism about the uniformity of mental phenomena.64 This kind of scepticism is referred to by the Jaina commentator, Silâhka, when he said: "Scepticism is best since it is difficult to gauge the thought processes of another" .65 The fourth form of knowledge is retrocognition (pubbenivâsânussatinâna, su ming chih), by which one is able to verify one's own history.66 It is said: "When the mind is supple and pliant on attaining the fourth jhana, one recalls one's manifold past existences, one birth, two births, and so on for many periods of the evolution and dissolution of the world in the following manner: 'I was in such a place with such a personal and family name, such a status, having such and such food. Dying there, I was bom here.' In this manner one recounts his various past existences in all their aspects and details." This perception is compared to that of a person who, after going on a journey from village to village, is able to recall all the details of his journey.67 This knowledge of one's own past existences is followed by the knowledge of the decease and survival of other beings. "With his clear and paranormal clairvoyant vision, he sees beings dying and being reborn, the low and the high, the fair and the ugly, the good and the evil, each according to his karm a." 68 This knowledge is sometimes called the 'divine eye' (dibbacakkhu, t'ien yen), by which the Buddha could perceive objects that were beyond the horizon of normal vision.69 Disciples of the Buddha also claimed to have had visions of celestial figures beyond the reach of normal vision.70 Perceptions of this nature may have served as a basis for inferences drawn by the Buddha and his disciples.

By means of the knowledge of the past existences and the knowledge of the decease and survival of beings, the Buddha was able to verify the problem of rebirth. In Buddhism, the propositions about the phenomenon of rebirth are inductive inferences based on the data of direct experience.71 The Buddha is represented as criticizing the Jaina ascetics for not personally verifying the truth or falsity of the theory of survival and moral retribution.72 The Buddha and his disciples, in contrast, were said to have personally verified the doctrine of survival and moral retribution; thus their inductive inferences with regard to the possibility of survival were more certain.

With knowledge of the destruction of defiling impulses, and also through the foregoing four forms of knowledge, one is able to verify the four Noble Truths and the origin and cessation of defiling impulses.73 This form of knowledge is peculiar to Buddhism, whereas the four other forms, together with psychokinesis, were developed by the contemplatives of pre-Buddhist India.74 Thus, having experienced particular instances of causation through sensory as well as extrasensory perception, the Buddha arrived at a general theory of 'causality' or 'causal uniformity,' which he considered to be a universally valid principle: "Whether the Tathagatas were to arise in this world or not, this nature of things, this state of things, this orderliness of things, this causal pattern remains." 7S The essence of the Buddha's enlightenment, as pointed out earlier, was said to consist of the realization of this causal uniformity (dhammata). This causal uniformity carries more certainty for Buddhists since it is not based on pure reasoning and since it is not a mere mental fabrication. It is based on actual experiences of individual causal situations. Thus, while 'causation' is considered part of experience, 'causality' or 'causal uniformity' is looked upon as an inductive inference based on particular instances of causation.

Acceptance of the validity of extrasensory perception and the employment of such means of knowledge in the verification of the truth about phenomena led the Buddhists to attach a greater degree of credibility and certainty to causal laws. Extrasensory perception was recognized as a valid form of perceiving and verifying not only mental phenomena but also physical phenomena that are not given to immediate sensory perception.

One of the most important attributes of the five forms of extrasensory perception is that they are all concerned with the past and the present. There is no reference to the future. It is believed that the "essential function which causality has been supposed to perform is the possibility of inferring the future from the past, or more generally, events at any time from events at certain assigned times." 76 Thus, after verifying a number of causal relations, such as between birth and decay and death, the Buddha made inductive inferences concerning the future. The knowledge of these causal situations was called the "knowledge of the [causal] processes" (dhamme hanam). Describing the inductive inferences made on the basis of such knowledge, it was said: " By seeing, experiencing, acquiring knowledge, and delving into these phenomena, he draws an inference (nayam neti) about the past and the future (<atTtanagate) as follows: 'All those recluses and brahmans who thoroughly understood the nature of decay and death, its cause, its cessation, and the path leading to the cessation of decay and death did so in the same way I do now; all those recluses and brahmans who in the future will thoroughly understand decay and death . . . will do so in the same way I do now'—
this constitutes his inductive knowledge (idam assa anvaye nanam)." 77 This represents an inference from one specific instance to a general law or uniformity.

The foregoing discussion should help to determine the nature and validity of the causal principle in early Buddhism. To summarize, knowledge of causation is obtained through experience, and knowledge of causal uniformity is obtained through inference (ianvaya) based on experience. The latter is especially necessary to understand the future; in other words, it accounts for predictability. This causal uniformity is considered by some modern philosophers to be a probability only, since sometimes we find instances that seem to violate this principle of uniformity. As a result, doubt has been cast on the validity of causal uniformity. This, as pointed out recently by H. Van Rensselar Wilson, owes to a confusion of standpoints, the epistemological and the ontological:
There are two kinds of problems connected with causation: epistemological problems and ontological or metaphysical problems. Although they are interrelated in many ways, it is possible for purposes of analysis to keep them relatively distinct, and in my opinion it is important that confusion between them be avoided so far as possible. It therefore strikes me as unfortunate to speak as though predictability (an epistemological concept) were synonymous with causal necessity (an ontological concept). Lack of causal necessity entails lack of predictability; but I see no reason to assume that lack of predictability entails lack of causal necessity. The fact that the epistemological difficulties in sociological, psychological, and many biological situations preclude our knowing what all the specific relevant causal factors are in a particular case does not warrant the conclusion that there are none. Present inability to specify the values of a variable can hardly be construed as evidence that no such values exist.78 It seems that the Buddha, while accepting the universal validity of the causal uniformity, avoided making it a strictly deterministic principle according to which the cause should produce the effect under any circumstances whatsoever. Such theories of strict determinism (niyati) were held by some of the thinkers prior to and during the Buddha's day (see chapter 2). It is against this background that we ought to evaluate the causal principle propounded by the Buddha. As a result of certain prejudices, some philosophers had refused to recognize the validity of certain causal factors operative in the causal process and thereby either tended to accept complete determinism, or, on the discovery of any instance of failure of this strictly determined process, indeterminism. This is very clearly brought out by a famous discourse in the Majjhima Nikaya called the Maha-kammavibhahga-sutta. In it the Buddha refers to some recluses and brahmans who, by a thorough application and concentration of mind, were able to see beings who had led an immoral life and had been reborn in an evil state. As a result of this telepathic insight, they concluded: "He who takes life, steals, . . . who is of wrong views, would be reborn in an evil state after death. They who know this have right knowledge; others are mistaken." To consider this conclusion to be "the only truth and that all else is false" 79 is, according to the Buddha, a very grave mistake. The difficulty of drawing such absolute conclusions is demonstrated by the Buddha. He points out that a person who sees a man reborn in a happy state after having led an immoral life comes to a conclusion diametrically opposed to the one given above. He maintains that there is no effect of good and bad deeds. The Buddha does not doubt the attainments of the person who perceives the phenomenon of rebirth and moral retribution;80 rather, he doubts the validity of the inference because certain aspects of the causal process were not taken into consideration. The Buddha points out that in the case of the person who had led an immoral life but was reborn in a happy state, there may have been counteracting tendencies. Perhaps he led a good life at some time during a past or present life, or perhaps he held right views at the moment of death.81 An interesting illustration of this problem is found in the Lonaphala-vagga of the Ahguttara Nikaya82 (see chapter 6 for a discussion of this text). Thus, according to the Buddha, strict determinism and indeterminism are both extremes that are products of prejudice and ignorance.83

Vi. The Causal Explanation Of Existence

THE CAUSAL PRINCIPLE discussed in the previous chapter was found to be operative in every sphere of existence. Later scholiasts refer to five broad spheres in which the causal process works. They are as follows:

1. The physical (inorganic) world (utuniyama).
2. The physical (organic) world (bTjaniyama).
3. The sphere of thought or mental life (cittaniyama).
4. The social and moral sphere (kammaniyama). 5. The higher spiritual life (dhammaniyama).1 This classification is, no doubt, based on the various statements made by the Buddha to explain man and his environment.

Physical Causation One of the problems that attracted the attention of the preBuddhist thinkers was the origin and development of the world.

The keenness of the Indian mind for cosmological speculation is well manifested in the large number of theories put forward during this period. We have seen how speculation starting as far back as the time of the Rgveda came to be systematized and assumed final form in the theories of the Upanisadic thinkers such as Uddalaka (see chapter 1). Most of these thinkers accepted a First Cause such as Being (sat) and explained the world as the final product of evolu tion, mostly by way of self-causation. Others, who conceived of this First Cause as a personal creator God, considered the world the creation of this omnipotent Being. As we have seen, most of these views were known to the Pali Nikayas and the Chinese Agamas.

For empirical and logical reasons the Buddha abstained from any discussion of the problem of the origin of the world (see chapter 9). He emphatically declared that "it is not possible to know or determine the first beginning of the cycle of existence of beings who wander therein deluded by ignorance and obsessed by craving." 2 Nevertheless, the Buddha found it necessary to give a rational explanation of the problem of evolution, especially to refute the claims of the Brahman caste to superiority which were based on the theory that Brahma had created the world. Thus, the Aggahha-suttanta, which discusses the evolution of the world process, was preached to explain the evolution of the existing social order, namely, the four castes.3 W ithout positing a First Cause such as the Being (sat) of Uddalaka, the Buddha, after stating that the beginning of the cycle of existence (samsara) is difficult to know or determine, described the world as being subject to a process of dissolution (samvafta, huai) and evolution (vivatta, pieri):
There comes a time,. . . when, sooner or later, after the lapse of a very long period of time, this world passes away (or is destroyed). And when this happens, beings (who have reached the end of their life span4) are reborn in the world of Radiance,5 and there they dwell; made of mind, they feed on rapture, are self-luminous, traverse the air, remain in glory, and thus they stay for a long time. There also come a time, . . . when, sooner or later, this world begins to reevolve. (When this happens, beings who have passed away from the world of Radiance,6 usually come to life as humans. And they too are made of mind, they feed on rapture, are selfluminous, traverse the air, abide in glory, and remain thus for a long time.)7" Whatever the credibility of the above description, it illustrates two im portant features of Buddhist cosmological speculation.8 First, it implies that the world in which we live is only a small part of an extensive universe. Although speculation about the origin and extent of the universe is generally discouraged in early Buddhism, the vastness of space and the immensity of time are never forgotten. In the vastness of cosmic space are found an endless number of worlds, of which this earth is only a very small part.

The passage quoted above implies that there can be mutual influence among these different worlds. When the earth undergoes dissolution, beings living there are reborn in another sphere, whence they come back to earth as it starts reevolving. In this manner, a continuity is maintained amidst dissolution and evolution, without a complete extinction of life. Such speculations enabled the Buddhists to avoid the question of the beginning of the world process and therefore of life.

Second, the passage emphasizes the immeasurable length of time between dissolution and evolution and between evolution and dissolution. The processes of evolution and dissolution take periods of time measured in eons {kappa, chieh). The duration of a single eon is such that it can be explained only by parables. One of them is, " [Suppose] there were a great mountain, one league in width, one league in length, and one league in height, a solid mass without chasms or clefts. And [suppose] a man at the end of every hundred years were to strike it once with a silk cloth. That mountain would be destroyed sooner than would an eon [pass]."9 Describing the relative beginning of the process of evolution
(vivatta, pien), the Agganna-suttanta says:
Now at that time, all had become one world of water, encircled by dense darkness. Neither moon nor sun appeared.

No stars or constellations were seen. Neither was night manifest nor day (neither months nor half-months), neither seasons nor years (neither male nor female).10 This passage is strongly reminiscent of the description in the NasadTya-sukta of the Rgveda.11 There, to explain the relative beginning of evolution, the Buddha, made use of current speculation in a way not inconsistent with his philosophy. Perhaps to accord with the description of the state of the world at the time of evolution, the Buddha maintained that beings of the world of Radiance who were self- luminous and capable of traversing the air were reborn on earth because no other beings could be expected to survive under earth's conditions.

Another stage in the process of evolution is described in the passage that follows the one quoted above :
And to those beings, . . . sooner or later, after a long period of time, the earth with its savor was spread out in the waters.12 (Even as scum forms on the surface of boiled milky rice that is cooling, so did the earth appear.)13 It became endowed with color, odor, and taste. Even as well-made ghee or pure butter, so was its color; even as the flawless honey of the bee, so sweet it was. Then . . . one of the beings with greedy disposition said: "Lo now! What will this be?" and tasted the savory earth with his fingers. He, thus tasting, became suffused with the savor and was overcome by craving.

Other beings who followed his example and tasted the savory earth with their fingers were also suffused with the savor and overcome by craving. Then those beings began to feast on the savory earth, breaking off lumps of it with their hands. And as a result, their self-luminosity faded away. Thereupon, the moon and the sun became manifest. When the moon and the sun became manifest, night and day became manifest. After this, the seasons and years became manifest. Thus far did the world evolve.

Comparing the foregoing description of the evolution of the world with the theories put forward by the earlier thinkers such as Mahîdâsa and Uddàlaka (see chapter 1), we discern one of the salient features of the Buddhist theory of evolution.

Whereas the theories of Mahîdâsa and Uddàlaka explained evolution as a pattern of self-causation, the Buddha explained evolution in terms of the causal formula, "when this exists, that exists or comes to be; . . This is especially illustrated by the last part of the preceding quotation in its Pah version, where, as in the general formula, the locative absolute construction is used: "sayam pabhàya antarahitàya candimasuriyà pàturahamsu; candimasuriyesu pàtubhütesu nakkhattàni tàrakarüpàni pàturahamsu" ("When self-luminosity faded away, the moon and the sun became manifest ;
when the moon and the sun became manifest, the stars and the constellations appeared.").

Although this account of the evolution of the world from a chaotic state is no more than a hypothetical description (as every description of the evolution of or origin of the world must be), it reveals, as T. W. Rhys Davids has observed, "a sound and healthy insight and is much nearer to actual facts than the Brahman legend it was intended to replace." 14 Its importance lies mainly in the fact that it gives a causal account of physical change.

The foregoing causal account of the evolution of the world should be supplemented by the causal account of the dissolution of the world, which is found in the Añguttara.15 The Buddha is explaining the impermanent nature of all component things {sañkhára, hsing) as introduction to the doctrine of renunciation.

He describes how the great earth would be destroyed by a cosmic catastrophe:
There comes a time, after many hundreds of thousands of years, when there is no rain. All vegetation, including the giant trees of the forests, is dried up by the heat of the sun and destroyed.16 After another long period, a second sun appears, as a result of which all the streams and water spouts dry up and disappear.17 With the appearance of a third sun, the great rivers18 are parched and dry up without leaving a trace behind. The huge lakes that are the sources of the great rivers are completely dried up when a fourth sun appears. The appearance of a fifth sun is the cause of the gradual drying up and disappearance of the four great oceans. The waters of the four great oceans recede a hundred leagues and [continue] until they reach seven hundred leagues. Then the waters remaining at a depth of seven palm trees19 gradually dry up so that their depth is up to a man's ankle. What is left is comparable to the puddles of water left in the footprints of cows during an autumnal rain.20 With the appearance of a sixth sun, both this earth and Sumeru, the king of mountains, begin to belch forth clouds of smoke. Lastly, when a seventh sun appears, the earth bursts into flames, becoming a single sheet of fire. Here too we find the general formula of causation applied to explain the gradual process of dissolution. The gradual increase in the number of suns appearing in the sky may be taken as a poetic way of describing an increase in the heat of the sun, which, it was believed, would cause the destruction or dissolution of the earth.

In addition to this description of the processes of evolution and dissolution of the world, we occasionally come across causal accounts of such physical events as earthquakes and drought.21 Occasional reference is also made in the Pali Nikayas and the Chinese Agamas to the causation of plant life. Most of these references occur in connection with explanations of the causation of the human personality, psychic events, or even moral behavior.22 Analogies are drawn between the growth of plants and the arising of the human personality. An explanation of the process by which a person comes to be reborn in an inferior existence, it is said: "Behavior is the field, consciousness the seed, and craving the moisture that cause beings who are deluded by ignorance and obsessed by craving to be reborn in an inferior existence." 23

Causation Of The Human Personality

One of the most important problems that the Buddha had to face as a result of denying a permanent self (atman) was how to explain the causation of the human personality and its continuance in samsara. It was pointed out that the process of rebirth of human beings had been directly verified by the Buddha and his disciples, who had developed extrasensory perception. The process of rebirth was thus not merely an explanation of certain problems connected with moral causation. The problem for him was to explain this fact of rebirth without positing a permanent and enduring entity, which he considered an unverifiable metaphysical principle.

In the Nikayas and the Agamas, the human personality is generally represented by the term ndmarupa or ming se and sometimes ming hsiang, where nama or ming represents the psychic personality and rupa or se {hsiang) stands for the physical personality. In further elaboration, man is explained in terms of six elements (dhdtu, chieh): of earth (pathavi, ti), water {apo, shui), heat
{tejo, huo), air (vayu, feng), space (akasa, k 'ung), and consciousness (vihhana, shih).24 The psychic personality is represented by one element, consciousness, while the physical personality is further analyzed to show that a permanent element such as a material soul {atman) posited by some of the Materialist thinkers does not exist.

But this analysis of the human personality into six elements was not as popular as another classification, which reduced the personality to five aggregates {khandha, yin). The physical personality is represented by one aggregate, form {rupa, se), while the psychic personality is further analyzed into four aggregates: feeling or sensation {vedarid, shou), perception {sahna, hsiang), dispositions
(sahkhara, hsing), and consciousness (vihhana, shih). This fivefold classification predominated in the early discourses because the Buddha used it to refute the conception of the more psychic self or soul that was accepted by the Upanisadic thinkers.25 With regard to these five aggregates, we agree with the explanation presented by C. A. F. Rhys Davids: "There is here no order in function and evolution." 26 The four immaterial aggregates merely represent the different aspects of the psychic personality, which in the earlier classification was denoted by the element of consciousness (vihhana, shih). The five aggregates have been called the aggregates of grasping {upadanakkhandha, shou yin21) because they represent the five things to which a person clings as his personality.28 It is specifically stated that these five aggregates of grasping, constituting what may be called the personality, are causally conditioned (paticcasamuppanna, yin yuan sheng).29 According to the early Buddhist texts, a being is conceived when three conditions are satisfied. First, there should be coitus of the parents; second, the mother should have her period; third, a gandhabba (hsiang yin) should be present.30 The first of these accounts for the seed that forms the physical personality (rupa, se).
This is clearly implied in the Maha-hatthipadopama-sutta, where it is said that the body, which is physical and which is derived from the four great existents, is born from the parents and is sustained by liquid and gross food.31 The second and third conditions describe the circumstances under which the seed provided by the parents will grow in the mother's womb. Even though the seed is provided by the union of the parents, if the mother does not have her period and if a gandhabba is not present, that seed will not germinate.32 Of these two circumstances, the first is purely a temporal one. The latter is very significant in that it determines the nature of the psychic personality of the new individual. Gandhabba
{hsiang yin) in this context is identified with 'consciousness' {vihhana, shih),33 which is the psychic factor that survives physical death and which, in association with the fetus or the biophysical factors in the womb, helps in the development of the new personality. This consciousness is said to serve as food {ahara, shih)
for beings who are conceived as well as for those seeking birth
(sambhavesi).3* The word vihhana (shih) is here used in an eschatological rather than a psychological sense.

In his analysis of the early Buddhist conception of vinnâna,
E. R. Saratchandra has rejected the view that rebirth constituted a central tenet of early Buddhism.35 He questions the interpretation given in the Mahâ-nidâna-suttanta36 to the statement " depending on consciousness arises the psychophysical personality"
(vinhânapaccayâ nâmarüpam). He thinks that it is due to the influence of the non-Buddhist belief in rebirth and insists that it not be taken as representative of the original Buddhist position.

Saratchandra sees a contradiction in the reply in the Mahânidâna-suttanta to the question whether the psychophysical personality would grow to maturity if consciousness did not enter the mother's womb. He maintains that "The very contradiction inherent in the explanation shows it up as a later intrusion. It is said that, if vinnâna did not descend into the mother's womb, the growth of nâmarüpa would be prevented. If nâmarüpa here stands for the whole individual composed of mental and physical factors, we should have to regard vinnâna as something over and above nâmarüpa, a position which is not consistent with the rest of the Buddhist teaching. Nâmarüpa, whenever it stood for the individual, always included vinnâna as well." 37 This view seems to us to be based on a superficial understanding of the early Buddhist texts.

In the Ahguttara it is stated that "conception (gabbhassâvakkanti, sheng mu t'ai) is dependent on the six elements, and when there is conception, there is the psychophysical personality (nâmarüpa).'"38 The six elements consist of the five physical elements and consciousness. Hence, conception is the conjunction of the two aspects of the personality, the physical and the psychic. This idea is stated more clearly in the Samyutta: "That which is thought of, that which is reflected upon, and that which is dwelt upon—that becomes the basis [literally, "object," ârammana] for the establishment of consciousness. Where there is a basis, consciousness is provided with a foothold. When consciousness is established and develops, then there is conception of a psychophysical personality
(nâmarüpassâvakkanti, ju yu ming sê)."39 All the statements just quoted emphasize that consciousness is dependent on a physical personality and that without the conjunction of consciousness and the physical personality there cannot be a psychophysical personality. While not denying the influence of the physical on the psychophysical personality, the Buddha emphasized the importance of consciousness or the psychic personality because only it has a past history. It is the personality that survives physical death and, in conjunction with the new biophysical contributions of the parents, gives rise to a relatively new psychophysical personality. Therefore, when it is said that "depending on consciousness there is the psychophysical personality
(vihhanapaccaya namarupam)," it means only that the 'surviving consciousness' is a factor in determining the nature of the psychophysical personality {ndmarupa). Hence, the question in the Mahanidana-suttanta—"Were vinhana, . . . not to enter the mother's womb, would body and mind be constituted therein?"—may rightly be considered "a case of folklore speech adopted by the Suttanta teaching," as C. A. F. Rhys Davids thought it to be. This conception of vinhana as a transmigrating entity is not, as Saratchandra tries to interpret the explanation of Mrs. Rhys Davids, an intrusion into Buddhist thought from folk religion.

While the term vinhana {shih) was used in passages describing the process of rebirth, it was also employed in the description of the attainment of enlightenment, because enlightenment culminates in the cessation of rebirth.40 In the Samyutta we come across two passages describing the passing way, immediately after the attainment of enlightenment, of two of the Buddha's disciples, Godhika and Vakkhali.41 They "attained perfect release, with consciousness finding no support or basis {appatitthitena vihhanena parinibbuto).,' This statement has been brushed aside by Saratchandra as another popular interpretation.42 That the word vinhana in this context is used in an eschatological sense is further proved by the Chinese translation of this passage. The statement appatitthitena vihhanena has been rendered into Chinese as wu yu shen shih ("there is no rebirth consciousness")43 or as pu chu shih shen ("without continuity in rebirth consciousness").44 It is significant that the phrase shen shih has been used to render the idea expressed by the phrase hsiang yin ( = gandhabba), which, as pointed out above, denoted an eschatological concept.4S
The combination of shih, the term for 'consciousness,' with shen is very interesting, especially because shen is also used to translate the word atta {atman) meaning 'soul.'46 This conception of vinhana {shih) no doubt created problems, even during the time of the Buddha. We find one of the Buddha's immediate disciples, SatT, affirming that "it is this consciousness that transmigrates without change." Asked by the Buddha what he means by this
'consciousness,' SatT replies: " It is that which speaks, feels, and experiences the effects of good and bad deeds."47 He was thus admitting the existence of a subject or agent within the psychophysical personality, the agent of all the actions as well as the enjoyer of all the experiences. The reasons for the Buddha's refusal to contribute to such views have been discussed earlier (see chapter 1). This, therefore, is a very clear attempt to interpret the Buddha's teaching as being not much different from those of the Upanisads, an interpretation that has its modern advocates.48 Rejecting the idea of a permanent consciousness that functions as the subject or agent, the Buddha insisted that he had "in many ways spoken of consciousness as being causally produced and that apart from causes there would be no arising of consciousness."49 The use above of the term 'consciousness' (vihhana, shih) as a connecting link between two existences invalidates Saratchandra's view that the concept of vihhana in an eschatological sense is not integral to the early teachings. Moreover, while explaining the connection between consciousness and the psychophysical personality, the Ch'eng wei shih lun maintains that vijhana in this context refers to the eighth vijhana (according to the Yogacarins), i.e.,
alaya-vijhana or 'store-consciousness,' because the other seven forms of consciousness are not continuous.so The fact is, the term vihhana (shih), like the term sahkhara
(hsing), is used in the early Buddhist texts in a wide variety of meanings that Saratchandra has failed to distinguish. At least three important uses of vihhana can be clearly distinguished. First, it is used to denote psychic phenomena in general, synonymous with the terms citta (hsin), 'mind,' and mano (i), 'thought.'51 Second, it is used to describe a complete act of perception or cognition (see below); and third, it stands for the connecting link between two fives, a form of consciousness that later came to be designated
'rebirth consciousness' (patisandhi-vihhana). The first of these refers to psychic fife in general, and the last two represent two important aspects of consciousness.

It is of interest to find out whether there is any connection between the last two uses of vihhana, namely, the psychological and the eschatological. 'Consciousness' in an eschatological sense is almost always associated with 'dispositions' (sahkhara, hsing). The nature of the sahkharas is exemplified by a statement in the Ahguitara Nikaya that one who has attained " the state of concentration free from cogitative and reflective thought can comprehend with his mind the mind of another, and by observing how the mental sahkharas are disposed in the mind of that particular individual, [he can] also predict that he would think such and such a thought at a later time." 52 As is evident from that passage, the subject is not aware of this thought process, which is cognized by the telepathic insight of another. This, no doubt, is a reference to unconscious mental processes. This unconscious mental process constituted the 'stream of becoming' (bhavasota or vihhanasota) and maintained continuity between two lives without interruption, but itself existed in a state of flux. A person who has developed extrasensory faculties is said to be able to "perceive a man's unbroken flux of consciousness established both in this world and in the next." 53 This is the same consciousness that was referred to as gandhabba (hsiang yin), the consciousness that is said to enter the mother's w om b; it is even, according to the twelvefold formula of causation (see below), the consciousness that conditions the psychophysical personality.

In the last case, when it is said that dispositions condition consciousness, it means that the dispositions (sahkhara, hsing), by conditioning consciousness, or more correctly the unconscious process, determine the nature of the psychic personality of the newly born individual. But these dispositions are ultimately the results of perceptive activity. This is clearly implied in a passage in the Samyutta that discusses the difference between a dead man (mato kalakato, wei szu) and a man who has entered the state of mental concentration characterized by the cessation of perception and sensation (sahhavedayitanirodham samapanno, ju mieh cheng shou).
" In the case of a dead man, his dispositions, bodily, verbal and mental, cease to exist and are pacified54; life has come to an end, breath is calmed, and the senses are destroyed. But in the case of a man who has attained the state of cessation of perception and sensation, even though his dispositions have ceased to exist and are pacified, his life has not come to an end, breath is not calmed, and the senses are not destroyed." 55 According to this account, although the senses of the man who has attained the state of cessation of perception and sensation are intact, because there is a temporary cessation of perceptive activity he does not accumulate any dispositions. The obvious conclusion is that dispositions are the results of perceptive activity.56 Not only the tendencies in the conscious mind but even those in the unconscious process are the results of perception. Therefore, as early as the time of the Nikayas, the conclusion was reached that "mind is luminous by nature and it is defiled by adventitious defilements." 57 This, no doubt, was the germ of the theory of the Yogacarins stated in the Lahkavatara-sutra.SB But the question of when this mind, pure and luminous, first came to be defiled by adventitious elements was more or less a question about the origin of samsaric existence.

Therefore, from the standpoint of early Buddhism, it was a problem that came under metaphysics proper.

'Consciousness' (vinnam, shih) can be described as something that is conditioned as well as something that conditions. On the one hand, consciousness arises because of conditions (paccayam paticca, sui so yuan sheng), for example, the contact of sense organs and sense objects (see below). On the other hand, as discussed above, it serves as a cause in that it conditions the psychic personality of the newly born individual. Thus, the problem of perception, as well as the problem of rebirth, which the Upanisadic thinkers solved by positing an immutable and perduring soul, were given causal explanations in the early Buddhist texts.

Causation Of The Perceptual Process

The process of perception, which the Upanisadic thinkers also explained on the basis of a metaphysical self (atman), received a causal explanation in the hands of the Buddha. For him, this was a problem of prime importance because he realized that all the misery and unhappiness in the world were due to the evils associated with sense perception. The Buddha thus found it necessary to explain clearly how sense perception takes place. He realized that a proper understanding of the sensory process would give insight into the origin of suffering as well as into the way one can attain freedom from suffering. Hence, in the Samyutta Nikaya, the higher liftfbrahmacariya) lived under the Buddha is said to be aimed at understanding the sense organ, the sense object, and sense contact, i.e., sense perception, because it is sense perception that leads to suffering.59 The theory of sense perception is represented in the twelvefold formula of causation by sal'ayatana (Hu ju chu). The term ayatana
(ju chu), which, to use a word from modern psychology, means 'gateway,'60 denotes both the sense organ and the sense object.

The origin of perception from the subject-object relationship is described in diverse ways in many places in the Nikayas and the Agamas.61 One of the most important statements follows:
Depending on eye . . . and visible form arises visual consciousness ; meeting together of the three is contact (phassa) ;
because of contact arises feeling or sensation (vedana); what one feels, one perceives (,sanjanati); what one perceives, one reflects on (vitakketi); what one reflects on, one is obsessed with (papahceti); what one is obsessed with, due to that, concepts characterized by such obsessed perceptions (papancasahhasahkha) assail him with regard to visible form cognizable by the eye, belonging to the past, the future, and the present."62 The implications of this passage have been interpreted in various ways by scholars who have written on the problem of perception in Buddhism.63 The latest on the subject is by Nanananda, who, in his small but excellent work, Concept and Reality in Early Buddhist Thought, has given a very clear exposition of this passage.64 He rightly points out its significance in distinguishing three important stages in the process of perception. The formula begins on a very impersonal note and follows the pattern set out in the general formula of causation ("When this exists, that exists or comes to be" ). This impersonal manner of description is found only up to the point of feeling or sensation (vedana).
Then the mode of description, the grammatical structure of the sentences changes to a very personal tone suggestive of deliberate activity. Note the use of the third-person verb: "W hat one feels or senses, one perceives; what one perceives, one reflects on; what one reflects on, one is obsessed with." Thus, immediately after feeling (vedana), the process of perception becomes one between subject and object. The feeling comes to be looked upon as belonging to a subject. This marks the intrusion of the egoconsciousness, which thereafter shapes the entire process of perception, culminating in the generation of obsessions.65 These obsessions, according to the commentator, are threefold: craving (tanha), conceit (mana), and dogmatic views (ditthi).66 The final stage in this process of perception seems to be different from the preceding two stages. It is no longer a mere contingent process, nor is it an activity deliberately directed, but an inexorable subjection to an objective order of things. At this final stage of perception, he who has hitherto been the subject now becomes as it were a hapless object.

This analysis of the process of perception is of tremendous importance for several reasons. First, it replaces the theory of an eternal and unchanging entity (like the atman) considered to be the subject, with a causal account of the process. Second, while tracing the origin of ego-consciousness to the deliberate activity of the mind, it also accounts for the phenomenon of free will, without which a theory of moral responsibility is untenable. It shows that up to the point of feeling or sensation one is governed by a natural flow of events, a flow in turn governed by the causal pattern. But immediately after that begins deliberate activity, which can lead one either to subjection to the objective order of things, that is, to enslavement to things of the world, or to freedom from bondage to such things through the elimination of egoconsciousness (ahamkara or mamamkara).
Let us examine the problem of free will in Buddhism. The reconciliation of free will with causality has been a perennial problem in philosophy. With regard to the problem of free will in Western philsophy, it is pointed out that the advocates of free will depend on the apparent indeterminacy of the future as compared with the determinacy of the past,67 because what is foreseen is considered to be fated. Jayatilleke presents another view in explanation of the Buddhist position. He distinguishes between physical and psychological causation and maintains that since causality "is a probability and not a necessity when psychological factors are involved," one can admit freedom of will.68 With regard to the first view, it has been well argued that dependence on future indeterminacy as the basis of a theory of free will is the result of ignorance. That is because "it is plain that no desirable kind of free-will can be dependent simply upon our ignorance; for if that were the case, animals would be more free than men, and savages than civilized people." 69 If we are able to recollect some of our past volitions, volitions that have changed the course of our lives, then we would certainly feel we were free in the past. Similarly, we might be free in the future even if we are able to perceive our future volitions. Therefore, the definition of freedom that "our volitions shall be as they are result of our own desires, not of an outside force compelling us to will what we should rather not will" 70 seems to be consistent with the teachings of Buddhism. This is possible only if we recognize the causal status of our dispositions and desires, a recognition that points to a Buddhist contribution to Indian thought when viewed in light of the theories propounded by the naturalistic schools current in India during the Buddha's day (see chapter 2).

Jayatilleke quotes two statements from the Pali Nikayas in support of his view that causality "is only a probability, not a necessity, when psychological factors are involved." The first is,
"A person who knows and sees things as they are, need not make an effort of will (saying) 'I shall become disinterested'; it is in the nature of things (dhammata) that a person who knows and sees becomes disinterested." 71 This statement implies that causality reigns supreme in the sphere of psychological life. As opposed to this, Jayatilleke quotes another statement that if a person "being ardent, gains knowledge and insight, and because of it, praises himself and disparages others," he will not progress in spiritual development.72 Comparing these two statements, one in which causality seems to work and the other in which the same causal process seems to have failed, Jayatilleke concludes that causality is a probability when psychological factors are involved.

Acceptance of such indeterminism in the sphere of psychological causation would seem to go against the Buddha's theory of the uniformity of mental phenomena (cittaniyama; see above). But a careful examination will show that these two statements explain two different causal situations. According to the first statement, causality is a law valid in the sphere of psychological life. In the second example, the individual's disposition, that is to say, his inclination to be satisfied with the knowledge he has gained, appears to have interfered with the natural process and therefore produced a result that is different from what it would otherwise have been. Thus, the difference between the two examples is that in the case of one a certain causal factor, namely, the disposition to be satisfied, is absent and in the case of the other, it is present. Only if we dismiss the importance of this disposition as a causal factor can we maintain that causality in the present case is merely a probability, not a necessity.

On the contrary, the examples above illustrate very clearly that causality is not incompatible with free will so long as psychological factors such as dispositions are given causal status. In fact, the incompatibility of causation with free will becomes a problem when causation is confined to physical phenomena alone, denying its validity and the causal efficiency of psychic phenomena. It was the knowledge that causality was effective in the past, is effective in the present, and will be effective in the future that enabled the Buddha and his disciples to put an end to suffering and thereby attain perfect happiness and peace. This may have been a very good reason for the inclusion of ignorance of the past as well as of the future under the category of ignorance (avijja, wu ming).73

Causation Of Moral Behavior

We have seen that before the rise of Buddhism several different theories of moral causation had been put forward by Indian thinkers. The eternalists of the Vedic and Upanisadic traditions held the view that man is both the doer (kartr) of the actions and the enjoyer (bhoktr) of the consequences (see chapter 1). This theory was based on a metaphysical self (atman) believed to reside in the individual. Hence they concluded that whatever happiness and suffering a man experiences owes to self-causation.

On the other hand, there were the nihilists, who denied any form of moral causation or responsibility. First, the Materialists advocated a strictly determined law such as the principle of inherent nature (svabhava) (see chapter 2). Second, the AjTvikas, led by Makkhali Gosala, believed in fate (niyati) and therefore could not grant m an's responsibility for his actions (see chapter 2).

The Theists, who transferred man's responsibility for his actions to an omnipotent God (issara), were criticized by the Buddha as denying moral responsibility (see chapter 1).

Opposed to these different schools of thought were the Jainas, who considered moral behavior as being completely determined. The present, they believed, is completely determined by one's past behavior (pubbekatahetu). Karma for them was an inexorable law that could not be escaped (see chapter 2).

Rejecting all these views as unsatisfactory, the Buddha gave a causal account of human behavior. Behavior, according to him, consists of three forms, bodily (kaya, shen), verbal (vaci, k'ou), and mental (mano, i), and he emphasized the psychological aspect. Once, when explaining what immoral behavior is, the Buddha maintained that both bodily and verbal behavior has mind as the basis.74 On another occasion, he was interrogated by the Jaina ascetic DTghatapassT, who believed that bodily punishment (kayadanda, shen fa) is more blameworthy than mental punishment (manodanda, i fa). The Buddha turned the discussion from punishment (danda, fa) to action (kamma, yeh) and maintained that mental behavior should be considered more blameworthy in the commission or perpetuation of evil.75 This is a clear example of the Buddha's emphasis on the psychological aspect of behavior rather than on external behavior manifested by way of body and speech. Hence, he defined behavior (kamma, yeh) as volition (cetarid, szu).'16 The Buddha's emphasis on the psychological aspect of behavior led some of his contemporaries to think that he believed that "Bodily behavior is unreal. So is verbal behavior. Only mental behavior is true or real." 77 One such person was the ascetic Potaliputta, but venerable Samiddhi corrected this misrepresentation. Samiddhi pointed out that a man "experiences pain, after having committed volitional acts (sahcetanikam kammanri, tso yeh), bodily, verbal as well as mental." 78 When the discussion between Potaliputta and Samiddhi was reported to the Buddha, he rebuked Samiddhi: "This foolish person Samiddhi has given a categorical answer to a question that demands a conditional or analytical reply." 79 After thus accusing Samiddhi, the Buddha explained the problem: " Having committed a volitional act leading to pleasurable feeling with the body, speech, and mind, one experiences pleasurable feeling. [The exposition continues with volitional acts, leading to painful and neutral feeling.]" Even from this analytical answer given by the Buddha it is evident that volition is the basis for the three forms of behavior, bodily, verbal, and mental. Thus, the two statements—(1) that mind is the basis of bodily and verbal behavior, and (2) that volition is the basis of all three forms of behavior, bodily, verbal, and mental—were made at different levels and should not be confused. The former ranks the three forms of behavior according to degree of importance, while the latter describes the psychological motives or springs of behavior, bodily, verbal, and mental.

Let us consider the cause of moral behavior. The Buddha did not present an overall postulate to account for the causation of moral behavior, mainly because his was a theory of conditionality rather than a doctrine of strict determinism. Instead of providing an all-inclusive theory of moral causation, as the Jainas did (see chapter 2), he gave answers to various questions in specific contexts. Therefore, it is possible to find several causal explanations of behavior in the early Buddhist texts.

In the Ahguttara, the problem of the causation of behavior is raised. The Buddha answers that "contact (phassa, keng lo) is the cause of behavior {kamma, yeh).80 This statement can be interpreted in two ways. First, taking phassa {keng lo) as sense contact, it may be interpreted as a broader or more general cause of behavior. This is because mental tendencies, such as craving {tanha, yu), that may be considered specific causes of behavior are only results of contact {phassa). This is exemplified by the statement that the cause of craving "is an agreeable object. [Because] in the case of a person who reflects wrongly on an agreeable object, craving that has not yet arisen arises and craving that has already arisen increases." 81 Second, we may take phassa
{keng lo) in a more physical sense to represent a stimulus-response sort of causal explanation, where reflex movement is followed by sensory excitation. This may be illustrated by the example of a person who, while crossing a road, jumps up either because of a twinge in his stomach or because a car happens to back fire.82 Another example is " an innocent little baby lying on its back
[who] quickly draws back its hand or foot if it has touched a live ember." 83 Apart from this more general cause (or the physical cause of sensory stimulation, whichever we may take it to be), there are certain other motives, such as craving {raga, fan), hate or aversion {dosa, wei), and confusion {moha, ch'ih), that are more or less conscious tendencies that serve as causes of behavior.84 Generally these causes are thought to produce evil or immoral behavior.

Hence, morally good behavior is produced by mental tendencies that are the opposites of those mentioned above. These fall into the category of volitions {cetana, szu), which determine the gravity of an action. Buddhism emphasizes the elimination of these springs of action. Hence the importance of mental culture.

In addition to these conscious motives, there are unconscious motives that determine the behavior of man. They are represented in the early Buddhist texts by the term 'disposition' (sahkhara). Dispositions are accumulated either consciously (sampajano) or unconsciously (asampaja.no).85 It was pointed out earlier that in the special formulation of the causal principle the term sahkhara (,hsing) stands for 'unconscious dispositions.' They also account for the problem of moral responsibility. The cause of behavior given in the Samyutta Nikaya is a specific instance of such unconscious motives. There it is pointed out that a man, when told that such and such things, for example, deadly poison, lead to disaster, will naturally avoid such things because he "desires to live
(jivitukama, ch'iu sheng), recoils from death (amaritukama, yen szii), desires happiness (sukhakama, ch'iu lo), and recoils from pain (dukkhapatikkula, yen k 'u)."86 Once a person is informed that such and such a thing is harmful to him, his behavior in the presence of that thing will be determined by these unconscious drives.

The Buddha gave different causal explanations for different problems, which shows, in the words of a modern writer, "the sensitivity to the different sorts of questions that can be asked about human actions and the different sorts of answers that are appropriate." 87 It also indicates the Buddha's reluctance to posit an overall theory of motivation that might create confusion by elevating an answer to a limited question to the status of a general postulate.

While human behavior is itself produced by causes, it is followed by the correlated consequences. This correlation between action (kamma, yeh) and consequence (phala, pao or vipáka, i shu) constitutes the doctrine of kamma in Buddhism. An examination of some of the texts that deal with the problem of moral behavior and responsibility reveals that it is generally founded on the doctrine of rebirth. This is evident from the Cula-kammavibhahgasutta, which maintains that a person who kills living creatures or has no compassion for them will, because of that behavior, be reborn in an evil state. If he were not reborn in an evil state, and if he returned to life as a human, . . . he would be short-lived.88 This implies that the doctrine of moral responsibility, like the doctrine of rebirth, is properly "verified" by the development of extrasensory powers (see chapter 5). On the basis of data available through such forms of telepathic insight as clairvoyance
(icut'upapatanana, sheng szu chih), inductive inferences were drawn. The texts reveal two kinds of correlations drawn between action
(,kamma, yeh) and consequence (vipaka, i shu). These may be grouped as specific and general correlations.89 A list of specific correlations is found in the Cula-kamma-vibhahga-sutta. They include the following:
A person who kills living creatures . . . tends to be shortlived, while a person who refrains from taking life . . . tends to be long-lived. A person who harms creatures . . . tends to be sickly, while one who refrains from harming creatures . . . tends to be healthy. One who is irritable . . . tends to be ugly, and one who is not irritable . . . tends to be handsome. A
person who is jealous . . . tends to be weak, while one who is otherwise . . . tends to be powerful. A person who is miserly . . . tends to be poor, while a person who is liberal . . . tends to be rich. A person who is humble . . . tends to be reborn in a good family, while one who is haughty . . . tends to be reborn in an evil family. He who does not consult the religious teachers for advice on what is good and bad . . . tends to be ignorant, while one who does so . . . tends to have great wisdom.90 From these specific correlations further generalizations were made that also had to be verified by telepathic insight. Thus, we find the theory that a person who leads an immoral life will be reborn in an evil state. These inductive inferences cannot be taken as the basis of absolute laws implying complete determinism. This is explicitly stated in the Maha-kammavibhahga-sutta, where the Buddha refers to some recluses and Brahmans who, by thorough application and concentration of mind, were able to see beings who had led an immoral life and were reborn in an evil state. As a result of this telepathic insight, they concluded, "He who takes life, steals, . . . who is of wrong views, will be reborn in an evil state after death. They who know this have right knowledge. Others are mistaken." 91 The Buddha rejected this conclusion as a very grave mistake. He pointed out that a person who sees a man reborn in a happy state after leading an immoral life comes to a conclusion that is diametrically opposed to the one given above.

The Buddha did not doubt the attainments of the person who perceives the phenomenon of rebirth and moral responsibility ;92 rather, he doubted the validity of the conclusion because certain aspects of the causal process were not taken into consideration. In the case of a person who led an immoral life but was reborn in a happy state, there may have been counteracting tendencies:
perhaps he led a good life during previous lives or in the present, or perhaps he held right views at the moment of death.93 An interesting illustration of this problem is found in the Lonaphala-vagga of the Ahguttara Nikaya.94 The Buddha says that if a person maintains that "Just as this man does a deed, so does he experience it," 95 then the living of the holy life would be meaningless; there would be no opportunity for the complete destruction of suffering. But if one accepts the theory, "Just as this man does a deed that would be experienced in a certain way, so does he experience its consequences," 96 this makes the religious life meaningful, and there is opportunity for the complete destruction of suffering. The former statement implies a form of complete determinism in the sphere of moral responsibility, comparable to the theory accepted by the Jainas. This may be why the Buddha considered it harmful to the religious life and the achievement of the goal. We agree with Woodward and Hare in GS that "this does not controvert the doctrine of the deed, but means that the particular kind of action does not find its exact replica in fulfilment." However, we cannot agree with their reason for that conclusion: "times and men and things are always changing." 97 The reasons for our disagreement may become clear in the following analysis.

The sutta describes similar deeds committed by two different people. For instance, one person may do a trifling evil deed, for which he ends up in hell. Someone else may do a similar trifling evil deed and experience the consequences in this life, not afterward.98 Thus, two people commit identical evil deeds but reap different consequences in different ways. In the case of the first person the consequence is magnified and is reaped in another evil existence. In the case of the other the consequences of the same evil deed are not powerful enough to lead him to an evil state after death, but are experienced in this life or are not even felt. The reason for this is not that "times and men and things are always changing" but that there are differences between the people committing these deeds. This is confirmed by the sutta itself: " Here a certain person has not properly cultivated his body, behavior, thought and intelligence (is inferior, insignificant), and his life is short (and miserable).99 With such a person . . . even a trifling evil deed done leads him to hell. . . . In the case of a person who has proper culture of body, behavior, thought, and intelligence [who is superior and not insignificant], and is endowed with long life, the consequences of a similar evil deed are experienced in this very life (and much of it or even a modicum of it would not be seen)." 100 This passage illustrates a salient feature of the Buddhist theory of moral responsibility. The effect (phala, pao) of a deed (kamma,
yeh) is not determined solely by the deed itself but also by the nature of the person who commits the deed and, we may add, by the circumstances in which it is committed. Several interesting metaphors are given in the sutta quoted above to illustrate this point, one of which may be summarized as follows: "If a man throws a grain of salt into a little cup of water, the water in that cup would become salty and undrinkable owing to that grain of salt.

But if a man were to throw a similar grain of salt into the river Ganges, because of the great mass of water therein, it would not become salty and undrinkable." 101 This illustrates further the danger of drawing absolute conclusions on the basis of generalizations.

From the description in the Maha-kammavibhahga-sutta it appears that some of the Buddha's contemporaries, although they had developed extrasensory powers by which they could verify the decease and survival of beings, had neglected certain important aspects of the causal process in drawing inferences. The views of those who denied moral responsibility have already been analyzed in detail (see chapter 2). They differ so sharply from the Buddhist theory that no confusion is possible. But the theory of complete determinism in moral responsibility adopted by the Jainas (see chapter 2) was very often confused with the Buddhist theory of moral causation. The main difference between the two is that the Jaina interpretation of karma is based on a theory of complete determinism (niyati), whereas the Buddhist conception is founded on the theory of causation (paticcasamuppada, yin yuan fa). The Buddha did not hold that everything is completely determined by one's past behavior (pubbekatahetu, yin pen tso). Acquisition of merit in the past (pubbe ca katapunnata) is only one of the factors that, along with " life in an appropriate surrounding" (patirupadesavasa) and "proper self-application" (in this life) (attasammapa-nidhi), contribute to an auspicious or good life.102 Moreover, according to the statement in the Maha-kammavibhanga-sutta, even an evildoer could be reborn in a happy state of existence if he held right views at the moment of death or had done good deeds in an earlier existence. Taking a specific instance of the causation of the human personality, the Buddha pointed out that "action or behavior (kamma) is the field, consciousness (vinnana) the seed, and craving (tanha) the moisture that lead to the rebirth of a being." 103 Therefore, behavior is only one of the causes that determine the nature of one's future life. The Milindapanha (p. 268) distinguishes things of the world according to their mode of genesis, for example, arisen on account of kamma (kammanibbatta), arisen on account of causes (hetunibbatta), and arisen on account of season {utunibbatta). Even in this case, the possible existence of counteracting causes is not ruled out. Thus, it is not complete determinism but conditionality that is the basis of the Buddha's theory of moral responsibility.104 Only in so far as behavior contributes, in this manner, to the determination of m an's future life, does a man have "kamma as his own, kamma as his matrix, kamma as his kin, kamma as his refuge." In this way kamma is said to divide beings as inferior and superior.105 Therefore, according to Buddhism, there is no need to expiate for past actions or to avoid performing actions in the future. W hat Buddhism emphasizes is the avoidance of evil actions, cultivation of morally good actions, and the purification of the mind106 as the way to attain perfect happiness.

Causation Of Social Phenomena

One of the major social philosophies dominating the life of the Indians before the rise of Buddhism was the caste system. Enunciated in the Purusa-sukta of the Rgveda, the system of four social hierarchies was said to be divinely ordained.107 The Pali Nikayas and the Chinese Agamas abound with refutations of this social theory.108 Of special significance is the Agganna-suttanta, which, in its refutation, presents an evolutionary account of the world. At this time the Brahman caste claimed the highest position in this social hierarchy: "Only a Brahman is of the highest social class; other classes are low. Only a Brahman is of white complexion; others are dark. Only Brahmans are of pure breed, not the nonBrahmans. Only Brahmans are genuine children of Brahma, born of his mouth, offspring of Brahma, created by Brahma, heirs of Brahma." 109 In the Agganna-suttanta, after giving an evolutionary account of the world, the Buddha explains how the different social grades came into existence. Following is a summary of this description.

Then differences of sex appeared; households were formed; and the lazy stored up the rice instead of gathering it each evening and morning; and the rights of property arose and were infringed. And when lusts were felt and thefts committed, the beings, now men, met together and chose men differing from them in no way except in virtue (dhamma) to restrain the evildoers by blame or fines or banishment. These were the first khattiyas. And others they chose to restrain the evil dispositions that led to the evildoing. And these were the first Brahmans, differing only in virtue (dhamma). Then certain others, to keep their households going, and maintain their wives, started various occupations. And these were the first vessas. And some abandoned their homes and became the first recluses (samana). But all these were alike in origin, and the only distinction between them was in virtue.110 The Buddha thus insisted that caste and other divisions in society were occupational in origin and maintained that one did not have to follow a particular caste merely because he was bom to parents who followed that caste. Moreover, to refute both the Brahmans' claim to superiority and the universal validity of the caste system, the Buddha cited existing societies with only two castes, masters (ayya, ta chia) and slaves (dasa, nu), and pointed out that even theirs were not rigid social divisions because a slave could become a master and a master a slave.111 From the spiritual standpoint, the Buddha considered moral life the factor that determines the status of beings; from the social and economic standpoints, he held that wealth creates the differences: "A sudra acquiring wealth and fame can command the services of even the ksatriyas, brahmanas and vaisyas."112 Apart from this Vedic theory of determinism in social life, there is another philosophical theory of natural determinism in the sphere of social life that does not seem to have exerted much influence on contemporary society. It is an application of the fatalistic theory of natural determinism propounded by the Ajivikas (see chapter 2). This social philosophy is attributed to Purana Kassapa, who was one of the leading Ajlvika teachers. He believed that human beings belong to one of six types of existence or species
(abhijati): the black species (kanhabhijati), the blue species (mlabhijati), the red species {lohitabhijati), the yellow species (haliddabhijati), the white species (sukkabhijati), and the pure white species
{paramasukkabhijati).113 Malalasekera and Jayatilleke have raised doubts that these colors denoted differences in physical complexion and suggest that the classification implies genetically different physical and psychological types.114 But considering the various types of people included under the different categories, we cannot say whether the classification implies genetically different physical and psychological types. The groups are described thus:

1. Black species: Butchers, raisers of fowl, hunters, fishermen, dacoits, executioners, and all who adopt a cruel mode of living.
2. Blue species: Monks leading ascetic lives115 and other people who accept the doctrine of moral responsibility.
3. Red species: The Niganthas (Jainas) who wear only one robe.
4. Yellow species: Laymen who wear white robes and the disciples of the naked ascetics.
5. White species: Male and female Ajlvika disciples.
6. Pure white species: Nanda, Vaccha, Kisa, Sankicca and Makkhali Gosala.

Cursory examination of the list above reveals that these groups do not represent categories differing in physical appearance but denote people graded according to the degree of moral advancement judged by Ajlvika standards. Thus, even the lay disciples of the Ajivikas were considered superior to the disciples of other religious teachers. The Ajivikas in general are considered to be of the white species, while the Ajlvika teachers belong to the pure white species. We have seen that the Brahmans considered themselves white, while all the rest, even the ksatriyas and the vaisyas, who belonged to the same stock and were thought to be fair-skinned as opposed to the dark aborigines, were considered black. The AjTvikas adopted this principle, increasing the number to six groups denoted by colors.

All these "typologists" believed that man is bom into the various groups as a result of fate and that a person is incapable of altering it by his own will or effort. Thus, both these schools of thought, the Brahmans insisting on the divine ordination of social rank, and the Fatalists emphasizing strict determinism, presented social philosophies that were static.

Dismissing these views, the Buddha gave an evolutionary account of human society. He pointed out that "living in an appropriate surrounding" (patirupadesavasa) was a factor that contributed to the moral and spiritual advancement of the individual.

Being concerned for the welfare of living beings, the Buddha could not neglect their life in society. As a social reformer, therefore, he was led to analyze the causes of social evils and suggest remedies. Just as Buddhahood is the goal of those who have renounced the world in search of the perfect happiness of nibbana, so is universal kingship (cakkavatti-rajja, chuan lun sheng wang) the ideal for the layman who chooses to live the secular life to perfection. Therefore, in Buddhism, the social or secular philosophy is on several occasions set forth by a universal monarch.

A universal monarch is said to be a person who has conquered the whole world, not by force but by virtue.116 But that state cannot be attained hereditarily. The Cakkavatti-sThanada-suttanta relates an incident about the accession of a son to the throne of his father, who was a universal monarch. Immediately after his accession, the glories of a universal monarch that the son was supposed to have inherited disappeared. Very much dejected, the son reported this to the father, who said: "The glories of a universal monarch cannot be considered a paternal heritage (pettikam dayajjam,fu ch'ari)."111 The implication is that one cannot even claim material possessions, not to speak of moral and spiritual distinction, on the grounds of one's birth. This is certainly a reaction against the claims of the Brahmans to superiority on the basis of their birth into that particular caste. Moreover, it emphasizes that even the highest secular position has to be earned, not inherited. According to the Lakkhana-suttanta the state of a universal monarch can be attained by a person who leads a virtuous life.118 The duties of a universal monarch are to impart moral instruction and to look after the moral as well as the material advancement of the people. For this it is necessary to analyze the causes of social evils and attempt to remedy them. In the Cakkavatti-sThanada-suttanta we come across an instance where a universal monarch made such an analysis. He found that "As a result of the nonaccruing of wealth to the destitute, poverty increased; when poverty increased, there was a rise in thefts; when thefts increased, there was escalation of violence; when violence was rampant, there was an increase in m urder; when murder increased, lying became com m on; when lying became common, the life span as well as the comeliness of human beings diminished." 119 The suttanta goes on to describe how all the social evils, including stealing, improper sexual behavior, hate, jealousy, disrespect of parents, elders, and teachers, were caused as a result. This is a strictly causal account of social evils, and it is interesting to note that poverty and maldistribution of wealth were considered major causes. According to the Agganna-suttanta, the king was first appointed when these evils first appeared in society, and his duty was to uproot their causes and prevent such evils from arising again.120 Since maldistribution of wealth was one of the main causes of social evils, it was the duty of the king to find ways by which people could obtain wealth,121 for material or spiritual prosperity could not be achieved through praying (patthanahetu) or begging (dyacanahetu).122 But the acquisition of wealth alone was not the solution, for some people, while protecting their own share, would undoubtedly try to appropriate what belonged to others.123 This implies that maldistribution of wealth is not the only cause of social evils. Equally important causes, such as greed, are mental tendencies found in the destitute and rich alike. For this reason, the universal monarch must instruct the people in spiritual (dhamma) advancement as well as material (attha) advancement.124 In fact, Buddhism emphasizes the mental tendencies that are the causes of social evils. This is because material progress alone cannot bring about the changes necessary for the moral advancement of man, although it is a prerequisite. Psychological tendencies such as greed and aversion must be gradually eliminated. Although these psychological tendencies depend on external things or sense data (nimitta), pleasurable (sukha) or unpleasurable (patigha), they arise primarily from lack of understanding or improper reflection
(ayoniso manasikara) on the objects that produce these evil tendencies in m an.125 Hence the importance of knowledge and mental concentration as means to the elimination of causes that give rise one's own suffering and that of others.

Causation Of Spiritual Phenomena

The Buddha criticized three of the existing theories of causation on the grounds that they were harmful to religious life (see chapters 1, 3). Acceptance of the belief that one's happiness and suffering is determined by an external agent such as God meant the surrender of one's freedom and ability to work out one's own salvation. In opposition to this theory of theistic determination, the Buddha held that "purity and impurity depend on oneself, and nobody can purify another." 126 Neither the theory of determinism in moral responsibility advocated by Mahavlra (chapter 2) nor that adopted by the Ajivikas (chapter 2) left room for individual freedom. The theory of indeterminism, on the other hand, led to the denial of the efficacy of religious life because one could not be sure of what would happen during the next moment.

Acceptance of a theory of causal dependence, not only in individual and social life but also in the physical world, enables one to put an end to suffering by removing the causes that produce it. Therefore, the Buddha maintained that there are causes for the defilement, and hence the purity, of m an.127 The Bodhisattvabhumi explains how the processes of defiling (sahklesa) and of purifying (vyavadana) take place according to the ten causes (dasabhir hetubhih).128 It was pointed out that wrong understanding of, or reflection on, the perceptual world produces attachments or aversions that lead to most of the suffering in the world. Proper reflection (yoniso manasikara) implies reflection according to the genesis (yoni) of things, that is to say, reflection on the causality of things. The purpose is to avoid the two extreme views of eternalism (sassata-vada) and annihilationism (ucchedavada), which are said to promote evil tendencies such as egoism. Knowledge of causality should go hand in hand with restraint of the senses (indriyasamvara, hu ch'u ken), which enables one to cut at the roots of craving. Thus, the religious life is directed at cutting the tangle of wrong views and developing insight (pannavimutti, hui chiai t'o).129 The outcome of this release is attainment of the knowledge of emancipation, the knowledge that one has "put an end to rebirth, that the higher life has been lived to its perfection, and that there is no hereafter." 130 This final knowledge is not attained by a beginner all of a sudden.

The Buddha declared: "I do not say that one can win final knowledge at the very outset; it is attained by a gradual discipline, a gradual mode of action and conduct." 131 The stages of the attainment of this final knowledge are described in the Nikayas and the Agamas.132 Briefly, they consist in practice of the virtuous fife (ariyena sTlakkhandhena, sheng chieh chu), followed by restraint of the senses (indriyasamvara, hu ctiu ken). When one is confronted by a sense object, he does not allow evil tendencies such as covetousness and displeasure to flow in; thus he restrains the senses. He then develops mindfulness (satisampajahha, cheng chih ch'u ju) and strives to eliminate the "five impediments." This leads him to the first stage of the jhana. By developing the mind further he is able to reach the fourth jhana, where the mind is so serene and supple that he is able to develop the sixfold higher knowledge (abhihha, chih t'ung). Three of these six are essential, but not necessary, for knowledge of emancipation. (1) Retrocognition (pubbenivdsdnussatinana, su ming chih) allows him to verify the fact of preexistence. According to the Bodhisattvabhumi, this knowledge is essential for the realization that the theory of etemalism (sasvatavada) posited by some of the sramanas and brahmanas is invalid.133 (2) Clairvoyance (cufupapatahana, sheng szu chih) enables him to verify the decease and survival of beings and the doctrine of karma. (3) Knowledge of the destruction of defiling impulses (asavakkhayahana, lou chin chih) is necessary for verifying the four Noble Truths. "As he thus knows and sees, his mind is emancipated from the inflowing impulses of sensuous gratification (kam'asava, yu lou), of personal immortality (bhav'asava, yu lou), and of ignorance (avijj'asava, wu ming lou). Along with this emancipation arises the knowledge that emancipation has been attained." 134 The Samyutta Nikaya gives a strictly causal account of the various stages of the path to enlightenment.135 This was in opposition to the view of Purana Kassapa that there is no cause or condition for the lack of knowledge and insight or for the presence of knowledge and insight.136 So far, the discussion has dwelt on the various causal patterns pertaining to the world of normal experience and the realm of spiritual life, or the attainment of freedom. But how does a Buddha or any other saint who has already attained freedom fit into this scheme? How is his behavior determined?

It was pointed out earlier that most of the suffering man experiences in this world is due to the way his perceptual process is conditioned. Among other things, the understanding of this perceptual process, followed by the stopping of evil impulses or defilements (asavakkhaya, lou chin), constitutes knowledge and freedom. Looking at the various aggregates constituting the psychophysical personality as being nonsubstantial (anatta) and preventing ego-consciousness from assailing one when perception takes place, a learned Aryan disciple has revulsion for (nibbindati) the physical form (rupa), feeling or sensation (vedana), perception
(sahha), dispositions (sahkhara) and consciousness (vihhana). Having revulsion, he is not attached (nibbindam virajjati) ; being nonattached, he is freed (viraga vimuccati) ; in the person who is thus freed there arises the knowledge of freedom (vimuttasmim vimuttam iti hanam hoti): "Destroyed is birth (khvna jati); lived is the higher fife (vusitam brahmacariyam); done is what ought to be done (katam karatuyam); and there is no future existence (naparam itthattaya fi)." 137 This means that the elimination of ego-consciousness by the development of insight can change the normal process of perception. With the attainment of mental concentration or restraint
(samvara), one is able to prevent the influx of impurities such as attachment (raga) and aversion (patigha). Thus, in an enlightened one, perception simply does not generate obsessions and the consequent suffering; instead, as a result of his not grasping onto things, including his own personality, as being substantial, he becomes detached (viraga). Detachment produces freedom (vimutti),
through which one may attain stability (thitata) of mind so as not to be agitated by gain (labha) or loss (alabha), reputation (yasa) or disrepute (ayasa), blame (ninda) or praise (pasamsa), happiness
(sukha) or suffering (dukkha)—the eight wordly phenomena (atthaloka-dhamma) by which one is constantly assailed in this life.138 The highest point of 'blessedness' (mahgala) is achieved, according to the Maha-mahgala-sutta,139 by "one whose mind is not overwhelmed when in contact with wordly phenomena (lokadhamma),
is freed from sorrow, taintless and secure." Such a person feels secure and at peace in the midst of all the destruction and confusion prevailing in this world. This form of behavior is described as "going against the current" (patisotagamT), but it is still a causal process where each state is conditioned by a previous state. It is called "going against the current" because, unlike an ordinary man, the person who has attained emancipation (sammadahhavimutto) does not allow any attachment (raga) to arise in him when he perceives a pleasurable object, even though he experiences a pleasurable feeling.140 Hence for him there is no grasping; in the absence of grasping he is not smeared by the world (anupalitto lokena).141 When he is not smeared by the world he remains in a state of perfect happiness (paramasukha).

It appears from the foregoing analysis that the causal process is operative in all spheres, including the highest state of spiritual development, namely, nirvana. But the later scholars attempted to distinguish two spheres, one in which causation prevailed and the other which is uncaused. This latter view was, no doubt, the result of a confusion in the meanings of the two terms, sahkhata ('compounded') and paticcasamuppanna ('causally conditioned').

We have already pointed out (see chapter 5) that sahkhata and paticcasamuppanna, although used to refer to the phenomenal world, connote two different meanings. The former, it was found, refers to anything that is 'compounded', that is 'organized,'
'planned,' or 'put together,' therefore, conditioned by the dispositional tendencies (sahkhara) of man. On the contrary, paticcasamuppanna refers to that which is 'naturally conditioned,' i.e.,
'causally conditioned.' For this reason, although both terms were used to describe the phenomenal world, only the former in its negative form (asahkhata, wu wei) is used to define nirvana. Since in the Buddha and the arahants " all dispositions have been completely calmed" (sabbasahkharasamatha), that state of freedom is called the 'element of the unconditioned or uncompounded' (asahkhatadhatu), and is always defined in terms of the absence of attachment (ragakkhaya), of aversion (dosakkhaya), and of confusion
(mohakkhaya).142 It was never described as 'the uncaused' or 'the in-dependent' (appaticcasamuppanna). This means that nirvana is a state where there is 'natural or causal happening" (paticcasamuppada), but not 'organized,' or 'planned' conditioning (sahkharana). This fact was completely overlooked by the scholars who were indiscriminate in defining sahkhata as a synonym (paryaya) of paticcasamuppanna.143 The Twelvefold Formula of Causation So far we have been discussing the Buddha's explanations of different causal situations. In addition to these different analyses, we come across a recurring, twelvefold formula in the early Buddhist texts, a formula that was intended to explain important questions about man and his destiny. This special formulation of the causal principle, which dominates the early Buddhist texts, is stated thus :
When this exists, that exists or comes to be; on the arising of this, that arises. When this does not exist, that does not exist or come to be; on the cessation of this, that ceases. That is to say:
on ignorance depend dispositions; on dispositions depends consciousness; on consciousness depends the psychophysical personality ;
on the psychophysical personality depend the six
'gateways' ;
on the six 'gateways' depends contact; on contact depends feeling [or sensation] ; on feeling depends craving; on craving depends grasping; on grasping depends becoming; on becoming depends birth; on birth depend aging and death.

In this manner there arises this mass of suffering.144 Because this formula dominates the early Buddhist texts, many scholars have considered it to be the only aspect of causation discussed in Buddhism.145 The preceding account of the various spheres in which the principle of causation operates, as well as the discussion in chapter 5, shows that this is not the case. Moreover, some of these scholars have maintained that the purpose of the special formulation is to explain the origin and cessation of suffering (idukkha). Keith says : "We can now see the limited character of the Chain of Causation, it is intended to explain the coming into being of misery . . . ." 146 This evaluation seems to take into account only one aspect of the special formulation, to the neglect of the other important aspects. It is possible to maintain that the ultimate purpose of the special formulation is to explain the origin and cessation of suffering. But other important issues are also involved.

We have already seen how some of the Upanisadic thinkers, who were able to verify the continuity of the human personality either rationally or intuitionally, came to believe in eternalism (sassatavâda, chang chieri), which they defended by a metaphysical theory of self-causation (see chapter 1). On the other hand, Materialists and Âjïvikas denied self-causation and adopted a theory of external causation, which led them to believe in annihilation
(uccheda, tuan) of the human personality at death and also of karma
(see chapter 2).

The Buddha, for whom karma and rebirth were realities, was reluctant to contribute to any one of these metaphysical theories.

The empiricist approach of the Buddha prevented him from positing an unverifiable soul to explain the continuity of the individual after death. On the other hand, he was far removed from the materialist approach denying the continuity of the individual and his moral responsibility. Thus, the problem he confronted was to explain the working of karma and the process of rebirth without falling into the two extreme metaphysical theories of self-causation and external causation. As Jayatilleke points out, the raison d'être of the special formulation of the causal principle " lies in the necessity to give a causal account of the factors operating in maintaining the process of human personality and thereby of suffering." 141 This is clearly expressed in a passage from the Samyutta : " In the belief that the person who acts is the same as the person who experiences [the result] . . . he posits eternalism; in the belief that the person who acts is not the same as the person who experiences [the result] . . . he posits annihilationism. Avoiding both these extremes, the Tathàgata preaches the doctrine in the middle. On ignorance (avijjâ,
wu ming) depends dispositions (sank heir a, hsing). . . . In this manner there arises this mass of suffering." 148 The theory of causation is placed not only against these two theories but also against two other metaphysical theories, a combination of self-causation and external causation and fortuitous origination (see chapter 2). Keith has written that by opposing the Buddhist theory to all preBuddhist theories, the foregoing passage in the Samyutta Nikâya places Buddhist doctrine in a difficult position. That is because " all these issues belong to the realm of the indeterminates." Therefore, he concludes, " We obtain nothing more than the vague general assertion that things as compounded come into being under the effect of causes, but we have to put beside this the doctrine that we do not know anything definite as to their operation; . . . ," 149 The four theories against which the Buddhist theory of causation was preached represent a fourfold scheme. These four alternatives were dismissed by the Buddha with the words, " Do not [ask]
thus" (ma h ' evatn), because he considered them to be indeterminate
(avyakata, wu chi),ls0 and therefore to be set aside. They are indeterminate because categorical answers to the first two alternatives
(and therefore also to the third and fourth alternatives, which represent the assertion and denial, respectively, of the combination of the first two) lead to metaphysical theories to which the Buddha was reluctant to contribute. Without being a partisan of any one of these metaphysical views, the Buddha adduced empirical causal explanations. Thus, it is unfair to equate the Buddha's theory of causation with those of the pre-Buddhist teachers, as Keith does.

Further, the formulation of the special theory giving empirical causal explanations of the birth and development of the individual eliminated other metaphysical problems such as creation by God, First Cause, and even Final Cause. This was observed by Buddhaghosa, who raised the question, " Is ignorance [which comes first in the explication of the special formulation] like the primordial matter (pakati) of the Sankhya school of thought (pakativadmam), an uncaused first cause of the world?" And he gives the following reply: " It is not uncaused. The cause of ignorance has been declared when it was said 'On account of the defilements (asava) ignorance arises.' " 151 In support of the view that ignorance is not without a cause (akaranam) he quotes a passage from the N ikayas:152 " The first beginning of ignorance is not known [for us to maintain that]
' before this there was no ignorance; at this point there arose ignorance.' But the fact that ignorance is causally produced can be known." 153 This means that the special formulation cannot be designated a 'chain' of causation because no absolute beginning is envisaged. On the contrary, it is better represented by a circle, without beginning. Thus, the special formulation has come to be known as vatta-katha,154 In the Pali Nikayas and the Chinese Agamas the special application of the causal formula is introduced in many ways. Sometimes it is introduced directly by the statement, " I will preach to you, O monks, the doctrine of causation." 155 Other times, the formula is presented in explanations of such things as aging and death
(jaramarana, lao szu), the four forms of nutrition (ahara, shih), the five aggregates (khandha, yin) constituting the individual, the causality of moral behavior, or amidst criticism of some current philosophical theories.156 Doubts have been raised about how the general formula
( " When this exists, that exists ….") came to be prefixed to the statement of the twelvefold formula. Thomas believes that the coupling of the two was a later addition.157 The philosophical importance of the general formula of causation, as well as the place accorded it in early Buddhism (chapter 5), does not warrant such an assumption.158 In the passage quoted above the Buddha is represented as demonstrating his intellectual powers by referring to his knowledge of the arising and passing away of the psychophysical personality. The general formula of causation was something that he discovered with his attainment of enlightenment.

Therefore, when he had to explain the arising and passing away of the psychophysical personality, he seems to have adopted the more instructive method of stating the formula first and then applying it to explain the causation of this personality. This is quite a logical procedure. Moreover, in most of the sutras of the Samyukta Agama,
where the theory of the twelve factors is discussed the general formula precedes it, even though this does not occur in the Pali counterparts.159 Considering the large number of passages in the twelfth fascicle of the Samyukta Agama (which roughly corresponds to the Nidana Samyutta of the Samyutta Nikaya), where the general formula has been prefixed to the theory of twelve factors, it would be difficult to reject them as late compositions, as Thomas does. On the contrary, the prefixed version may even be an earlier version, and the practice of prefixing the general formula may have been abandoned when it was taken for granted that the special formulation represented an application of the general formula.

Several modern scholars have made important analyses of the twelvefold formula. A brief account of this formula concludes our analysis of the causal principle in early Buddhism.

Ignorance (avijja, wu ming) heads the list of twelve factors.

But, as pointed out earlier, it is not presented as the beginning of a process but as the most important factor to eliminate in seeking enlightenment and hence in disrupting the worldly process. It is explained in various ways. Ignorance is said to determine the dispositions (sahkhara, hsing), in the sense that in the absence of correct knowledge about the nature and destiny of the individual, one's dispositions are determined in a way detrimental to one's future. These dispositions give shape to one's consciousness
(vihhana, shih), which in turn tends to determine one's current psychophysical personality (namarupa, ming se) as well as the psychophysical personality one inherits in the next life. Depending on the psychophysical personality, there arise the six senses (saTayatana, liu ju chii). The activity of the senses leads to contact (phassa, chu),
which brings about feeling or sensation (vedana, shou). This psychological process generates deliberate activity, and the result is the arising of craving (tanha) (or its opposite, revulsion, patigha, which ultimately can be traced back to craving). Craving is said to be of three types: craving for sense pleasures (kama-tanha, yu ai), for existence ( bhava-tanha, yu ai), and for nonexistence (vibhava-tanha, wu yu ai). Craving leads to grasping or clinging (upadana, ch'u),
which culminates in becoming ( bhava, yu) in the sense of rebirth
(punabbhavabhinibbatti, t'eng laiyu). Becoming is followed by birth (jati, sheng), with its associated suffering.

Thus, as C. A. F. Rhys Davids concludes: " In the central links we have the working out of the process of sentience, culminating in the central links—sense, feeling, desire—and representing a fresh ebullition, a new source of causal force reaching on into the next birth. There its resultant is renewed sentience, eventually again to be darkened by the inevitable disease-decay-death—a centre of effects in sentience due to causes in the past." 160 These past causes have been simplified and given in abstract form, while the present is analyzed in detail from conception to grasping for another life.

Thus it is difficult to agree with Beckh, who maintains that the idea of the 'chain' cannot be spread over three lives.161 Several attempts have been made to compare the special formulation of the causal principle with the Sankhya series, based mainly on such slender evidence as the similarity of terms used.162 Jacobi and Pischel believe that the theory is derived from Sankhya. Keith sees close parallels. Senart finds borrowings only in the first two terms, arguing, " if ignorance is, as in Buddhism, empiric, it has no claim to head the list of terms." 163 The nature of ignorance and the reason for its placement at the head of the formula have been discussed earlier. These views may have originated in a misinterpretation of terms used in the special formulation as well as a wrong assessment of the purpose for which the theory was formulated. First, the Sankhya theory purports to explain the evolution of the world from the primordial source (prakrti). No such thing is envisaged by the Buddhist theory, which is mainly intended to explain the problem of rebirth and moral responsibility, especially in relation to the individual. Second, since the Sankhya accepted a theory of self-causation (satkaryavada), to them each factor in the causal series is produced out from the other. But such a relation is not proposed in the Buddhist theory (chapter 1). Keith's misinterpretation of the causal formula prevented him from agreeing with Oltramare, who gave a reasonable analysis of how the theory came to be propounded.164 Keith wrote that the suggestions made by Oltramare are ingenious but too coherent and logical to be primitive.165

Vii. Later Developments

ON THE BASIS of the foregoing analysis, we maintain that the teachings preserved in the Pali Nikayas and the Chinese Agamas show no significant difference, at least with regard to the conception of causality. Sometime after the parinibbana of the Buddha, the Buddhists attempted to systematize the teachings scattered throughout the Nikayas and the Agamas. This resulted in the emergence of the different schools of Abhidharma, each possessing an Abhidharma Pitaka of its own. At least the two major Abhidharmika schools, Theravada and Sarvastivada, have preserved their Abhidharma Pitakas, which vary considerably.

The Theravada form together with its literature, both canonical and commentarial, was introduced into Ceylon during the reign of the great Indian Emperor Asoka (3rd century B.C.). There it developed in relative isolation, and thus many of the original ideas could be preserved from amalgamation with the new conceptions and theories propounded by the later schools of Buddhism, at least until the time of Buddhaghosa. The Sarvastivada school became popular mainly in northwestern India. Which of these schools represents the earliest phase of Buddhism has been a subject of much controversy. Stcherbatsky maintains that " the Vaibhasikas are the only continuators of one of the oldest schools, the Sarvastivadins. They derive their name from the title of a huge commentary upon the Kanonical works of this school and follow in philosophy generally the same lines as did the original school." 1 Murti also sees no major difference between the two schools: " The Theravada and the Sarvastivada, in spite of some important differences, may be considered as representing one metaphysical standpoint." 2 It has been pointed out that as regards the conception of dharma, the Theravada and Sarvastivada differ considerably. Two of the most important ideas associated with the conception of dharma that developed later are (1) the theory of moments ( ksana),
and (2) the theory of atoms (paramanu). These two theories are conspicuous by their absence in the earlier phase of Theravada
(before Buddhaghosa). In the Theravada tradition we first encounter a theory of moments in the works of Buddhaghosa. This is clearly evident from a statement found in Buddhaghosa's AtthasaliriT. Commenting on the 'present' (paccuppanna), he says,
" Herein, the continuous present (santati-paccuppanna) finds mention in the commentaries; the enduring present {addha-paccuppanna) in the suttas. Here some say that 'the thought existing in the momentary present (khana-paccuppanna) becomes the object of telepathic insight.' " 3 This implies that the theory of moments was not among the doctrines embodied in the Sutta Pitaka or the commentaries preserved at Mahavihara, the center of Theravada Buddhism in Ceylon. If so, the incorporation of this theory in the Pali Commentaries may be taken as the work of Buddhaghosa. Even the theory of atoms (paramanu) was not found in the Sutta Pitaka or the commentaries of the Theravadins. Buddhaghosa seems to have made a halfhearted attempt to introduce it also into the Theravada tradition.4 Thus, two of the most im portant theories, which created innumerable philosophical problems for the Sarvastivadins as well as the Sautrantikas, were not found in the pre-Buddhaghosa Theravada tradition. This is the main reason why we consider the original Theravada distinct from the Theravada embodied in the commentaries of Buddhaghosa.s It was pointed out in chapter 4 that with the acceptance of the theory of moments, the Sarvastivadins had to explain the problem of continuity. This they did by accepting a theory of 'own nature'
(svabhava). In fact, Yasomitra (who had leanings towards the Sautrantika school), commenting on the Abhidharmakosa, maintained that " by 'own nature' means by the 'self.' " 6 That is why all the other Buddhist schools criticized the Sarvastivada teachings as heretical.

The Sarvastivada theory of 'own nature' left its impressions on the Sarvastivada theory of causation, too. As pointed out above
(chapter 3), the Sarvastivadins distinguished between cause (hetu) and condition {pratyaya) because they accepted the substantialist standpoint that cause and the effect are connected by their 'own nature' (svabhava). The Sarvastivadins themselves admit that they are 'substantialists' (sadvadi).7 This is almost identical with the theory o f 'everything exists' (sabbam atthi) rejected by the Buddha because he thought it would lead to a belief in eternalism {sassataditthi) (chapter 1). Thus, not only was the theory o f 'own nature' identical with the theory o f'substance' or 'self' (atman), as pointed out by Yasomitra, but it also tended toward eternalism {sasvatadrsti); hence the view of the Sarvastivadins that things (i.e., 'own nature') exist during past, present, and future.

If so, it is difficult to agree with Murti that the Sarvastivada
(or more exactly, Vaibhasika) theory of causation is a nonidentity theory (asatkaryavada).8 The evidence adduced above goes against the view that the Sarvastivadins perceived a complete difference between a cause and its effect. For them to have considered cause and effect as completely different entities would have made their theory of 'own nature' meaningless. In fact, as will be pointed out later, the Sautrantikas affirmed a difference between cause and effect " because there was no 'own nature' (svabhava) connecting them."
In the very first stanza of the first chapter of the Mulamadhyamaka-karikd, Nagarjuna refers to four types of causal theories: (1) self-causation (svata-utpatti), (2) external causation (paratautpatti), (3) both (i.e., self-causation and external causation, dvdbhyam), and (4) noncausation (ahetu). In the second stanza he refers to the Buddhist theory of four causes or causal correlations {pratyaya). Murti seems to consider that this theory of four pratyayas comes under the category of external causation, probably because some Buddhist schools, while accepting the theory of pratyayas, considered the cause as being different from the effect.

If M urti were right, we would except Nagarjuna, after stating the four types of causal theories (stanza 1) to criticize the first type, self-causation. Immediately after enumerating the four pratyayas, Nâgârjuna analyzes the nature of the causal relations and says: " The 'own nature' of the existents (bhâva) is not found in the causes
(pratyaya)." 9 Thus, the reference here is to the theory presented by the Sarvâstivâdins, who, while accepting the theory of four pratya-yas, also believed that cause and effect are related to each other by way of 'own nature.' Therefore, to Nâgârjuna, the Sarvâstivàda theory of causation is a theory of self-causation, not a theory of external causation.

The theory of causality propounded by the Sânkhya school is generally known as the 'identity theory' (satkârya-vâda). This is because, according to the Sânkhya school,prakrti is the 'primordial m atter' out of which the world evolved, and this prakrti persists in the products of evolution, too. Thus, the cause and the effect are identical in essence because they are 'made o f' prakrti. Now this prakrti is sometimes called svabhâva ('own nature').10 This shows the very close resemblance of the Sarvâstivâda theory to that of the Sânkhya.

The Sânkhya conception of evolution seems to be a systematic exposition of the ideas presented by thinkers such as Uddâlaka during the Upanisadic period, with the difference, as pointed out by Sankara (chapter 1 ), that the Being (sat) of Uddâlaka is sentient while the prakrti of the Sânkhya is insentient. This difference is also, to some extent, reflected in their theories of causation. In the theory of Uddâlaka the 'cause' was looked upon more as a 'sentient being' ;
hence causation is one o f 'self-causation' (svayamkrta-vâda). In the Sânkhya school the 'cause' is considered 'insentient,' and therefore causation consists of 'self-generation' or 'generation out of itself' (svatotpatti- or satkârya-vâda).11 Yet in both cases the basis is the same. The former recognizes a substantial agent (âtman), and the latter affirms a substance (svabhâva = prakrti) by which the identity of the cause and the effect is maintained.

We have seen how the Upanisadic theory of self-causation was criticized and rejected by the Buddha. One of the main arguments adduced by the Buddha was that this view leads to belief in a permanent and eternal self or soul. The same can be said of the Sânkhya theory, which leads to a permanent and eternal substance. The implication of the Sarvâstivâda theory is not very different. The Sarvâstivâdin admission that cause and effect are related by way of 'own nature' (svabhâva) implies that this 'own nature' is the 'substance' (dravya) that survives through the past, present, and future and is therefore permanent and eternal. This is why they maintained that substance (dravya) exists (asti) during the past, present, and future. Therefore, it is possible to conclude that the Sarvastivada theory of causation, along with that of the Sankhya, falls into the category of 'self-causation' that came to be known as the 'identity theory' {satkaryavada), but not under the category of the 'nonidentity theory' {asatka.ryava.dd), as Murti seems to believe. Moreover, as pointed out earlier, the Sarvastivadins claimed to be 'substantialists' (sad-vadi).

If the Sarvastivada theory of causation is a parallel form of the identity theory {satkaryavada) of the Sankhya school, which of the Buddhist schools accepted a theory similar to the nonidentity theory {asatkaryavada) of the Vaisesika school ? Explaining the conception of dharma presented by the Sautrantikas, we pointed out that it is based on the theory of moments. The Sautrantikas recognized only two moments, nascent (utpada) and cessant (vyaya),
and rejected the static moment {sthiti-ksana). Since each moment was considered to be different from the-other, and since no underlying substratum (like the svabhava of the Sarvastivadins) was recognized, they maintained that there was only a series of moments that succeeded one another, the causation of each individual moment being reduced to invariable antecedence. W hat is perceived as duration is only the series of successive moments with a continuous flow. It was pointed out that the Sautrantikas had to solve yet another problem, the origin or the beginning of the series (chapter 4). It was to explain this problem that the Sautrantikas said that a thing being nonexistent comes into existence {abhutva bhava utpada, pen wu chin yu sheng).

The view that a thing being nonexistent comes into existence seems to have been the basis of the theory of causation that came to be known as the 'nonidentity theory' {asatkaryavada). As pointed out above, M urti's attribution of this theory to the Sarvastivadins seems to have no basis. All the available evidence indicates that it was the Sautrantikas who advanced such a theory. A statement in the Siksasamuccaya runs thus: "A thing, being nonexistent, comes into existence, and having come into existence, passes away, because it has no 'own nature' (svabhava)." 12 The theory of abhutva bhava utpada, therefore, contradicts the Sarvastivada conception of dharma-svabhâva. This is further exemplified by CandrakTrti's statement: " Thus heat is said to be without 'own nature' [= substance], because fire itself is associated with causes and conditions.

Fire, by being previously nonexistent and coming into existence later, is contingent or causally produced." 13 Finally, the Sphutârthâbhidharmakosa-vyâkhyâ definitely attributes this theory to the Sautrântikas.14 The Sautrântikas, whose theory of abhütvâ bhâva utpâda is almost identical with the asatkaryavada of the Vaisesikas, were questioned as to why the sesame seed should produce oil, not any other substance, though they are equally nonexistent in the causal entity. Their reply was that there cannot be any questioning with regard to the ultimate laws of nature, which are unthinkable and beyond the scope of speculation.15 After examining in detail the arguments for and against the Buddhist theory of momentariness, Mookerjee says: "From the elaborate exposition of the theory of causation with its confused tangle of criticism and counter criticism, . . .one cannot resist the impression that the Sautrântika has failed, in spite of his logical acumen and wealth of dialectic, to carry any conviction. The fact of the matter is that causation is as unintelligible in the theory of flux as in the theory of permanent cause." 16 Thus, it was left to Nâgârjuna and Sankara to expose this, and they very successfully made use of their dialectics to prove the inherent contradictions both in the theory of satkarya (production of a potentially existing effect) and in the conception of asatkarya (production of a previously nonexistent effect). Murti gives a lucid account of the Mâdhyamika criticism of asatkaryavada, which we need not repeat here.17 When Das Gupta said that " the effect according to the Buddhists was nonexistent, it came into being for a moment and was lost," 18 he was not confusing the non-Buddhist theory with the causal theory of the Buddhists, as Jayatilleke seems to think.19 He was referring to a theory of causation actually held by one of the schools of Buddhism, the Sautrântikas. The wrong impression conveyed by Das Gupta is that this theory was accepted by all the early Buddhist schools.

An attempt has been made by de la Vallée Poussin to equate the Sautrântika theory of abhütvâ bhâva utpâda with the conception of causation in the Pali Nikâyas. He quotes a statement pertaining to causation from the Majjhima Nikâya and places it side by side with the Sautrântika statement of causation.20 The Majjhima Nikâya statement runs thus: " In this manner, these dhammas, being nonexistent come to be" (evam khila me dhammâ ahutvâ sambhonti).21 By placing these two statements together de la Vallée Poussin seems to be trying to show that the Sautrântika theory is similar, if not identical, with the theory of causation in early Buddhism.

It is true that the two statements abhütva bhâva utpada and ahutvâ sambhonti convey the same idea. But this similarity is only superficial and should not be allowed to obscure the fact that the two concepts have rather different substructures. Mookerjee, as mentioned earlier, has pointed out all the difficulties presented by the theory of momentariness, especially with regard to the conception of causation. The Sautrantikas came to adopt the theory of abhütvâ bhâva utpada because of their acceptance of the theory of momentariness. But a theory of momentariness appears nowhere in the Nikâyas and the Agamas (chapter 4). Nor do we find in them any metaphysical speculations on the problem of time. Therefore, the phrase ahutvâ sambhonti, in the Nikâyas, can be considered a straightforward empirical statement involving no speculation about momentariness. It simply states that a dhamma that did not exist before comes into existence (when the necessary conditions are present). Thus, the objections raised against the Sautrântika conception of abhütvâ bhâva utpâda do not apply to the Nikâya conception of ahutvâ sambhonti. For instance, Sântaraksita refers to criticism of the theory of momentariness by Bhadanta Yogasena thus: " Since there cannot be causal efficiency, either successively or simultaneously, the belief in momentariness is vain. When no peculiarity can be brought about [in the cause] by the auxiliaries, the series is rightly held to be undifferentiated [i.e., there is no occasion for diversity ; it would produce the same seed-series instead of the dissimilar sprout-series.]" 22 But this kind of criticism cannot be leveled against the teachings in the Nikâyas and the Agamas, where there is a recognition of empirical things, impermanent but still existing for some time (chapters 4, 5), not necessarily momentary. Causes, therefore, are observable facts existing for some time, and they can act successively or simultaneously because they are not momentary.

H. V. Guenther writes that the statement in the Majjhima Nikaya (i.e., ahutva sambhonti), in spite of its high authority, is rejected by the author of the Milindapahha.23 This is because of the Milinda statement; "natthi keci sahkhara ye abhavantd jayanti," 24 But Guenther has failed to notice that the very statement from the Majjhima Nikaya is asserted by the author of the Milinda-. "yam ahutva sambhonti hutvapativigacchati esapurima kotipahhayati."23 Moreover, the two statements natthi keci sahkhara ye abhavantd jayanti and ahutva sambhonti are semantically different. The words ahutva and abhavantd refer to a difference in time. While ahutva refers to the past, abhavantd refers to the present or even the future. Thus, the first statement means that "there are no dispositions
[produced] that are not [susceptible to] arising," the reason being that when the necessary conditions are present the effect would be produced; and the second statement means that "whatever, being nonexistent, comes to be and having been, passes away—such is the apparent beginning."
Nor does the phrase ahutva sambhonti imply the metaphysical question whether the effect is not inherent in the cause. This is attested to by a statement in the Siksasamuccaya: "Here, O King, the subjective heat element arises; it does not come from somewhere, nor does it, when ceasing, go into accumulation somewhere." 26 On the other hand, the statement "natthi keci sankhara ye abhavantd jay anti" does not imply that the effect is inherent or immanent in the cause. The examples quoted in the Milinda clearly state that the causes exist and that depending on these causes the effect is produced. For example, in the case of a house that did not exist before, it is said that there was wood in the forest, clay in the earth, and as a result of exertion on the part of men and women in handling these materials, there arose the house.27 Just as the identity theory (satkaryavada) leads to a belief in permanence, so does the non-identity theory (asatkaryavada) lead to a belief in annihilation or the absence of continuity. The Buddha faced this identical situation, which is evident from the Kaccayanagotta-sutta.23 There he rejects both atthita and natthita because they would lead to belief in permanence (sassata) and annihilation (;uccheda), respectively.

From the analysis above it will be evident that Buddhist schools such as Sarvastivada and Sautrantika, as a result of the problems created by the theory of momentariness, adopted causal theories that were metaphysical in character. The analysis of experience into indivisible moments was a dominant feature of the philosophical atmosphere in which Nágárjuna lived. Hence the situation was extremely complicated for Nagárjuna, so whatever new interpretation he gave to the causal theory propounded by the Buddha was prompted by circumstances. Though presented with a choice of metaphysical theories of causation presented by both Buddhist and non-Buddhist schools, Nagárjuna was drawn to the doctrines embodied in the Prajñápáramitá literature. He therefore sought a way to justify the teachings embodied there, and his dialectic seems to be an attempt to provide a philosophical foundation for these doctrines. Hence a word about these doctrines may be in order.

The concept of Buddha is the most important topic in the Prajñapáram itá literature. Buddha Gotama was a historical person.

The Sutta Pitaka affords us ample evidence of that.29 He influenced the lives and thought of the people of India during his time to such an extent that superhuman qualities came to be attributed to him, not only after his death but even while he was alive. These qualities—intellectual, moral, and even physical—soon raised him to the position of a deva in the eyes of his followers. The result was that the followers themselves became puzzled as to the real nature of the Buddha's personality. When the question regarding the Buddha's personality was raised, the Buddha himself answered that he was neither a manussa, nor a gandhabba, nor a yakka, nor even a deva or a brahma, but that he was only a Buddha.30 Similar questions were being raised even two hundred and fifty years after his death, dining the reign of Asoka, the Maurya, in the third century B.C.31 Thus it became one of the most im portant topics of discussion in the history of Buddhist thought.

The passing away of the Buddha created a big vacuum in the lives of his followers and admirers. The Mahaparinibbana-suttanta, which relates the incidents in the last days of the Buddha's life seems to imply this. To perpetuate the memory of the Buddha, the Buddha himself recommended to his followers four places of pilgrimage.32 The desire of the faithful followers to have the Buddha as an object of worship contributed to the development of the conception of an eternal spiritual body (dharmakaya) of the Buddha.

Although in the early Sütra tradition the question whether the Buddha exists after death was regarded as a metaphysical question and was left unanswered, speculation regarding the immortality of the Buddha continued, and in the M ahayana tradition the Buddha came to be looked upon as one who "remains forever"
(sada sthitah).33 According to the Saddharmapundarika, Gautama Buddha did not really die after eighty years' sojourn on earth. He can never die. He is immortal, and his parinirvdna is only an illusion.34 If the Buddha is supramundane and immortal, his physical body (rupakaya) could not represent his real nature. Therefore, the Vajracchedika maintains, "The Tathagata is not to be recognized by means of the marks on his body." 35 Buddha is the embodiment of dharma.36 Thus the real body of the Buddha is the spiritual body (dha.rmak.dya).31 The Buddha's real body is not only spiritual but cosmic as well. While the spiritual body (dharmakaya) is identified with all the constituents of the universe (sarvadharma), it is considered to be the same as absolute reality (tathata).38 This monistic philosophy, which is a culmination of the speculation on the nature of the Buddha, is the basic theme of the Prajnaparamita literature. Running through that literature is a conflict between absolute reality, the dharmakaya, considered to be nondual (advaya), and phenomenal reality, which is a plurality.

To resolve this conflict we find the Vajracchedika adopting the standpoint that reality is beyond description.39 This was the religio-philosophical tradition that caught the fascination of Nagarjuna. In his attempt to resolve the conflict between the ultimate and phenomenal realities, Nagarjuna seems to have adopted a novel technique. Instead of merely pointing out the conflict between the ultimate and the phenomenal, he used the dialectical method to eliminate the phenomenal from the discourse. As pointed out earlier, the Buddhist as well as the non-Buddhist schools of his day provided Nagarjuna with the opportunity of developing his dialectic by presenting two contradictory theories of causation. By resorting to the dialectical method he attempted to expose the inherent contradictions in these theories of causation. Since experience was reckoned in terms of moments and since the theory of moments stood in the way of a satisfactory explanation of the process of causal production (utpada), Nagaijuna, unlike the Buddha, gave up the appeal to experience. He was quite aware that pratTtyasamutpada was the central tenet of Buddhism and that the Buddha's enlightenment consisted in the discovery of the causal principle. Therefore, in setting forth his dialectic, he retained one aspect of the theory of causation recognized by early Buddhism, the idea of relativity
(chapter 5). Then he raised the principle of causation from the empirical level to that of absolute reality. This shift of emphasis emerges very clearly from the interpretation that the Madhyamikas gave to the Kaccayanagotta-sutta of the Samyukta.40 The early Buddhist theory of causation was called the middle path because it steered clear of the two extremes represented by the theories of existence (atthita) and nonexistence (natthita). The Buddha rejected these two views because he thought they would lead to the belief in eternalism (sassata) and annihilationism
(uccheda), respectively. The Kaccayanagotta-sutta, which gives the analysis above, is specifically mentioned by Nágárjuna in the Mülamadhyamaka-karika,41 Analyzing the two extremes of existence (astitva) and nonexistence (nástitva), Nágárjuna comes to the same conclusion: " [The theory that everything] exists means adherence to eternalism. [The theory that] nothing exists is annihilationism. Therefore, the wise do not adhere to either of the views of existence and nonexistence."42 But in the Mádhyamika literature we come across two versions of the analysis found in the Kaccayanagotta-sutta. The first is the Kasyapaparivarta of the Ratnaküta-sütra, which is profusely quoted in the Madhyamakavrtti of Candrakirti, and the other is the Madhyamakvrtti itself. A comparison of the two throws much fight on the difference between early Buddhist and Mádhyamika theories of causation. In the Kasyapaparivarta, where the interlocutor is not Kaccáyana but Kášyapa, it is said: " '[Everything] exists,'
Kášyapa, is one extreme. '[Everything] does not exist' is the second extreme. In between these two extremes, Kášyapa, is the middle path, and it is the correct perception of things." 43 The Kasyapaparivarta then describes this middle path (madhyama pratipad)
in terms of the twelvefold causal formula in its progressive and regressive orders. This description is very close to the one found in the Pali Nikáyas as well as in the Chinese Ágamas.

The same passage occurs in the Madhyamakavrtti and was identified by de la Vallée Poussin. But there it appears with an addition. Unlike the earlier references, there the middle path
(madhyarría pratipad) is qualified by several other epithets such as
'formless' (arüpya), 'nonindicative' (anidarsana), 'supportless'
(apratistha), noumenal (anabhasa), signless (aniketa), and nonconceptual (avijñaptika),4-* most of which are generally applied to the transcendental reality, nirvana. But the definition of the middle path as consisting of the twelvefold causal formula is omitted. The Kasyapaparivarta, like the Nikayas and the Ágamas, rejects the two metaphysical theories and gives a causal account of the phenomenal reality. But in the Madhyamakavrtti these two views are criticized from the standpoint of ultimate reality. While the middle path in the former is empirical and phenomenal, the middle path in the latter is transcendental. In fact, the general tendency in the Madhyamakavrtti is to identify causality (pratTtyasamutpada) with the transcendental reality.45 The perfect Buddha, The foremost of all teachers I salute. He has proclaimed The Principle of [Universal] Relativity. 'Tis [like] blissful [Nirvana], Quiescence of Plurality. There nothing disappears, Nor does anything appear. Nothing has an end, Nor is anything eternal. Nothing is identical (with itself), Nor is anything differentiated. Nothing moves, Neither here nor there.4®
Thus, it seems that the doctrine of origination was denied by the Madhyamikas only from the standpoint of the transcendental reality. Candrakirti makes a statement to this effect: " From the Transcendentalist's standpoint it is a condition where nothing disappears, [nor something new appears], etc., and in which there is no motion. It is a condition characterized by the eight [abovementioned] characteristics [such as] 'nothing disappears,' etc."47 The use of such epithets to describe the state of nirvana is not rare in the early Buddhist texts.

The enumeration of the eight attributes—'nondisappearing'
(anirodham), 'nonarising' (anutpadam), etc.—as characteristics of causality may have been prompted by a statement in the early Buddhist texts. This statement is also found in the Madhyamakavrtti: " Whether the Tathagatas were to arise or not, this nature of phenomena exists."48 In the early Buddhist texts this statement implied merely the objective validity of the causal principle (see chapter 5). The elevation of the causal principle from the phenomenal to the transcendental level seems to have created many problems for the Madhyamikas, as is evident from the great attention paid to it by Candrakirti.49 Since the theory of causation was formulated to account for the arising (utpada) and passing away (nirodha) of things, the question was raised how we can deny events such as disappearance. Candrakirti is represented as saying that Nagáijuna composed the Madhyamaka-sastra to explain that problem. Therein he shows that there is a difference between the real meaning (neydrtha) and the conventional meaning (mtdrtha) of the scriptures.50 But that does not appear to solve the problem because it too is an appeal to the transcendent. The transcendental standpoint, which is so emphatically stated in the Prajfiaparamita literature, was adopted by the Madhyamikas to reject all forms of views (drsti). The Atthaka-vagga of the Sutta-nipata contain many discussions of the problems connected with metaphysical views (ditthi). In one of them, it is true, the Buddha maintains that "There is only one truth; there cannot be a second." 51 But the problem is whether the Buddha was referring to an Ultimate Reality, a transcendental Absolute, on the basis of which all other theories are considered to be false. In other words, did the Buddha adopt a transcendentalist point of view in his analysis of phenomenal reality? Did he maintain that " Reason involves itself in deep and interminable conflict when it tries to go beyond phenomena to seek their ultimate ground?" 52 This is a crucial problem that needs detailed analysis; we take it up later (chapter 9) so as not to interrupt the argument here.

The adoption of the transcendentalist standpoint is noticed in the Kasyapaparivarta as well as in the Madhyamakavrtti,53 where the extremes of permanence (nitya) and impermanence
(anitya), of 'substantiality' (atman) and 'nonsubstantiality' (ariatman), of 'defilement' (samklesa) and 'purity' (vyavaddna), are rejected as being unreal from the standpoint of the transcendental reality (paramartha). Commenting on these statements in the Kasyapaparivarta, Murti says: "Dialectic is engendered by the total opposition between two points of view diametrically opposed to each other. And the required opposition could have been provided by the atma-view of the Brahmanical systems and the anatma-vada of earlier Buddhism." 54 Later M urti declares that
"As a matter of dialectical necessity then did the Buddha formulate, or at least suggest, a theory of elements. The M ahayana systems clearly recognise this dialectical necessity when they speak of pudgala-nairatmya—the denial of substance—as intended to pave the way for Absolutism. Sunyata is the unreality of the elements as well (dharma-nairatmya)." 55 We have pointed out (chapter 4) that M urti's assumption that the Buddha suggested a theory of elements as a matter of dialectical necessity is contradicted by a statement made by Candrakirti himself in the Madhyamakavrtti.
W hat is more important to us at present is M urti's view that the atmavada of the Brahmanical systems and the anatmavada of earlier Buddhism provided the required opposition for the development of the Madhyamika dialectic. Murti seems to think that the pudgala-nairatmya (nonsubstantiality of the individual) presented by the Buddha and the early Buddhists constituted one extreme, opposed to the atmavada, the other extreme. But what of the dharma-nairatmya (nonsubstantiality of the elements) of the Mahayanists? It is not a form of anatmavada"! The statement of Murti above seems to be based on the assumption that the Buddha, and therefore the early Buddhists, formulated only a theory of elements, not a theory of the nonsubstantiality of elements (dharmanairatmya), which enabled the Madhyamikas to bring about a
'Copernican revolution' in Indian philosophy. Our investigation has shown that this view is untenable (chapter 4).

In the earlier part of this chapter we indicated what constituted the thesis and antithesis that enabled Nagaijuna to formulate his dialectic. The metaphysical theories of causation presented by the Sankhya and Sarvastivada constituted the thesis, i.e., the assertion of substance; and the causal theories of the Vaisesika and Sautrantika provided the antithesis, i.e., the negation of substance.

Although the latter view denying substance may appear to be similar to the early Buddhist theory, as we have pointed out there is a major difference in that it leads to a denial of causation, thus coming very close to the anatmavdda of the Materialists, which was rejected by the Buddha himself. Thus, when Nágárjuna wrote, the philosophical atmosphere was so polluted by speculative metaphysics that either he had to accept causality, and along with it the belief in 'substance,' or he had to reject 'substance' (the early Buddhist position) and along with it causality. This was the dilemma faced by Nágárjuna, and, as pointed out earlier, he resorted to the transcendental standpoint to reject all metaphysics.

How did he achieve this? We saw earlier that from the standpoint of the transcendental everything in the phenomenal or the conditioned (samskrta) world was considered unreal in that everything is relative. Substance (atma) and 'no-substance' (ariatma)
are relative; so are permanence (nitya) and impermanence (anitya),
defilement (samklesa) and purity (vyavadana). Nágárjuna emphasized this aspect of causation to deny the reality of the phenomenal.

Relativity implies a denial of self-existence (svabhava), hence the absence of reality (šunyata). This aspect was emphasized by Nágárjuna. He found that speculation was based on concepts. He took each concept and showed how it is relative. Thus, by showing the antinomial conflict, he demonstrated the futility of speculative metaphysics. This was the purpose of his dialectic. In the philosophical atmosphere in which he lived, he could not maintain that something arose as a result of causes, because immediately the question would have been raised whether that which arose inhered in the causes or not. Therefore, he was compelled to give up the idea that causation explains 'arising' (utpada) and 'passing away' (nirodha); instead, he maintained that causation explains only relativity.

But even if pratityasamutpada were to be interpreted as a
'theory of relativity,' would there not be an antithesis, the 'theory of nonrelativity' (apratltyasamutpada)? If so, even pratityasamutpada had to be given up as an extreme. This Nágáijuna was not prepared to do. Therefore, he denies that anything is 'unrelated'
(apratTtyasamutpanna),56 and raises pratityasamutpada to the level of transcendental reality, thereby avoiding any interpretation of it as an extreme (anta). By doing so, he seems to divorce the theory (i.e., pratityasamutpada) from the things the theory was intended to explain (i.e., the relative or the conditioned,pratltyasamutpanna).
This seems to be the main difference between early Buddhist and Madhyamika conceptions of causation. Whereas in early Buddhism the theory of causation was employed to explain all types of causation available in the world of experience, including nirvana (chapter 6), in Madhyamika thought it was employed to explain only the relativity of the phenomenal, the theory itself being considered transcendental.

Viii. Causal Correlations : Another Facet Of Development

THE D EFIN ITIO N of a cause {hetu, pratyaya) as the sum total of several factors led to further developments in the Buddhist theory of causation (see chapter 3). During the period of the Abhidharma, the Buddhists began to analyze each of these several factors to determine the exact relationship between them. In the Theravada such speculations are embodied in the Patthana, while in the other schools of thought these analyses are found in almost every text.

The theory of causal correlations {pratyaya, yuan) mentioned in the Abhid.harmak.osa seems to be the nucleus from which the more elaborate theories developed. That most of the schools started with the theory of four correlations is attested by the im portant place accorded it in the different schools. In the Theravada Abhidhamma these four are listed among the first five, the samanantara-pratyaya being counted as two, the anantara- and samanantara-paccayas. In the Sarvastivada and Madhyamika schools the number was fixed at four.1 Hence, when the Yogacarins wanted to account for certain relations that are not covered by these four, they subdivided one of them, the hetu-pratyaya (primary cause; see chapter 3). The Theravadins, who were not restricted by such limitations, went on multiplying the number freely until they formulated a theory of twenty-four relations.

We have pointed out that the Yogacara school enumerated seven characteristics of the primary cause {hetu-pratyaya, yin yuan), one of which was subdivided into twenty forms (chapter 3). Of the seven characteristics of the hetu-pratyaya, six were related to the six hetus (causes) enumerated by the Sarvastivadins. The other, prabheda, has twenty subdivisions, of which the first ten are mentioned in the Madhyantavibhaga-bhasya of Vasubandhu and the last ten are treated in the Bodhisattvabhumi and the Ch'eng weishih lun.2 Thus, the four pratyayas represented a very broad classification of causes, and their subdivisions provide a detailed analysis of all the different causes.

In his translations of the A-pi-ta-mo chu-she lun and the Ch'eng wei shih lun, de la Vallee Poussin has discussed in detail the various pratyayas formulated in these texts.3 A critical analysis of the theory of twenty-four paccayas of the Theravadins has been made by Nyanatiloka.4 We do not propose to re-cover their ground. Our attempt here will be to compare the theory of paccayas presented by the Theravadins on the one hand and the theories presented by the Sarvastivadins and the Yogacarins on the other, to determine whether there is any correspondence between them.

Heiu-paccaya or 'primary cause' is the first of the twenty-four forms of causal correlation enumerated in the Patthana. It occupies a place of similar importance in the Sarvastivada and Yogacara teachings. In the philosophy of early Buddhism, psychological motives such as greed (lobha, t'an), hate or aversion (dosa, wei), and confusion (moha, ch'ih) are referred to as the roots (mula, ken),5 in the sense of primary causes, of evil behavior. The Patthana cites these psychological motives as examples of primary causes (hetupaccaya),6 and Buddhaghosa maintains that a thing can be a primary cause in the sense of being the root (mulatthena).7 These three motives are compared to the roots of a tree, which feed and nourish the other parts of the tree.8 Just as greed, hate or aversion, and confusion are the primary causes of evil (akusala, pu shan), so their opposites are the primary causes of good (kusala, shan).9 According to the Yogacara school, the 'store-consciousness'
(alaya-vijnana), which serves as a receptacle of the "seeds" (bija) such as dispositions (vasana), is the primary cause of the seven forms of active consciousness (pravrtti-vijhana), which are the effects.10 But to the Yogacarins, alaya-vijnana includes both good and bad tendencies,11 although according to the Abhidharmasamuccaya, which represents a formative stage in the evolution of Yogacara thought, only the good tendencies ( kusala-vasana) are considered to be the primary causes.12 The arammana-paccaya or the alambana-pratyaya {suo yuan yuan) is the objective cause or condition. Discussing the alambanapratyaya, Yasomitra says: " There are two kinds of causal relations, namely, that which produces (janaka) and that which does not produce (ajanaka). The alambana-pratyaya does not produce because it is only an objective support." 13 Here the reference is to the perceptual 'image' produced by the object (yisaya), rather than the object itself. Since the 'image' has already been produced by the object, it need not be produced again in the mind by the object, and therefore it serves only as objective support. Hence the distinction between the alambana-satka (i.e., the six forms of vijndna) and the visaya-satka (the six objects).14 For the manifestation of mental phenomena, some kind of objective support is a sine qua non. Buddhaghosa maintains that there is nothing in this world that will not become an object of consciousness.15 While the five forms of sense consciousness that are produced by external stimuli serve as objective support for the five forms of sensory perception, all forms of mental coefficients, all terms expressive of concepts, and nibbana are related to mind by way of objective support. The Yogacara school, which did not accept the reality of the external object, nevertheless recognized this relation. They believed that consciousness (vijndna) contains within itself the ingredients of the subject-object relation and represents one stage in the evolution of consciousness.16 Adhipati-paccaya is the dominant cause. It represents the efficient cause because it exerts influence over the effect.17 For example, the six internal bases of cognition (the eye, etc.) are related to the six forms of cognition in this manner.18 Pali Abhidhamma distinguishes two forms of the dominant cause: (1) objective dominance
(arammanadhipati) and (2) coexistent dominance {sahajatadhipati).19 The first accounts for the impressions created by external objects on the mind. The external world presents us with various agreeable and disagreeable objects. These impressions determine to a great extent the nature of our cognitions. N ot only the impressions but also the nature of the sense organs themselves affect the character of the cognitions. But apart from these objective presentations and the nature of the sense organs, there are certain motives that dominate our consciousness, which are said to arise along with consciousness. Intention, will, energy or effort, reason, and investigation fall in this category and are considered coexistent dominant conditions.20 In the ultimate analysis even such mental concomitants appear to be engendered by external objects. But because of the dominating or "overpowering" influence of these motives a distinction seems to have made between objective and coexistent dominance.

The Sarvastivadins and the Yogacarins give a much wider meaning to adhipati-pratyaya (tseng shang yuan). According to the Sarvastivada it is a comprehensive and universal cause.21 The Yogacarins go so far as to include the other three causes, hetu, samanantara, and alambana, under this category.22 While the other three causes explain specific relations, adhipati-pratyaya accounts for any possible relations. Hence we find the Sarvastivadins identifying it with karana-hetu.22 The difference between the Theravada and Sarvastivada conceptions may be explained th u s: The Theravadins, whose speculations were not restricted by the limitations imposed by other schools, continued to expand the original theory of four pratyayas, enunciating new causes as occasion demanded.

Therefore, it was not necessary to accept a cause that could accommodate anything not falling under the other three causes. As a result, their definition of adhipati-paccaya was limited. On the contrary, the Sarvastivadins who accepted the theory of four pratyayas and formulated a theory of six hetus, defined the adhipati-pratyaya so that anything not accounted for in these two theories could be included in it. The Ch'eng-wei-shih-lun states that adhipati-pratyaya exerts influence in four ways, namely, by being a generating cause
(sheng), a sustaining cause (wei or chu), an accomplishing cause (cheng), and a cause of acquisition (fe).24 Thus, all primary and subsidiary causes fall into this category.

Samanantara-paccaya or samanantara-pratyaya ( teng wu chien yuan) is defined as the proximate or contiguous cause. The Theravada tradition perceived two forms of this pratyaya, although they are not strictly distinguished (see above). The formulation of this correlation may have been necessitated at first by the rejection of the idea of annihilation (uccheda, tuan). But with the development of the theory of momentariness during the period of the Abhidharma, its importance in accounting for the rapid succession of momentary phenomena came to predominate (chapter 4). With the formulation of this relation, the Sarvâstivàdins, the Sautrântikas, who formulated a theory of momentariness, and the later Theravâdins, who accepted this theory—all were able to explain the continuity of momentary phenomena, primarily the mental. According to the definitions given by all the schools of thought, a phenomenon that serves as a cause for an immediately succeeding phenomenon, without pause, can be called an immediately contiguous cause.25 Abhinirvrtti-karana (or he tu [sheng chH nêng tío]) mentioned in the Yogâcàra treatises also emphasizes the immediate production of the effect and is therefore called the proximate cause (âsannah pratyayah).26 Next in the list of twenty-four paccayas of the Theravâdins is sahajata-paccaya, or the conascent cause. In the Patthâna it is defined as " that which arises to help or assist the arising of another phenomenon." 27 The example of the lamp is quoted to illustrate this relation. When a lamp is lighted, the light accompanies the lighting of the lamp. When the lamp is burning, it burns together with its heat and fight. In this case, the lamp relates itself to fight and heat by way of conascence.

This corresponds to sahabhü-hetu (chü yu yin) in the Sarvâstivàda classification. The Sphutârthâ quotes an example from the early texts as an illustration: "These three limbs of the path accompany right view. Along with them have arisen feeling, sensation, volition, etc." 28 De la Vallée Poussin translates it as "cause mutuelle" .29 This relation seems to refute the idea that a cause should always be temporally prior to its effect. An effect will appear when the necessary factors summarized by the cause have been fulfilled—not necessarily after the cause.30 This relation implies that factors mutually support each other to give rise to the effect and continue to do so even after the effect has come into existence.31 In this respect it is similar to the co-relative cause (anhamanha-paccaya, see below). This, according to the Abhidharmasamuccaya, is an aspect of the primary cause (hetu-pratyaya) and is described as
'assistance' (sahâya, chu pan).32 It is further explained as the relation between phenomena that "arise together and exist without deficiency, like the primary and derived elements." 33 N ot all relations are genetic or 'intrinsic.' In many cases it is possible to discern interdependence rather than genetic connection.

The reciprocal or co-relative cause (ahhamahha-paccaya) was formulated to account for such connections. The idea was first expressed in the Upanisads: " The body is founded on breath, and breath is founded on the body." 34 In the early Nikayas and the Agamas it is maintained that such a relation occurs between consciousness (vihhana, shih) and the psychophysical personality
(,namarupa, ming si). The relation is compared to that between two reeds that stand leaning against one another; if one were to be taken away the other would certainly fall.3S The example of the three sticks ( tidanda) is usually quoted in the Pali Abhidhamma to illustrate this relation.36 The Yogacarins consider this relation a characteristic of the primary cause, and they call it 'coexistence' (sampratipatti, teng hsing). Coexistence is explained as the function of a phenomenon that exists with another phenomenon and serves it by way of objective support, such as the mind and mental concomitants.37 According to H aribhadra's classification (chapter 3), it coincides with sabhaga-hetu (hsiangyingyin), formulated by the Sarvastivadins.38 In the She ta ch'eng lun, Asanga is represented as maintaining that the two forms of consciousness, alaya-vijhana and pravrtti-vijnana, are reciprocal causes (anyonyapratyaya = ahhamahha-paccaya).39 The interdependence here does not mean genetic interrelation but, rather, mutual interdependence among existents, " a static set of reciprocal dependencies like that among the parts of a steel frame." 40 The dependence cause (nissaya-paccaya) is described as the ground or basis for the existence of some other phenomenon.41 This relation is slightly different from the two preceding (sahajata and ahhamahha) paccayas. For example, the earth is the dependence cause or the basis on which a tree can grow. But the earth does not arise with the tree, as in the case of the conascent cause (sahajatapaccaya), nor does the earth depend on the tree for its existence, as in the case of the coexistent or reciprocal cause (ahhamahhapaccaya). In psychology, the six 'gateways' (ayatana) of sense perception serve as dependence causes for the six forms of cognition
(vihhana).
The counterpart of this cause is the dhrti-karana (ch'ih neng iso) of the Yogacarins, a subdivision of the primary cause (hetupratyaya). According to their definition, the earth is related in this manner to the beings who live therein42 because the earth holds them and prevents them from falling.43 A separate cause corresponding to this does not appear in the Sarvastivada classification. But it may be possible to include it under adhipati-pratyaya,
which in the Yogacara tradition functions as a supporting cause (pratistha, chu).

Next is the 'sufficing cause' (upanissaya-paccaya), which Buddhaghosa defines as "excessive dependence" .44 It represents a powerful means or inducement.45 According to the Patthana, there are three forms of the sufficing cause. They are (1) the objective sufficing cause (arammanupanissaya), (2) the immediate sufficing cause (anantarupanissaya), and (3) the natural sufficing cause (pakatupanissaya).46 The first is similar to the dominant influence of the object (arammanadhipati); the second is similar to the immediate contiguous cause (samanantara-paccaya). The importance of the third lies in the fact that it explains moral and spiritual advancement. Because of sufficing causes such as faith (saddha),
one gives alms, observes the moral rules, performs uposatha functions, develops meditative powers and insights, etc.47 In a certain way this relation is similar to sarvatraga-hetu
(pien hsing yin) of the Sarvastivadins. It may be argued that any phenomenon serving as a powerful inducement for certain forms of behavior, moral or immoral, persists until the goal to which that behavior is directed is achieved. In this sense upanissaya paccaya resembles sarvatragahetu, for according to the latter a false view held by a man dominates all his behavior, bodily, verbal, and mental. His behavior becomes infused with the false view and is made disagreeable to others.48 False view, while serving as a strong inducement, runs through his entire behavior. Haribhadra has pointed out similarities between sarvatraga-hetu and the characteristic of the primary cause described as " opposition" (paripantha,
chang ai).49 The preexistent or prenascent condition (purejata-paccaya)
recognizes the prior existence of some phenomenon as a condition for the production of another phenomenon. Helping or supporting the arising of a thing by its prior existence is the function of this cause.so Among the list of hetus or karanas put forward by the Yogacarins, none corresponds exactly with this cause. But aksepahetu (chao yin neng tso) or 'projecting cause,' which is a subdivision of the primary cause,51 in certain respects resembles the preexistent condition. Aksepa-hetu accounts for the problem of action at a distance, hence is defined as 'remote cause' (vidurah pratyayah).
Ignorance (avidya) produces old age and death (jaramarana) and is therefore a remote cause.52 According to the Bodhisattvabhumi, a seed producing another of its kind is a remote cause, because the intermediary stage represented by the tree is not given.53 But the Theravada description seems to imply the continued existence of the cause even after the effect has come into existence. Thus, the only similarity between the two relations is that they both recognize a time lag between the cause and the effect.

That which supports the continued sustenance of a phenomenon that has already come into existence is said to be the postexistent or postnascent condition (pacchajata-paccaya).54 For example, the continued supply of the necessary quantity of moisture, etc. is necessary for an existing plant to grow to maturity. Otherwise there would be change in its growth (S 3.91 -92). In the same way, a personality, which has come into existence because of past causes, requires continued sustenance in the future. If the four kinds of food—material food, contact, volitions, and consciousness—do not feed this personality, it will not develop or continue to exist.55 This definition is quite similar to that of the nutriment cause (iahara-paccaya, see below) and may therefore be compared to the cause of stability (sthiti-karana, chu neng tso), which again is a subdivision of the primary cause.56 According to the Patthana, any phenomenon that causes its resultant to accept its inspiration so that the latter can gain greater and greater advancement is called the habitual-recurrence condition or cause (.asevana-paccaya).57 The term asevana is used in the sense of habituation by constant repetition. If a man develops thoughts of loving kindness (metta) once, he will be enabled to develop the same thoughts with a greater degree of perfection later.

An important characteristic of this relation is that it exists among things of the same order, among likes.

The same relation is expressed by a characteristic of the primary cause termed 'increase' (pusti, z).58 It is explained as "the good, bad and, neutral dharmas previously cultivated that cause greater and greater efficiency of the dharmas, good, bad, and neutral, respectively, to be produced in the future." 59 Haribhadra maintains that this is similar to the samprayuktaka-hetu.60 According to the Sphutârthâ, there are five characteristics of hetu, one of which is upabrmhana. It is defined in the same way pusti is defined in the Abhidharmasamuccaya.61 The need to account for the problem of moral responsibility gave rise to the relation of kamma (kamma-paccaya). The problem of the causation of moral behavior and responsibility has been discussed earlier (chapter 6). The importance of this problem may have induced the Àbhidhammikas to formulate a special relation to account for it. According to the Theravâda Abhidhamma, kamma here refers to the particular function of the volitions.62 It is a reflection of the statement in the early Buddhist texts that kamma is merely volition (see chapter 6). Two forms of kamma relations were distinguished by the Abhidhammikas: (1) the asynchronous (nânakkhanika), and (2) the conascent (sahajata).63 The psychophysical personality that arises in this existence is due to the dispositions (sankhâra) or volitions (ce tana) of the past life. This is the asynchronous kamma relation because the dispositions or the volitions belong to the past. On the other hand, there are certain thoughts, good (kusala) or bad (akusala), that arise along with the volitions. Such volitions are related to the thoughts by way of the conascent kamma relation.64 In several respects, the kamma relation resembles vipâka-hetu
(i shu yin) of the Sarvâstivâdins. Like the kamma relation, vipâkahetu emphasizes the volitional aspect of karma. Like the asynchronous kamma relation, it partakes of the idea of projection
(aksepakatvd) of the effect and recognizes a time lag between the cause and the effect.65 Haribhadra has equated vipaka-hetu with a characteristic of the hetu-pratyaya given in the Abhidharmasamuccaya as 'grasping' (parigraha, she shou).66 The characteristic of 'grasping' is explained by the example of "bad and defiling tendencies causing the belief in a [permanent] soul." 67 But a closer relationship exists between the asynchronous kamma relation and dksepa-karana (chao yin nêng tso), the projecting cause enunciated by the Yogâcàrins.68 In the asynchronous kamma relation, kamma signifies a particular energy. It does not cease, though the volition may cease to be evident, but exists in a latent form. As soon as it obtains a favorable opportunity, and when the other necessary conditions are available, it produces the effect.69 The nutriment-cause (âhâra-paccaya) is one that is prefigured in the Pali Nikayas and the Chinese Agamas. The Nikayas and the Agamas refer to four things—material food, contact, volition, and consciousness—that serve as nutrition for beings who are born and those seeking birth (chapter 6). But the Abhidhammikas specify the function of food (ahara). They maintain that "even though food has the power to generate [some effect], the primary function of food is to support or sustain [what has already come into existence]." 70 This view is clearly implied in the sthiti-karana (chu neng tso).71 The faculties (indriya)—such as faith (saddha), energy (viriya),
mindfulness (sati), concentration (samadhi), and knowledge (pahha.)—that control the behavior of man come under the category of controlling conditions (indriya-paccaya). During the time of the Abhidhamma twenty such faculties were enumerated.72 Because of the importance of these faculties in determining the behavior of an individual, the Patthana has formulated this special kind of cause. But the idea of dominance (adhipati) implied in this relation makes it quite similar to adhipati-paccaya (see above). Therefore, the Sarvastivadins and the Yogacarins may have been satisfied with the formulation of adhipati-pratyaya.
The stages on the path to a goal are considered by the Abhidhammikas as causes (paccaya) because each stage has the power of clearing the ground and assisting the attainment of the succeeding stage. Such causes or conditions are called the path conditions (magga-paccaya).73 This relation resembles prapana-karana {ting chih neng tso),74 which is illustrated by the example of the path leading to nirvana.15 The CKeng wei shih lun refers to it as an aspect of adhipati-pratyaya (an accomplishing cause, cheng, see above).

Sampayutta-paccaya, or the 'association condition,' accounts for the synthesis of phenomena that are analyzed into different parts for the sake of examination. Statements referring to the association of ideas are not rare in the Nikayas and the Agamas.76 This relation corresponds in many respects to samprayukta-hetu (t 'ung lui yin), formulated by the Sarvastivadins. The Theravada and Sarvastivada definitions are similar. The former maintains that association takes place in four ways: (1) having one base (ekavatthuka), (2) having one object (ekarammana), (3) arising simultaneously (ekuppada), and (4) ceasing together (ekanirodha).77 Speaking of the samprayukta-hetu, the Sphutartha says, "it is deter mined by its function of having one object." 78 Haribhadra says the same thing with regard to a characteristic of the primary cause that he describes as sampratipatti (teng hsing), although he prefers to identify samprayukta-hetu with the characteristic pusti (tseng 2).79 Atthi-paccaya, or the 'presence condition', is defined in the Theravada Abhidhamma as "that which renders service by being a support to another through presence." 80 This may appear to be a redundance, but the importance of this relation becomes clear when we consider the early Buddhist notion of 'cause.' It has been pointed out that a cause is the sum total of several factors (chapter 3). Taking the example of a plant, it was pointed out that there are three factors essential for its arising. The presence of earth and moisture is essential, not only for the arising of the sprout, but also for its later development. It is this aspect of the 'presence' of certain conditions that is emphasized in this relation.

Atthi-paccaya seems to correspond to sahakari-karana (t'ung shih neng tso)81 or sahakari-hetu (t'ung shih yin).82 Sahakari-hetu, or the supporting cause, is a subdivision of the primary cause and is defined as the concurrence (samagrT, ho ho) of various factors,83 thus emphasizing the need for the presence of several conditions.

Thus, nearly eighteen of the twenty-four causal correlations enumerated in the Patthana have counterparts in the Sarvastivada and Yogacara theories. We have not been able to find parallels for six of the relations enumerated by the Theravadins. However, in addition to those mentioned above, the Yogacara list contains thirteen more relations for which parallels are not traceable in the Theravada Abhidhamma.

One of the relations enumerated in the Patthana that has no parallel in the other schools is vipaka-hetu. It does not, by definition, correspond to the vipaka-hetu of the Sarvastivadins. The former, unlike the latter, does not emphasize the importance of volitional activity. In fact, the Patthana maintains that a phenomenon that aids, without exerting any effort, the arising of another phenomenon is called vipaka-paccaya, or the 'resultant condition.'84 Jhana-paccaya, or the 'contemplation condition,' was formulated by the Theravadins to explain the process of concentration.

The factors that allow the mind to sustain concentration are such causes (paccaya). Some of these factors are initial application ([vitakka), sustained application {vicara), pleasurable interest (pTti),
joy (somanassa), indifference (upekkha), and one-pointedness of mind (cittassa ekaggata).8S
While the relation by way of association (sampayutta) illustrated the homogeneous nature of consciousness, the relation by way of dissociation (vippayutta) explains the distinction between mental and physical phenomena. It purports to refute the view of the idealists that material elements are mere projections of the mind. While explaining the interdependence of mental and physical phenomena, it helps to keep them apart, thus affirming the realist standpoint of the Abhidhammikas.

An im portant aspect of the causal situation left unexplained by the presence condition (atthi-paccaya) is expressed by the 'absence condition' (natthi-paccaya). The presence condition emphasizes only the presence of certain conditions or factors for the arising of a thing. But there are certain factors that should disappear to make room for the appearance of the effect. In the example of the seed, we found that the presence of three conditions was necessary. If the sprout is to come into existence, the seed has to give way, but the other two conditions may still have to be present and continue to support the sprout. It is this disappearance and making room for the effect to manifest itself that is emphasized in the absence condition.86 The 'abeyance condition' (vigata-paccaya) and the 'continuance condition' (avigata-paccaya)87 are defined in the same way as the absence and presence conditions, respectively. The formulation of these conditions may have been prompted by the desire to eliminate the belief in a static reality, which may be implied by the absence and presence conditions. The abeyance condition emphasizes gradual disappearance, and the continuance condition avoids the static existence implied by the presence condition.

The following several paragraphs give a brief description of the thirteen forms of correlation discussed in the Abhidharmasamuccaya that have no parallels in the Theravada list. (1) Utpatti-kdrana
(sheng neng tso)88 is the producing or generating cause. It is defined, like the supporting cause (sahakdri-kdrana), as the complex (samagri) of causes that gives rise to the effect. It is illustrated by the relationship between the complex of causes such as the visual organ, etc. and visual consciousness (caksurvijnana). Haribhadra says that it is the cause of production because it gives rise to an effect that did not exist earlier.89 The Ch'eng wei shih lun, which does not refer to the first ten causes enumerated in the Abhidharmasamuccaya, includes the producing cause under the category of adhipati-pratyaya by pointing out that production (janana,
sheng) is one of the modes by which the adhipati-pratyaya manifests its activity.

(2) Prakasa-karana (chao neng tso)90 is the revealing cause.

It is like the lamp, which reveals objects (or colors) by destroying the darkness. (3) Vikara-karana (pien huai neng tso),91 or the cause of alteration, brings about a change in another phenomenon. It is illustrated by the example of fire, which alters the nature of anything inflammable. Haribhadra points out that this is a cause that changes one series to another, as, for example, fire changes the series called "wood" to the series called "charcoal." 92
(4) Viyoga-karana (fen li neng tso)93 is a cause of separation as a sickle is in relation to what is to be cut. It cuts into two what is connected or conjoined. (5) Parinati-karana (chuang pien neng tso)9* is a cause of transformation as is a skill in the metalworking art in relation to gold and silver. This refers only to the transformation of a basic material; hence it differs from the cause of alteration, which implies a complete change. (6) Sampratyayakarana (hsin chiai neng tso),95 is the cause of agreement, as smoke is to fire, because what is not manifest can still be known by comparison or inference. (7) Sampratyayana-karana (hsien liao neng tso)96 is a cause of making known or proving, for example, a proposition, a reason, and an example. (8) Vyavahara-karana (sui shuo neng tso)91 is the cause of reference or denomination, which is the basis of speech. Speech depends on names (nama), perception (samjha), and views (drsti), which therefore are the causes of reference. Here nama names the object, samjha perceives it, and drsti adheres to it. Thus, all forms of speech are determined by names, perceptions, and views (namasamjhadrstihetuka).

(9) Apeksa-karana (kuan tai neng tso) is the cause of expectation. It is illustrated by the relation between hunger and thirst to the search for food and drink, respectively. (10) Avaha-karana (yin fa neng tso) is defined as the coinciding or agreeing cause because it is supposed to bring about results that are in conformity (anukula)
with the causes. It is illustrated by the example of proper service to royalty leading to the gaining of the confidence of the royalty.

(11) Pratiniyama-karana (ting piehneng tso) is the cause of specialized activity. The dispositions (samskara), in so far as they possess a special force to produce their fruits, are called pratiniyamakarana. For example, a birth in any one of the five realms is determined by the appropriate causes producing birth in that realm, and this is thought to emphasize the diversity of causes. (12) Virodhikarana (hsiang wei neng tso) is the cause of opposition or an obstructing factor, such as the relation of hail to corn. (13 )Avirodhikarana (pu hsiang wei neng tso) is merely the absence of obstruction, hence a negative cause.

A passage in the SumahgalavilasinTof Buddhaghosa is strongly reminiscent of the analysis of the pratyayas in the Abhidharmasamuccaya and other treatises of the Yogacara school. There Buddhaghosa describes the various powers and types of knowledge possessed by the Buddha.98 One of them consists of the knowledge that ignorance (avijja) is related to the dispositions (sahkhdra) in nine possible ways, as a causes o f:

1. Production or genesis (uppado hutva, cf. utpatti-karana).
2. Natural happening (pavattam hutva, cf. pravrtti, abhinirvrtti).
3. Objectivity (nimittam hutva, cf. arammana, alambana).
4. Endeavoring or striving (ayuhanam hutva).
5. Association (samyogo hutva, cf. samprayuktaka'sampayutta).
6. Obstruction (palibodho hutva, cf. paripanthato, virodhikdrana).
7. Arising (samudayo hutva). 8. Primary support (hetu hutva). 9. Dependence (paccayo hutva).
   The marked similarity between this and the analysis in the Abhidharmasamuccaya seems to show that Buddhaghosa was aware of the developments taking place in northern India during his time. It also indicates that Buddhaghosa's interpretation of the Theravada texts was very much colored by these ideas. In fact, the cause of obstruction (palibodha), number 6 in this list, was not recognized in the Theravada tradition, because in that tradition a 'cause' was understood to be something that helps or supports (upakaraka) the arising of another thing, not something that obstructs.99

Ix. Conclusion

THAT TH E BUDDHA left certain metaphysical questions unexplained (avyakata, Sk. avyakrta) has engaged the attention of many scholars. T. R. V. Murti after having examined all the previous theories, advanced a theory that has exerted so much influence on m odem scholars that it seems to be considered the last word on the subject. As an exposition of the Madhyamika philosophy, M urti's The Central Philosophy o f Buddhism is unquestionably the best we have. But we contend that although Murti has presented an authoritative account of Madhyamika philosophy, his interpretation of early Buddhism is not in the least satisfactory. We have pointed out (chapters 4, 7) how Murti wrongly attributed certain theories, such as the theory of real elements, to the early Buddhists and even to the Buddha himself. In this chapter we propose to show that M urti's theory regarding the silence of the Buddha does not have any basis.

In The Central Philosophy o f Buddhism, Murti starts by saying,
" It is our contention that the Madhyamika dialectic is anticipated in essentials by the Buddha. The Madhyamikas have but systematically formulated his suggestions and drawn out their implications" (p. 36). The conclusion being thus preconceived, Murti goes on to present the different views expressed by modem scholars and then interprets the ten (in later Mahayana, fourteen) unexplained questions in a manner that supports his conclusion. He selected only a few sections from the early Nikayas for his explanation of the unexplained questions and overlooked many of the earliest and most important portions of the Nikayas, which throw much light on this problem—for example, the Atthaka-vagga of the Suttanipata. As a matter of fact, the selections that directly refer to the ten unanswered questions afford very little help in revealing the reasons for the Buddha's silence, except the Culla Malunkya-sutta,1 which emphasizes pragmatic reasons.

Let us state the ten questions as they occur in the early sources.

They are grouped by topic.

Duration o f the universe 1. The world is eternal.

2. The world is not eternal.

Extent o f the universe 3. The world is finite.

4. The world is infinite.

Nature o f the soul 5. The soul is identical with the body. 6. The soul is different from the body.

Destiny o f the saint (arahant)

7. The saint exists after death. 8. The saint does not exist after death. 9. The saint does and does not exist after death.
8. The saint neither exists nor does not exist after death.

Jayatilleke has made the most recent analysis of these ten questions. He differs from Murti in the way he distinguishes the different types of questions, and gives different reasons for the Buddha's silence on them.2 He points out that the first four questions have no answer because of the limitations of empiricism (see chapter 6). He finds the other six questions logically meaningless and maintains that they resemble the solution of the Logical Positivists. They differ from the solution of the Logical Positivists, he points out, as follows: " The Buddhist while saying that it is meaningless to ask whether one exists in (hoti), does not exist in (na hoti), is born in (upapajjati), is not born in (na upapajjati) in Nirvana, still speaks of such a transcendent state as realizable." 3 Jayatilleke's source for this interpretation is a statement in the Sutta-nipata: "The person who has attained the goal is without measure; he does not have that with which one can speak of him." 4 This brings us face to face with one of the most crucial problems in early Buddhism—the interpretation of nirvana.s In the first place, the person referred to in the quotation above (atthahgata)
could either be one who has attained the goal (artha) in this very life (ditthadhamma), in which case it does not pose much of a problem, or one who has passed away (astha-gata), in which case Jayatilleke's interpretation creates difficulties. Jayatilleke's interpretation seems to assume the existence of a transcendental state realizable after death, a state that is not describable. This appears to be the same theory that the Buddha considered to be metaphysical, namely, " The Tathagata exists after death" (hoti tathagato parammarana). The second clause in the quotation above, yena nam vajju tarn tassa natthi [he does not have that with which one can speak of him], if interpreted literally, would mean that a being exists in nirvana after death, but that no concepts can be used to describe him. In this, Jayatilleke's view comes close to that of Murti.

We have seen that the Buddha rejected the Upanisadic belief in atman. It was pointed out in chapter 1 that one of the ways the Buddha analyzed this concept resembles the analysis of such concepts by the Logical Positivists, who believed that the grammatical structure of a sentence is not a trustworthy guide to its meaning. If the Buddhists had interpreted the foregoing statement in the Sutta-nipata literally, to mean the existence of a being in nirvana (after death), there is no reason why they should have rejected the Upanisadic theory of atman, for according to the Upanisads atman was also a transempirical reality. Thus we maintain that the reason for which the Buddha rejected the Upanisadic conception of atman, whether empirical or transcendental, was the same for which he rejected the belief in the survival of a saint after death. For to accept a transempirical or transcendental state, to be realized by the enlightened one after death, is to reject the very basis of early Buddhist epistemology, namely, empiricism. The reason is that an enlightened one realizes that he has put an end to craving and grasping and has eliminated any kind of future existence (khina jati, sheng i chin; naparam itthattaya, kengpu shou y u 6).

The statement, "The person who has attained the goal is without measure" (atthahgatassa na pamanam atthi) seems to convey the idea that there is no way of knowing (pamana = pramana, a source of knowledge) the state of the enlightened one after his death.

Hence there is no sense in applying concepts to describe him. Thus, here, as with the question of the extent and duration of the universe, there appears to be an epistemological problem, the limitation of empiricism. That, we believe, is why the Buddha was silent on the status of the enlightened one after death. It accounts also for the Buddha's rejection of the two metaphysical theories: the concept of atman and the theory of the existence of a transcendental state attained after the death of the enlightened one, a state that seems to be identical with the Brahman of the Upanisads. The silence of the Buddha was thus due to his awareness of the limitation of empiricism, rather than of concepts.

What, then, is the nirvana that he always spoke of? It is a state of perfect mental health (aroga), of perfect happiness (parama sukha), calmness or coolness (sTtibhuta), and stability ( anenja), etc. attained in this life, or while one is alive.7 It is the nibbuti attained by every arahant, male and female, as described in the Thera-gatha and the TherT-gatha. After attaining this state, a person enjoys perfect happiness until the end of his life.

As mentioned earlier, speculation regarding the fate of the enlightened one after parinibbana came to occupy a very important place in Buddhism, especially after the passing away of Gotama Buddha (chapter 7). The faithful follower was not satisfied until he was convinced that the Buddha, after his parinibbana, continued to exist in some transcendental form. This was the kind of speculation that came to the forefront in Mahayana and that culminated in the conception of the Absolute (variously termed tathata, dharmakaya, etc.) in the Prajnaparamita literature.

The belief in a transempirical reality may have received support from the speculations of the Yogacarins, who considered the highest knowledge to be the nondiscriminative consciousness attained in the highest state of samadhi by a Yogi. This nondiscriminative consciousness may have been considered a foretaste of the transempirical state to be attained by the saint after death. Significantly, the Buddha was said to have attained this latter state, called nirodhasamapatti, just before he passed away. When he was in this state some even thought that he had attained parinibbana.8 As the texts indicate, he emerged from this trance before he finally passed away. (This attainment of nirodhasamapatti by the Buddha when he needed rest is often referred to in the texts; the attainment of that state just before passing away could also have been an attempt by the Buddha to overcome the physical pain that came upon him just before death.)
Nirodhasamapatti is the same as sahnavedayitanirodha, " the cessation of perception and feeling," which is the highest state of meditation attained by the Buddha and the other arahants. That the Buddha attained nirodhasamapatti just before his parinibbana may have led the later Buddhists to believe that the Tathagata continues in a transempirical state after death. Since in nirodhasamapatti both perception and feeling ceased to exist, it could not be described in positive terms such as perception (sahha) or feeling (yedana). On the other hand, complete annihilation (uccheda) did not occur, for a person who had attained this state could, at will, return to a normal.state of consciousness. W hat is im portant is that nirodhasamapatti does not constitute enlightenment and freedom. It is an im portant stage of mental concentration attained by the Yogi whence he is able to develop not only the insight necessary to achieve enlightenment but also the renunciation that leads to freedom. A prerequisite of the state of complete freedom (nirvana) is insight, which is also attained as a result of concentration, which in turn consists of the sixfold higher knowledge (see chapter 5). These two processes, insight and concentration, culminate in nirvana,
understanding of and freedom from the world. These two processes are represented in Figure 3.

The first process enables one to gain calmness and therefore freedom of mind (cetovimutti, hsin chiai t 'o); the second leads to understanding and freedom through insight (pahhavimutti, hui chiai t'o). Both these processes therefore seem to converge when one attains enlightenment. Hence, in early Buddhism, enlightenment is said to consist of freedom of mind and freedom through insight, which result in the realization that one has put an end to birth ( khina jati, sheng i chin), that one has lived the higher life (vusitam brahmacariyam,fan hsing i wei), that one has done what is to be done ( katam karamyam, so tso i p'an), and that there is no future existence (naparam itthattaya, keng pu shou yu). An enlightened person can go about in the world without being smeared by the world (chapter 6). It is this state of nirvana that the Buddha says is realizable. Such behavior is considered transcendental (lokuttara), because it contrasts with the ordinary behavior of men, who are engrossed in the darkness of ignorance (avijja).

As for the state of the enlightened one after death, there is no

Fig. 3: The Process of Spiritual Development Leading to Nirvana way of knowing. Therefore the silence of the Buddha with regard to these questions seems to have been prompted by the limitations of empiricism—the very same reason the Buddha refused to answer questions about the extent and duration of the universe. The Buddha is recorded as maintaining that there is no further existence for one who has attained enlightenment (see above). If he had pressed this view, he would no doubt have been criticized as a
'nihilist,' which in fact he was at one stage.9 On the other hand, if he had maintained that the enlightened one attains a transcendental state after death, he would have earned the title of 'etemalist' and would not have been very different from the pre-Buddhist teachers whose doctrine he had categorically rejected. Similarly, one cannot have the experience of a personality (JTva) divorced from the physical body (sarTra) and a personality identical with the physical body.

Thus, according to our understanding of the early Buddhist texts, the silence of the Buddha regarding these ten questions is due entirely to the limitations of empiricism, and not to the inability of concepts to describe a transcendental reality.

One could, of course, raise the question, W hat is the meaning or implication of the phrase " transcends logic" {atakkavacara, Sk.

atarkavacara) occurring in the early Buddhist texts? Referring to his discovery, causality (paticcasamuppada), the Buddha said that it is deep, immeasurable and "transcending logic." 10 Did he mean that concepts, which are the tools of logical thinking, are inadequate to express reality? It does not seem to be so, because analysis of the conceptual thinking in one of the oldest parts of the canon, the Atthaka-vagga of the Sutta-nipata, does not lead to such a view.

A careful study of the oldest suttas indicates not that the concepts themselves are inadequate to express reality, but that the way our minds are conditioned when using such concepts creates difficulties in understanding reality through such means. Let us take, for example, the concept that is most controversial among Buddhists, the concept o f 'self' (atta, Sk. atman). The mind of the pre-Buddhist thinker was conditioned in one of two ways when using the concept of 'self': either (1) he believed that there is a permanent, transempirical entity (as in the Vedic tradition), or (2) he believed that there is no personality, no personal identity, and therefore no continuity except in the material particles that constitute one's physical body
(M aterialist tradition). The concept o f 'self' thus appears to have been used as each person wanted it to be used—in other words, according to each individual's inclination. Says the Sutta-nipata
(781):
"When, inclination prompts and self-will reigns, shall men desert their cherished views?

—Their outlook shapes their speech." 11 The interference of one's likes and dislikes in the use of concepts, according to the Buddha, leads to dogmatic beliefs (ditthi)
and hence endless logomachies.12 The Buddha realized that by calling for definitions he was able to prevent people from using concepts in this manner. He often followed this method in his teaching, as in the case of the monk Sati, who held the view, according to the Buddha, that " it is this very consciousness that transmigrates, not another." The Buddha immediately asked, " What now Sati is this consciousness?" (katamam tarn Sati vinnanam), to which Sati answered, " He who is the speaker, experiencer, and who experiences the consequences of the good and bad actions in such and such places." 13 Here Sati was referring to an agent, a subject like the 'self' posited in the Upanisads. The Buddha considered that a heresy (ditthigatam) and went on to explain consciousness as a causally conditioned phenomenon (paticcasamuppannam) (see chapter 1). But the denial of a permanent entity as posited by the Upanisadic thinkers did not lead the Buddha to the other extreme of denying personal continuity, as did the Materialists. The atmavada of the Vedic tradition gave rise to the andtma-vada of the Materialists. The Buddha considered this andtma-vada itself another extreme, because while denying the transempirical or nonempirical 'self,' the Materialists also denied empirical consciousness, which for the Buddha was a reality. Therefore, the Buddha continued to use concepts such as 'I ' (aham), 'you' (tvam), and 'self'
(atta), but without either implying the existence of a transempirical reality or denying personal continuity.

Another group of concepts seems to have expressed only some aspects of empirical reality because they were used to clothe one's own metaphysical assumptions. These were the pre-Buddhist concepts of causation such as 'self-causation' (sayam katam, tsu tso)
and 'external causation' (param katam, f a tso) (chapters 1, 2). The Buddha found these concepts to be limited and inadequate to express reality. This was not because reality as he saw it was indescribable or transcendental but because people used these concepts to express only a part of reality, the part that fit their metaphysical predilections. Thus, the man who was prone to believe in a transempirical soul (atman) as a reality would explain causation in terms of
'self-causation.' On the other hand, the Materialists, who denied the reality of psychic phenomena, looked upon causation as a mere external causation. The syncretist Jainas accepted both 'self-causation' and 'external causation,' along with their metaphysical assumptions. The Buddha was unable to use the existing concepts and employed entirely new concepts to explain such situations. The use of the term paticcasamuppada (Sk. pratityasamutpada), a purely Buddhist term, to denote the causal situation illustrates this problem very clearly.

All this may lead to the following conclusions. Rejecting an Absolute (such as the Brahman or Atman of the Upanisads) or a transempirical reality, the Buddha confined himself to what is empirically given. Following a method comparable to that adopted by the modern Logical Positivists, he sometimes resorted to linguistic analysis and appeal to experience to demonstrate the futility of metaphysics. As a result of his empiricism he recognized causality as the reality and made it the essence of his teachings. Hence his statement: " He who sees causality sees the dhamma." 14

Abbreviations

A AA Abhs Abvn A D A D V A it A r A K A kb A M As.P A Š S A V Bbh BCPA BH SD Brh B sk B SO A S

Ahguttara N ikaya ManorathapuranT, Ahguttara-atthakatha Abhidharmasamuccaya AbhidhammatthavibhaviriT-tTkd AbhidharmadTpa Vibhasdprabhdwtti, commentary on the AbhidharmadTpa A it arey a A r any aka Abhidharmakoša A bh idharmakoša-bhdsya Asia M ajor Astasahasrika Prajňdpáramitd A rya-sdl istamba-sutra Atharvaveda Bodhisattvabhumi Bodhicaryávatara-paňjika Buddhist H ybrid Sanskrit Dictionary Brhadaranyaka Upani$ad (see Principal Upanisads) Buddhist Sanskrit Bulletin o f the School o f Oriental and African Studies (University of London) Chdndogya Upanisad CKang a-han ching Chdndogya Upanisad Bhasya CKéng wei shih lun Chung a-han ching Ceylon Journal o f the Humanities Ta cK éng a-pi-ta-mo chi lun Chung pien fén pieh lun

Ch CKang ChB CKéng Chung CJH CL CPFPL

Abbreviations

C S L	A-pi-ta-m o chü-shê lun
C S S L	A-pi-ta-m o chü-shê shih lun
D	DTgha Nikâya
D A	Sumahgalavilâsinï, Digha-atthakathâ
D C B T	D ictionary o f Chinese Buddhist Terms
Dh	Dhammapada
DhA	Dhammapada-atthakathâ
DhsA	AtthasalinT, Dhammasahganï-atthakathâ
D ial	Dialogues o f the Buddha (see DTgha N ikâya)
D ivy	Divyâvadâna
G V	Gandvyüha
G S	The B ook o f Gradual Sayings (see Ahguttara Nikâya)
H O S	H arvard Oriental Series
It	Itivuttaka
J	Jataka
J A O S	Journal o f the American Oriental Society
JA SB	Journal o f the Asiatic Society o f Bengal
JB	JaiminTya Brâhmana
J B O R S	Journal o f the Bihar and Orissa Research Society
J D L U C	Journal o f the Departm ent o f L etters, University o f Calcutta
J P T S	Journal o f the Pali Text Society
J R A S	Journal o f the R oyal Asiatic Society o f Great Britain and Ireland
K atha	Katha Upanisad (see Principal Upaniçads)
Kaus	KausTtakT Upanisad (see Principal Upanisads)
K osa	L. de la Vallée Poussin's French translation o f C SL
K P	Knsyapaparivarta, o f the Ratnaküta-sUtra.
K S	The Book o f the Kindred Sayings (see Sam yutta Nikâya)
Kvu	Kathâvatthu
KvuA	Kathâvatthu-aühakathâ
L P S S C	Liao pên sheng szü ching
L V	Lalitavistara
M	Majjhima Nikâya
M A	Papancasüdani\ M ajjhima-atthakathâ
M bh	Mahâbhârata
M C B	M élanges Chinois et Bouddhiques
M iln	Milindapcmha
M K	M üla-madhyamaka-kârikâ (see M âdhyam ikavrtti)
M K V	M âdhyam ikavrtti, Prasannapadâ
M L S	M iddle Length Sayings (see M ajjhima N ikâya)
M und	M undaka Upanisad (see Principal Upanisads)
M V B B	Madhyântavibhâga-bhâsya
N K	NyâyakusumânjalT
P C P L	Pien chung pien lun
P E W	Philosophy East and West
Pieh i tsa	Pieh i tsa a-han ching
P IP C	Proceedings o f the Indian Philosophical Congress

188 Ps	Patisambhidâmagga
P S T C C	P 1u-sa ti ch'ih ching
PTS	Pali Text Society, London
P T SD	Pali-English Dictionary, PTS
P U	Principal Upanisads, ed. S. Radhakrishnan
R S	R atnakütasütra
R V	Rgveda
S	Sam yutta N ikâya
SA	Sâratthappakâsinl] Sam yutta-atthakathâ
Sadmp	SaddharmapundarTka-sûtra
Sakv	Sph utârth âbh idharmakosa-vyâkhyâ
ÉB	èatapatha Brahmana
S BE	Sacred Books o f the East
SD S	Sarvadarsanasamgraha
S D M S	Saddarsanasamuccaya
Siddhi	L. de la Vallée Poussin's French translation o f Ch'êng wei shih lun
Siks	Siksasamuccaya
Skr	Sûtrakrtâhga, commentary on Süyagadahga by Silânka
Sn	Sutta-nipata
SnA	Sutta-nipata-atthakatha
S T C L	Shê ta ch'êng lun
Su	Süyagadahga
Sv	Évetâsvatara Upanisad (see Principal Upanisads)
Svpr	Suvarnaprabhasottama-sütra
Tait	Taittirvya Upanisad (see Principal Upanisads)
TB	Taittinya Brahmana
T C L	Ta ch'êng a-pi-ta-mo tsa chi lun
TD	TaishU Shinshu D aizôkyô
Thag	Theragâthâ
Thig	Therigatha
Tikap	Tikapatthâna, with Buddhaghosa's com m entary
T S	Tattvasamgraha, with panjika o f KamalasTla
Tsa	Tsa a-han ching
Tseng	Tseng i a-han ching
Tvps	Tattvopaplavasimha
UC R	University o f Ceylon Review
Ud	Udana
UdA	ParamatthadîpanT, Udâna-atthakathâ
VA	Samantapâsâdikâ, Vinaya-atfhakathâ
Vajr	Vajjracchedikâ Prajhaparamita
Vbh	Vibhahga
VbhA	Sammohavinodcml\ Vibhahga-atthakatha
Vin	Vinaya P itaka
Vism	Visuddhimagga
W ZK S& O	Wiener Zeitschrift für die Kunde Süd- und Ostasiens und Archiv für indische Philosophie

Notes

N O T E O N C H In E S E R Efer E N C Es

All references to the Chinese Buddhist texts are based upon the Taisho Shinshu D aizokyo unless otherwise stated. The volume number o f the Taisho Shinshu D aizokyo (abbreviated TD) is given first, followed by the page number and the
colum n, indicated by a, b, or c. W ithin parentheses is given the title o f the work in abbreviated form together with the fascicle number and the number of the sutra within that fascicle. Thus, TD 1.443c (Chung 4.2) is to be read: Taisho Shinshu D aizokyo, volum e 1, page 443, column 3, which is Chung a-han ching (Madhyama Agam a), fourth fascicle, second sutra.

Preface

1. T. I. Stcherbatsky, The Conception o f Buddhist Nirvana (Leningrad:
   Academy o f Sciences o f the USSR, 1927), pp. 2 3 -2 4 ; T. R. V. Murti, The Central Philosophy o f Buddhism (London: George Allen & Unwin, 1960), p. 66.
2. Bu-ston, H istory o f Buddhism (Chos-Hbyung), translated from the Tibetan*
   by E. Obermiller, 2 vols. (Heidelberg; Otto Harrassowitz, 1931-1932), 2:52ff.; Murti, Central Philosophy, pp. 123ff.; T. I. Stcherbatsky, Buddhist Logic, 2 vols.*
   (Leningrad: Academy o f Sciences o f the USSR , 1930-1932), 1 :3ff.
3. Stanislaus Schayer, " Pre-CanonicalBuddhism ," Archiv Orientalni(Prague)
   7 (1935): 131-132.

CHAPTER I

1. M . Hiriyanna, Outlines o f Indian Philosophy (London: George Allen &
   Unwin, 1956), p. 31.

      2. R V 1.115.3-4.
      3. R. T. H. Griffith, The Hymns o f the Rigveda (Benares: E. J. Lazarus & Co.,
1889), 1:200.
      4. R V 1.115.4.

      5. B. A. W. Russell, Our Knowledge o f the External W orld (London: George
Allen & Unwin, 1926), p. 220.
      6. See, for example, R V 5.83.
7. A. A. M acDonnell, A Vedic Reader fo r Students (L on d on : Oxford University Press, 1917), pp. 106ff.
      8. L. S. Stebbing, A Modern Introduction to Logic (London: M ethuen & Co.,
1962), pp. 260-261.
      9. F. R. Tennent, " Causality," ER E 3:261.
      10. Stebbing, Introduction to Logic, p. 292.
      11. Ibid., p. 261. Stebbing quotes Professor T. P. N unn, who said that even
the average student o f physics today is still an anthropomorphist at heart.
      12. R V 1.65.3.
      13. Ibid. 10.121.1.
      14. Ibid. 4.23.9.
      15. Ibid. 1.163.5.
      16. Ibid. 5.62.1.
      17. Ibid. 2.27.10.
      18. Ibid. 4.42.1-2; 5.69.4; 6.67.5; 7.41.7.
      19. A. B. Keith, The Religion and Philosophy o f the Vedas and TJpanisads
(Cambridge, M ass.: Harvard University Press, 1925), p. 83. He says: "From the
physical it is an easy step, to the conception of the Rta not merely in the moral
world, . . . but also in the sphere o f sacrifice."
      20. Ibid. pp. 246ff.
      21. See R V 7.86.
      22. H. Liiders, Varum, 2 vols. (Gottingen: V anderihoeck and Ruprecht,
1951, 1956), 2:405.
      23. Keith, Religion and Philosophy, pp. 454ff.
      24. Ibid. p. 84: "The stress laid on the conception o f the Rta in the sphere o f
the sacrifice . . . seems certainly to be no more than a reflex o f its importance at
once in the physical and the moral sphere."
      25. A V 9 .1 A 1 .
      26. K. N . Jayatilleke, Early Buddhist Theory o f Knowledge (London: George
Allen & Unwin, 1963), p. 443.
      27. § B 2.1.4.10.
      28. S 2.18; 1.134.
      29. TD 1.76a (CKang 12.1).
      30. TD 2.81a (Tsa 12.6); 2.86a-c (Tsa 12.20-21); 2.93c (Tsa 14.1).
      31. 5 2.18.
      32. TD 1.76a (CKang 12.1); TD 2.81a (Tsa 12.6).
      33. Skr 1: fols. 30-3 1 .
      34. 5 2 .1 8 .
      35. TD 2.81a (Tsa 12.6).
      36. 5 2 .1 8 .
      37. TD 2.81a (Tsa 12.6).
      38. 5 2.18.
      39. TD 2.81a (Tsa 12.6).
      40. Sv 1.2.

        41. R V 10.129.2.
        42. B. M. Barua, A H istory o f Pre-Buddhistic Indian Philosophy (Calcutta:
University o f Calcutta Press, 1921), p. 8.
        43. R V 10.190.
        44. Barua, p. 12.
        45. R V 10.129.
        46. S B 10.4.2.2.
        47. See Keith, pp. 442ff.
        48. TB 2.2.7.1.
        49. SB 11.5.8.1.
        50. Ibid. 4.1.1.22; see SB E 26 :253.
        51. Ibid. 13.7.1.1.
        52. A it I r 2.1.3.
        53. A. E. Taylor, Elements o f M etaphysics (London: Methuen & Co., 1961),
p. 159.
        54. A it A r 2.3.8.3; SBE 1:233.
        55. Ibid. 2.1.3: "atháto retasah srstih, prajápate reto deva devanám reto
varsam varsasya reto osadhaya osadhlnam reto'nnam annasya reto prajáh prajanam
reto hrdayam hrdayasya reto mano manaso reto vag vaco retah karma tad idam
karma krtamayam puruso brahmano lokah."
        56. Ibid. 2.1.8: "apa ity apa iti tad idam apa evedam vai mülam adas tülam
ayam pitaite putra yatra ha kva ca putrasya tat pitur yatra va pitus tad va putrasyety
etat tad uktam bhavati."
        57. Commentary on the A it Á r, by Sáyana Ácárya, edited by Rajendralala
Mitra, Bibliotheca Indica (Calcutta, 1876), p. 188: "karyakaranayor mrdghatayor
aty antabhedádarsanát. ' *
        58. Sv 1.12: "Etad jñeyam nityam evatmasamstham,
                          nátah param veditavyam hi kiñcit;
                          bhokta bhogyam preritaram ca matvá,
                          sarvam proktam trividham brahmam etat."
See R. E. Hume, The Thirteen Principal Upani$ads (London: Oxford University
Press, 1921), p. 396.
        59. Katha 2.8; see Hume, Upanisads, p. 349.
        60. M und 1.4-5; Hume, Upanisads, pp. 366-367.
        61. i?r/z 3.8.8.
        62. Ch 4.1.3.
        63. Év 1.1.ff.; M und 1.1.6.
        64. Ch 6 .2 .1 -4 ; Hume, Upanisads, p. 241.
        65. Ch 6.8 .3 -4 ; Hume, Upanisads, p. 245.
        66. ChB, p. 331: "Tatas ca pradhánam sáñkhyaparikalpitam jagatkaranam
pradhánasyácetanatvábhyupagamat. Idam tu sac cetanamTksitrtvat."
        67. ChB, p. 322: " sad ity astitámátram."
        68. S 2.17; TD 2.85c (Tsa 12.19).
        69. S 2.17: "Lokanirodham kho . . . yathabhütam sammappaññáya passato
ya loke atthitá sa na hoti." TD 2.85c (Tsa 12.19).
        70. SA 2.32: "atthitan ti sassatam."
        71. S 2.20; TD 2.85/c (Tsa 12.18).

72. S. Das Gupta, A H istory o f Indian Philosophy (Cambridge: Cambridge University Press, 1963), 1:257-258.
73. M K 1.1: M K V , pp. 13ff.
74. D. 3.117ff.; TD .72c (Ch'ang 12.1). 75. D 3.137-138: "sayam kato atta ca loko ca." See also Ud 69. The Chinese version o f the Pasadika-suttanta does not refer to the soul but merely says: "The world is self-caused" (TD 1.76a [Ch'ang 12.1]).
75. D 3.80; TD 1.36b (CKang 6.1). 77. Jayatilleke, Theory o f Knowledge, p. 475. 78. See chapter 6. 79. Sv 2.7; Hume, Upanisads, p. 278.
76. Brh 4.4.5.
77. i v 5.12: "sthulani suksmani bahuni caiva rupani dehl svagunair vrnoti" ;
    Hume, Upanisads, p. 407.
78. S 1.134; TD 2.327c (Tsa 45.6). 83. £v 5.7: "gunanvayo yah phalakarmakarta, krtasya tasyaiva sa copabhokta. sa visvarupas trigunas trivartma, pranadhipas samcarati svakarmabhih."
    Hume, Upanisads, p. 407. In translating samcarati, instead o f the phrase " roams about" we have used "wanders along," since the latter better expresses the meaning o f samcarati.
79. 55 1.1.2.2.
80. Skr 1: fols. 3 0 -3 1 : "yadi purusakarakrtam sukhadyanubhuyeta tatah sevaka-vanik-karsakadmaip samane purusakare sati phalapraptivaisadr§yam phalapraptis ca na bhavet. Kasyacit tu sevadi vyaparabhave pi visistaphalavaptir drsyata iti. A to na purusakarat kincit asadyate."
81. See chapter 2. 87. D 3.138; S2.19ff.; Ud 69; TD 2.86b (Tsa 12.21); 2.93c (Tsa 14.1).
82. S 2.18ff.; TD 2.86a (Tsa 12.21). 89. S 2.20: "so karoti so patisamvedTyatiti kho kassapa adito sato sayam
    katam dukkhan ti iti vadam sassatam etam pareti" ; TD 2.85c (Tsa 12.18).
83. K S 2.x.
84. S 2.13ff.: "N a kallo pahho ti bhagava avoca. Aharetiti na aham vadami.

Aharetiti caham vadeyyam tatr assa kallo pahho 'ko nu kho bhante ahareti'ti. Evan caham na vadami. Evam mam avadantam yo evam puccheyya' kissa nu kho bhante vinhanaharo'ti esa kallo pahho" ; TD 2.102a (Tsa 15.8).

92. A. J. Ayer, Language, Truth and Logic (London: Victor Gollancz, 1962),
    p. 51. For a detailed discussion o f this problem, see Jayatilleke, Theory o f Knowledge, pp. 277ff.
93. S 2.33: "tatravuso ye te samanabrahmana kammavada sayam katam
    (sukha-) dukkham pannapenti tad api phassa paccaya. . . . te vata annatra phassa patisamvedissantiti n'etam thanam vijjati" ; TD 2.94a (Tsa 14.1); also TD 1.76a
    (CKang 12.1), where the same argument is adduced against the theory o f the selfcausation of the world.

      94. D avid Hume, A Treatise o f Human N ature, ed. L. A. Selby-Bigge (O xford:
Clarendon Press, 1888), p. 252.
      95. Taylor, M etaphysics, p. 159, says that this was the attitude adopted by
Parmenides and his Eleatic successors.
      96. Brh 4.4.19; Katha 4.10-11.
      97. Paul Deussen, Philosophy o f the Upanisads, tr. A. S. G eden (Edinburgh:
T. & T. Clark, 1906), p. 156.
      98. Brh 3.8.7; 4.4.15; Sv 6.5; K ath a2A 4; Brh 4.4.20.
      99. Ch 6.1.4-6.
      100. D 3.28; M 2.222; ^ 1.173; TD 1.435b (Chung 3.3).
      101. R V 10.129.
      102. Ibid. 10.72.
      103. Ibid. 10.129.1.
      104. This is the method by which Aristotle arrived at the conception o f God
as the 'U nm oved M over'; see J. H. Randall, Aristotle (New York: Columbia
University Press, 1960), p. 137f.; the same argument was adopted by Thomas
Aquinas; see F. C. Copleston, Aquinas, Pelican Philosophy Series (London, 1955),
chap. 3; B. A. W. Russell, H istory o f Western Philosophy, rev. ed. (London:
George Allen & Unwin, 1961), pp. 446ff.
      105. R V 10.90.
      106. Barua, Pre-Buddhistic Indian Philosophy, p. 34.
      107. R V 10.121.
      108. F. D. K. Bosch, The Golden Germ: An Introduction to Indian Symbolism
(The Hague: M outon, 1960), p. 53.
      109. R V 10.121.
      110. Ibid. 10.82.4; Griffith, H ym ns o f the R igveda, 4:260.
      111. P. Edwards, and A. Pap, A M odern Introduction to Philosophy (New
York: Free Press, 1963), p. 507.
      112. Com m entary on 10.82.1: " nacaikasyajanyajanakabhavo virudhyate
tapobalena sariradvayasvlkarat." Rgveda, edited by M ax Muller, 6:172.
      113. R V 10.82.5: "dhata vidhata paramota samdrk."
      114. TB 3.1.9 J ; see also SB 7.4.2.5.
      115. SB 1.6.3.35ff.; 7.1.2.1ff.; St. Augustine believed that the preservation
o f the world was due to continuous procreation (ERE 4:229).
      116. SB 2.5.1.Iff.
      117. Ibid. 11.2.3.1.
      118. JB 3.17.2.
      119. Tandy a M aha Brahmana 21.3.7.
      120. TB 8.1.3-4.
      121. See Jayatilleke, Theory o f Knowledge, pp. 178ff.
      122. See M ahldasa's conception o f causation discussed earlier.
      123. A itA r 2.1.8.
      124. Ibid. 2.3.6.2.
      125. N yaya Varttika, p. 457: "yatha vasyadi buddhimata tak§nadhi§-
thitam acetanatvat pravartante, tatha pradhana-paramanu-karmani acetanani
pravartante. Tasmat tanyapi buddhimat karanadhifthanam." These arguments

have been very clearly stated by C. Bulcke, The Theism o f N yaya-V aisesika, Its
Origin and Early Development (Calcutta: Oriental Institute, 1947), pp. 36fT.; see
also G. M. Bhattacharyya, Studies in N yaya-V aisesika Theism, (Calcutta: Sanskrit
College, 1961), pp. 53ff.
      126. Brh 1.4.1-5.
      127. Ib id 2.5.19; 4.4.25; Ch 3.14.4.
      128. Jayatilleke, Theory o f Knowledge, p. 477.
      129. M und 1.1: "Brahma devanam prathamah sambabhuva,
                       visvasya karta bhuvanasya gopta."
      130. Sv 1.14; also 2.15; Hume, Upanisads, p. 396.
      131. Sv 1.1: "adhisthitah kena sukhetaresu vartamahe . . . vyavastham ."
      132. Ibid. 1.3.
      133. Ibid. Is: 3.7, 20; 4.7; Tsvara: 4.10; 6.7.
      134. Ibid. 4.14; 5.13: visvasya srastaram; 4.17: esa devo visvakarma.
      135. Ibid. 6.7, Tam lsvaranam paramam mahesvaram,
                    tain devatanam paramam ca daivatam.
      136. See NydyakusumdhjalT o f Udayana.
      137. Sv 5.6.
      138. D 1.17f.; TD 1.90c {Ch'ang 14.1).
      139. M 1.222; A 1.173f.; 7 5.228; TD 1.435b (Chung 3.3).
      140. Thomas, E. J., History o f Buddhist Thought (London: Kegan Paul,
Trench, Triibner & Co., 1959), p. 90.
      141. Text in D 1.18; M 2.227; translation in D ial 1.32.
      142. TD 1.90c {Ch'ang 14.1); see also TD 1.69a {Ch'ang 11.1); 1.547b {Chung
19.1).
      143. D 3.28: "issarakuttam brahmakuttam aggannam pannapenti." In the
Chinese version {TD 1.69a: Ch'ang 11.1), the terms brahma and issara appear to
have been amalgamated.
      144. D A 3.830.
      145. D 1.18; TD 1.90b-c {Ch'ang 14.1).
      146. The arguments adduced by various philosophers to refute the intuitional
m ethod o f verification are discussed by Edwards and Pap, Introduction to Philosophy,
p. 458.
      147. D 1.18; TD 1.90c {Ch'ang 14.1); see also TD 1.547b {Chung 19.1).
      148. Brh 1.4.3.
      149. D 1.18f.; TD 90c {Ch'ang 14.1).
      150. 7 5.238: "Issaro sabbalokassa sace kappeti jTvitam,
                     iddhivyasanabhavan ca kammam kalyanapapakam,
                    niddesakarl puriso issaro tena lippati."
      151. M 2.222, 227; TD 1.444a, 444c-445a {Chung 4.2).
      152. A 1.174; TD 1.435b {Chung 3.3).

CHAPTER II

1. Dictionary o f Philosophy and Psychology, ed. I. H. Baldwin (N ew York:
   M acm illan & Co., 1940-1949), 4 vols., 2 : 137f.
2. D. M. Riepe, The Naturalistic Tradition in Indian Thought (Seattle:
   University o f W ashington Press, 1961), p. 57.
3. Jayatilleke, Theory o f Knowledge, p. 71.
4. M bh, chaps. 215, 217.
5. They include: M. Hiriyanna, "Svabhävaväda or Indian Materialism,"
   in Indian Philosophical Studies (M ysore, 1957), 1.7 Iff.; G. Tucci, "A Sketch of Indian Materialism, P IP C , 1925, pp. 3 4 -4 3 ; Riepe, Naturalistic Tradition, pp. 53-78;
   W . Ruben, "Über denTattvopaplavasimha des Jayaräsi Bhatta, eine agnostizistische Erkenntniskritik," in W ZKS& O 1 (1958): 141-153; B. Chattopadhyaya, Lokäyata,
   A Study on Ancient Indian M aterialism (New D elhi: People's Publishing H ouse
   1959).
6. S. D as Gupta, A H istory o f Indian Philosophy, 3 vols. (Cambridge:
   Cambridge University Press, 1963), 3:528; Chattopadhyaya, L okayata, p. 45.
7. Jayatilleke, Theory o f Knowledge, p. 70.
8. Walter Ruben, Die Philosophen der Upanisaden (Bern: A. Francke Ag.

Verlag, 1947), p. 163.

9. W. K. C. Gutherie, H istory o f Greek Philosophy (Cambridge: Cambridge University Press, 1962), 2:145.
10. Analyzing the ontological speculations o f Parmenides, whose conception o f "Being" may be considered similar to the conception o f sat in the Upanisads, J. Burnet says that while som e consider Parmenides the "father o f Idealism," he himself considers Parmenides the "father o f Materialism." See his Early Greek Philosophy (London: A. & C. Black, 1958), p. 182.
11. Sv 1.2.
12. Jayatilleke, Theory o f Knowledge, p. 71. 13. M bh 12.215, 217; SD S, p. 13; Tattvopaplavasimha, p. 88. 14. Jayatilleke, Theory o f Knowledge, p. 444. 15. M bh 12.215,217. 16. D 1.55; TD 1.108b {CKang 17.1); S D S M p . 306; SD S, p. 2; Mbh 12.215. 17. Tvps, p. 1: "prthivyäpastejoväyur iti tattväni." 18. S D M S , p. 306: "prthivyädisamhatyä tathä dehaparinateh madasaktih
    surangebhyo yadvattadvaccidätmani" ; SD S, p. 2: "tatra prthivyadmi bhütäni
    catväri tattväni, tebhya eva dehäkäraparinatebhyah . . . caitanyam upajäyate."
13. In Buddhist texts: D 1.55: " attä rüpT catum m ahäbhütiko" ; TD 1.108b
    (Ch'ang 17.1); M 1.426: "tarn jtvam tarn sariiam " ; TD 1.804a {Chung 60.6). In Jaina texts: Su 2.1.9: "tajjTvatacchanraetti."
14. Tvps, p. 1: "prthivyäpastejoväyur iti tattväni, tat samudaye sarirendriyavisayasamjnä." Gunaratna, in his TarkarahasyadTpikä (p. 307) quotes a similar passage where vi$aya is placed before indriya, with an addition, tebhyas caitanyam, that Jayatilleke considers to be a reference to an emergent atrnan (see Theory o f Knowledge, p. 81, n. 2).
15. Jayatilleke, Theory o f Knowledge, p. 71.
16. Tvps, pp. 87ff. 23. Ibid., pp. 70f. 24. Ibid., p. 106.
17. SU 2.1.10: "sato natthi vinäso asato natthi sambhavo." 26. Skr 2: fol. 17; see Jayatilleke, Theory o f Knowledge, p. 70.
18. Tvps, p. 106: "tadä sarvabhävänäm nityatvam äpadyate vinäsäsambhavät."
19. Ibid., p. 87: "avicalitarupena purvaparakaryam na janyate; also ito pi dahanadhumayoh hetuphalabhavanupapattih yatha avicalitadahanarupasya purvaparanekakaryavirbhavakatvam na pratipadyate."
20. Ibid., p. 106: "na hi vandhyasutasyabhave sahetukatvam nirhetukatvam va vicarayanti santah."
21. Anvayarthaprakasika, p. 390: "svabhava iti svabhava vadino nastikavisesas carvakadayah." 31. Tattvabodhinl, p. 461: "svabhavavyatirekena karyakaranabhavanirupanayogat svabhavah karanam iti carvakah."
22. SamksepasarTrika, 1.528. 33. See S D S , p. 3; Jayatilleke, Theory o f Knowledge, p. 72. 34. Jayatilleke, Theory o f Knowledge, p. 77.
23. Nyayam ahjari, p. 109: "bhuyodarsanagamya pi na vyaptir avakalpate sahasraso pi tad drste vyabhicaravadharanat bhuyo drstva ca dhum o'gnisahacariti gamyatam agnau tu sa nastiti na bhuyodarsanadgatih" ; see Jayatilleke, Theory o f Knowledge, p. 78.
24. SD S, p. 13. 37. SV. Sankara's Bhasya, p. 276.
25. Hiriyanna, Indian Philosophical Studies, p. 73.
26. S D S , p. 13: " svabhavad eva tad upapatteh." 40. Tvps, p. 88. 41. Bibliotheca Indica, 7:276: "svabhavo nam a padarthanam pratiniyata saktih, agner ausnyam iva."
27. N K , p. 59: " sarvasya bhavatah svabhavatvanupapatteh. na hy ekam
    anekasvabhavam nama vyaghatat."
28. N K (Prakasa), p. 59: "ekaniyato dharmah svabhava ity ucyate, tad yad
    sarvasya sambhavet tada svabhavatvam asadharanatvam nopapadyate."
29. N K , p. 57: "niyamasyaivapek§arthatvat."
30. N K (BodhanT), p. 57: "idam eva hi karyasya karanapeksitvam nama,
    yat tasmim saty eva bhavatl ti Yadi niyamatiriktaupakaro nanglkriyate tarhi svabhavavada eva siddhah s y a t . . . dahanadTsu satsv eva dhum adyo bhavantrtTdrsah svabhavavadas cet sa asmabhir i§yata eva."
31. Ibid., p. 56.
32. N K , p. 56.
33. BodhanT, p. 56: "anapek§atvapratipatyartham nityagrahanam." 49. It has been pointed out that the Materialists not only accepted direct perception (pratyaksa) as the only valid source o f knowledge but also denied causation. See S. K. Belvalkar, and R. D. Ranade, H istory o f Indian Philosophy,
    vol. 2, The Creative Period, (Poona: Bilvakunja Publishing H ouse, 1927), p. 459.
34. TarkarahasyadTpika, p. 13: "kah kantakanam prakaroti taiksnyam,
    vaicitrabhavam mrgapaksinam ca, . . . svabhavatah sarvam idam pravittam ."
35. Hiriyanna, Indian Philosophical Studies, p. 74.
36. Tvps, p. 88: " sarve bhavah svabhavena svasvabhavavyavasthiteh . . .
    vyavrttabhaginah. "
37. According to the Milinda Panha (p. 4), this seems to be the view expressed
    by the Materialist Purana Kassapa, who, in reply to the question, "W ho rules the world? (ko lokam paleti)," said, "The earth rules the world (pathavi lokam paled )."
38. A. L. Basham, History and Doctrines o f the AjTvikas (London: Luzac &
    Co., 1951), p. 226.
39. Ibid., pp. 224-239; Jayatilleke, Theory o f Knowledge, pp. 142ff. 56. Basham, AjTvikas, p. 226. 57. D 1.53. The Chinese version of this sütra attributes this teaching to Pakudha Kaccäyana; see TD 1.108c (Ch'ang 17.1). See also TD 2.44a (Tsa 7.16).
40. Basham, ÄJTvikas, p. 225; Jayatilleke, Theory o f Knowledge, p. 144. 59. A Practical Sanskrit Dictionary, ed. A. A. M acD onnell (London: Oxford University Press, 1924), p. 328, col. 2; A Sanskrit-English D ictionary, ed. M. MonierW illiams (Oxford: The Clarendon Press, 1899), p. 1047, col. 2.
41. Cf. Mario Bunge, Causality: The Place o f the Causal Principle in M odem Science (Cambridge, M ass.: Harvard University Press, 1959), p. 13.
42. Jayatilleke, Theory o f Knowledge, p. 144.
43. Sü 1.1.2.3. 63. Skr 1: fol. 30. 64. D 1.53; TD 1.108c (Ch'ang 17.1). 65. DA 1.161; SA 2.341; / 5.237.
44. H oem le appears to follow Buddhaghosa in translating sahgati as "environm ent" ; see ERE, s.v. "ÄjTvikas."
45. D A 1.161. 68. TD 1.108c (CKang 17.1); cf. TD 2.44a (Tsa 7.16). 69. M 2.222; TD 1.443c (Chung 4.2).
46. TD 1.769b (Chung 54.2); cf. M 1.265f. (Maha-tanhasahkhaya-sutta). 71. TD 2.44a (Tsa 7.16).
47. A Dictionary o f Chinese Buddhist Terms, ed. W. E. Soothill and L. Hodous
    (London: Kegan Paul, Trench, Triibner & Co., 1937), s.v. hsiang hsu; Index to the M ahayanasüträlahkära, ed. G. M. Nagao, 2 pts. (Tokyo: Nippon Gakujutsu Sinko-kai, 1958), 2.225. A Sanskrit-Chinese Dictionary o f Buddhist Technical Terms Based on the M ahävyutpatti, ed. U. Wogihara (Tokyo: Sankibo, 1959), p. 70, gives
    prahandhah as an equivalent.
48. D A 1.161. 74. Belvalkar and Ranade, H istory o f Indian Philosophy 2:457. 75. See also B. C. Law, "Influence o f the Five Heretical Teachers on Jainism and Buddhism," JA SB (N ew Series) 15 (1919): 134.
49. Basham, AjTvikas, p. 225.
50. Jayatilleke, Theory o f Knowledge, p. 144. 78. Skr 1: fol. 31: " Samgaiyam ti samyak svaparinämena gatih. Yasya yadä
    yatra yat sukhaduhkhänubhavanam sä sangatih. Niyatis tasyäm bhavam sangatikam. Yatas caivam na puni§akärädikrtam sukhaduhkhädi, atas tat tesäm parinäm am niyatikrtam sangatikam ity ucyate." To illustrate this statement, Silänka
    (loc. cit.) quotes a verse that emphasizes complete determinism rather than indeterminism : "Präptavyo niyatibaläsrayena y o 'rthah, so 'vasyam bhavati nraäm subhäsubho vä.

bhütänäm mahati krte'pi hi prayatne, nabhavyam bhavati na bhävino'sti näsah."
Cf. "Jat tu jada jena jaha jassa ya niyamena hodi, tat tu tada tena taha tassa have idi vado niyadivado nu' Gommatasara, quoted in Belvalkar and Ranade, H istory o f Indian Philosophy, p. 455, n.

79. Skr l: f o l. 31: "niyater eve ti, etac ca dvitlyaslokante 'bhidhasye [i.e.,
    a.tSu 1.1.2.3]."
80. B. C. Law, Buddhistic Studies (Calcutta and Simla, 1931), pp. H i. 81. Basham, AjTvikas, p. 226. 82. D A 1.161: "yena hi yatha bhavitabbam so tatheva bhavati, yena na bhavitabbam so na bhavatiti." Cp. A. J. Ayer, The Concept o f a Person and Other Essays (London: Macmillan and Co., 1963), p. 235.
81. Law, Buddhistic Studies, p. 78.
82. Jayatilleke, Theory o f Knowledge, p. 142. 85. A V 19.53, 54; 19.54.3: "K alo ha bhutam bhavyan ca . . . ajanayat."
83. Sv 1.2.
84. Basham, AjTvikas, p. 225.
85. D 1.54: "donamite sukhadukkhe pariyantakate samsare n'atthi hayanavaddhane n 'atthi ukkamsavakamse. Seyyathapi nam a suttagule khitte nibbethiyamanam eva phaleti, evam eva bale ca pandite ca sandhavitva samsaritva dukkhassantam karissanti." It has not been possible to trace this important description o f salvation in the Chinese Agamas.
86. M 1.81-82.
87. J 5.228: "Tesu ahetukavadT ime satta samsarasuddhika ti mahajanam ugganhapesi" ; ibid., 5.237.
88. M bh 12. 215, 217. In 217.17, it is maintained that all creatures spring
    up from the elements— earth, water, space, fire, and wind.
89. Jayatilleke, Theory o f Knowledge, pp. 146-147.
90. M bh 12.215.19-20:
    "Anistasya hi nirvrttir anirvrttih priyasya ca, laksyate yatamananam purusarthah tatah kutah. Anislasyabhinirvrttim istasamvrttim eva ca, aprayatnena pasyamah kesancit tat svabhavatah."
91. Jayatilleke, Theory o f Knowledge, pp. 146-147. 95. In D 1.53; TD 1.108c (Ch'ang 17.1), it is attributed to PakudhaK accayana.
92. Ibid 1.56: "Tattha n 'atthi hanta va ghateta va sota va vinnata va vinnapeta va. . . . Y o pi tinhena satthena sTsam chindati na koci jlvita voropeti, sattannam eva kayanam antarena satthavivaram anupatati." A similar repudiation o f a personal agent and responsibility is found in the Bhagavad-gTta 2 : 19ff.
93. M bh 12.217.14. 98. TD 1.108a-b (CKang 17.1). 99. Basham, AjTvikas, p. 17.
94. M bh 12.217.36.
95. D 1.53-54; 1.55; 1,52-53; 1.56.
96. M 1.40Iff.
97. Ibid. 1.78 \T D 1.736c {Chung 49.3); see also A 2.31. However, Jayatilleke is reluctant to consider them as synonym s; see Theory o f Knowledge, p. 94.
98. TD 1.736c {Chung 49.3).
99. Pali-English Dictionary, ed. T. W. Rhys Davids and W. Stede (London:
    PTS, 1959), s.v. natthikaváda.
100. Basham, ÁjTvikas, p. 18.
101. Ibid., p. 23.
102. M 2.222: "sace . . . satta sañgatibhavahetu sukhadukkham patisamvedenti, addhá . . . niganthá pápasañgatiká yam etarahi dukkhá tibba kafuká vedan!

vediyanti." As pointed out earlier, to represent the term abhijatihetu o f the Pali version, the Chinese version has yin wei ming* cf. TD 1.443c (Chung* 4.2).

109. J 5.237: "UdTrana ce sañgatyá bhavayam anuvattati, akama akaramyam va karanTyam va pi kubbati, akamakaramyasmim kuvidha papena lippati."
110. JA 1.214; 3.20; 5.198, 459; Vism 238.
111. The conception o f svabhava discussed above and the theory of svabhava propounded by the Sarvástivadins appear to be different. The svabhava o f the Materialists and the AjTvikas refers to a pattern o f change and comes closer to the conception o f dhammata in early Buddhism (see below). But the svabhava o f the Sarvástivadins signifies an underlying eternal substance in phenomena— as opposed to their characteristics (lak$ana), which are subject to change according to the causal pattern. The term svabhava seems to have been used in the latter sense in the
     Sankhya school, where it is used as a synonym o f prakrti, as opposed to the three
     derivatives o f prakrti. See J. A. B. van Buitenen, "Studies in Sankhya," JA O S 76 (1956): 156.
112. M 1.185; TD 1.464c (Chung 1.2). 113. M 1.324: " dhammata esa . . . difthisampannassa puggalassa: kiñcápi
     tathárüpim apattim ápajjati yathárüpáya apattiya vutthánam paññáyati, atha kho nam khippam eva satthari va viññüsu va sabrahmacárisu deseti vivarati uttanlkaroti, desetva vivaritva uttánTkatvá áyatim samvaram ápajjati. Seyyathápi daharo kurnaro mando uttánaseyyako hatthena va pádena va añgárarp akkamitvá khippam eva patisam harati. . .
113. Ibid.
114. A 5.3, 313; TD 1.485b-c (Chung 10.2).
115. A 3.200; TD 1.486c (Chung 10.7); see also TD 2.129a (Tsa 18.6).
116. S 2.20: "añño karoti añño patisam vedfyatfti. . . param katam dukkhan ti.iti vadam ucchedam etam pareti" ; TD 2.85c (Tsa 12.18).
117. J 5.228: "ucchedavadT ito paralokagata nam a n 'atthi ayam loko ucchijjatiti ganhápesi."
118. D 1.55: "bale ca papdite ca káyassa bhedá ucchijjanti vinassanti na honti parammarana" ; TD 1.108b-c (Ch'ang 17.1).
119. We find yet another use of the term uccheda when the Buddha himself admitted that he could be called an ucchedavadT because he advocated the destruction o f the three roots o f evil: greed, hatred, and confusion; see Vin 1.235.
120. M 3.180; A 1.139; TD 1.504a-c {Chung 12.2).
121. For a m odem interpretation o f this problem, see Ayer, Concept o f a Person, p. 127:
     "even if someone could remember the experiences of a person who is long since dead, and even if this were backed by an apparent continuity of character, I think that we should prefer to say that he had somehow picked up the dead man's memories and dispositions rather than that he was the same person in another body. The idea of a person's leading a discontinuous existence in time as well as in space is just that much more fantastic. Nevertheless, I think that it would be open to us to admit the logical possibility of reincarnation merely by laying down the rule that if a person who is physically identified as living at a later time does have the ostensible memories and character of a person who is physically identified as living at an earlier time, they are to be counted as one person and not two."
122. S 2.17: "Lokasamudayam kho kaccayana yathabhutam sammappan-

naya passato ya loke natthita sa na hoti" ; TD 2.85c (Tsa 12.18). In S 2.17, atthitam
(<atthikavada) is identified with sabbam atthi, and natthitam (natthikavada) with
sabbam natthi. Sabbam atthi and sabbam natthi are mentioned in S 2.77, along with
two other metaphysical theories, sabbam ekattam (monism) and sabbam puthuttanri
(pluralism). The commentator points out that sabbam atthi and sabbam ekattarfi
belong to the etem alist school and that the other tw o are theories o f annihilationism
(see SA 2.76). This explains why there was much opposition to the Sarvastivada
theory o f sarvam asti from the other schools o f Buddhist thought.
      124. H. Jacobi, "Jainism" E R E 7 :465ff.; J. Sinha, A H istory o f Indian
Philosophy (Calcutta: Central Book Agency, 1952-1956), 2:197ff.; U . Misra,
H istory o f Indian Philosophy (Allahabad: Tirabhukti Publications, 1957), 1:229ff.;
D as Gupta, H istory o f Indian Philosophy 1:305ff.; M. Hiriyanna, Outlines o f Indian
Philosophy (London: George Allen & Unwin, 1956), pp. 157ff.; S. Mookerjee,
The Jaina Philosophy o f Non-Absolutism (Calcutta: University o f Calcutta Press,
1944), pp. 25ff.
      125. Jayatilleke, Theory o f Knowledge, p. 261; Sv 5.11.
      126. Sinha, H istory o f Indian Philosophy, 2:197.
      127. Tattvarthadhigama-sutra, 5.29: "utpadavyayadhrauvyayuktam sat."
      128. Misra, H istory o f Indian Philosophy, 1:300.
      129. Jayatilleke, Theory o f Knowledge, pp. 146ff.
      130. Su 1.1.3-3:
            N a tarn sayam kadam dukkham, kao annakadam ca nam,
            suham va jai va dukkham, sehiyam va asehiyam,
            sayam kadam na annehim vedayanti pudho jiya,
            samgaiyam tarn taha tesim, ihamegesi ahiam.
      131. Skr 1: fol. 3 0 -3 1 .
      132. Ibid.: "niyater eve ti etac ca dvifiyaslokante'bhidhasye [i.e., at Su
1.1.2.3]."
      133. Su 1.1.2.4: evam eyani j amp am t a bala pandiamanino,
                       niyayaniyayam samtam ayanamta abuddhiya.
      134. Skr 1: fol. 3 1 -3 2 :

"yato niyayaniyayam samtam iti sukhadikam kiricit niyatikrtam—avasyambhavyudayaprapitam tatha aniyatamatmapurusakaresvaradi prapitam sat niyatikrtam evaikantenasrayanti, ato'jananab sukhaduhkhadi karanam abuddhika buddhirahita bhavantiti. Tatha hi arhatanam kincit sukhadufikhádi niyatita eva bhavati—tat karanasya karmanah kasmimšcid avasare 'vasyambhávyudayasadbháván niyatikrtam ity ucyate, tathá kiňcid aniyatikrtaň ca—puru§akarakalesvarasvabhávakarmádikrtam, tatra kathaňcit sukhaduhkhádeh purusakárasádhyatvam apy ášňyate, yatah kriyatah phalam bhavati, kriýá ca purusakaráyatta pravartate, tathá coktam: na daivam iti saňcitya tyajed udyamam atmanah, anudyamena kas tailam tilebhyah práptum arhati? Yat tu samáne punisavyapare phalavaicitryam dusanatvenopanyastaiji tad adusanam eva, yatas tatrápi purusakaravaicitryam api phalavaicitrye káranam bhavati. Samáne vá puru$akáre yab phalábhávah kasyacid bhavati so'drstakrtah. Tad api ca asmabhih karanatvenášitam eva. Tatha kálo pi karta, yato bakulacampakásokapunnáganágasahakárádmám višista eva kale puspaphaládyudbhavo na sarvadeti, yac coktam kálasyaikarúpatváj jagadvaicitryam na ghatata iti, tad asmán prati na dúsanam, yato'smabhir na kála evaika kartrtvenábhyupagamyate api tu karmá pi, tato jagadvaicitryam ity ado§ab."

135. Cf. Uttaradhyayana-sutra 33.16; SBE 45.194, 196, n.2; E R E 7:468b.
136. Pubbekatahetu: M 2.217; A 1.174; yin pen tso: TD 1.442c (Chung 4.2);
     yin su ming tsao: TD 1.435a {Chung 3.3).
137. M 2.214: "yam kincáyam purisapuggalo patisamvedeti sukham vá dukkham vá adukkhamasukham vá sabbam tam pubbekatahetu; iti puránam kammánam tapasá vyantibhává, navánam kammánam akaraná áyatim anavassavo áyatim anavassavá kammakkhayo kammakkhayá dukkhakkhayo dukkhakkhayá vedanakkhayo vedanakkhayá sabbam dukkham nijjinnam bhavissatiti. Evamvádino . . . N iganthá" ; TD 1.442c (Chung 4.2).
138. B. M. Barua, A H istory o f Pre-Buddhistic Indian Philosophy (Calcutta:
     University o f Calcutta Press, 1921), pp. 385-386.
139. Ibid., p. 386, n. 1. 140. Ibid. p. 391. 141. Ibid., p. 393.
140. Ibid., p. 394. 143. Jayatilleke, Theory o f Knowledge, p. 151.
141. 7 5.238: "Sace pubbekatahetu sukhadukkham nigacchati, poranam katam papam tam eso muccate inam, poránam inamokkho kuvidha pápena lippati."
142. M 2.222: "S a c e ,. . . sattá pubbekatahetu sukhadukkham patisamvedenti, addhá, . . . Niganthá pubbe dukkatakammakarino yam etarahi evarupa dukkhá tippá katuká vedaná vediyanti. . . . tathágato pubbe sukatakammakan, yam etarahi evarupá anásavá sukhá vedaná vedeti" ; TD 1.443c {Chung 4.2).
143. M 2.214ÍT.; TD 1.442c {Chung 4.2); Jayatilleke, Theory o f Knowledge,
     p. 461.
144. Šv 5.11.
145. Jayatilleke, Theory o f Knowledge, p. 67.
146. UdA 345: "yasmá attánaň ca lokaň ca nimminantá issarádayo na
     kevalam sayam eva nimminanti, atha kho tesám tesám sattánam dhammádhammam sahakárikáranain labhitvá va, tasmá sayam kato ca param kato ca attá ca loko cati ekaccánam laddhi."
147. Jayatilleke, Theory o f Knowledge, p. 261.
148. D 1.28-29, 40; 3.33, 138; 5 2.20, 33-35, 113ff.; Ud 69, 70; TD 2.81a
     (Tsa 12.6).
149. UdA 345: "adhiccasamuppanno ti yadicchâya samuppanno, kenaci
     kâranena vinâ uppanno ti adhiccasamuppannavâdo dassito. Tena ahetukavâdo pi sangahito hoti.,'

Chapter M

1. SA 2.6 : "Paticcasamuppâdan ti paccayâkâram. Paccayâkâro hi annamahnam paticca sahite dhamme uppadeti. Tasmâ paticcasamuppâdo ti vuccati."
2. Vism, p. 521 : "uppajjamâno ca saha samâ ca uppajjati na ek'ekato na
   pi ahetuto ti samuppanno."
3. SA 2.41 : "paccaye nissâya samuppannam." 4. Gv, p. 89 ; Bbh, pp. 110, 204, 304, 396. 5. SA 2.41: " yathâ vuttânam etesam jarâm aran'âdînam paccayato va paccayasamuhato va idappaccayatâ ti vutto." Cf. Vism 518.
4. B H SD , p. 114, col. 1. The reference in Vin is 1.5. 7. S 2.26: "tâthatâ avitathatâ ananhathatâ idappaccayatâ ayam vuccati paticcasam uppâdo" ; TD 2.84b (Tsa 12.14).
5. M K V , p. 9: " asmin sati idam bhavati asyotpâdâd idam utpadyata iti
   idampratyayatârthah prafîtyasamutpâdârtha iti."
6. Vin 1.5; D 2.36-37; M 1.167; S 1.136: "duddasam idam thânam yadidam idappaccayatâ paticcasamuppâdo."
7. Cf. Bbh, p. 110, line 23, and TD 30.905b 23; Bbh, p. 204, line 25, and
   TD 30.921b 3.
8. See S 1.134; TD 2.327c {Tsa 45.6). 12. SA 1.193. 13. Vism, p. 521.
9. TD 2.84b, 85b ( Tsa 12.14, 17). 15. S 1.134; TD 2.327c (Tsa 45.6).
10. A Sanskrit-English Dictionary, ed. M . M onier-W illiam s (Oxford: The*
    Clarendon Press, 1899); D C B T , ed. W. E. Soothill and L. H odous (London:
    Kegan Paul, Trench & Triibner, 1937) ; Tibetan-English D ictionary, ed. H; A Jeschke*
    (London: Kegan Paul, Trench & Triibner, 1958), s.v. rkyen; A Tibetan-English Dictionary with Sanskrit Synonyms, ed. Saratchandra D as (W est Bengal: W est Bengal Government Press, 1960), s.v. rkyen.
11. L. de la Vallée Poussin, "Philosophy" (Buddhist), E R E 9 :848.
12. L. de la Vallée Poussin, "Identity" (Buddhist), E R E 7:99.
13. Vism 532: "paccayo hetu kâranam nidânam sambhavo pabhavo ti âdi atthato ekam vyanjanato nânam." AA 2.154: "nidânam hetu sankhâro paccayo
    rûpan ti pi hi etâni kâranavevacanân'eva." Referring to hetu and paccaya, Buddhaghosa says : " evam atthato ekam pi vohâravasena vacanasilitthatâya ca tatra etam
    ubhayam pi vuccati" (VA 1.185).
14. S 2 .3 0 -3 1 ; see also D 2.217, 259 ; M 3.71. 21. Sakv, p. 188: "hetu pratyayo nidânam kâranam nim ittam lingam
    upanisad iti paryâyâh."
15. Edgerton, in B H S D , p. 375, points out that B um ouf, K em , and Foucaux have rendered them thus.

      23. In the Nikayas: M 1.442, 444 ; 2.45, 74; A 1.55, 66, 200 (ko hetu ko
paccayo); in the Agamas: TD 2.57c (Tsa 9.9); 2.343b (Tsa 47.10); Sadm p, p. 8;
Svpr, pp. 6, 11; L V , pp. 120, 128.
      24. Af 1.294; >1 1.87; TD 1.50a (Ch'ang 8.2).
      25. Sakv, p. 188: "hetunam pratyayanam ca kah prativisesah. na kascid
ity aha. tatha hy uktam Bhagavata. dvau hetu dvau pratyayau samyagdr$ter
utpadaya. katam au dvau. paratas ca gho$o'dhyam am ca yoniso-m anasikara iti."
      26. L. S. Stebbing, A Modern Introduction to Logic (London: Methuen &
C o., 1962), pp. 2 7 0 -2 7 1 ; J. S. M ill says: "N othing can better show the absence o f
any scientific ground for the distinction between the cause o f a phenomenon and
its conditions than the capricious manner in which we select from among the conditions that which we choose to denominate the cause" (A System o f L ogic, 2 vols.
[London: Longmans Green & Co., 1872], 2:380).
      27. S 1.134; TD 2.327c (Tsa 45.6).
      28. The Chinese text reads: "arising as a result o f the harmony o f causes."
      29. It may be objected that an effect is not destroyed or does not disappear
if the cause is removed or destroyed; for example, when the seed, which has given
rise to the sprout, is destroyed, the sprout itself is not destroyed. But this can be
maintained only by taking the seed alone to be the cause. If we consider the cause
to consist o f factors such as fertility o f the soil and moisture, with the destruction
o f all the factors that constitute the cause the sprout would fail to grow.
      30. S 1.134: "yatha annataram bijam khette vuttam viruhati,
                    pathavirasan ca agamma sinehan ca tad ubhayam,
                    evam khandha ca dhatuyo cha ayatana ime,
                    hetum paticca sambhuta hetubhanga nirujjhare."
      31. Mill, System o f Logic, 1:383. B. A. W. Russell argues thus: " If the
inference from cause to effect is to be indubitable, it seems that the cause can hardly
stop short o f the w hole universe. So long as anything is left out, som e thing may be
left out which alters the expected result. But for practical and scientific purposes
phenomena can be collected into groups which are causally self-contained, or nearly
so. In the com m on notion o f causation the cause is a single ev en t. . . . But it is
difficult to know what we mean by a single even t; and it generally appears that, in
order to have anything approaching certainty concerning the effect, it is necessary
to include many more circumstances in the cause than unscientific commonsense
would suppose" Our Knowledge o f the External World (London: George Allen &
Unwin, 1926), pp. 229-230, 235.
      32. M 1.259; cp. TD 1.767a (Chung 54.2).
      33. Vism 520.
      34. This seems to be the position adopted by the Logical Positivists. Stating
the Positivist view, A. J. Ayer sa y s: "Another point in which our formula is somewhat
at variance with popular notions o f causality is that it does not differentiate between
a 'cause' and the accom panying 'conditions', the presence o f which is considered
to be necessary for the cause to produce the effect" (Foundations o f Empirical
Knowledge [London: M acmillan 1962], p. 181).

       35. For an analysis o f the contents o f the Patthâna, see Bhikkhu N yanatiloka,
A Guide through the Abhidhamma P itaka, (Colom bo: A ssociated Newspapers o f
Ceylon, 1957), pp. 118-127; D . J. Kalupahana, "The Philosophy o f Relations in
Buddhism 1 and 2," UCR 20 (1962): 19-54, 188-208.
       36. Vism, p. 235; Tikap 1.11: "paticca etasmâ etiti paccayo." Sumangala,
in Abhidhammatthavibhâvim-tïkâ, p. 133, gives a similar definition.
       37. Vism, p. 235; Tikap 1.12: "m ül'atthena hetu."
       38. M K 1.5 : "utpadyante pratltyem ln it'ime pratyayah kila" ; see M K V , p. 81.
       39. Tikap 1.5. The Abhs refers to it as the 'cause o f stability' or sthitikârana
(p. 28), and Dharmapala includes it under adhipati-pratyaya (TD 31.40a; CKeng 7).
       40. The Abhs seems to take it as dhrti-karana.
       41. Siddhi, p. 459; TD 31.41c (C E eng 8); and Abhs p. 28; TD 31.671 (C L 3).
       42. Ayer, Empirical Knowledge, p. 181 : "this use o f a special name to refer
to one am ong a number o f 'jointly sufficient conditions' m ay serve to single out an
event as one in which the speaker is particularly interested! ;] it does not correspond
to any difference o f function with regard to the production o f the effect."
       43. A K 2.49; TD 29.30a (C SL 6); K osa, p. 245.
       44. A K 2.61 ; TD 29.36b (C SL 7); Kosa, p. 299.
       45. Takakusu observes this distinction when he refers to the six hetus as the
chief causes and the four pratyayas as the four subcauses in the Sarvâstivâda theory
o f causality (The Essentials o f Buddhist Philosophy, 3rd ed. [H onolulu: University
o f Hawaii Press, 1956], p. 71).
       46. Sakv, p. 188: "atha katamasmim sutre sad hetava uktlh. sarvo hy
abhidharmah sütrârthah sûtranikasah sütravyâkhyânam iti. antarhitam tat sütram
iti Vaibhâsikàh."
       47. The Sanskrit version is in Akb 464; Hsiiang Tsang's translation appears
in TD 29.153c, 2 1 -2 2 (C SL 29); and Paramârtha's translation is in TD 29.305b,
13-14 (C SSL 22). See K osa, chap. 9, p. 241, n. 3, for the reconstruction o f the passage
by de la Vallée Poussin.
       48. Depending on the Tibetan translation, Stcherbatsky takes hetu and
pratyaya to mean 'first cause' and 'second cause,' respectively (The Soul Theory o f
the Buddhists [Petrograd : Academ y o f Sciences o f the U S SR , 1919], p. 836).
       49. Stebbing, Introduction to Logic, p. 262, says: "Further, in determining
which o f the various occurrences that are present is to be taken as the cause, com m onsense selects what is striking." See also p. 271.
       50. Kokuyaku Issaikyd (Pi fa n pu), 26/2:479; TD 2.57c (Tsa 9.9).
       51. In the Pali Nikâyas: M 1.108ff. ; S 2.72ff.; 4.32, 67, 166f.; in the Chinese
Agamas: TD 1.604b (Chung 28.3); 2.78b (Tsa 11.10); 2.87c (Tsa 13,3); and in
later Buddhist Sanskrit texts: LV , p. 176: "caksus prafîtya rüpatah caksurvijnanam
ihopajàyate."
       52. Vism, p. 599 : " tattha cakkh'âdïni cha dvârâni, rüp'âdîni cha ârammanâni
nàmassa sâdhâranapaccayo."
       53. Stebbing, Introduction to Logic, pp. 266, 271.
       54. Padmanabh S. Jaini, éd., Abhidharmadïpa, (Patna: K. P. Jayaswal
Research Institute, 1959), Introduction, pp. 95ff.
       55. For more examples o f this point, see Stebbing, Introduction to Logic,
pp. 270-271.

56. Sakv p. 188: "hetünäm pratyayänäm kah prativisesah. na kascid ity aha."

        57. Ibid.., p. 703: "hetur äsannah pratyayah; viprakr§tas tu pratyaya eva;
janako hetuh, pratyayas tu älambanamätram ity apare; paryäyäv etäv ity apare."
        58. N ettippakarana, p. 78: "D ve dhammä janayanti: hetu ca paccayo ca."
        59. Ibid.: "asädhäranalakkhano hetu, sädharanalakkhano paccayo."
        60. Ib id .: "yathä ankurassa nibbattiyä bTjam asädhäranam pathavl äpo ca
sädhäranä. Ankurassa hi pathavl äpo ca paccayo."
        61. Ibid.: "iti sabhävo hetu, parabhävo paccayo, ajjhattiko hetu, bähiro
paccayo, janako hetu, pariggähako paccayo, asädhärano hetu,sädhärano paccayo."
        62. Nänam oli, Guide, p. 111, n. 456/2.
        63. Äbhs, pp. 2 8 -2 9 ; TD 31.671b (CL 3).
        64. TD 31.4 6 7 b -c (PC PL 2).
        65. T D 31.41b (C K m g 8); Siddhi, pp. 453-463.
        66. Abhs, p. 28; TD 31.671b (C L 3).
        67. T D 31.713a (TC L 4).
        68. A K 2.61: "hetväkhyäh pancahetavah."
        69. TD 31.713a (TC L 4).

Chapter Iv

1. Stebbing, Introduction to Logic, p. 258.
2. B. K. M alinowski, "Magic, Science and Religion," in Religion, Science and R eality , ed. J. N eedham (London: Sheldon Press, 1925), pp. 3 0 -3 1 .
3. Stebbing, Introduction to Logic, p. 259. 4. Vin l . l f . ; Ud 1.
4. S 2.25: "paticcasamuppädan ca vo bhikkhave desissämi paticcasamuppanne ca dham m e" ; TD 2.84b (Tsa 12.14).
5. TD 1.562c-563a (Chung 21.3); Chachakka-sutta; M . 3.280-287.
6. Cf. B. A. W. Russell, Human Knowledge, Tte *Scö/?e *and Lim its* (London:
   Allen & U nwin, 1948), p. 439: "What passes for knowledge is o f two kinds: first, knowledge o f facts; second, knowledge of the general connections between facts."
7. Edward Conze, Buddhist Thought in India (London: George Allen &
   Unwin, 1962), p. 92.
8. M agdalene Geiger and Wilhelm Geiger, Pali Dhamma, vernehmlich in der Kanonischen Literatur (M unich: Bavarian Academ y o f Science, 1921); Conze, Buddhist Thought, pp. 92ff.
9. One is in D A 1.99 and DhA 1.22; the other is in DhsA 38. 11. Geiger and Geiger, Pali Dhamma, p. 4. 12. See T. I. Stcherbatsky, The Central Conception o f Buddhism and the M eaning o f the Term iDharm a) (London: Royal Asiatic Society, 1923), which is
   mainly devoted to an analysis o f the conception o f dharma in the Abhidharmakosa o f Vasubandhu.
10. Stcherbatsky, Central Conception, pp. 2ff.; idem, Buddhist Logic, 2 vols.*
    (Leningrad: A cadem y o f Sciences o f the U SSR , 1930), 1 :3ff.; C. A. F. Rhys Davids, Buddhism, A Study o f the Buddhist Norm, H om e University Library o f M odem*
    K nowledge (London: Wilhams & Norgate, 1912), pp. 49, 5 Iff.; Geiger and Geiger, Pali D hamm a, p. 4.

       14. 5 3.41, 67, 180; 4.28, 85, 106ff.; TD 2.66a (Tsa 10.6).
       15. 5 3.43, 139; 4.68; TD 2.8b (Tsa 2.4).
       16. 5 3.67; TD 2.66a (Tsa 10.6).
       17. Stcherbatsky, Central Conception, pp. 3 7 -3 8 .
       18. Stcherbatsky, Buddhist Logic, 1 :4 -5 .
       19. D hsA , p. 17, also p. 4; Sakv, p. 12.
       20. C. A. F. Rhys Davids, Buddhism, p. 47.
       21. D . J. Kalupahana, "Aspects o f the Buddhist Theory o f the External
W orld and the Emergence o f the Philosophical Schools o f Buddhism," C JH 1
(1970): 97ff.
       22. S 2.3f.; TD 2.85a-b (Tsa 12.16).
       23. M 1.185; TD 1.464b (Chung 1 2 ).
       24. S 2.33; TD 1.76a (Ch'ang 12.1).
       25. S 4.15 ; TD 2.91 a - b (Tsa 13.16); see also D . J. Kalupahana, "A Buddhist
Tract on Empiricism," P E W 19 (1970): 65ff.
      26. P. S. Jaini, ed., A D , Introduction, pp. 29ff.
       27. Cp. Dasuttara-suttanta o f the DTgha N ik â ya ; also the Ahguttara Nikâya.
       28. Dhs 125, 206-207. In Sarvâstivâda, the term cittaviprayukta referred to a
category o f dharmas drawn up later on; see A D , Introduction, pp. 93ff.
       29. Stcherbatsky, Central Conception, p. 37.
       30. Sahkara-hhâsya on Brahma-sUtra 2.2.20: " Kçanabhangavâdino'yam
abhyupagama, uttarasmim ksana utpadyamâne pürvah kçano nirudhyata iti. N a
caivam abhyupagacchatâ pürvottarayoh kçanayor hetuphalabhàvah sakyate sampàdayitum, nirudhyamânasya niruddhasya va pürvaksanasyâbhâvagrastatvâd
uttarakçanahetubhâvânupapatteh. '"
       31. T S 1.153. Bhadanta Yogasena is said to be a Buddhist o f the HTnayâna
school; see S. M ookeijee, The Buddhist Philosophy o f Universal Flux (Calcutta:
University o f Calcutta Press, 1935), p. 39.
       32. Yamakami Sôgen, System s o f Buddhistic Thought (Calcutta: University
o f Calcutta Press, 1912), p. 134.
       33. Ibid.
       34. Stebbing, Introduction to Logic, pp. 265ff. Following C. D . Broad (M ind
and Its Place in Nature [London: Kegan Paul & Co., 1962] p. 432), she says: "W hat
changes are the states ; what does not change is the thing o f which the states are the
states."
35. Dictionary o f Philosophy and Psychology, J. H. Baldwin, ed., s.v. " element."
       36. Stcherbatsky, Central Conception, pp. 25-26.
       37. Ibid., p. 37.
       38. Kosa 5.52ff.; TD 29.104c (C SL 20); A D V , pp. 259-260; T S 1.504; Sakv,
p. 44, 4 7 Iff.; see also de la Vallée Poussin, "Sarvâstivâda," M C B 5.1-157; AD ,
Introduction, pp. 117ff.
       39. T S 1.504: "tathà kâritre'vasthite bhâvo vartamânas tatah pracyuto'tïtas
tad aprâpto'nâgata iti."
       40. TS, pp. 504-505.
       41. Cf. F. H. Bradley, Appearance and R eality (Oxford : The Clarendon Press,
1951), p. 110.

42. T D 29.104b (C SL 20).
43. Ninian Smart, Doctrine and Argument in Indian Philosophy (London:
    George Allen & Unwin, 1964), p. 183.
44. M K V , p. 40; in the Pali Nikayas: S 2.25; and in the Chinese Agamas:
    TD 2.84b (Tsa 12.14).
45. T. I. Stcherbatsky, The Conception o f Buddhist Nirvana (Leningrad:
    Academ y o f Sciences of the USSR, 1927), p. 123, n. 4.
46. Stcherbatsky, Central Conception, p. 25.
47. TD 29.104c (C SL 20).
48. Stcherbatsky, Central Conception, p. 5. In fact, this is a quotation from
    W . M. M cGovern, A M anual o f Buddhist Philosophy (London: Kegan Paul, Trench, Triibner, and Co., 1923), and is a reference to TD 2.91b (Tsa 13.17). For a detailed
    discussion o f these texts and the Sarvastivada theoiy, see D . J. Kalupahana,
    "Sarvastivada and Its Theory o f sarvam asti," UCR 24 (1966): 94ff.
49. Stcherbatsky, Central Conception, pp. 5 -6 .
50. TD 2.91b (Tsa 13.18).
51. This interpretation of the passage in the Sam yukta Agama would mean
    that Stcherbatsky is not justified in maintaining that the "Theravadins have suppressed it because it did not agree with their particular tenets" (Central Conception,
    p. 5, n. 13).
52. In the Nikayas: S 3.70-73; in the Agam as: TD 2.65c-66a (Tsa 10.5),
    which refers to the past skandhas only.
53. See Nalinaksa Dutt, A spects o f M ahay ana Buddhism and Its Relation to H m ayana, Calcutta Oriental Series 23 (London: Luzac & Co., 1930), pp. 22 8 -2 2 9 ;
    and idem, "The Place o f the Aryasatyas and PratTtyasamutpada in Hmayana and M ahayana," JB O R S 11 (1929-1930): 119, which indiscriminately attributes it to
    all the Hmayana schools.
54. Stcherbatsky, Central Conception, p. 25.
55. 5 1.135; Kv 66\Mtin, p. 28; TD 2.327b (Tsa 45.5); 454c (Pieh-i Tsa 12,5).
56. M 1.136; T D 1.764c (Chung 54.1); see also Khandhasamyutta o f Samyutta N ikaya .
57. TD2AK. (Tsa 1.15).
58. M K , chap. 15, "Svabhavapariksa."
59. M K 24.9: "ApraGtyasamutpanno dharmah kascin na vidyate," M K V ,
    p. 505.
60. M K 15.2: "svabhavafi kptoko nama bhavisyati punah katham."
61. M K V , p. 260: "Rrtakas ca svabhavas ceti parasparaviruddhatvad asangatartham eva tat. Iha hi svo bhavah svabhava iti vyutpatter yah krtakah padarthah salokenaivasvabhavaitivyapadisyatetadyathaapam ausnyam . . . . yas tv akrtakah svabhavas tad yatha agner ausnyam . . . sa hi te§am padarthanantarasamparkajanitatvat svabhava ity ucyate."
62. Ib id .: "tad evam akrtakah svabhava iti lokavyavahare vyavasthite vayam idanlm brumo yad etad ausnyam tad apy agneh svabhavo na bhavatiti grhyatam kjtakatvat. Iha manindhanadityasamagamad araninighar§anades cagner hetupratyayasapeksataivopalabhyate, na cagnivyatiriktam ausnyam sambhavati, tasmad ausnyam api hetupratyayajanitam, tatas ca krtakam, kftakatvac capam au$nyavat svabhavo na bhavatiti sphutam avaslyate."
63. M K 15.7: "Kâtyâyanâvavâde câstî (ti) nâstîti cobhayam, pratisiddhaip bhagavatâ bhâvâbhâvavibhâvinâ."
64. M K V , p. 269 : "idan ca sütram sarvanikâyesu pathyate." 65. S 2.17; TD 2.86a {Tsa 12.18). 66. For a general discussion o f these problems, see Bradley, Appearance and R eality, pp. 38-3 9 .
65. Stcherbatsky, Central Conception, pp. 4ff.
66. M K 15.10, Astîti sâsvatagrâho nâstîty ucchedadarsanam. 69. Kathavatthu, pp. 115ff. 70. T. R. V. Murti, The Central Philosophy o f Buddhism (London: George Allen & Unwin, 1960), p. 26; see also Stcherbatsky, N irvana, p. 41.
67. M 1.228; S 3.133; 4.401; A 1.286; Thag 678; Dh 279.
68. L. de la Vallée Poussin Théorie des Douze Causes, p. 111. H e was quoting
    it from A 1.286.
69. TD 2.66b-67a (7j û 10.7); 2.668c 23.4); also 1.9b {Ch'ang 1.1).
70. A D p. 104.
71. Ibid., p. 281 : "Vartamânâdhvasampâtât sâmâgryângaparigrahât, labdhasakteh phalâksepah kâritram abhidhlyate."
72. TD 29.27c (C SL 20) : Kosa 2.228 : "ksanikasya hi dhannasya stbitim vinâ bhaved vyayah."
73. Sakv , p. 33: "ksanikânâm nâsti desântaragamanam, yatraivotpattih tatraiva vinasah."
74. TD 29.27c (C SL 20). 79. Ibid.
75. Cf. B. A. W. Russell, Analysis o f M ind chap. 5; Stebbing, Introduction to Logic, pp. 282ff.
76. TD 29.27c (C SL 20): Kosa 2.228.
77. r s 1.11. The question: "Ksanikâh sarve samskârâh, asthirânâm kutah kriyà." (Cf. Stcherbatsky, Buddhist Logic 1:119.) The answer: " bhütir yesâm kriyà saiva karakam saiva cocyate."
78. T S 1.11: "Ksanikàh sarve samskârâh" ; B C A P 376: "ksanikâh sarvasamskârâb" ; see also TD 29.27c {CSL 20). This statement in the Nikâyas and the À gam as always reads : sahhe sahkhdm aniccâ.
79. S 2.94-97; TD 2.81c-82a (Tsa 12.7-8).
80. O. H. de A. Wijesekera, The Three Signata (Kandy: Buddhist Publication Society, 1960), pp. 2 -3 .
81. SA 2.99. 87. Sakv, p. 33. 88. SA 2.99. 89. S 3.38; TD 2.607c (Tseng 12.15). 90. A 5.187: "Y am kind bhutarn . . . tad aniccam ." 91. M 1.350; 3.108; A 5.343-346: "yam abhisankhatam tad aniccam." 92. S 1.186; Thag 1215: "yam kind parijTyati sabbam aniccam ." 93. Ps 1.191: "uppâdavayattena aniccâ." D 2.157, 199; S 1.191; 3.146:
    " aniccâ vata sankhârâ uppâdavayadham m ino" ; TD 2.153c {Tsa 22.1).
82. A 1.286.
83. M K 24.19: " apraiîtyasamutpanno dharmah kascin na vidyate" ; see
    M K V , p. 505.
84. Vin 1.41 : "Y e dhammâ hetuppabhavâ tesam hetum tathâgato âha, tesan ca yo nirodho evamvâdr M ahâsam ano."
    See also Lankâvatâra-sütra, p. 444; ÂryapratTtyasamutpâda-sütra, p. 26.
85. M 1.228; S 3.133; 4.401 ; A 1.286; TD 2.66b-67a (Tsa 10.7); 668c (Tseng 23.4); see also TD 1.9b (Ch'ang 1.1). E. Lamotte has collected many references to this statement in the early as well as later Buddhist texts (see L'Enseignement de V im alakïrti, Bibliothèque de M uséon, vol. 51 [Louvain : M uséon, 1962], p. 165, n. 51).
86. D 2.198. 99. S 1.134. The Chinese translators o f this passage on sahkhâra (TD 2.327b :
    Tsa 45.5) were aware o f this general use when they rendered the phrase sahkhârapuhja as yin chu, which means a 'collection o f aggregates,' and avoided using the character hsing, which is generally used to translate sahkhâr a ( 'disposition').
87. P T S D p. 233, s.v. Khandhâ. 101. Murti, Central Philosophy, pp. 4 9 -5 0 .
88. M K V , pp. 4 1 -4 2 ; S 3.142:
    " PhenapimJüpamam riïpam vedanà bubbulüpamâ, marTciküpamâ sannâ sankhârâ kadalüpamâ, m âyopaman ca vinhànam dTpitâdiccabandhunà."
    TD 2.69* (Tsa 10.10).
89. 5 1.134; TD 2.327c (Tsa 45.6).
90. 5 3.96; TD 2.14a (Tsa 2.25); 5 3.103; TD 2.14b-15a (Tsa 2.26).
91. C. A. F. Rhys Davids, Buddhism, p. 82. 106. M K V , p. 443: "N a hi Tathagatalj kâcid apy âtmanah skandhânàm vâstitvam prajnâpayanti."
92. D ictionary o f Philosophy and Psychology, J. H. Baldwin, ed., s.v. "phenom enon."
93. See A. J. Ayer, Language, Truth and Logic (London: Victor Gollantz,
    1962), pp. 32, 53.
94. Stanislaus Schayer, "Pre-Canonical Buddhism," Archiv Orientalni
    (Prague) 7 (1935): 127ff.
95. S 2.26; TD 2.84b (Tsa 12.14). 111. S 2.77. This passage could not be traced in the Chinese Agamas. 112. SA 2.76.

Chapter V

1. S 2.25 : "U ppâdà va tathâgatànam anuppâdâ va tathàgatànam thitâ va sa dhatu dham m atthitatà dhammaniyàmatà idappaccayatâ" ; TD 2.84b (Tsa 12.14). See also M K V 40; A S S 73; de la Vallée Poussin, Théorie des Douze Causes,
   pp. 111-112. Cf. Russell, Our Knowledge o f the E xternal World, p. 217: " It is to be
   observed that what is constant in a causal law is not the object or objects given, nor yet the object inferred, both o f which may vary within wide limits, but the relation between what is given and what is inferred." See also idem, M ysticism and Logic (London: George Allen & Unwin, 1963), p. 142.

     2. Ud, p. 1: "Yada have patubhavanti dhamma,
                 atapino jhayato brahmanassa;
                 ath'assa kankha vapayanti sabba,
                 yato pajanati sahetudhammam.''
     3. 5 2.25; TD 2.84b {Tsa 12.14).
     4. M l . 190-191 a n d 5 3 .1 2 0 ; TD 1.467a(Chung7.2 )and 16.815 b (.L P S S C) ; see
also A S S , p. 70.
     5. 5 4.330; 5 2 .1 7 ,2 0 ,2 3 ; 7Y) 2.85c {Tsa 12.18, 19).
     6. M 1.262-64; 5 2.28, 70, 96; Ud,, p. 2.
     7. In M K V , p. 9 the locative absolute construction is given thus: "asmin sati
idam bhavati, asyotpadad idam utpadyate."
     8. M ahavastu, 2.285, which seems to contain the only com plete statement in
Buddhist Sanskrit tex ts; in all the other places only the first part is found.
     9. TD 2.67a (Tsa 10.7), 100a (Tsa 14.16). Instead o f the character sheng
sometimes we find the character ch'i; see TD 2.85b (Tsa 12.17).
     10. TD 1.562c {Chung 21.3); 2.713c-714a {Tseng 30.2).
     11. C. A. F. Rhys Davids, "Paticcasamuppada," ERE 9:672; idem, Sakya
or Buddhist Origins (London: Kegan Paul, Trench & Triibner, 1931), p. 154, n. 2.
     12. M ario Bunge, Causality: The Place o f the Causal Principle in M odern
Science (Cambridge, M ass.: Harvard University Press, 1959), p. 37.
     13. Epigraphia Indica (Delhi: Government o f India), vol. 21, 1931, p. 197.
See W. D . W hitney, A Sanskrit Grammar (Leipzig, 1879), p. 174, for the difference
in meaning between idam and asau.
     14. Bunge, C ausality, pp. 3 5-36.
     15. 5 2.26: "Avijjapaccaya bhikkhave sankhara. Iti kho bhikkhave ya tatra
tathata avitathata anannathata idappaccayata, ayam vuccati bhikkhave paticcasam uppado." TD 2.84b {Tsa 12.14).
     16. A. B. Keith, Buddhist Philosophy in India and Ceylon (Oxford: The
Clarendon Press, 1923), p. 96.
     17. TD 2 .8 5 b -c {Tsa 12.17). This short but very important sutra does not
appear in a PaH version.
     18. In another place, TD 2.217c {Tsa 30.16), fa hsing, hsing meaning 'nature,'
appears instead o f fa chieh.
     19. 5 2.104-107, called N agara-sutta. The Chinese Buddhists seem to have
considered this a very important sutra, as is evident from the existence o f three
separate translations apart from the two entries in the Sam yukta {TD 2.80b-81a)
and E kottara {TD 2.718a-c) Agam as. The three translations are (1) attributed to
Chih Chien {TD 16.826b), (2) by Hsiian Tsang {TD 16.827b), and (3) by Fa-hsien
{TD 16.829a).
     20. 5 2.105-106; TD 2.80c {Tsa 12.5). Also Lahkavatara-sutra, pp. 143-144.
     21. SA 2.41: " so tehi tehi paccayehi anunadhikehi eva tassa tassa dhammassa
sambhavato tathata ti." Cf. Vism, p. 518.
     22. N . Tatia, "Paticcasamuppada," Nava Nalanda M ahavihara Research
Publication 1 (1957): 179.
     23. P T S D , p. 296, s.v. tathata.
     24. B H S D , p. 248, s.v. tathata.

25. SA 2.41 : "sâmaggim upagatesu paccayesu muhuttam pi tato nibbattânam
    dhammânam asambhavâbhâvato avitathatâ." Cf. Vism, p. 518.
26. Bunge, C ausality, pp. 39f. See also H. Feigl and W . Sellars, Readings in Philosophical Analysis, Century Philosophy Series (N ew York: Appleton-CenturyCrofts, 1949), p. 523.
27. SA 2.41 : "annadhammapaccayehi annadham m ânuppattito anannathatâ
    ti."
28. Bunge, C ausality, p. 38; Russell, Our Knowledge o f the External World,
    p. 234.
29. Russell, Our Knowledge o f the External W orld, p. 217.
30. SA 2.41 : "Yathâ vuttânam etesam jarâm aranâdlnam paccayato va paccayasamûhato va idappaccayatâ ti vutto."
31. Bunge, Causality, p. 103.
32. Ibid.
33. Keith, Buddhist Philosophy, p. 116.
34. P. Oltramare, La Formule Bouddhique des D ouze Causes: Son Sens Original et Son Interprétation Théologique (Geneva: Georg, 1909), p. 48.
35. Since the time o f Hume, the formula o f constant conjunction has been regarded as exhausting the meaning o f causation. It is considered to be the correct statement o f the causal bond. Thus, A. J. Ayer maintains that " every general proposition o f the form 'C causes E ' is equivalent to the proposition o f the form 'whenever C, then E '," (Language, Truth and L ogic, p. 55). The empiricists have held that "the meaning o f causation consists in the statement o f an exceptionless repetition, a constant union between cause and effect." This is what W illiam James criticized as " a world o f mere withness, o f which the parts were strung together by the conjunction 'and'" (Pragmatism [New York, Toronto, and London: Longmans Green & Co., 1943], p. 156).
36. Bunge, Causality, p. 42.
37. Jayatilleke, Theory o f Knowledge, p. 449. See also Keith, Buddhist Philosophy, p. 96, for a criticism o f this formula. Contrary to the view o f Keith, the statement bears the stamp o f comprehensiveness rather than imperfection.
38. Cf. the definition o f paticcasamuppâda by Buddhaghosa, Vism, p. 520ff.

and the definition o f paccaya in Vism, p. 532-533 : "yo hi dham m o yam dhammam apaccakkhâya titthati vâ uppajjati vâ so tassa paccayo ti vuttam hoti. Lakkhanato pana upakârakalakkhano paccayo; yo hi dhammo yassa dhammassa thitiyâ va uppattiyâ va upakârako hoti, so tassa paccayo ti vuccati."

39. R atnàvatï 1.48: "Asmin satîdam bhavati dlrghe hrasvam yathâ sati." 40. Ibid. 1.49, "Hrasve sati punar dïrgham na bhavati svabhàvatah."
40. S 2.150, "Yâyam . . . âbhâdhàtu ayam dhâtu andhakâram paticca pannàyati. Y âyam . . . subhadhâtu ayam dhâtu asubham paticca pannâyati. Yâyam . . .
    âkâsancâyatanadhâtu ayam dhâtu rüpam paticca pannâyati." TD 2.116c (Tsa 17.1).
41. Stebbing, Introduction to Logic, p. 264.
42. Jayatilleke, Theory o f Knowledge, p. 449.
43. M . R. Cohen and E. Nagel, An Introduction to Logic and Scientific M ethod
    (London: Routledge & Kegan Paul, 1949), p. 270.
44. Sn, p. 139.
45. Stcherbatsky's volum inous work, Buddhist L ogic , is based primarily on the "system o f logic and epistem ology" formulated by Dinnaga and Dharmakifti.
46. Jayatilleke, Theory o f Knowledge, pp. 443, 457. 48. Russell, Our Knowledge o f the External W orld, p. 226; idem, M ysticism and Logic, p. 143; P. Edwards and A. Pap, A M odern Introduction to Philosophy (N ew York: Free Press, 1963), pp. 122-123.
47. D avid Hume, A Treatise o f Human N ature, ed. L. A . Selby-Bigge (Oxford :
    The Clarendon Press, 1888), p. 252.
48. Ibid., pp. 252-53. Italics mine. 51. Ibid., p. 34. Italics mine. 52. Ibid., p. 39. Italics mine.
49. This was m ore or less prompted by a desire to refute the belief in 'substance' and belief in the continued existence o f objects when n o t perceived.
50. Russell, M ysticism and Logic, p. 132.
51. Dh 33. 56. D 1.78ff.; TD 1.553b-c (Chung 19.3). See also TD 1.86a (Ch'ang 13.1).
52. D 1.79; M 2 . 19: "So dibbâya sotadhàtuyà visuddhaya atikkantamânusikâya ubho sadde sunàti, dibbe ca mànuse ca, ye düre santike ca" ; TD 1.86a (Ch'ang 13.1).
53. M 1.502; TD 1.670b (Chung 38.2).
54. See Jayatilleke, Theory o f Knowledge, pp. 4 3 9 -4 4 0 . 60. D 1.80-81; TD 1.86a (Ch'ang 13.1); see also D 1.213; TD 1.101c-102a
    (Ch'ang 16.1).
55. A 1.170-171.
56. Cf. "muscle-reading," in R udolf Tischner, Telepathy and Clairvoyance,
    tr. from the German by W. D . Hutchinson, International Library o f Psychology
    (London: Kegan Paul & Co., 1925), p. 4.
57. A 1.170-171: "yathâ imassa bhoto m anosankhârâ panihita imassa cittassa anantarâ am unnâm a vitakkam vitakkissatîti. So bahun ce pi âdisati tath'eva tarn hoti no annathâ."
58. Russell, Our Knowledge o f the External W orld, p. 222. 65. Skr 1 : fol. 35 : " paracetovrttînâm duranvayâd ajnânam eva sreyah." 66. Paul Demieville, in "Sur la M émoire des Éxistences Antérieures," has m ade a comparative study o f the material available in the N ikâyas and the Agam as on retrocognition : Bulletin de VEcole Française d'E xtrêm e Orient (Hanoi) 27 (1927) :
    283-298.
59. D 1.82; TD 1.86b (Ch'ang 13.1). 68. Ibid. 69. D 2.20. 70. Ibid. 1.152. 71. C. D . Broad, who believes that the question o f the possibility o f human survival o f bodily death is partly empirical and partly philosophical, says: "It is empirical in the sense that if it can be clearly formulated and shown to be an intelligent question, the only relevant way to attempt to answer it is by appeal to specific observable facts. . . . The relevant observable facts are som e o f those investigated by psychical research and in particular certain phenomena o f trancem edium ship" : Human Personality and the Possibility o f its Survival (Berkeley and L os Angeles: University o f California Press, 1955), p. 1.
60. M 1.93; TD 1.587b-c (Chung 25.2). 73. D 1.83-84; T D 1.86c (Ch'ang 13.1). 74. Vism, p. 411. 75. S 2.25; TD 2.84b (Tsa 12.14). 76. Russell, M ysticism and Logic, p. 144. 77. 5 2 .5 8 ; Cf. TZ) 2 .9 9c-100a (Tsa 14.14-15): " So im inâdham m enaditthena viditena akâlikena pattena paryogâlhena afTtânâgate nayam neti : ye kho keci afTtam addhânam sam açâ va brâhmanâ va jarâmaranarp abbhannaipsu, jarâmaranasamudayam . . . jaxâmarananirodham . . . jarâm araçanirodhagâm inim patipadam . . .

seyyathâpaham etarah i. . . Ye hi pi keci anâgatam addhânam samanâ va brâhmanâ vâ jarâmaranam abhijânissanti. . . seyyathâpaham etarahîti idam assa anvaye nânam ." The Chinese version gives a very brief description o f the contents of the Pali passage. The phrase fa chu chih ( = anvaye hâna) may be interpreted as "the knowledge o f the continuity [of the nature] o f things."

78. H . Van Rensselar W ilson, "On Causation," in Determinism and Freedom in the A ge o f M odern Science, ed. Sidney H ook, 2nd impression (N ew York: N ew
    York University Press, 1965), pp. 225ff.
79. M 3.207ÎL; TD 1.707b (Chung 44.2). 80. Ibid. ; see also Jayatilleke, Theory o f Know ledge, p. 463.
80. M 3.214-215; TD 1.708b (Chung 44.2). 82. A 1.249 ; the Chinese version is included in the M adhyam a Âgam a, TD
    1.433a (Chung 3.1), which is called the "sutra o f the salt-sim ile."
81. This is the implication o f the famous Kaccâyanagotta-sutta o f the Samy u tta ; see S 2.17; TD 2.85c (Tsa 12.19).

CHAPTER VI

1. D A 2.432; D h sA, p. 272; cf. C. A. F. Rhys Davids, Buddhism, p. 119.
2. S 2.178ÎT.: " anam ataggo'yam . . . samsâro pubbâkoti na pannâyati avijjânTvaranânam sattânam tanhâsamyojanânam sandhâvatam samsaratam." TD
   2.485c (Pieh-i Tsa 16.1); see also TD 2.241b (Tsa 34.1). It is interesting to note that the locution anamatagga in the Pali version, which means "inconceivable is the beginning" (ana = negative prefix; m ata = past participle o f man or mny "to think,"
   and agga = agra, "beginning" ), occurs in m ost o f the Buddhist Sanskrit texts as
   anavarâgra (see B H SD , p. 21). In the M adhyam akakârikâ (11.1), Nâgâijuna says:
   "Pürvâ prajnâyate ko tir nety uvâca mahâmunih, sam sâro'navarâgro hi nâsyâdir nâpi pascim ah."
   Anavarâgra has generally been translated "without beginning and end" (BH SD,
   p. 21). This is quite clear from Nâgârjuna's use o f nâsyâdir nâpi pascimah, on the basis o f which Jacques M ay translates anavarâgra as " sans début ni terme"
   (Candrakïrti, Prasannapadà M âdhyam akavrtti, Paris: Adrien M aissoneuve, 1959,
   p. 170). But the Àgam a version seems to imply "without prior limit" when it renders anavarâgra as wu yu pên chi (TD 2.214b; Tsa 34.1). Again comparing anamatagga and anavarâgra, it seems that the former is in keeping with the teachings o f early Buddhism in that it implies the difficulty o f knowing or determining the beginning, hence an epistemological rather than an ontological problem. This view carries the support o f later Theraväda commentators (see SA 2.156: " anamataggo ti anu am ataggo, vassasatam vassasahassam nänena anugantvä pi anam ataggo aviditaggo, nässa sakkä ito va etto vä aggam jänitum aparicchinnapubbaparakoliti attho" ). But the latter implies a definite denial o f a beginning and is in keeping with the metaphysics o f M ahäyäna.
3. This sutra is found in D 3.80ff.; TD 1.36b (Ch'ang 6.1). The Chinese
   version, unlike the Pah version, specifically states its purpose.
4. The portion within parentheses is not found in the Chinese version.
5. The etym ology o f the term äbhassara has presented problems. The P T S D
   derives it from ab hä + svar> "to shine," and defines it as " shining in splendor" (p. 103). The Chinese translators seem to have followed a traditional explanation when they rendered it as kuang yin , "bright speech" {kuang = äbhä, yin = svara,
   "syllable").
6. Here the Chinese version adds: " On account o f the exhaustion o f merit
   [ = puhhakkhayal] and the termination of the life span [ = äyukkhayal]."
7. In the Chinese version the portion within parentheses is given after another passage.
8. See G. P. Malalasekera and K. N . Jayatilleke, Buddhism and the Race Question (Paris: Unesco, 1958), pp. 32ff.
9. S 2.181-184; TD 2.242a-243b (Tsa 34.8-15). 10. The portions in parentheses are not found in the Chinese version. 11. 10.129.1-4. 12. According to the Chinese version, the earth gushed forth or bubbled up like a fountain and was like cheese or honey in appearance.
10. This simile is not found in the Chinese text. 14. D ial 1.107. 15. A 4.100-103; TD 2.736b {Tseng 34.1). 16. According to the Chinese version, it is at this stage that the small rivers, water spouts, and even the four great rivers dry up, which phenom ena occur, according to the Pali version, after the appearance o f the second sun.
11. According to the Chinese version, the appearance o f the second sun is follow ed by the drying up o f all vegetation.
12. W hile the Pah version refers to five rivers, the G an g!, Y am una, Aciravati, Sarabhu, and Mahl, the Chinese version has only four, the Gangä, Sindhu, Slta and Oxus.
13. The account in the Chinese version starts with seven hundred leagues and goes down to one league.
14. This simile is not found in Chinese. 21. Earthquakes: A 4.312; TD 2.753c {Tseng 37.5); also TD 1.477b {Chung 9.1); drought: A 3.243.
15. In connection with causality of the human personality: S 1.134; TD
    2.327b {Tsa 45.6); 455a {Pieh-i Tsa 12.6); S 3.54; TD 2.9a {Tsa 2.7); A 1.223f.;
    psychic events: A 1.135, 136; 3.404; moral behavior: A 1.134-135; TD 1.647b
    {Chung 34.6); Sn 77; Thag 363, 388.
16. A 1.223-224: "Iti kho Änanda kammam khettam vinnanam bijam tanhä sineho avijjanlvarananam sattanam tanhasamyojananam hlnaya dhatuya vinnanam patitthitam." W e have not been able to trace this passage in the Agamas, but it seems to have been known to the compiler o f the Arya-salistamba-sutra (p. 84).
17. M 3.239: "chadhaturo'yam . . . puriso ti." TD 1.690b (Chung 42.1).

A lthough m an is said to be composed o f six elements, the Pali version enumerates only five, om itting the element o f water. For an idealistic interpretation o f the theory o f six elements, see Stanislaus Schayer, "Pre-Canonical Buddhism," Archiv Orientalni (Prague), 7 (1935): 130.

25. Ch 8.8.Iff., where the atman is progressively defined starting with the theory that it is the physical body. But this view is rejected in favor o f the view that the atman is best represented by the mind in the turiya (fourth) state.
26. C. A. F. Rhys Davids, Buddhist Psychology, p. 41.
27. Sometimes rendered chi eng yin (see TD 1.464c [Chung 7.2], 788a-b
    [Chung 58.1]), but better translated as chlu yun; see TD 2.499c (N o. 102).
28. M 1.299: "Panca kho m e . . . upadanakkhandha sakkayo vutto Bhagavata" ; TD 1.788a (Chung 58.1).
29. M 1.191: "Paticcasamuppanna kho pan'im e . . . panc'upadanakkhandha" ; TD 1.467a {Chung 1 2 ).
30. M 1.265: "Tinnam kho bhikkhave sannipata gabbhassavakkanti hoti" ;
    TD 1.769b {Chung 54.2).
31. M 1.184; TD 1.464b {Chung 7.2); see also D 1.76. 32. M 1.265-266. 33. See D 2.63; TD 1.61b {Chlang 10.2). 34. M 1.262; TD 1.767c {Chung 54.2). 35. E. R. Saratchandra, Buddhist Psychology o f Perception (Colom bo: The
    Ceylon University Press, 1958), pp. 18ff.
32. D 2.63; TD 1.61b {CKang 10.2). 37. Saratchandra, Buddhist Psychology o f Perception, p. 20. 38 . A 1.176: "Channam . . . dhatunam upadaya gabbhassavakkanti hoti, okkantiya sati namarupam . . . ." TD 1.435c {Chung 3.3) reads: "Because o f the harm ony o f the six elements, there is conception. Because o f the six elements, the six senses com e into being." It is stated very often that the six senses depend on the psychophysical personality {namarupapaccaya sarayatanam ). Therefore, it seems that in the Chinese version the six elements represent the psychophysical personality (the riamarupa).
33. S 2.66: "yan ca . . . ceteti yan ca pakappeti yan ca anuseti arammanam etam hoti vinnanassa thitiya, arammane sati patittha vinnanassa hoti, tasmim patitthite vinnane virulhe namarupassavakkanti hoti" ; TD 2.100b {Tsa 14.19).
34. M 1.184; TD 1.658a {Chung 36.2).
35. G odh ik a : S \ .\2 2 ; T D 2.286b{T sa 39.11); 383a {P ieh -iT sa 2 .8);Vakkhali:
    S 3.124; TD 2.347b {Tsa 47.25).
36. Saratchandra, Buddhist Psychology o f Perception, p. 20, n. 65. 43. TD 2.383a-12 {Pieh-i Tsa 2.8), which is closer to the Pali version. 44. TD 2 .3 4 7 b -ll {Tsa 47.25).
37. TD 2.603a {Tseng 12.3). The parallel passage is found in M 1.265-266. 46. TD 1.596b {Chung 26.4); M 1.1 {M ulapariyaya-sutta). The term shen is never used alone, in the Chinese Àgamas, to denote consciousness that survives death.
38. M 1.256: "tad ev idam vinnânam sandhâvati samsarati anannam ." TD
    1.766c (Chung 54.2) specifically states that it is the consciousness that does not change or alter. M 1.258 : "yvâyam . . . vado vedeyyo tatra tatra kalyânapâpakânam kammànam vipàkam patisamvedetîti." TD 1.767a (Chung 54.2) says that consciousness is " the doer as well as the causer to do" ; see also M iln, pp. 54-56ff.
39. K S 3.viii. E. Ziircher points out that "The Chinese (not unreasonably)
    were unable to see in the doctrine o f rebirth anything else than an affirmation o f a survival o f a 'soul' (sheri) after death" ; see The Buddhist Conquest o f China (Leiden:
    E. J. Brill, 1959), p. 11.
40. M 1.256-257: "Nanu maya anekapariyâyena paticcasamuppannam vinnâ^am vuttam, annatra paccayâ natthi vinnânassa sambhavo ti." TD 1.766c (Chung 54.2).
41. TD 31.17a-b (Ch'eng 3); see also Siddhi, pp. 199-200. 51. S 2.94: "cittam iti pi mano iti pi vinnânam, iti pi," where the words are used synonym ously; TD 2.81c (Tsa 12.7). The Pali-English D ictionary (PTS) renders the terms citta and mano as "thought" and "m ind," respectively. But citta is used in the early Buddhist texts as a generic term, and the Chinese translators have used the character hsin meaning "mind" to render this term. On the other hand, mano,
    which is only a faculty, has been translated into Chinese as z, "thought."
42. A 1.171. 53. D 3.105 (vinnânasota); S 4.128 (bhavasota). D 3.105: "purisassa ca vinnânasotam pajànâti ubhayato abbocchinnam idhaloke patitthitan ca paraloke patitthitan ca." TD 1.77b (Ch'ang 12.2).
43. The bodily dispositions are defined as exhaling and inhaling. Verbal dispositions are reflection and investigation. M ental dispositions are explained as perception and feeling (sahhâ, vedana), but the Chinese version seems to imply perception and volition (hsiang szü = sahnà and cet ana).
44. 5 4.294; TD 2.150a-b (Tsa 21.10).
45. This implies that a saint w ho has com e out o f a trance in which perception and feeling had ceased (sahhâvedayitanirodha) and whose perceptive faculties are once more active accumulates sahkkàras and is therefore not different from an ordinary man, or at least is liable to fall away from sainthood. This m ay have prompted the Sarvâstivâdins to uphold that a saint could fall away from the state he had attained. See Vasumitra's Nikdyâlam bana-sastra, tr. J. M asuda, Origin and D octrines o f Early Indian Buddhist Schools, in A M 2 (1925): 27. But according to early Buddhism, the difference between an ordinary man and a saint (arhat) is that the dispositions o f the latter are inoperative because he has attained the state o f "pacification o f all dispositions" (sabbasahkhârasamatha).
46. A 1.10: "Pabhassaram idam . . . cittam tan ca kho àgantukehi upakkilesehi upakkilittham." E. Lamotte has collected m ost o f the references to the conception o f "luminous mind" in Buddhist literature; he points out that the Sarvâstivâda Vaibhâçikas disagreed with the Vibhajyavâdins (Theravadins?) on this problem. See L 'Enseignement de VimalakTrti, Bibliothèque du M useon, vol. 51.
    (Louvain, 1962), pp. 52ff. ; also A. Bareau, Les Sectes Bouddhiques du P etit Véhicule,
    (Saigon: École française d'Extrême-Orient, 1955), pp. 6 7 -6 8 .
47. Lahkavatara-sutra, p. 222.
48. S 4.138; this passage could not be traced in the Agamas.
49. M unn, Norm an L., Psychology: The Fundamentals o f Human A djustm ent,
    4th ed. (London: George G. Harrap & Co., 1961), p. 507.
50. M 1.190; TD 1.467b (Chung 7.2); in the special form ulation o f the causal principle the two factors namarupa and saVayatana explain this relationship.
51. M 1.111-112: "cakkhuh ca paticca rupe ca uppajjati cakkhuvinnanam, tinnam sangati phassao, phassapaccaya vedana, yam vedeti tarn sanjanati, yam sanjanati tarn vitakketi, yam vitakketi tain papanceti, yam papahceti tato nidanam purisam papancasannasankha samudacaranti atltanagatapaccuppannesu cakkhuvihneyyesu rupesu." TD 1.604b {Chung 28.3).
52. C. A. F. Rhys Davids, Buddhist Psychology; Saratchandra, Buddhist Psychology o f Perception.
53. Nanananda, Concept and Reality in Early Buddhist Thought (Kandy:
    Buddhist Publication Society, 1971).
54. Ibid., p. 5. 66. S n A ,p .4 3 . 67. Russell, Our Knowledge o f the External W orld, p. 238.* 68. Jayatilleke, Theory o f Knowledge, pp. 453-454.*
55. Russell, Our Knowledge o f the External W orld, p. 238.
56. Ibid., p. 239. 71. A 5.313; see TD 1.485b-c {Chung 10.2).
57. M 1.195; TD 2.759b-c {Tseng 38.4). 73. D 3.275; TD 2 .85a-b {Tsa 12.16).
58. M 2 .2 5 -2 6 ; TD 1.720 {Chung 47.2).
59. M 1.373: " Imesam kho aham Tapassi tinnam kammanam evam pativibhattanam evam pativisitthanaiji manokammam mahasavajjataraip pannapemi papassa kam m assa kiriyaya papassa kammassa pavattiya, no tatha kayakammam no tatha vadkam m am ." TD 1.628b {Chung 32.1).
60. A 3.415; " Cetana'haip bhikkhave kammam vadam i; cetayitva kam m am
    karoti kayena vacaya m anasa." TD 1.600a {Chung 27.5).
61. M 3.207: " M ogham kayakammam mogham vacTkammaiji manokamm am eva saccan ti." TD 1.706b {Chung 44.2).
62. Ibid.
63. M 3.207: "Imina . . . Samiddhina moghapurisena Potaliputtassa paribbajakassa vibhajja byakarahiyo panho ekansena byakato." T D 1.707a, lines 6 -7
    {Chung 44.2).
64. A 3.415: "Phasso . . . kammanam nidanasam bhavo" ; TD 1.600a {Chung 27.5). Here the term keng lo is used to translate phassa; see also TD 1.435c
    {Chung 3.3).
65. A 1.200.
66. R. S. Peters, Concept o f M otivation (London: Routledge & Kegan Paul,
    1960), p. 15.
67. M 1.324. 84. A 1.134: "T m 'imani . . . nidanani kammanam samudayaya. Katam ani tTni? L obho . . . doso . . . m oho nidanam kammanam sam udayaya." TD 1.483c {Chung 3.6).
68. A 2.158.
69. S 4.172ff.; TD 2.313b (Tsa 43.9). In the first instinct mentioned, the character sheng, which means 'birth', 'arising,' etc., m ay be taken in this context as 'life' or 'existence.' In the second instinct, while the Pali term am aritukam a is used as a synonym for jTvitukama, the Chinese version clearly implies 'aversion to death.' The first two instincts enumerated com e very close to what Freud called the life instinct, i.e., the drive for self-preservation— see A. J. C. Brown, Freud and the Post-Freudians
    (London: Pelican Books, 1963), p. 27f. The last two instincts represent another idea, which Freud called the pleasure principle. H e sa id : " It seems that our entire psychical activity is bent upon procuring pleasure and avoiding pain, that it is automatically
    regulated by the PLEASU R E PR IN C IPLE," Introductory Lectures on PsychoAnalysis, translated from the German by Joan Riviere (London: G eorge Allen &
    Unwin, 1949), pp. 298-299.
70. Peters, Concept o f M otivation, p. 50. 88. M 3.203; TD 1.705a (iChung 44.1). 89. Jayatilleke, Theory o f Knowledge, p. 460.
71. M 3.203ff.; TD 1.705a {Chung 44.1). 91. M 3.210; TD 1.707bf. {Chung 44.2).
72. Ibid, see Jayatilleke, Theory o f Knowledge, p. 463. 93. M 3.214-215; TD 1.708b {Chung 4 4 2 ).
73. A 1.249. The Chinese version is included in the M adhyam a A gam a ; see
    TD 1.433 {Chung 3.1).
74. "yatha yatha'yam puriso kammam karoti tatha tatha tarn patisamvediyati."
75. "yatha vedanlyam ayam puriso kam m am karoti tatha tatha'ssa vipakam patisamveditayi." The Chinese version does not show this difference in the two statements. In G S 1.227, the Pali passage is translated: "Just as this man does a deed that is to be experienced, so does he experience its fulfilm ent." W e feel that this does not convey emphatically enough the difference between the two statements.

The words yatha vedanlyam should be taken as emphasizing the way it would be
experienced.

97. G S 1.227, n. 2. 98. A 1.249: "idha . . . ekaccassa puggalassa appam attikam pi papam kam m am katam tarn enam nirayam upaneti. Idha pana . . . ekacassa puggalassa tadisam eva appamattikam papam kammam katam ditthadhamme c'eva vedaniyam hoti nanu pi khayati bahud eva." TD 1.433 {Chung 3.1). The section within parentheses is not found in the Chinese version.
98. W oodward seems to have had difficulty rendering appadukkhavihdri (see
    G S 1.228, n. 1). F ollow ing the Chinese passage, "shou ming shen tuan" {TD 1.433a) and the translation o f the phrase appamdnaviharT as shou ming shen cKang {TD
    1.433b), we have interpreted appa as referring to length o f life and dukkha as denoting the quality o f life.
99. The passage within parentheses is not found in the Chinese version.
100. A 1.250; TD 1.433a-b {Chung 3.1). 102. Sn 260: " Patirupadesavaso ca pubbe ca katapunnata, attasammapanidhi ca etaip m angalam uttam am ."
101. A 1.223-224; A SS , p. 84.
102. The passage in the Dhammapada (127), quoted in the Milindapanha
     (p. 150) and in the Divyavadana (p. 532), states that there is no place visible in this universe, the m idst o f the ocean or even in mountain cave, from which one could ward off the consequences o f evil actions ("na antalikkhe na samuddamajjhe na pabbatanam vivaram pavissa, na dissati yo jagatippadeso yatthatthito mucceyya papakam m a"). Taken in isolation, this statement may imply complete determinism in m oral responsibility. But considered along with the other statements cited above, it should only be taken as emphasizing the invariability o f the causal pattern when there are no obstructing conditions (see the discussion in M iln, pp. 150ff.).
103. M 3.203; A 3.72, 186; 5.88, 288: "Kammassaka satta kammadayada kam m ayoni kam m abandhu . . . ." TD 1.704c (Chung 44.1).
104. D 2.49: " Sabbapapassa akaranam kusalassa upasampada, sacittapariyodapanam etam Buddhanasasanam."
105. R V 10.90. 108. The data available in the Pali N ikayas have been subjected to exhaustive analyses; see R . Fick, Social Organization in N orth-E ast India in Buddha's Tim e,
     translated by S. Mitra (Calcutta: University o f Calcutta Press, 1920) and Malalasekera and Jayatilleke, Buddhism and the Race Question.
106. D 3.81; TD 1.36c (Ch'ang 6.1). 110. D ial 1.106. 111. M 2.149; TD 1.664a (Chung 37.3). The two countries mentioned are Y ona and Kamboja.
107. M 2.84; TD 2.142b (Tsa 20.12). 113. A 3.383. W e have not been able to trace this passage in the Chinese Agam as. The term 'species' (abhijati) in this context is used in a very narrow sense
     to include human beings only, whereas in the philosophy o f M akkhali G osala it was used in a broader sense to include "all beings," although Buddhaghosa has explained it in terms o f the six human types (DA 1.162).
108. M alalasekera and Jayatilleke, Buddhism and the Race Question, pp.

3 8 -3 9 .

115. Kandakavuttika, which is explained by the com m entator as kantakavuttikd (AA 3.395). See G S 3.273.
116. D 1.89: " So imam pathavim sagarapariyantam ada^dena asatthena dham m ena abhivijiya ajjhavasati." TD 1.82b (Ch'ang 13.1).
117. D 3.60; TD 1.39b (Ch'ang 6.2). The glories o f a universal monarch are
     said to consist o f seven jewels; see D 3.142; TD 1.493b (Chung 11.2).
118. D 3.142ff.; TD 1.493a (Chung 11.2).
119. D 3.68: " Iti kho . . . niddhananam dhanam anuppadiyamane daliddiyam vepullam agamasi, daliddiye vepullagate adinnadanam vepullam agamasi, adinnadane vepullagate sattham vepullam agamasi, satthe vepullagate paijatipato vepullam agamasi, panatipate vepullagate musavado vepullam agamasi, musavade vepullagate tesam sattanam ayu pi parihayi vapno pi parihayi." TD 1.40c-41a (Ch'ang 6.2).
120. D 3.95f.; TD 1.3 8 b -c (Ch'ang 6.1).
121. D 3.65; TD 1.40b (Ch'ang 6.2).
122. A 3.47-48.
123. D 3.92; TD 1.38b (Ch'ang 6.1).
124. D 3.155ff.
125. A 1.200.
126. Dh 165: " suddhï asuddhl paccattam nânnam anno visodhaye." 127. 5 3.69ff.; TD 2.21a (Tsa 3.23). 128. Bbh, pp. 99ÊT. 129. D 2.251-252; TD 1.34a (Ch'ang 5.1).
127. M 1.184 : "khînâ jati vusitam brahmacariyam katam karanlyam nâparam itthattâyâ ti" ; TD 1.658a {Chung 36.2).
128. M 479,480; 3.1 : "N âham . . . âdiken'eva annârâdhanam vadàmi, api ca
     anupubbasikkhâ anupubbakiriyâ anupubbapatipadâ annârâdhanâ hoti." TD 1.652a
     {Chung 35.3).
129. In the N ikâyas: D 1.70, 172, 181 ; A f 1.180-181, 269, 346; A 2.210; in the Àgamas: TD 1.657b {Chung 36.2); 733 {Chung 49.1).
130. Bbh, p. 67.
131. M 1.181-184; TD 1.657b-658a {Chung 36.2).
132. S 2.29ff.
133. Ibid. 3.126. 137. M 1.184; TD 1.658a {Chung 36.2).
134. D 3.260. 139. Sn 268. 140. It, pp. 3 8-39.
135. A 2 .37-39; TD 2.2 8 a -b (Tsa 4.14); 467 a -b (Pieh-i Tsa 13.18); 7 1 7 c718a (Tseng 3 ).
136. 5 4.362. 143. See D hs 193; also Sakv 174: "samskitatvam pratïtyasamutpannatvam iti prayàyâv etav. Sametya sambhüya pratyayaih krtam samskrtain. Tam tam pratyayam pratitya sam utpannam pratTtyasamutpannam 'ti."
     144*. 5 2 .2 8 ; M 1.262f.; *d . TD* 2.98a-b (Tsa 14.6).
137. Keith, Buddhist Philosophy, pp. 96 -1 1 4 ; de la Vallée Poussin, Théorie des Douze C auses; E. J. Thom as, H istory o f Buddhist Thought (London: Kegan Paul, Trench, & Triibner, 1959), pp. 5 8 -7 0 , 7 8 -8 0 ; Stcherbatsky, Central Conception; Oltramare, La Formule Bouddhique des Douze C auses; P. M asson-Oursel,
     "Essai d' interprétation de Théorie des Douze Conditions," Revue de VH istoire des Religions, 71 (1915): 3 0 -4 6 .
138. Keith, Buddhist Philosophy, p. 109; Thom as, H istory o f Buddhist Thought, p. 58.
139. Jayatilleke, Theory o f Knowledge, p. 451. 148. 5 2.20; TD 2.85c (Tsa 12.18).
140. Keith, Buddhist Philosophy, pp. 110, 112.
141. 5 2 .1 9 -2 0 ; TD 2.86a (Tsa 12.20). Jayatilleke points out that in speaking o f logical alternatives that m ay be false, the Nikâyas use the phrase no h'idam, "it is not so" not ma K evam , "do not ask thus" ; see Theory o f K now ledge, p. 293.
142. VbhA, pp. 132-133 : "K im pakativâdïham pakati viya avijjâ' pi akaranam mulakaranam lokassa ti. N a akaranam, asavasamudaya avijjasamudayo ti hi avijjaya karanam vuttaip" ; cf. M 1.55.
143. A 5.113.
144. VbhA, p. 133.
145. S. Z. Aung and C. A. F. Rhys Davids, tr., Compendium o f Philosophy
     (London: PTS, 1910), p. 262.
146. S 2.1, 25: "Paticcasamuppadam vo bhikkhave desissam i" ; TD 2.84b
     (Tsa 12.14), 85a (Tsa 12.16).
147. S 2 .5 -1 1 ; TD 2.101a-b (Tsa 15.2).
148. Introduction o f the special application o f the causal form ula in connection with aging and death: S 2.11; TD 2.101c (Tsa 15.7); the four form s o f nutrition:
     S 2.28; the five aggregates: S 2.19; TD 2.85c (Tsa 12.19); also 86a (Tsa 12.20); the causality o f m oral behavior: S 2.17f.; 76f.; TD 2.85c (Tsa 12.18); criticism o f current philosophical theories: Thomas, H istory o f Buddhist Thought, p. 60, n. 1.
149. C. A . F. Rhys Davids goes to the other extreme, m aintaining that the abstract statement is earlier and that the applied form ula is later and was "crudely fitted" ; S akya , p. 152.
150. TD 2.83c, 84b, 85a-c, 8 6 a -b (Tsa 12.11-20). 160. C. A. F. Rhys Davids, "Paticcasamuppada," 673b. 161. Keith, Buddhist Philosophy, p. 108. 162. Ibid, pp. 99f. 163. Ibid., p. 106.
151. Oltramare, L a Formule des Douze Causes, pp. 2 8 -2 9 .* 165. Keith, Buddhist Philosophy, p. 107.*

C H A P T E R V N

1. Stcherbatsky, Conception o f Buddhist N irvana, pp. 2 3 -2 4 .
2. Murti, Central Philosophy, p. 66.
3. D hsA, p. 421: "Santatipaccuppannan c 'ettha atthakathasu agatam,
   addhapaccuppannam sutte. Tattha keci khanapaccuppannam cittam cetopariyananassa arammanam hoti ti vadanti."
4. D . J. Kalupahana, "Schools o f Buddhism in Early C eylon," C J H 1 (1970):
   159ff.
5. ibid.
6. A D V , p. 258: "Y ah khalv e§a prathamo v ad l sarvastivad'akhyah, esa
   khalu . . . sadvadT."
7. Sakv, p. 362: "svabhavata ity atmatah."
8. Murti, Central Philosophy, pp. 133-134.
9. M K 1.3: "N a hi svabhavo bhavanam pratyay'adisu vidyate."
10. SeeJ. A. B. van Buitenen, " Studies in Sankhy a (1),^" J A O S 76 (1956): 156. 11. W e have not strictly observed this distinction in rendering svatotpattivada as 'self-causation'.
11. Sik$dsamuccaya, p. 248: "iti hy abhutva bhavati bhutva pativigacchati
    s vabhavarahitatvat. ' '
12. M K V , p. 263: "atha ev ausnyam agner hetupratyaya-pratibaddhatvat
    purvam abhutva pascad utpadena krtakatvac ca na svabhava iti yujyate."
13. Sakv, p. 294: "utpadas ca namabhutvabhavalaksanah. Sautrantikanayenotpattir dhannasya tadammtanaiva bhavati ti. A lso, na casau purvam utpadat kascid astiti Sautrantikamatena. "
14. T S 1.155: "NiyatacintyasaktTni vastuniha pratiksanam, bhavanti nanuyojyani dahane dahasaktivat."
15. S. M ookeijee, The Buddhist Philosophy o f Universal Flux (Calcutta:
    University o f Calcutta Press, 1935), p. 54.
16. Murti, Central Philosophy, pp. 170ff.
17. S. D as Gupta, A H istory o f Indian Philosophy (Cambridge: Cambridge
    University Press, 1963), 1:257.
18. Jayatilleke, Theory o f Knowledge, p. 453.
19. K osa 2.228, n. 1. 21. M 3.25.
20. T S 1.153: "Kramena yugapac capi yatas te'rthakriyakrtah,
    na bhavanti tatas tesam vyarthah ksanikatasrayah, sahakarikrtas caiva yada natisayah kvacit, sarvada nirvisesaiva tada santatir i$yate."
21. H . V. Guenther, Philosophy and Psychology in the Abhidhamma (L ucknow :
    Buddha Vihara, 1957), p. 259, n. 1; see also Kosa 2.228.
22. M iln, p. 52.
23. Ibid., pp. 51-52.
24. Siksasamuccaya, p. 248: "Tatra maharaja adhyatmikas tejodhatur utpadyate, na kutascid agacchati nirudhyamano na kvacit sancayam gacchati." For
    a discussion o f the significance o f this statement, see A D , pp. 266ff., where it is said to be a quotation from the Paramarthasunyata-sutra.
25. M iln, p. 53.
26. S 2.17, which Nagarjuna quotes in support o f his theory; see M K 15.7. 29. I am indebted to my friend Upali Karunaratna, Assistant Editor o f the Encyclopaedia o f Buddhism, for providing me with som e o f the reference material on this subject.
27. A 2 .38-39.
28. Kvu, KvuA 18.1-4.
29. D 2.140. 33. S ad m p ,-p .2 1 . 34. Ibid. 272. 35. Vajr, p. 56. 36. D ivy, p. 158, 396; Siks, p. 159. 37. Vajr, p. 43; A sP , p. 94. 38. A sP, p. 153, 307, 512. 39. Vajr, pp. 32—33. 40. S 2.17; TD 2.85c (!Tsa 12.19).
30. M K 15.7, "Katyayanavavade castiti nastiti cobhayam, pratisiddham bhagavata bhavabhavavibhavina.,,
31. Ibid., 15.10.
32. K P , p. 90: "astiti kasyapa ayam ek o'ntah nastiti ayam dvifiyo'ntah yad
    etayor dvayor antayor madhyam iyam ucyate kasyapa madhyama pratipad dharmanam bhutapratyavek§at."
33. M K V , p. 270, n. 1: "Astiti kasyapa ayam eko'nto nastiti kasyapa ayam ek o 'ntah, yad enayor dvayor madhyam tad arupyam anidarsanam apratistham anabhasam aniketam avijnaptikam iyam ucyate kasyapa madhyama pratipad dharmanam bhutapratyavek$eti."
34. Thus, the Bodhicarydvatara, a later text o f the M adhyam ika school, seems to maintain that "things that are causally produced belong to the 'conventional' (sam vrti)," but not causahty itself; see p. 352.
35. M K V , p. 11: "Anirodham anutpadam anucchedam asasvatam
    anekantam ananartham anagamam anirgamam yah prafltyasamutpadam prapahcopasamam sivam desayamasa sambuddhas tarn vande vadatam varam."
    See also Stcherbatsky, Nirvana, pp. 9 1 -9 2 .
36. M K V , p. 11: "aryajnanapek§aya nasmin nirodho vidyate. yavan nasmin
    nirgam o vidyate. ity anirodhadibhir astabhir visesanair visisyate." See Stcherbatsky, Nirvana, p. 90.
37. M K V , p. 40: "Utpadad va tathagatanam anutpadad va tathagatanam
    sthitaivaisa dharmanam dharmata." In spite o f the occurrence o f this passage in the N ikayas and the Agamas, as well as in the Arya-salistam ba-sutra and several other texts, Stcherbatsky believes that "the second tathagatanam must be dropped
    and the first understood with Madhyavacarya, Sarvadars (anasamgraha) p. 21.8, in the sense o f tathagatanam m a te " (Nirvana, p. 123, n. 4). H aving thus misconstrued this im portant statement, in spite o f its acceptance by all the schools o f Buddhism, Stcherbatsky suggests a translation that is not at all consistent with the teachings o f early Buddhism. In the M adhyam ikavrtti, the passage was quoted by Candraklrti to illustrate the opponent's view that the early teachings did accept the fact o f
    'arising' (utpada) and 'passing away' (nirodha), which the M adhyamikas were trying to refute.
38. M K V , pp. 39-40. 50. M K V , pp. 4 0 -4 1 . See also A 1.60, TD 2.592c (Tseng 9.9), which need a
    different explanation.
39. Sn 884. 52. Murti, Central Philosophy, p. 40. 53. K P , pp. 8 2 -8 8 ; M K V , p. 358.
40. Murti, Central Philosophy, p. 27.
41. Ibid., pp. 49-50. 56. M K 2 A A 9 .
    CHAPTER V in 1. A K 2.61: "catvarah pratyaya ukta" ;
    M K 1.2: "catvarah pratyaya hetus calambanam anantaram, tathaivadhipateyan ca pratyayo nasti pancamafi."
42. M V B B , p. 31; see also MadhyantavibhagatTka, pp. 8 4 -8 9 . Bbh, pp. 97-9 8 .

In the Ch'eng wei shih lun; TD 31.41b (Ch'eng 8); Siddhi, pp. 453ff.

3. Kosa 2.299ff. and Siddhi, pp. 453ff., respectively. See also L. de la Vallée
   Poussin, Théorie des Douze Causes (Ghent: University o f Ghent, 1913), pp. 51-67.
4. Nyanatiloka, A Guide through the Abhidhamma Pifaka (Colom bo: The A ssociated Newspapers o f Ceylon, 1957), pp. 118-127.
5. D 3.214; TD 1.50a (Ch'ang 8.2). 6. Tikap 1.27. 7. Vism, p. 235. 8. Abvn, p. 137. 9. Tikap 1.27; D 3.214; TD 1.50a (CKang 8.2).
6. TD 31.40a (C tieng 7); Siddhi, p. 436; Madhyantavibhaga-tTka, pp. 3 2 -3 3 :
   translation by Freedman, p. 42.
7. MadhyantavibhagatTka, p. 33. 12. Abhs9 p. 28.
8. Sakv, p. 218: "dvidha hi pratyayah, janakás cájanakás ca. alambanapratyayas cajanakah. álambanamátratvád.''
9. T. I. Stcherbatsky, The Central Conception, p. 59, n. 1.
10. Vism, p. 533; see also A bvn, p. 138.
11. Vijnaptimatratasiddhi, p. 22.
12. T. I. Stcherbatsky, Buddhist Logic, 2 vols. (Leningrad: A cadem y o f
    Sciences o f the U SSR , 1930), 1 :139; MadhantavibhagatJka, Freedm an translation, p. 109.
13. Cf. Kosa 2.299ÍT.; Siddhi, pp. 227ff., 436ff.
14. Tikap 1.13.
15. Ibid. 21. See K osa 2.299, n.
16. TD 31.41a (C Keng 7); Siddhi, p. 448.
17. A K 2.62d: "karapahetuh adhipatipratyaya ucyate." 24. TD 31.41a (CKeng 7); Sida&i, pp. 4 4 8-449. 25. Tikap 1.3; A K 2 .6 2 a -b ; Kosa 2.233f., 236, 245; Abhs, p. 29; TD 31.671c
    (CL 3.1); 7 1 4 a -c (TCL 4); 40c (CKeng 7); Siddhi, pp. 437f.
18. A bhs, p. 28; TD 31.671c (CL 3.1); 41b (C/zV/ig 8); Siddhi, p. 456.
19. Tikap 1.14: "uppajjamano saha-uppajjamánabhávena upakárako paccayo sahajátapaccayo" ; p. 138: "attano uppattiya . . . sahuppannanam sahuppadabhá vena paccay o . ' '
20. Sakv, p. 188. 29. Kosa 2.245. 30. Russell, Our Knowledge o f the External W orld, p. 219.
21. iSVzfcv, p. 188: "anyonyánuparivartanaikakrtyárthena hy e§a vyavasthápyate."
22. A bhs, p. 28; T D 31.671b (CL 3.1). 33. Ibid., p. 29: "ye dharmáh sahabhavenotpadyante nányatamavaikalyena tad yatha bhütáni bhautikañ ca."
23. Tait 3.7.1: "Práne sariraip pratisthitam, sanre pranah prati§t:hitah" ;
    see also Kaus 3 .7 -8 ; 4.20.
24. S'2.114. 36. A bvn, p. 138.
25. Abhs, p. 28; TD 31.671b (CL 3.1). Abhs, p. 29: "ye dharmah sahabhâvenâlambanam sampratipadyante tad yathâ cittam caitasikâs ca."
26. TD 31.713c (T C L 4). 39. TD 31.135b (ST C L 1);E. Lamotte, L a Som m e du Grand Véhicule cTAsanga
    (Louvain: Bareaux du M uséon, 1932), p. 46.
27. Bunge, Causality, p. 91.
28. Vism, p. 535.
29. A bhs, p. 28; TD 31.671b (CL 3.1): "dhrtikâranam tad yathâ prthivT
    sattvânâm."
30. 72) 31.713b (CL 4).
31. Vism, p. 536. 45. N yanatiloka, Guide through the Abhidhamma P itaka, p. 102. 46. Patthana, 1.15. 47. See D . J. Kalupahana, "The Philosophy o f Relations in Buddhism ,"
    U CR 20 (1962): 38fif.
32. Safcv, p. 189: "mithyadrsteb purusapudgalasya yac ca kàyakanna tad dis ter yac ca vâkkarma yâ cetanà yah pranidhih ye ca samskârâs tad anvayâh. sarve py ete dhannâ anistavyâya samvartante'kantavyâyâpriyatvâyâmanâpatvaya."
33. TD 31.713c (T C L 4).
34. Tikap 1.17: "Pathamataram uppajjitvâ vattamânabhâvena upakârako
    dham m o."
35. A bhs, p. 28; TD 31.671c (CL 3.1); Bbh> p. 97; TD 31.41c (CKeng 8);
    Siddhi, p. 456.
36. Abhs, p. 28; TD 31.671c (CL 3.1).
37. Bbh, p. 97.
38. Tikap 1.17: "Purejâtânam . . . dham m ânam upatthambhaka^thena upakârako dham m o."
39. Ibid.
40. Abhs, p. 28; TD 31.671b (CL 3.1); M V B B , p. 31. 57. Patthana, 1.17. 58. A bhs, p. 29. Reference to this characteristic is m issing in the list given in Abhs, p. 28. TD 31.671b (CL 3.1).
41. Ibid., p. 29 : "pürvabhâvitânâm kusalâkusalâvyâkrtânâm dharmânâip yâ aparânte uttarottarâ pustatarâ pustamâ pravrttih."
42. T D 31.713c (T C L 5.1). 61. S akv, p. 29: "jananân nisrayât sthânâd upasthambhopabnnhaijât; also upabnnhanahetur vrddhihetutvât."
43. Tikap 1.45: "kamman ti cetanâ kam m an c'eva" ; Abvn, p. 139.
44. Ibid. 1.18. 64. Ibid. 1.45-46.
45. Sakv, p. 189. 66. TD 31.713c {TC L 5.1). 67. Abhs, p. 29; TD 31.671c (CL 3.1). 68. Abhs, p. 28; TD 31.671c (CL 3.1); Bbh, p. 97; TD 31.41c (CKeng 8);
    Siddhi, p. 456.
46. Tikap 1.46: "avasesapaccayasamâyoge sati phalam uppâdeti."

        70. Abvn, p. 139: "sati hi pi janakabhave upattham bhakattam eva aharassa
padhânakiccam ."
        71. Abhs, p. 28; TD 31.671b (CL 3.1); M V B B , p. 31.
        72. Vbh, p. 5.
        73. Tikap 1.18.
74. Abhs, p. 28 (TD 31.671b [CL 3.1])} which m ay be read as sam âpattikârana; M V B B, p. 31 reads prâpti-kârana; see TD 31.454b (C P F PL 2).
        75. p. 28: "màrgo nirvànasya."
        76. 5 1.25, 38; 4.37; M 1.38; TD 1.575c (Chung 23.5).
        77. Tikap IA 9.
        78. iSûfcv, 189: "ekâlambanakrtyena hy e$a vyavasthâpyate."
        79. 77) 31.713c (T C L 4).
        80. TV/cap 1.20; Vism, p. 540.
        81. p. 28; TD 31.671c (CL 3.1).
        82. 77) 31.41c (CKeng 8); Siddhi, p. 456.
        83. ¿M y, p. 28; TD 31.671c (CL 3.1).
        84. Tikap 1.18; also Abvn, p. 139.
        85. Tikap 1.51.
        86. Ibid. 1.21.
        87. Ibid.
        88. D e la Vallée Poussin has reconstructed the term as janana-hetu; see
Siddhi, p. 453, note c.
        89. TD 31.713b {TCL 4).
        90. Siddhi, p. 453, note c, gives vyahjana-hetu.
        91. Ibid. : dhvamsana-hetu.
        92. TD 31.713b {T C L 4).
        93. Siddhi, p. 453, note c: chedana-hetu.
        94. Ibid. : vikara-hetu.
        95. Ibid. : jhapana-hetu.
        96. Ibid. : avagama-hetu.
        97. Ibid. : anuvyavahara-hetu.
        98. Lb4 1.101.
        99. Fw m ,p. 235.

CHAPTER IX

1. M 1.426; TD 1.804a-b (CW zg 60.6).
2. Jayatilleke, Theory o f Knowledge, pp. 475f. 3. Ibid. 4. Sn 1076: "Atthangatassa na pamânam atthi yena nam vajju tam tassa
   natthi."
3. One o f the latest expositions o f the Buddhist conception o f nirvana based entirely on the Pali Nikâyas is by Rune E. A. Johansson, The Psychology o f Nirvana (London: George Allen & Unwin, 1969).
4. M 1.184; TD 1.658a {Chung 36.2).
5. See Johansson, Psychology o f Nirvana, for discussion o f all the aspects o f nirvana attained in this life.

Notes 227 8. 2)2.156.

9. Vin 1.235; 3 2 ; A 4.174, 182f.
10. M 1.167. 11. Translation by Chalmers {Buddha's Teachings, 2 2 0 5 37.187). 12. S>z 862ff. 13. M 1.256fr.; TZ) 1.766bff. {Chung 54.2). 14. M 1.190-191 : "Y o paticcasamuppadam passati so dham m am passati" ;
    TD 1.467a {Chung 7.2).

Bibliography

Original Sources

PALI TEXTS AND TRANSLATIONS
Abhidhammatthasahgaha, ed. T. W. Rhys Davids, JP T S , 1884; tr. S. Z. Aung and
C. A. F. Rhys Davids, Compendium o f Philosophy, London: PTS, 1910.
Abhidhammatthavibhâvim-tîkâ, ed. D . Pannasara and P. W imaladhamma, Vidyodaya Tika Publication, vol. 1, Colom bo: M ahabodhi Press, 1933.

Ahguttara N ikâya, ed. R. Morris and E. Hardy, 5 vols., L ondon: PTS, 1885-1900;
tr. F. L. W oodw ard and E. M . Hare, The B ook o f the Gradual Sayings, 5 vols., London: PTS, 1932-1936.

Atthasalim, ed. E. Müller, London: PTS, 1897.
Dhammapada, ed. S. Sumangala, London: PTS, 1914.
DTgha N ikâya, ed. T. W . Rhys Davids and J. E. Carpenter, 3 vols., London: PTS,
1890-1911; tr. T. W . and C. A. F. Rhys D avids, Dialogues o f the Buddha,
3 vols., L ondon: PTS, 1899-1921.

Itivuttaka, ed. E. Windish, London: PTS, 1889. Jataka, ed. V. Fausboll, 6 vols., and index, London: PTS, 1895-1907.
Kathâvatthu, ed. A. C. Taylor, 2 vols., London: PTS, 1894-1897.
Kathâvatthu-atthakathâ, ed. J. Minayeff, JP TS, 1889.
M ajjhima N ikâya, ed. V. Trenckner and R. Chalmers, 3 vols., L ondon: PTS,
1887-1901; tr. I. B. Horner, M iddle Length Sayings, 3 vols., London: PTS,
1954-1959; tr. R. Chalmers, Further D ia lo g u eso f the Buddha, 2 vols., London:
PTS, 1888.

M anorathapüram, ed. M . W alleser an d H . K opp, 5 vols., L ondon: PTS, 1924-1926. Milinda-panha, ed. V. Trenckner, London: PTS, 1928; tr. I. B. H om er, MilindcCs Questions, 2 vols., London: PTS, 1963-1964.
Nettippakarana, ed. E. Hardy, London: PTS, 1902; tr. Bhikkhu N ânam oli, Guide,
London: PTS, 1962.

Papahcasûdanl\ ed. J. H . W oods, D . K osam bi, and I. B. H om er, 5 vols., London:
PTS, 1922-1938.

ParamatthadTpam, commentary to Udâna, ed. F. L. W oodward, L ondon: PTS, 1926.
Patisam bhidàm agga, ed. A. C. Taylor, 2 vols., London: PTS, 1905-1907.
Sam antapâsâdikâ, ed. J. Takakusu and M . Nagai, 7 vols., L ondon : PTS, 1924-1947.
SammohavinodanT, ed. A. P. Buddhadatta, London: PTS, 1923.
Sam yutta N ik â ya , ed. L. Feer, 6 vols., London: PTS, 1 884-1904; tr. C. A. F. Rhys
     D avids and F. L. W oodward, The Book o f the Kindred Sayings, 5 vols.,
     London: PTS, 1917-1930.
Sâratthappakâsim , ed. F. L. W oodw ard, 3 vols., London: PTS, 1929-1937.
Sumahgalavilâsinl\ ed. T. W . Rhys Davids, J. E. Carpenter, and W . Stede, 3 vols.,
     London: PTS, 1886-1932.
Sutta-nipata, ed. D . Anderson and H. Smith, London: PTS, 1913; ed. and tr. R.
     Chalmers, Buddha's Teachings, Harvard Oriental Series, vol. 37, Cambridge,
     M ass.: Harvard University Press, 1932.
Sutta-nipâta-atthakathâ, ed. H. Smith, 2 vols., London: PTS, 1916-1917.
Tikapatthcma and Commentary, ed. C. A. F. Rhys Davids, 3 vols., L ondon: PTS,
     1921-1923.
Thera-then-gâthâ, ed. H . Oldenberg and R. Pischel, L ondon: PTS, 1883; tr. C. A. F.
     Rhys Davids, Psalm s o f the Early Buddhists, 2 vols., L ondon : PTS, 1903-1913.
Udâna, ed. P. Steinthal, London: PTS, 1948.
Vibhahga, ed. C. A. F. Rhys Davids, London: PTS, 1904.
Vinaya P itaka, ed. H. Oldenberg, 5 vols., London: PTS, 1879-1883; tr. T. W. Rhys
     D avids and H. Oldenberg, SBE, vols. 13, 17, 20, London: Oxford University
     Press, 1881-1885.
Visuddhimagga, ed. C. A. F. Rhys D avids, 2 vols., L ondon: PTS, 1920-1921.

SANSKRIT AND PRAKRIT TEXTS AND TRANSLATIONS
AbhidharmadTpa with Vibhâsâprabhâvrtti, ed. P. S. Jaini, Patna: K . P. Jayaswal
Research Institute, 1959.

Abhidharm akosa-bhâsya, ed. Pralhad Pradhan, Patna: K . P. Jayaswal Research
Institute, 1967.

Abhidharmasamuccaya, ed. Pralhad Pradhan, Santiniketan : Visva Bharati, 1950. A itareya Aranyaka, ed. A. B. Keith, Anecdota Oxoniensa 9, London: Oxford
University Press, 1909; Commentary by Sâyana Âcârya, ed. Rajendralala M itra, Bibliotheca Indica, Calcutta: Asiatic Society o f Bengal, 1876.
A lam banaparïksâ, restored from the Chinese and Tibetan by A . N . Sastri, with an English translation, Adyar Library Series 32, Adyar: Adyar Library, 1942; Tibetan text and German translation by E. Frauwallner, Wiener Zeitschrift fu r die Kunde des M orgenlandes, vol. 37, Vienna: Holder, 1930; Tibetan and Chinese texts with French translation, "Examen de l'objet de la connaissance . . by S. Yam aguchi in collaboration w ith Henriette Meyer, Journal Asiatique, Paris: Société Asiatique, 1929, pp. 1 -6 5 .
A nvayârthaprakâsikâ, commentary on Sartiksepasârïrikâ, ed. Bhau Sastri Vajhe,
K asi Sanskrit Series 2, Benares: Chawkhamba Sanskrit Series Office, 1913.
Arya-pratïtyasam utpâda-sUtra. See rya-sâlistam ba-sütra.

Ärya-sälistamba-sütra, restored from quotations and Tibetan and Chinese translations by L. de la Vallée Poussin, in Théorie des Douze Causes, Ghent: University o f Ghent Press, 1913; by N . A. Sastri, A dyar L ibrary Series 76, Adyar : Adyar Library, 1950, together with restorations o f Pratïtyasam utpâdavibhcmganirdesa and Ârya-pratïtyasam utpâda-sütra.

Astsâhasrikâ-prajhâpâramitâ, ed. Rajendralala Mitra, Calcutta: Asiatic Society o f Bengal, 1888.

A tharvaveda, ed. R. R oth and W. D . W hitney, Berlin: F. Dum m ler, 1885.
The BhagavadgTta, ed. and tr. S. Radhakrishnan, London: G eorge A llen &
Unwin, 1948.

Bodhicaryâvatâra, Sanskrit and Tibetan texts, ed. V. Bhattacharya, Bibliotheca Indica 280, Calcutta: Asiatic Society o f Bengal, 1960; with panjikä ed. L. de la Vallée Poussin, Bibliotheca Indica Calcutta: Asiatic Society o f Bengal, 1902-1914.

BodhisattvabhUmi, ed. U . W ogihara, T okyo: The Seigo K enkyu-kai, 1930-1936. B rahm aputra, ed. with translation and notes by S. Radhakrishnan, L ondon:
George Allen & Unwin, 1960.

Brhadâranyaka Upanisad. See Principal Upanisads.

Buddhacarita, ed. and tr. E. H. Johnston, Punjab University Oriental Publication, Calcutta: Baptist M ission Press, 1935.
Chandogya Upanisad. See Principal Upanisads. Sahkara-bhâçya, ed. Kasinatha Sastri Agase, Anandasrama Sanskrit Series 14, Poona: Anandasrama Press,
1913.

Dasabhïanika-sûtra, ed. J. Rahder, Paris: Paul Geuthner, 1926. D ivyävadäna, ed. C. B. Cowell and R. A. N eil, Cambridge: Cambridge University
Press, 1886.

Gandavyüha, ed. D . T. Suzuki and H. Idzumi, K yoto; The Sanskrit Buddhist Text Publishing Society, 1934-1936.

GândhârT Dharm apada, ed. J. Brough, London: Oxford University Press, 1962.
Isa Upanisad. See Principal Upanisads. Jaimim ya Brahmana, ed. H. Oertel, Boston, Mass. : American Oriental Society, 1894.
K äsyapaparivarta o f the R atnaküta-sütra, ed. A. Stael-Holstein, Shanghai: Com m ercial Press, 1926.
Katha Upanisad. See Principal Upanisads.
KausTtakT Upanisad. See Principal Upanisads. Kena Upanisad. See Principal Upanisads. Lalitavistara, ed. S. Leffinann, H alle: Buchhandlung des W eisenhauses, 1902-1908.
Lahkâvatâra-sütra, ed. B. Nanjio, Bibliotheca Otaniensis 1, K yoto: The Otani University Press, 1923; tr. D . T. Suzuki, London: R outledge and Kegan Paul, 1932.

M ädhyam ikavrtti (M adhyam akakarikäs), ed. L. de la Vallée Poussin, Bibliotheca Buddhica 4, St. Petersburg: The Imperial Academy o f Sciences, 1903-1913; tr. chaps. 1, 25, T. I. Stcherbatsky, The Conception o f Buddhist Nirvana,
Leningrad: Academ y o f Sciences o f the USSR, 1927; chaps. 2 -4 , 6 -9 , 11, 2 3 -2 4 , 2 6 -2 7 , Jacques M ay, Candrakïrti, Prasannapadà M âdhyam akavrtti, Paris: Adrien M aisonneuve, 1959; chap. 10, Stanislaus Schayer, R ocznik Orjentalistyczny, vol. 7, Warsaw: Lvov, 1930; chaps. 5, 12 -1 6 , idem, Ausgewählte K apitel aus der Prasannapadä, Cracow: N akadem Polskiej Akademji

      Um iejetnosci, 1931 ; chap. 17, E. Lam otte, Mélanges Chinois et Bouddhiques,
      vol. 4, Brussels: Imprimerie Sainte Catherine, 1936; chaps. 1 8-22, J. W . de
      Jong, Cing Chapitres de la Prasannapadä, Paris: Librairie Orientaliste Paul
      Geuthner, 1949.
M adhyântavibhâga-bhâsya, ed. G. M. N agao, T okyo : Suzuki Research Foundation,
       1964.
M adhyântavibhâga-tïkâ, ed. S. Yam aguchi, N agoya: Haijinkaku, 1934; tr. T. I.
      Stcherbatsky, Bibliotheca Buddhica 30, Leningrad: Academy o f Sciences o f
      the U SSR , 1936; chap. 1, D . L. Freedman, Exam ination o f the M iddle Path
      and the E xtrem es, Amsterdam, 1937.
M ahäbhärata, Bhandarkar Oriental Research Institute, Poona, 1933.
M undaka Upanisad. See Principal Upanisads.
Nikâyâlam bana-sâstra o f Vasumitra, tr. J. M asuda, "Origin and Doctrines o f Early
      Indian Buddhist Schools," A M 2(1925): 1-75.
N yâyakusum âhjalî o f Udayana, with prakasa o f Vardhamana and bodham o f
      Varadarâja, ed. Padmaprasâda Upadhyaya, 2nd ed., part 1, Benares: Chawkhamba Sanskrit Series Office, 1950.
Nyäyam ahjarTo f Jayanta B hatta, ed S. S. N . Sukla, K asi Sanskrit Series, Benares:
      Chawkhamba Sanskrit Series Office, 1936.
Pratïtyasamutpüda-sUtra, ed. N . P. Chakravarti, Epigraphia Indica 21, Delhi:
      Government o f India, 1931, pp. 179-199; see also Sarnyukta Ägam a
      (fragment) and Ärya-sälistamba-sütra.
PratTtyasamutpäda-vyäkhyä o f Vasubandhu (fragment), ed. G. Tucci, JR A S, N ew
      Series, 1930, pp. 61 Iff.
RatnâvalTof Nagârjuna, ed. and tr. G. Tucci, J R A S , N ew Series: 1934, pp. 3 0 7 -325;
       1936, pp. 237-252.
Rgveda with Säyana's commentary, ed. M ax M üller, 6 vols., L ondon: W . H. Allen,
       1849-1874; tr. R. T. H. Griffith, H ym ns o f the Rigveda, 2 vols., Benares;
      E. J. Lazarus & Co., 1889; Vedic R eader fo r Students, A . A. M acD onnell,
      London: Oxford University Press, 1917.
Saddarsanasamuccaya o f Haribhadra, with G unaratna's TarkarahasyadTpikä, ed.
      L. Suali, Bibliotheca Indica, Calcutta: A siatic Society o f Bengal, 1905.
SaddharmapundarJka-sütra, ed. H. Kern and B. Nanjio, St. Petersburg: The Imperial
       Academ y o f Sciences, 1912. Rom anized text ed. U . W ogihara and C. Tsuchida,
       Tokyo: The Seigo Kenkyu-kai, 1934-1936.
Sam ksepasarvrika, with the commentary Tattvabodhim\ ed. Surya Narayana Sukla,
       Sarasvati Bhavana Texts 69, Benares: Government Sanskrit Library, 1937.
Sam yukta Ä gam a (fragment), ed. Chandrabhäl Tripäthl, Fünfundzwanzig sütras des
      Nidäna Sam yukta, Berlin : Deutsche Akadem ie der W issenschaften zu Berlin,
       1962.
Sarvadarsanasamgraha, ed. V. S. Abhyankar, P oona : Bhandarkar Oriental Research
       Institute, 1924.
Satapatha Brähmana, ed. V. S. Gauda, C. Sharma, and S. V. Sastri, 2 vols., Benares:
      Acyuta Granthamäla Käryälaya, 1922-1937.

Siksâsam uccaya, ed. C. Bendall, St. Petersburg: The Imperial A cadem y o f Sciences,
      1879-1902.
Sphutârthâbhidharmakosa-vyâkhyâ, ed. U . W ogihara, T okyo: The Publication
      A ssociation o f Abhidharmakosavyâkhyâ, 1932-1936.
Sütrakrtâhga (Süyagadanga) with commentary o f Silànka, ed. A. S. Suri and
      C. Ganindra, 2 vols., Bhavanagara, Bombay: Vijayadeva Sura, 1950-1953;
      tr. H. Jacobi, Jaina Süîras, SB E 45, Oxford: The Clarendon Press, 1895.
Suvarnaprabhâsottama-sütra, ed. Johannes N obel, Leipzig: Otto Harrassovitz, 1937.
Süyagadanga. See Sütrakrtâhga.
Svetâsvatara Upaniçad. See Principal Upaniçads. Sankara's bhâsya (commentary),
      ed. E. Rôer, Bibliotheca Indica 7, Calcutta: Asiatic Society o f Bengal, 1850.
Taittirxya Brâhmana (o f the Black Yajurveda), with the commentary o f Sâyana
      Àcârya, ed. Rajendralala Mitra, 3 vols. Bibliotheca Indica, Calcutta: A siatic
      Society o f Bengal, 1859.
Tandy a M ahâ Brâhmana, with the commentary o f Sâyana Àcârya, ed. A nandacandra
      Vedântavâgfsa, 2 vols., Bibliotheca Indica, Calcutta: A siatic Society o f Bengal,
      1870-1874.
Tattvabodhim. See Sam ksepasâfïrikâ.
Tattvasamgraha o f Sântaraksita, with pahjikâ o f Kamalasïla, ed. E. Krishnamacharya, 2 vols., Gaekwad Oriental Series 30, 31, Baroda: Oriental Institute,
      1926.
Tattvopaplavasim ha o f Jayarâsi Bhatta, ed. S. Sanghavi and R. C. Parikh, Gaekwad
      Oriental Series 87, Baroda: Oriental Institute, 1940.
Upanisads, ed. andtr. S. Radhakrishnan, The Principal Upanisads, London: George
      Allen & Unwin, 1953; tr. R. E. Hume, The Thirteen Principal Upanisads, 2nd
      ed. London: Oxford University Press, 1934.
Vajracchedikâ-prajhâpâramitâ, ed. and tr. E. Conze, Serie Orientale Rom a 13,
      Rom e: Instituto italiano per il M edio ed Estremo Oriente, 1957.
Vijhaptimâtratâsiddhi, Vimsatikâ et Trimsikâ, ed. S. Lévi, Paris: Cham pion, 1925;
      tr. into French by S. Lévi, M atériaux pour IÉ tude du Systèm e Vijhaptim àtra,
      Paris: Paul Geuthner, 1932.
Vimalakïrtinidesa, tr. E. Lamotte, L'Enseignement de Vim alakïrti, Bibliothèque
      du M uséon, vol. 51, Louvain: M uséon, 1962.

CHINESE TRANSLATIONS
A-pi-ta-m o chü-shê lun (Abhidharmakosa), tr. H süan Tsang, Taisho N o. 1558;
French translation by L. de la Vallée Poussin, U Abhidharmakosa de Vasubandhu, 6 vols., Paris: Paul Geuthner, 1923-1925.
A -pi-ta-m o chü-shê shih lun (.Abhidharmakosa), tr. Paramârtha, Taisho N o. 1559.
Ch'ang a-han ching (Dïrgâgama), tr. Buddhayasas, Taisho N o. 1. Ch'êng wei shih lun ( Vijhaptimâtratâsiddhi), D harm apâla's version, tr. H süan Tsang,
Taisho N o. 1585 ; French translation by L. de la Vallée Poussin, Vijhaptim âtratâsiddhi, La Siddhi de Hiuan-Tsang, Buddhica, vols. 1, 5, Paris: Paul Geuthner, 1928-1929.

Chung a-han ching (.M adhyamâgama), tr. Gautama Sanghadeva and Sangharaksa,
Taisho N o. 26.

Chung lun (M âdhyam aka-sâstra), tr. KumarajTva, Taishd N o. 1564.
Chungpien fên pieh lun (.M adhyântavibhâga-sâstra), tr. Paramârtha, Taishd'No. 1599.
Fu shuo chiu ch'êng yü ching (.Nagaropama-sütra), tr. Fa-hsien, Taishd N o. 715.
Fu shuo ta ch'êng tao ch'ien ching (Â rya-sâlistam ba-sütra), tr. unknown, Taishd N o . 712.

Fu shuo tao ch'ien ching (Ârya-sâlistamba-sütra), tr. unknow n, Taishd N o. 709. Ju lêng-chia ching (.Lahkâvatâra-sütra), tr. Bodhiruci, Taishd N o. 671. Liao pên sheng szü ching (Ârya-sâlistamba-sütra), tr. Chih Chien, Taishd N o. 708.
P ei to shu hsia szü w ei shih êrh yin yuan ching (Nagaropama-sütra), tr. Chih Chien,
Taishd N o. 713.
Pi-p'o-shih fu ching (VipassT-sutta), tr. Fa-hsien, Taishd N o . 3.
Pieh i tsa a-han ching (Sarnyuktâgama), tr. unknown, Taishd N o. 100.
Pien chung pien lun (M adhyântavibhâga-sâstra), tr. Hsüan Tsang, Taishd N o. 1600. P'u-sa ti ch'ih ching (Bodhisatîvabhümi), tr. Dharmaksema, Taishd N o. 1581. Shê ta ch'êng lun (M ahâyânasamgraha), tr. Hsüan Tsang, Taishd N o. 1594; French
translation by E. Lamotte, La Somme du Grand Véhicule d'Asahga, Louvain:
Bureaux du M uséon, 1932.

Ta ch'êng a-pi-ta-m o chi lun (Mahâyânâbhidharmasamuccaya), tr. Hsüan Tsang,
Taishd N o. 1605.
Ta cKêng a-pi-ta-m o tsa chi lun (Mahâyânâbhidharmasamuccaya-bhâsya), tr. Hsüan
Tsang, Taishd N o . 1606.
Ta ch'êng she-li-so-tan-mo ching (Ârya-sâlistamba-sütra), tr. Dânapâla, Taishd N o . 711.

Taishd Shinhsu D aizdkyd, ed. J. Takakusu and K . W atanabe, T okyo : D aizô Shuppan
Com pany, 1924-1934.

Tsa a-han ching (Sam yuktâgama), tr. Gunabhadra, Taishd N o. 99.
Tseng i a-han ching (Ekottarâgama), tr. Gautama Sanghadeva, Taishd N o. 125.
Yuan c ïïi ching (Pratïtyasam utpâda-sütra), tr. H süan Tsang, Taishd N o. 124. Yuan ch'i sheng tao ching (Nagaropama-sütra), tr. Hsüan Tsang, Taishd N o. 714.
Secondary Sources REFERENCE WORKS
Buddhist H ybrid Sanskrit Grammar and Dictionary. Ed. F. Edgerton. 2 vols. N ew H aven and London: Yale University Press, 1953.

Bibliography on Buddhism. Ed. Shinshu Hanayama. T okyo: The Hokuseido Press,
1961.

A Chines e-English Dictionary. Ed. H. A. Gües. 3 vols., L ondon: B. Quaritch, 1912. A Chinese-English Dictionary. Ed. R. H. M athews. American ed., Cambridge,
M ass. : H arvard University Press, 1960.

Concise O xford Dictionary. Ed. H. W. Fowler and F. G. Fowler. 4th ed. by E. M acintosh. Oxford: The Clarendon Press, 1959.

Concordance to the Principal Upanisads. Ed. G. A. Jacob. D elhi : M otilal Banarsidass, 1963 reprint.

C ritical P ali D ictionary. Ed. Y. Trenckner, D . Anderson, H. Smith, and H. Hendriksen. vol. 1. Copenhagen : The Royal D anish Academ y o f Science and Letters, 1924-1948.

A Com parative Catalogue o f Chinese Agam as and P ali N ikayas. Ed. Akanum a

      Chizen. N agoya: Hajinkaku, 1929.
A D ictionary o f Chinese Buddhist Terms. Ed. W. E. Soothill and L. H odous. L o n d on :
      Kegan Paul, Trench, Triibner & Co., 1937.
Dictionary o f Pali Proper Names. Ed. G. P. Malalasekera. Indian Text Series.
      London: Murray, 1937-1938.
Dictionary o f Philosophy and Psychology. Ed. J. H . Baldwin. 4 vols. N ew Y ork:
      Macmillan Co., 1940-1949.
Encyclopaedia o f Buddhism. Ed. G. P. Malalasekera. Colom bo: Governm ent o f
      Ceylon, 1961.
Encyclopaedia o f Religion and Ethics. Ed. J. Hastings. 13 vols. N ew York: Charles
      Scribner's Sons, 1908-1926.
Grammata Serica Recensa. Ed. B. Karlgren. Stockholm : M useum o f Far Eastern
      Antiquities, 1959.
Index to the Lahkavatara Sutra. Ed. D . T. Suzuki. K yoto: The Sanskrit Buddhist
      Texts Publishing Society, 1934.
Index to the Mahaydnasutralahkara. Ed. G. M. N agao. 2 pts. T okyo: N ippon
      Gakujutsu Sinko-kai, 1958.
M ahavyutpatti. Ed. I. P. Minaev. Bibliotheca Buddhica 13. St. Petersburg: The
      Imperial Academ y o f Sciences, 1911.
Pali-English D ictionary. Ed. T. W. Rhys Davids and W . Stede. London: PTS, 1959.
Pali Tipitaka Concordance. Ed. A. K. Warder. London: PTS, 1956.
A Practical Sanskrit Dictionary. Ed. A. A. M acD onnell. L o n d o n : Oxford University
      Press, 1924.
A Sanskrit-Chinese D ictionary o f Buddhist Technical Terms B ased on the M ahavyutpatti. Ed. U . Wogihara. Tokyo: Sankibo, 1959.
A Sanskrit-English Dictionary. Ed. M. Monier-W illiams. Oxford: The Clarendon
      Press, 1899.
Tibetan-English D ictionary. Ed. H. A. Jeschke. London: K egan Paul, Trench,
      Triibner & Co., 1958.
A Tibetan-English D ictionary with Sanskrit Synonyms. Ed. Sarat Chandra Das.
      W est Bengal: W est Bengal Government Press, 1960.
Vedic Bibliography. 2 vols. ed. R. N . Dandekar. Bom bay: K am atak Publishing
      H ouse, 1946, 1961.
Vedic Concordance. Ed. M . Bloomfield. Cambridge, M ass.: Harvard University
      Press, 1906.

MONOGRAPHS AND ARTICLES

Anesaki, M . " On the Relation o f the Chinese Agam as to the Pali N ikayas." JR A S,
      N ew Series 1901, pp. 895-900.
- - - . "The Four Buddhist Agam as in Chinese." The Transactions o f the A siatic
      Society o f Japan 35(1908), pt. 3, pp. 1-149.
Ayer, A. J. Foundations o f Em pirical Knowledge. London: M acm illan & Co., 1963.
 . The Problem o f Knowledge. London: M acmillan & Co., 1956.
 . Language, Truth and Logic. London: Victor Gollantz, 1962.
 . The Concept o f a Person and Other Essays. L o n d o n : M acm illan & C o., 1963.
Bailey, H. W. " GandharT." B SO A S 9(1946): 764ff.

Bareau, A. Les Sectes Bouddhiques du P etit Véhicule. Saigon: École Française
      d'Extrême-Orient, 1955.
Barua, B. M . A H istory o f Pre-Buddhistic Indian Philosophy. Calcutta: University
      o f Calcutta Press, 1921.
 . "Pratïtyasam utpàda as a Basic Concept o f Buddhist T hought." B. C. Law
      Volume. Calcutta: Bhandarkar Oriental Research Institute, 1945.
Basham, A. L. H istory and Doctrines o f the ÂjTvikas. London: Luzac & Co., 1951.
Belvalkar, S. K. and Ranade, R. D . H istory o f Indian Philosophy. Vol. 2, The
      Creative Period. Poona: Bilvakunja Publishing House, 1927.
Bhattacharyya, G. M . Studies in N yâ ya -Vaisesika Theism. Calcutta: Sanskrit
      College, 1947.
Bosch, F. D . K. The Golden Germ, An Introduction to Indian Sym bolism . Indo-Iranian
      M onographs, no. 2. The Hague: M outon, 1960.
Bradley, F. H. Appearance and R eality. Oxford: The Clarendon Press, 1951.
Broad, C. D . The M in d and Its Place in Nature. London: Kegan Paul & Co., 1962.
 . Human Personality and the Possibility o f Survival. Berkeley and Los Angeles :
      University o f California Press, 1955.
Brown, A. J. C. Freud and the Post-Freudians. London: Pelican Books, 1963.
Bulcke, C. The Theism o f N yâya- Vais es ika, Its Origin and Early Development.
      Calcutta: Oriental Institute, 1947.
Bunge, M ario. Causality : The Place o f the Causal Principle in M odern Science.
      Cambridge, M ass: Harvard University Press, 1959.
Burnet, John. E arly Greek Philosophy. 4th ed. London: A. & C. Black, 1958.
Bu-ston. H istory o f Buddhism (Chos-Hbyung). Translated from the Tibetan by
      E. Obermiller. 2 vols. Heidelberg: Otto Harrassowitz, 1931-1932.
Chattopadhyaya, B. Lokâyata, A Study on Ancient Indian M aterialism . N ew Delhi:
      People's Publishing House, 1959.
Clark, W . E. "Som e Problems in the Criticism o f the Sources for Early Buddhist
      H istory." H arvard Theological Review 23(1930): 121-147.
Cohen, M . R. and N agel, E. An Introduction to L ogic and Scientific M ethod. L ondon :
      R outledge & Kegan Paul, 1949.
Copleston, F. C. Aquinas. Pelican Philosophy Series. London: Pelican Books, 1955.
Conze, Edward. Buddhist Thought in India. London: George Allen & Unwin, 1962.
D as Gupta, S. A H istory o f Indian Philosophy. Vol. 1. Cambridge: Cambridge
      University Press, 1963.
 . Indian Idealism. Cambridge: Cambridge University Press, 1933.

D e la Vallée Poussin, L. Théorie des Douze Causes. Ghent: University o f Ghent,
1913.

. "Philosophy" (Buddhist). E R E 9 : 848ff. . " Identity" (Buddhist). E R E l:9 9 f. Dem ieville, P. "Sur la Mémoire des Éxistences Antérieures." Bulletin de TÉcole Française d E x trê m e Orient (Hanoi) 27(1927): 283-298.
Deussen, Paul. The Philosophy o f the Upanisads. Tr. A. S. Geden. Edinburgh:
T. & T. Clark, 1906.

Dutt, N . A spects o f M ahay ana Buddhism and Its Relation to HTnayâna. Calcutta O riental Series 23. London: Luzac & Co., 1930.
. "Theravada School o f Buddhism ." JD L U C 8(1920): 130-140.

Edward, P. and Pap, A. A M odem Introduction to Philosophy. N ew York: Free
Press, 1963.

Ewing, A. C. The Fundamental Questions o f Philosophy. London: Routledge &
Kegan Paul, 1951.

Feigl, H. and Sellars, W. Readings in Philosophical Analysis. Century Philosophy
Series. N ew York: Appleton-Century-Crofts, 1949.

Fick, R. Social Organization in North-East India in Buddha's Time. Tr. S. Mitra.
Calcutta: U niversity o f Calcutta Press, 1920.

Franke, O. "Buddhist Councils at Räjagaha and V esali." J P T S , N ew Series, 1908; pp. 1-80.

Freud, Sigmund. Introductory Lectures in Psycho-Analysis. Tr. Joan Riviere.
London: George Allen & Unwin, 1949.

Friedmann, D . L. "The Creative Force in Buddhism." The Buddhist (Colom bo)
28 (1957).

Geiger, M agdalene and Wilhelm. Pali Dhamma, Vernehmlich in der Kanonischen Literatur. M unich: Bavarian Academy o f Science, 1920.
Glasenapp, H. von. The Doctrine o f Karman in Jain Philosophy. Tr. G. B. Gifford.

Bom bay: Trustees, Bai Vijibai Jivanlal Panalal Charity Fund, 1942.

Guenther, H. V. Philosophy and Psychology in the Abhidhamma. Lucknow: Buddha Vihara, 1962.

Gutherie, W. K . C. H istory o f Greek Philosophy. 2 vols. Cambridge: Cambridge University Press, 1962.

Hiriyanna, M . Indian Philosophical Studies. M ysore: K avyalaya, 1957. . Outlines o f Indian Philosophy. London: G eorge Allen & Unwin, 1956. H oem le, A. F. R. M anuscript Remains o f Buddhist Literature foun d in Eastern Turkestan. Vol. 1. Oxford: The Clarendon Press, 1916.

Höffding, H. A H istory o f M odern Philosophy. 2 vols. London: M acmillan & Co.,
1908.

H ook, Sidney (ed.). Determinism and Freedom in the A ge o f M odern Science. N ew
York: N ew Y ork University Press, 1958.

Hume, David. A Treatise o f Human Nature. Ed. L. A . Selby-Bigge. O xford: The Clarendon Press, 1888.

Husserl, E. Ideas: General Introduction to Pure Phenomenology. Tr. W. R. BoyceG ibson. London: George Allen St Unwin, 1931.
Iliade, M . Yoga, Im m ortality and Freedom. London: Routledge & Kegan Paul, 1958.

Jacobi, H. "Jainism." E R E 7 :465ff. James, William. Pragm atism . N ew York, Toronto, and L o n d o n : Longm ans Green &
C o., 1943.

Jayatilleke, K. N . E arly Buddhist Theory o f Knowledge. L o n d o n : George Allen &
Unwin, 1963.

. " Som e Problems o f Translation and Interpretation I." UCR 8(1949):
208-224.

Jennings, J. G. The Vedantic Buddhism o f the Buddha. London: Oxford University Press, 1947.

Joseph, H. B. W. An Introduction to Logic. Oxford: The Clarendon Press, 1925.

Kalupahana, D . J. "Prolegomena to the Philosophy o f Relations in Buddhism."
      UCR 19(1961): 169-194.
 . "Philosophy o f Relations in Buddhism I and II." U C R 20(1962): 19-54,
      188-208.
 . "Aspects o f the Buddhist Theory o f the External W orld and the Emergence
     o f the Philosophical Schools in Buddhism." C JH 1(1970): 9 3 -108.
 . "Schools o f Buddhism in Early C eylon." C JH 1(1970): 159-190.
Keith, A. B. Buddhist Philosophy in India and Ceylon. Oxford: The Clarendon Press,
      1925.
 . Religion and Philosophy o f the Vedas and the Upanisads. 2 vols. Harvard
     Oriental Series 31, 32. Cambridge, M ass.: Harvard University Press, 1925.
K em , H. M anual o f Buddhism. Strasbourg: K. J. Triibner, 1896.
Lamotte, E. H istoire du Bouddhisme Indien. Louvain: Publications Universitaires,
      1958.
 . L*Enseignement de Vimalakïrti. Bibliothèque du Muséon. V ol. 51. Louvain:
     M uséon, 1962.
Law, B. C. "Influence o f the Five Heretical Teachers on Jainism and Buddhism."
     J A S B , N ew Series, 15(1919): 134ff.
Lindquist, Sigurd. Siddhi und Abhihnà. Uppsala: U ppsala University, 1935.
Lüders, H. Varuna. 2 vols. Gottingen: Vanderhoeck & Ruprecht, 1951, 1956.
Malalasekera, G. P. and Jayatilleke, K. N . Buddhism and the R ace Question. Paris:
      U N E SC O , 1958.
M alinowski, B. K . "M agic, Science and Religion." Science, Religion and R eality.
      Ed. J. Needham . London: Sheldon Press, 1925.
M asson-Oursel, P. " Essai d'interprétation de Théorie des D ou ze C onditions."
      Revue de /'H istoire des Religions 71(1915): 30-46.
M asuda, J. " Origin and Doctrines o f Early Indian Buddhist Schools (N ikâyâlam -
     bana-sastra)." A M 2(1925): 1-75.
M cG overn, W. M. A Manual o f Buddhist Philosophy. London: Triibner, 1923.
M ill, J. S. A System o f Logic. 2 vols. London: Longmans Green & C o., 1872.
Mishra, Um esha. H istory o f Indian Philosophy. Vol. 1. Allahabad: Thirabhukti
      Publications, 1957.
M ookerjee, S. The Buddhist Philosophy o f Universal Flux. Calcutta: University o f
      Calcutta Press, 1935.
 . The Jaina Philosophy o f Non-Absolutism. A Critical Study o f Anekântavâda.
      Calcutta: Bharati M ahavidyalaya, 1944.
M unn, Norm an L. Psychology, The Fundamentals o f Human Adjustm ent. 4th ed.
      L ondon: George Harrap & Co., 1961.
M urti, T. R. V. The Central Philosophy o f Buddhism. London: George Allen &
      U nwin, 1960.
Nyanatiloka, Bhikkhu. A Guide Through the Abhidhamma Pitaka. C olom bo:
      A ssociated Newspapers o f Ceylon, 1957.
Oltramare, P. L a Formule Bouddhique des Douze C auses: Son Sens Original et son
      Interprétation Théologique. Geneva: Georg, 1909.
O sbom , Andrew D . The Philosophy o f Edmund Ifusserl. N ew York: International
      Press, 1934.

Peters, R. S. Concept o f M otivation. London: Routledge & K egan Paul, 1960. Radhakrishnan, S. H istory o f Indian Philosophy. 2 vols. London: George Allen &
Unwin, 1962.

Raju, P. T. "The Buddhistic Conception o f D harm a." J B O R S 21(1939-40):
192-202.

Randall, J. H. Aristotle. N ew York: Columbia University Press, 1960.

Renou, L. LTnde Classique. 2 vols. Paris: Imprimerie N ationale, 1947. Rhys Davids, C. A . F. Buddhism, A Study o f the Buddhist N orm . H om e University Library o f M odem Knowledge. London: W illiams & N orgate, 1912.

. Buddhist Psychology. London: George Bell & Sons, 1914. . SaJcya or Buddhist Origins. London: Kegan Paul, Trench, Triibner & Co.,
1931.

---------- . " Papccasam uppada." ER E 9:612R .
Riepi, D . M . The Naturalistic Tradition in Indian Thought. Seattle: W ashington
University Press, 1961.

Robertson, G. C. Elements o f Psychology. Ed. C. A. F. Rhys Davids. London:
J. Murray, 1896.

Robinson, Richard H. E arly M ddhyam ika in India and China. M adison: University o f W isconsin, 1967.

Ruben, Walter. D ie Philosophen der Upanisaden. Bern: A . Franke Ag. Verlag, 1947.

. "Uber den Tattvopaplavasimha des Jayarasi Bhatta, eine agnostizistiche Erkenntniskritik." W ZK S& O 1(1958): 141-153.
Russell, B. A. W . Analysis o f M ind. London: George A llen & Unw in, 1921.

. Human Knowledge, Its Scope and Lim its. L o n d o n : G eorge A llen & Unwin,
1948.

. H istory o f Western Philosophy. London: George A llen & Unwin, 1961. . M ysticism and Logic. London: George Allen & U nw in, 1963. . Our Knowledge o f the External World. L o n d o n : G eorge Allen & Unwin,
1926.

Saratchandra, E. R. Buddhist Psychology o f Perception. C olom bo: The Ceylon University Press, 1958.

Schayer, Stanislaus. "Pre-Canonical Buddhism." Archiv Orientalni (Prague) 7(1935):
121-132.

Stcherbatsky, T. I. Buddhist Logic. 2 vols. Leningrad: Academ y o f Sciences o f the
U SSR , 1930.

. The Central Conception o f Buddhism and M eaning o f the Term 'Dharma\
London: Royal Asiatic Society, 1923.

. The Conception o f Buddhist Nirvana. Leningrad: Academ y o f Sciences o f the U SSR , 1927.

. The Soul Theory o f the Buddhists. Leningrad: A cadem y o f Sciences o f the U SSR , 1919.

Stebbing, L. S. A M odem Introduction to Logic. London: M ethuen & Co., 1962. Takakusu, J. The Essentials o f Buddhist Philosophy. Ed. W ing-tsit Chan and Charles M oore. 3rd ed. H onolulu: University o f Hawaii Press, 1956.

Tatia, N . "Paticcasamuppada." Nava Nalanda M ahavihara Research Publication 1(1957): 177-239.

Taylor, A . E. Elem ents o f M etaphysics. London: M ethuen & Co., 1961.
Tennent, F. R. " Cause and Causality." E R E 3:261.
Thom as, E. J. H istory o f Buddhist Thought. L o n d o n : K egan Paul, Trench, Triibner &
      Co., 1959.
 . Life o f the Buddha. London: Kegan Paul, Trench, Triibner & Co., 1960.
Tischner, R. Telepathy and Clairvoyance. Tr. W. D . H utchinson. International
      Library o f P sychology. London: K egan Paul & Co., 1925.
Tucci, G. "A Sketch o f Indian M aterialism." P IP C , 1925, pp. 34-43.
V an Buitenen, J. A. B. "Studies in Sänkhya (1)." J A O S 76(1956): 153-157.
W ijesekera, O. H. de A . "Vitalism and Becoming: A Comparative Study." UCR
       1(1943): 4 9 -5 8 .
 . "Vedic G andharva and Pali G andhabba." U CR 3(1945): 73-105.
W intem itz, M . H istory o f Indian Literature. 2 vols. Calcutta: University o f Calcutta,
       1923.
W isdom , J. O. Foundations o f Inference in N atural Science. L o n d o n : M ethuen & Co.,
       1952.
Y am akam i Sögen. S ystem s o f Buddhistic Thought. Calcutta: University o f Calcutta
      Press, 1912.
Zürcher, E. The Buddhist Conquest o f China. 2 vols. Leiden: E. J. Brill, 1959.

Index Of Chinese Terms

A-pi-ta-mo chu-she lun M i i , M & i£i A-pi-ta-mo chu-she shih lun PT
ch'a na (ksana) -£']
ch'a pieh (prabheda) M. # ]
Chang a-han ching fk H &
chang ai (paripantha) ^ ch'ang chien (sassata-ditthi) ^ chao neng tso (prakasakarana) £& Vfchao yin neng tso (aksepakarana) 51 f t ch'eng ( 'accom plishing')
cheng chien (samma-ditthi) JL cheng chih ch'u ju (satisampajanna) SL ¿j cheng szu wei (yoniso manasikara) iL S ' I f t Ch'eng wei shih lun ^ ^
Index o f Chinese Terms 241 ch'i (uppäda)
chiai t'o (vimutti) fl?* K t chieh (dhätu)
chieh (kappa) i jj ch'ien p 'ien (thitassa annathatta) Ü
ch'ih (moha) ^ ch'ih nêng tso (dhrtikarana) lih chih ťu n g (abhiňňá) 1ŽL chin (vartamäna) 4^ ch'iu lo (sukhakäm a) ^ ch'iu sheng ( jTvitukäma) ^ £ chu ( 'sustaining') \ í ch'u (phassa) AS
ch'ü (upadána)
ch'ü (atlta)
chu nêng tso (sthitikärana) M ttL
chu pan (sahâya) M
chü yu yin (sahabhü hetu) ffl* 0 chuan pien (parinata) ít* ^ chuang lun sheng wang (cakkavattirajja) ít* lè* -2L chuang pien nêng tso (pariijatikârana) ít* $£ *$ Chung a-han ching Ť f5! ÈSL
Chung lun Ť Ira
Chung pien fén pieh lun Ť i 4 ^ chung tao (majjhimä patipadä) ^ á t fa (dhamma) fa (danda) fa chieh (dhammadhätu) & f a fa chu chih (anvaye ñaña)
fa hsing (dhammadhätu?) '/k 'tí fa pu i ju (anaññathata) & ^ fa -ka fa pu li ju (avitathatä) ^ fa tsu jan (dhammatä) >žr Í¡ fan hsing i wei (vusitam brahmacariyam) ÍT fei ching chieh (avisaya) & f a fei tsu fei ť a tso (asayamkäram aparamkaram) # |]
fen li néng tso (viyogakärana) ^ Sf*-
feng (vâyu) Â
fu ch'an (pettikam däyajjam) X . á Fw s/zwo c/izm ch'êng yü *ching* i f M
Fu shuo ta ch'êng tao ch'ien ching - f Fm «s/zwö tao ch'ien ching -f- M
hêng yu (sarvadâ asti) 'ti- ^f ho hui (saňgati, sannipäta) o" hsiang (sañña)
hsiang hsü (sañgatibhava?) ^0 ^
Index o f Chinese Terms hsiang i (laksanânyathàtva) M- hsiang wei nêng tso (virodhikârana) îfc Ifé» Vf hsiang yin (gandhabba) Ê
hsiang ying yin (sabhâgahetu) Æ« 0 hsien liao nêng tso (sampratyâyanakârana) gg Ht Vf hsin (citta) ^
hsin chiai nêng tso (sampratyayakârana) i t fft- Vf hsing (saiikhâra) V\
hu ch'u kên (indriyasamvara) M. Â . huai (samvatla) & hui chiai t'o (pannâvimutti) i l Æ?- huo (tejo) 'K
i (mano)
i ch'ieh ch'ung sheng, yu ming chih lui ^ ^ i chieh fa wu o (sabbe dhammà anattâ) - i77 i chieh yu (sarvam asti) - i fa (manodanda) 15 i shu (vipâka) ife i shu yin (vipâkahetu) ife 0 jo ^
jo yu 3êr jo yu tzu tsê yu pi, jo wu tzù tsê wu pi
(imasmim sati idam hoti, imasmim asati idam na hoti) y& ^ J^t* M Î& yêr H'] $L
244 ju chü (ayatana) ^ M.

ju fa érh (tathatä) ju i tsu (iddhividha) ^ Ä .

Ju léng-chia ching Aw í ? ^ $2.

ju mieh chéng shou (safmavedayitanirodham samäpanno) A . ^ iE
ju yu ming sé (nämarüpassävakkanti) A^ ^ 4 j i?L
kén (müla) kéng lo (phassa) kéng pu shou yu (näparam itthattäya) 'S* k'ou (vacl) u k'u (dukkha) kuang yin (äbhassara) Je, •§*
kuan tai néng tso (apeksäkarana) IÖL % tl* k'ung (suñña, äkäsa) S "
lai (anägata) 4^
L/ao pen sheng szu ching 7 A S
liu ju chü (sal'äyatana) liu sheng (chalabhijäti) Í % lou chin chih (asavakkhayañana) ^ £ lui i (bhävänyathätva) M -J£ mieh chin (vaya) ^ é ming hsiang (nämarüpa) l i l ming sé (nämarüpa) 4* i?L
Index o f Chinese Terms nêng tso (kârana) nêng tso yin (kâranahetu) ® nu (dâsa) -kX. pao (phala)
P ei to shu hsia szu wei shih êrh yin yuan ching H +
pên sheng pên chien (pubbantakappanâ) ^ ^ JL
pên wu (abhütvâ)
pên wu chin yu (abhütvâbhâva) pên wu chin yu sheng (abhütvâbhâva utpâda) ^ 4" i pi (asau) Í& Pi-p'o-shih fu ching F ify Æü Pzč/z ¿ tsa a-han ching « í ffî & M.
píen (anta) pien (vivatta)
Pien chung pien lun pien hsing yin (sarvatragahetu) fâ) ÍŤ ® pien i fa (viparinâmadhamma) ^ ^ pien huai nêng tso (vikârakârana) *M. Í¡L
pu chu shih shen (appatitjhita viññána) ^ të- M ^ pu hsiang wei nêng tso (avirodhikârana) 7 *fà î t* ÍÍÍL
P'u-sa ti ch'ih ching ^
pu shan (akusala) ^ ■$-
246 sê (rüpa) shan (knsala) -H- she shou (parigraha) 1% $1 Shê ta ch'êng lun ^ A ^ Ira shen (atta?) shen (kâya) A
shen fa (kâyadanda) % shen shih (viññana) 1^ sheng (jâti, uppatti, uppâda) sheng chiai chü (ariya sllakkhandha) S sheng i chin (khlnâ jâti) £ £-» Ä
sheng mu t'ai (gabbhassâvakkanti) 4 flê sheng nêng tso (uppattikârana) £
sheng s zu chih (cut'üpapâtafiâna) £ &
shih (viññana) ^
shih (âhâra) ^ shou (vedanâ) è shou m ing shen ch'ang (appamânavi-,
hari) # 4" -fr *ÿ:
shou ming sheng tuan (appadukkhavihâri)
shou yin (upâdânakkhandha) fér shui (âpo) ?K shuo wu lun chê (n'atthikavâda) *i ' t & Îèï* 45"
Index o f Chinese Terms shuo wu tso (akiriyavâda) -M- Vf shuo wu yeh (n'atthikaväda) Ü LM lU t shuo yu lun chê (atthikaväda) 4 t So ch'u ching #T M. &
so tso i p'an (katam karaniyam) #T Vf ^ so yiian yûan (âlambana-pratyaya) #T ^ su ming chih (pubbenivásánussatiñana) ÍS ^ % sui shun yiian ch'i (idappaccayatâ) 64 ;,IR
sui shuo nêng tso (vyavahârakârana) 64 ÜL V f sui so yüan sheng (paccayam pajicca) 64 #T £ szù (cetanä) &
Ta ch'êng a-pi-ta-m o chi lun ^ i£ï Ta ch'êng a-pi'ta-mo tsa chi lun Æ 4ír ièt Ta ch'eng she-li-so-tan-mo ching A 3fl ^ ^ 41- i f Æ && ta chia (ayya) A ť a hsin chih (cetopariyañana) Vl % ta fan (M ahä Brahma) Á. t'a so tso (param katam) 'fc #T ť a tsao (param patam) i4.

ť a tso (param katam) V ^V f tai i (anyonyathâtva) % fa ťa n (raga, lobha) f t tê ( 'acquisition') ^
têng chih nêng tso (prâpanakârana) -S» fiîl têng hsing (sampratipatti) i ř ÍŤ ťe n g lai yu (punabbhavâbhinibbatti) ^ 4^* 'ÎT
têng wu chien yüan (samanantarapratyaya) M fri ti (pathavi)
t'ien êrh (dibbasota) ^ ťien yen (dibbacakkhu) A ^ ting fen (niyati)
ting fen hsiang hsû chuan pien (niyatisangatibhâvaparinatâ) & '/f 40 it* ting pieh nêng tso (pratiniyamakàrana) ÍÍL
Tyfl ching # FT ^ M.

tsê $']
tsêng i (vrddhi, pu?ti) ^ &
Tseng i a-han ching dř FT ^
tsêng shang yüan (adhipati-pratyaya) -itso flF tso yeh (sañcetanikam kammam) Ü£ tsü hsing (svabhâva) é O-
tsü tsao (attakatam, sayam katam) ÍH ijfc tsü tsai (issara) "â £ tsü tso (attakatam, sayam katam) tsü tso ťa tso (sayam katan ca param patañ ca) Æ
Index o f Chinese Terms 249 tsun yu (issara) *£ tsun yu tsao (issaranimmana)
ts'ung ch'i (uppada)
ts'ung t'a wen (parato ghosa) ifc fifl tu ch'u (titth'áyatana) *ÉL Á* tuan (uccheda) tuan chien (ucchedaditthi) IL
t'ung lui yin (samprayuktakahetu) 0 4R 0 t'ung shih néng tso (sahakarikarana) ® t'ung shih yin (sahakárihetu) 0 ^ 0 tz'ü (idam)

tz'ü yu ku pi yu, tz'ü ch'i ku pi ch'i; tz'ü
  wu yu ku pi wu yu, tz'ü mieh ku pi
  mieh (imasmim sati idam hoti, imassa
  uppada idam uppajjati; imasmim
  asati idam na hoti, imassa nirodhá
  idam nirujjhati) jfc f t i t ít f t i t ít M f t i t it M ft '/¡t i t it 'Á*

wei (sankhata)
wei ( 'sustaining') '(£-
wei (dosa) A
wei i (avasthányathátva) Isl wei ming (niyati?) ^
wei szü (m ato kálakato)
wu ch'ang (anicca) ^
wu ming (avijja) *M *f* wu m ing lou (avijj'asava) M
250 wu o (anatta) fe wu tso (akiriya) f e Vf wu wei (asaňkhata) f e Ä wu yin tso (ahetu appaccayâ) f e 0 Vf wu yin wu yiian (ahetu appaccayá, adhiccasamuppanna) f e 0 fe wu yu ai (vibhavatanhä) fe wu yu pên chi (anavarâgra) f e f a wu yu shen shih (appatitthita viññána) f e ^ yeh (kamma) i ř yen (cakkhu) yen ch'u (cakkhusamphassa) ÄS yen k'u (dukkhapatikküla) ^ yen szu (amaritukäma) 9L
yen yin yiian sé yen shih sheng (cakkhuñ ca paticca rüpe ca uppajjati cakkhuviññánam) ^ 0 fL £ .
yin (hetu, upanisa) 0 yin (khandha) fé?

yin chien (ditthadhammaupakkamahetu) 0 JL
yin chü (sañkharapuñja) SI
yin fa nêng tso (ävähakärana) ? I 'SbÜ Lfä yin ho hui (sañgatihetu?) 0 & f t yin pên tso (pubbekatahetu) 0 f a Vf Index o f Chinese Terms yin su m ing tsao (pubbekatahetu) 0 ^ ijfc yin tsun yu tsao (issaranimmâi^ahetu) 0 ^ 4 yin wei ming (abhijätihetu, niyatihetu?) 0 ^ 4 " yin yuan (hetu-pratyaya) 0 ^ yin yüan ch'i (paticcasamuppäda) 0 yin yüan ch'i so sheng fa (paticcasamuppanna dhamma) 0 ffc £ y k yin yüan fa (paticcasamuppäda) 0 Ü yin yüan ho ho sheng (hetum paticca sambhütam) 0 4 ° ^
yin yüan hui êrh sheng (hetum paticca sambhütam) 0 yu (bhava, bhavati) yü (tanhä)
yu ai (bhavatanhä)
yü ai (kämatanhä) yu lou (bhav'äsava) ^ iÄ
yü lou (käm 'äsava)
yüan (pratyaya) Yüan ch'i ching & yüan ch 'i fa (paticcasamuppäda) M, ü Yüan ch*i sheng tao ching yüan sheng fa (paticcasamuppanna dhamma) 4