The Authenticity
of the Early Buddhist Texts

Bhikkhu Sujato
&
Bhikkhu Brahmāli

Abbreviations

References to EBTs outside the Pali canon are usually from Anālayo’s comparative study¹ or from suttacentral.net, unless otherwise stated.

Abstract

This work articulates and defends a single thesis: that the Early Buddhist Texts originated in the lifetime of the Buddha or a little later, because they were, in the main, spoken by the Buddha and his contemporary disciples. This is the most simple, natural, and reasonable explanation for the evidence.

Our argument covers two main areas:

1. The grounds for distinguishing the Early Buddhist Texts (EBTs) from later Buddhist literature;

2. The evidence that the EBTs stem from close to the Buddha’s lifetime, and that they were generally spoken by the historical Buddha.

Most academic scholars of Early Buddhism cautiously affirm that it is possible that the EBTs contain some authentic sayings of the Buddha. We contend that this drastically understates the evidence. A sympathetic assessment of relevant evidence shows that it is very likely that the bulk of the sayings in the EBTs that are attributed to the Buddha were actually spoken by him. It is very unlikely that most of these sayings are inauthentic.

Acknowledgements

We are grateful to a number of people for providing valuable feedback on a draft version of this work. Professor Richard Gombrich, Ven. Anālayo, Ven. Shravasti Dhammika, Ven. Ñāṇatusita, and Dr. Alexander Wynne all kindly provided detailed feedback. The quality of this paper is much improved as a result. Ven. Bhikkhu Bodhi also gave of his valuable time to provide a few comments.

In addition, the content of this work was presented by the authors as a series of classes at Bodhinyana Monastery in 2013, as well as at a seminar for the Australian Association of Buddhist Studies on March 13, 2013; in both these cases we benefited from many illuminating discussions and suggestions.

We also gratefully acknowledge the support and understanding of Ajahn Brahm and the supporters of the Buddhist Society of Western Australia, which has provided us with the time and resources needed to undertake this project.

Thank you!

Introduction

Are there any authentic Buddhist texts? If so, what are they? These are questions of tremendous spiritual and historical interest, about which there is a range of opinions that often appear to be irreconcilable. Traditionalists insist that the texts were “spoken by the Buddha” in the most literal of senses, while sceptics assert that we cannot know anything about the Buddha for certain, and further, that the notion of authenticity is irrelevant or pernicious. Rarely, however, has the question of authenticity been systematically investigated. Seeing the lack of an easily accessible summary of the evidence, we decided to assemble this survey.

There are two main aspects to our argument: (1) there is a body of Early Buddhist Texts (EBTs), which is clearly distinguishable from all other Buddhist scripture; (2) these texts originated from a single historical personality, the Buddha. These two aspects are closely linked, so we have not tried to separate them or present them in sequence.

We consider the doctrinal and linguistic evolution of the texts, grounded in their social and economic context. No particular methodology is preferred; rather, we aim to be inclusive, as we believe that diverse perspectives are useful, indeed essential. So we take into account the internal development and structure of the texts, as well as the results of comparative studies. Multiple independent lines of evidence from the EBTs converge on a point of origin geographically in Northern India and temporally around the 5th century bce.

But the literature converges on more than a time and a place. It converges on a man: the historical Buddha. It took an astonishing energy and dedication to create and sustain this literature. It must have been produced by an extraordinary historical event. And what could this event be, if not the appearance of a revolutionary spiritual genius? The Buddha’s presence as a living figure in the EBTs is overwhelming and unmistakable. It stands in stark contrast with all other Buddhist literature, where the Buddha has faded to legend. Yet none of the later texts could exist without the EBTs; they are the foundational literature from which everything else derives.

We pay special attention to the historical situation of India in the period after the Buddha as revealed through inscriptions, monuments, and writings by Indians and Greeks. Within a few decades of the Buddha, under the reign of the Nandas, the political situation completely changed, from the diverse patchwork of small states depicted in the EBTs to a mighty, unified empire. And from the Mauryan empire of Asoka, starting around 130 years after the Buddha, the sources speak of the existence and wide dominion of Buddhism. For this to be possible, Buddhism must have undergone an extended period of growth prior to the Mauryan empire. This is in agreement with the EBTs, which vividly and realistically depict social, geographical, linguistic, political, philosophical, religious, and other conditions that obtained in a period substantially prior to the Mauryan.

Our argument may be seen by traditional Buddhists as too obvious to need stating, and by some Buddhist scholars as being hopelessly optimistic and unprovable. We reject both positions. The variations and divergences in the Buddhist traditions, as well as the lack of direct manuscript or archaeological evidence for the life of the Buddha, are real and important considerations, which should raise questions for any serious inquirer as to what the historical foundations for the Buddhist religion really are. However, it is overly sceptical and unscientific to conclude from the lack of such direct evidence that we cannot say anything, and that we cannot reach firm conclusions. Science works from indirect and inferred evidence, and the preponderance of such indirect evidence points to the authenticity of the EBTs.

We are not denying the obvious fact that the texts bear all the marks of redaction and editing, and that they have been optimised for the oral tradition. There are even a few cases where the editorial hand seems to have added interpretations to existing ideas. But to assume from this that the literature as a whole has not conserved the central ideas propounded by its founder, or even that it was invented ad hoc by redactors, is to lose sight of the distinction between editing and composing. So when we say that the texts were “spoken by the Buddha”, we mean it in this non-literal sense. Clarifying the exact nature and degree of the editorial influences on the EBTs is one of the primary tasks of the student of Early Buddhism.

Anyone wishing to establish the thesis that the EBTs are inauthentic needs to propose an explanation that accounts for the entire range of evidence in a manner that is at least as simple, natural, and reasonable as the thesis of authenticity. To our knowledge, this has never even been attempted. Rather, sceptics content themselves with picking holes in individual pieces of evidence, which merely distracts from the overall picture, and discourages further inquiry. Their methods have much in common with denialist rhetoric (see section 7.4).

The field of study we are covering is vast and cannot be adequately represented in this short survey. Our aim is not to treat all items with the detail they deserve, but to outline the basic areas of interest, present as strong a case as we can for authenticity, and give further references for those who wish to learn more.

We have chosen to construct this work in a “modular” form, rather than as a sequential argument. Each section stands, more or less, on its own, so anyone can quickly check the relevant facts on any particular aspect. For this reason there is a fair amount of repetition, and the bibliographic information is gathered at the end of each section. We have kept the tone as direct as possible, and we apologise if at times we seem blunt. Of course, each statement can be qualified and nuanced, but it becomes tedious to preface every detail with “perhaps” or “evidence indicates that”.

We aim to give a fairly comprehensive and historically flat overview, rather than a survey of the contemporary state of the field. We believe that, while in certain respects the field of Buddhist studies has progressed, many of the arguments made by foundational scholars still hold good, and that they have sometimes been unfairly neglected or dismissed. The very longevity of these ideas suggests that they may still be pertinent long after more fashionable contemporary notions have been forgotten.

Inevitably, different people will find different arguments more or less persuasive. However, our thesis does not depend on individual details, but on the persuasiveness of the bulk of the arguments taken as a whole. We are dealing with events over two millennia ago, so it is of course possible, if not trivial, to cast doubt on any individual detail. It is only when the facts are seen together as contributing to a coherent narrative that they become compelling. We outline our theoretical perspective more in the last chapter.

The length of the sections is somewhat arbitrary, and does not necessarily correspond to the importance of the topic. In particular we are conscious of the inadequacy of the section on comparative studies. This field is perhaps the most important, and has on the one hand had a broad-based, lasting scholarly consensus, and yet is one of the fastest growing and most in need of further research. Virtually every study that has been done so far, covering many hundreds of EBTs, could be quoted in support of the authenticity of the texts. A survey of the field would in itself be a major research project. So we will content ourselves with indicating the broad outlines and a few examples.

Definitions

Authenticity:

An authentic text is one whose provenance is what it says it is. In this case this means that texts that purport to be the words of the historical Buddha and his immediate disciples were in fact spoken by them.

Early Buddhist Texts:

Texts spoken by the historical Buddha and his contemporary disciples. These are the bulk of the Suttas in the main four Pali Nikāyas and parallel Āgama literature in Chinese, Tibetan, Sanskrit, and other Indian dialects; the pātimokkhas¹⁴ and some Vinaya material from the khandhakas;¹⁵ a small portion of the Khuddaka Nikāya, consisting of significant parts of the Sutta Nipāta, Udāna, Itivuttaka, Dhammapada, and Thera- and Therī Gāthā. The “Suttas” in a narrow sense are those passages that are directly attributed to the Buddha himself (and to a lesser extent his direct disciples).

Non-EBTs:

Abhidhamma, Mahāyāna Sūtras, Buddha biographies, historical chronicles, as well as the majority of the Khuddaka Nikāya and the Vinaya Piṭaka. The Jātakas are non-EBT, but derive from stories that in some cases may even be earlier than the Buddha. Commentaries and other late texts may contain some genuine historical information alongside much later invention.

Thesis

1. That most of the EBTs are authentic.

2. That the EBTs were edited and arranged over a few centuries following the Buddha’s demise. The texts as we have them now are not a verbatim record of the Buddha’s utterances, but the changes are in almost all cases details of editing and arrangement, not of doctrine or substance.

3. That the inauthentic portions of these texts are generally identifiable.

4. That the above points are supported by a substantial and varied body of empirical evidence.

5. That the denial of authenticity is a product of excessive and unreasonable scepticism, not evidence.

Timeline

Our dating is based on the “median chronology”,^{16 17} which places the birth of the Buddha at around 480 bce. According to the  Theravāda tradition, the birth of the Buddha was in 623 bce and the Parinibbāna in 543 bce. According to the “long chronology” the dates are 563 bce and 483 bce respectively.¹⁸

Chapter 1
History & geography

1.1 Political geography

The political geography of Northern India changed rapidly after the period described in the EBTs. Within a few decades the diverse kingdoms had been unified under the Nanda dynasty, a process that is seen beginning in the EBTs, and which culminated in the Mauryan empire of Candagutta²⁴ and Asoka.²⁵ The EBTs must consequently belong to a period of history at least several decades prior to the reign of the Nandas. This locates them at or very near the historical Buddha. Before discussing some of the major changes that occurred in this interval, we will have a brief look at the main Greek source of knowledge of ancient India: the writings of Megasthenes.

1.1.1 Megasthenes

Megasthenes was the ambassador of the Greek king Seleucus i to Candagutta’s court at Pāṭaliputta. His writing on India, and those of other Greeks, survive in quotations in Greek in works by classical Roman historians such as Arrian, Strabo, and Pliny.²⁶

The political and economic scene he depicts is broadly in agreement with the archaeological evidence for the Mauryan period, but post-dates the descriptions in the EBTs. He describes the powerful kingdom of Magadha, with its magnificent capital Pāṭaliputta, which was an obscure village at the time of the EBTs (see section 1.1.5). Candagutta had inherited the vast dominions of the Nandas and his empire was the grandest known in India to that time.

Megasthenes does not refer directly to Buddhism, but does notice the samaṇas and brahmans, who are essential figures of religious life in the EBTs.²⁷ His description of samaṇas appears to be of Jains or Ājīvakas, although some of what he says may also apply to Buddhist monastics. Clement of Alexandria, writing around the 2nd century ce, connects Megasthenes’ samaṇas with Buddhists, saying:²⁸ “Among the Indians are those philosophers also who follow the precepts of Boutta (i.e., the Buddha), whom they honour as a god on account of his extraordinary sanctity.”

1.1.2 The expansion of the known world

Most prominently featured in the EBTs are the countries within which the Buddha taught, a list of which is found at dn 18.1 and dn 18.4. The combined area of these countries is roughly equivalent to the central part of the Ganges plain.²⁹ Occasionally this list is expanded to include other known kingdoms, all of which combined are known as the 16 great countries (mahājanapada). These countries stretch as far as the north-west of India, including Gandhāra.

Apart from one EBT reference to Greeks (see next section, 1.1.3) and occasional references to the Kambojans, possibly referring to the Persians,³⁰ passages on areas outside of this region are typically mythological.³¹ South India and Sri Lanka are not mentioned at all and appear to be unknown. Aparantaka, in the west of India, is regarded as a brutal, uncivilised place.³² Kaliṅga, in the east, is mentioned as a mythical wilderness at mn 56.14,³³ but as a civilised region in later passages.³⁴ Andhra (Pali: Andha),³⁵ in the south-east, is only mentioned in the Pali commentaries.

One passage, the introduction to the Pārāyana Vagga of the Sutta Nipāta, describes a long journey by brahman students seeking the Buddha, which begins somewhat south of the regions otherwise known to the Suttas.³⁶ But this passage, an unusual narrative section of the Khuddaka Nikāya, is probably somewhat late.³⁷

By the time of Candagutta, a mere century after the Buddha, and even more so at the time of Asoka, the known world had expanded to include Sri Lanka and South India, and possibly Burma.³⁸ Although Persia, and probably Greece too, were known in India from an early period, even prior to the EBTs (see next section), the knowledge of these distant countries expanded in the Mauryan period.³⁹ re 13⁴⁰ mentions several Greek kings by name,^{41 42} indicating extensive knowledge of the Greek world.⁴³

1.1.3 Greek connections

There is only one known reference to the Greeks in the EBTs.⁴⁴ This comments on the different social structure of the Greeks, saying they did not have the four castes (cf. re 13).⁴⁵

The mention of the Greeks has been regarded as a late interpolation into the Suttas at the time of Candagutta.^{46 47} Certainly, the relationship with the Greek world was a distant one in the time of the EBTs. However, there are several sources that mention or imply contact between Greeks and Indians in the time of the EBTs, or even earlier.^{48 49} Ctesias’ Indika was written only a few years after the Buddha passed away.⁵⁰ Herodotus, a contemporary of the Buddha, speaks of India many times, and says that Indian troops fought in the Persian army of Xerxes.^{51 52} Thus Indian troops fought against continental Greece at the beginning of the 5th century bce.⁵³ The Persian emperor Cyrus i had in fact conquered parts of north-western India as early as mid-6th century bce,⁵⁴ and the territorial expansion into India along the Indus continued under Darius i in the late 6th century.^{55 56 57} According to McEvilley, the vast Persian Empire, and especially its capital Persepolis, functioned as a cosmopolitan meeting place for people from distant countries, including Greeks and Indians.⁵⁸

Candagutta’s reign, by contrast, marks the period of close contact with the Greeks. Candagutta had a Greek ambassador at his court and probably married a Greek princess.⁵⁹ At the time of Asoka the geographical knowledge was even greater, with re 13 referring to the kings of Egypt, Cyrene (in present day Libya), Epirus (north-western Greece and Southern Albania), and Macedonia.^{60 61}

1.1.4 The 16 nations become part of the Māgadhan empire

According to the EBTs the north of India at the time of the Buddha, from the Ganges delta to Gandhāra, corresponding roughly to the area covered by the Ganges plain and large parts of the Indus plain, was divided into 16 great countries (mahājanapada), one of which was Magadha.⁶² This tradition is echoed in Jain (Bhagavati Sūtra) and Brahmanical (Mahābhārata) sources, which however give somewhat different lists of countries.⁶³ Where they differ, the EBT list covers a smaller area and appears to be earlier.⁶⁴ While the 16 nations are only mentioned occasionally, the political situation that they represent—of Northern India divided into a patchwork of minor principalities and republics—is inherent to the entire socio-political environment of the EBTs.

Only a few decades after the Buddha these nations had disappeared, as most of north and central India was unified in the kingdom of Magadha by the Nanda dynasty. We have a fairly good idea of the range of this empire. When Alexander invaded India in 326 bce he turned back at the mere rumour of the Nandas’ military might. The domain of the Nandas at this time must therefore have spread as far, or nearly as far, as Alexander reached, which was the Hyphasis River (the modern Beas River), to the north-west of the Kuru region. We also have the Hatthigumpha inscription of king Khāravela at Udayagiri in Kaliṅga, from the 2nd century bce, where the king speaks of extending the canal built by King Nanda.⁶⁵ Thus Nanda had not only conquered Kaliṅga but had accomplished substantial capital works there. The Nanda empire must therefore have encompassed virtually all of the former sixteen nations, with the exception of Kamboja and Gandhāra. This agrees with the accounts given in the Hindu Purāṇas, which consistently say that Mahāpadma Nanda subjugated India under one rule (ekacchatra),⁶⁶ and occasionally identify the nations which he conquered, including the Kurus, Kāliṅgas, Mithilas, Sūrasenas, etc.⁶⁷

The 16 nations, therefore, had disappeared a few decades after the Buddha’s death, which means the process must have started much earlier. This fits with the picture given in the EBTs. In the Buddha’s lifetime, King Bimbisāra of Magadha subjugated the kingdom of Aṅga to the east,⁶⁸ and the Māgadhan dispute with Kosala over Kāsi was settled in Magadha’s favour by Bimbisāra’s son Ajātasattu.⁶⁹ By the time the Buddha died, the EBTs thus portray Magadha as controlling the south bank of the Ganges from Campā to Benares, and also Ajātasattu as desiring to bring the Vajjian republic under his control.⁷⁰ The EBTs and independent sources thus agree on the gradual expansion of the Māgadhan state in the decades following the Buddha, and that the subjugation of the 16 nations was mostly complete before Alexander arrived.

Many of these events would have been witnessed by the Buddha’s direct disciples, and the rest by the first couple of generations of his followers. Yet not a single hint of any of this has made its way into the EBTs. This is not because they are ignorant of or unconcerned with politics; as we have seen, there are plenty of references to political and societal details in the EBTs. The simplest explanation is that the main content of the EBTs derives from the lifetime of the Buddha, during which the 16 nations were the dominant political feature, and this content was left unchanged despite the political upheaval that followed.

Once they had disappeared, the 16 nations never returned. The Nanda dynasty was succeeded by the Mauryan dynasty established by Candagutta.^{71 72 73 74} He further expanded the Nanda realms to the west and south.⁷⁵ This is further confirmed for the Asokan period, when the distribution of Asokan pillars and rock edicts shows that all of India, except the extreme south, was part of Magadha.⁷⁶

In the EBTs, political and military rivalry is confined to the 16 mahājanapadas.⁷⁷ But this soon changed: Alexander’s army mutinied rather than take on the Nandas. By the time of Candagutta, Greek sources show that Māgadhan military might was measured against that of southern and eastern Indian countries, in particular against Kaliṅga and Andhra, as well as other countries lying outside of the Ganges plain.⁷⁸ Kaliṅga was then conquered by Asoka in about 260 bce.⁷⁹ According to a number of sources, including re 13,⁸⁰ the  Mahākarmavibhaṅga,⁸¹ and the Mahāvaṁsa,⁸² Buddhism became established in Sri Lanka (Tāmraparṇi) in the Asokan period, apparently with the direct involvement of Asoka himself.

1.1.5 The transformation of Pāṭaliputta

Pāṭaliputta is unknown to any pre-Buddhist texts, such as the early Upaniṣads, and we have no evidence that it existed in this period. A list in the EBTs of the main cities of Northern India does not include Pāṭaliputta.⁸³ While the Buddha was alive, Pāṭaliputta was still called Pāṭaligāma (“the village of Pāṭali”; dn 16.1.19–28/smps 4.1–4.), and it would therefore have been very modest in size.^{84 85} Apart from a few late Suttas,⁸⁶ it can only be traced to this one passage. Towards the end of the Buddha’s life a city, the future Pāṭaliputta, was being built at or near Pāṭaligāma, and the Buddha predicts its future greatness, using the name Pāṭaliputta for the first time.^{87 88}

This fits with the archaeological evidence which indicates that large scale urban development in the Ganges plain, characterised by “town planning, fortification, monumental buildings, the use of bricks etc., does not appear much older than the fourth and even the third century bce”.⁸⁹

The picture found in the EBTs contrasts sharply with the situation about a century later. Various sources, including Greek ones, tell us that at the time of Candagutta Pāṭaliputta is the capital of Magadha and the greatest city in India.^{90 91} According to Jain and Brahmanical sources, Pāṭaliputta probably became the capital of Magadha under Udayabhadda, son of Ajātasattu, and remained as such until after the dissolution of the Asokan empire.⁹² Again, archaeological excavations at Patna/Pāṭaliputta indicate that this picture is historically accurate.⁹³

1.1.6 Absence of Candagutta and Asoka

Considering how regularly kings, including the kings of Magadha, are mentioned in the EBTs, and considering how celebrated Asoka is in other Buddhist literature, it is particularly noteworthy that neither Candagutta nor Asoka, both ruling an empire previously unparalleled in Indian history, is mentioned in the EBTs.⁹⁴ Perhaps the earliest mention of Candagutta in Pali literature is found in the Milindapañha.⁹⁵ And Asoka’s life, by way of contrast to the EBTs, is described at length in a number of Buddhist sources, such as the Dīpavaṁsa, the Mahāvaṁsa, the Samantapāsādikā, and the Aśokāvadāna.⁹⁶ Moreover, the Dīpavaṁsa (1.26–27 of Oldenberg’s translation), as well as certain Mahāyāna works such as the quasi-historical Mañjuśrimūlakalpa, portray the Buddha as predicting the advent of Asoka.⁹⁷ No EBT contains such a prediction.

1.1.7 No contradictory evidence

There is no reliable evidence, either literary or archaeological, that contradicts the picture given in the EBTs. If the texts that are set in the Buddha’s time were composed after the time of the Buddha, one would expect errors in the descriptions of the political situation. Such errors are found regularly in other Buddhist literature, for example in the depiction of Sakya as a kingdom, or in the exaggeration of the size of cities.

1.2 Social conditions

In contrast to the Brāhmaṇas and the Upaniṣads, which portray rural socio-economic conditions, the EBTs depict an urban life with flourishing trade.⁹⁸ As to the political situation, although small scale state formation can be traced in the Upaniṣads, it is only in the Buddhist literature that regional development and struggles between kings are depicted.⁹⁹

“Although the Upaniṣads contain only very few references to socio-political and economic conditions, they allow the inference that their actors lived in a predominantly rural society, dominated by a landed aristocracy and strong chieftains. These rājās lived with their little patriarchal courts in a permanent residence within a still very rural context. … The subsequent emergence of these strong mahājanapada kingdoms, which is identifiable mainly in the early Buddhist and Jain literature, therefore would have to be dated in the fifth or even early fourth centuries bce rather than in the sixth century as we have been used to do till now. A. Ghosh’s critical evaluation of the archaeological evidence of the early cities of these mahājanapadas (e.g. Vaisālī and Srāvastī) makes such an inference quite likely as none of them show genuine traces of urbanisation at about 500 bce, the only exception perhaps being Kausambi.”¹⁰⁰

Again, the conditions described in the EBTs fit well with the conditions expected at the time of the earliest form of Buddhism, that is, before Candagutta and Asoka, but after the early Upaniṣads.

1.3 Economic conditions and trade

In the EBTs we rarely find mention of trade outside the Ganges plain. Even luxury goods such as fine cloth were produced locally, especially in Kāsi/Vārāṇasī (an 3:39, mn 123.18), and then traded within North India. There are some exceptions to this. Kāsi sandalwood, which was used by the upper strata of society, including kings, most likely originated in southern India,¹⁰¹ but was processed in Kāsi and thus known as Kāsi sandalwood (mn 87.28, an 3:39).

This situation is perhaps to be expected given the political divisions in North India at the time, which may have complicated long-distance trade. There is only one mention of the existence of seagoing merchants in the EBTs (dn 11.85), but nothing is said of where they traded.

This picture fits with the scant historical information available for this time. According to McCrindle, Nearchos’ coastal voyage from the mouth of the Indus to the Persian Gulf, undertaken at the behest of Alexander the Great, was done with the express purpose of gathering information about this stretch of coast, and it resulted in the opening up of communication between Europe and Asia.¹⁰² This indicates that seafaring in this area, including sea trade, would have been very limited even 100 years after the Buddha.

The picture given by the EBTs and McCrindle contrasts with the situation in India at the time of Asoka, who ruled only a few decades after Alexander. By this time we hear of trade between India and far off regions such as the Mediterranean.¹⁰³

This situation matches quite well with trade as depicted in the Buddhist Jātaka literature,¹⁰⁴ which is normally assigned to a period in the centuries after the Buddha. The Jātakas mention trade by sea to Suvaṇṇabhūmi (possibly lower Burma), and also over desert to Sovīra (Rajasthan) and Baveru (Babylon).¹⁰⁵

1.4 The Universal Monarch

The EBTs refer occasionally to the notion of the Cakkavatti, a universal “wheel-turning” emperor who rules from sea to sea, justly, without violence. Since there is no evidence for a pan-Indian empire before the Buddha, it is sometimes argued that this must date from Asokan times.¹⁰⁶

But this conclusion is unwarranted.¹⁰⁷ The Mauryan empire probably had a different structure from the one connected to the Cakkavatti ideal, according to which vassal kings were ruled by the universal monarch.¹⁰⁸ Indeed, the Cakkavatti is clearly a mythic figure and an imagined ideal that sits ill with the realistic stories of Asoka and his ambiguous relation to violence.¹⁰⁹ Also, Asoka claims many titles for himself and Cakkavatti is not one of them. Moreover, the Cakkavatti is also found in Jain Sūtras and there is no evidence that the Jains had a close relationship with Asoka in the way the Buddhists did.

Rather, the antecedent for the Cakkavatti is the ancient Brahmanical Horse Sacrifice (aśvamedha), which establishes sovereignty from ocean to ocean.¹¹⁰ Such a notion would have arisen through both the magnification of the glories of former Aryan realms and by comparison with the powerful contemporary Persian empire. The Buddha appropriated the Vedic myth, removed the violent and coarse aspects, and retained the ethical ideal of righteous rule.¹¹¹ Parallels between the Horse Sacrifice and the Cakkavatti include:¹¹²

1. The “Horse Treasure” of the Cakkavatti is white with a black head (dn 17.1.13; mn 129.37); the sacrificial horse is white with a black head/forequarters.¹¹³

2. The Cakkavatti has 1000 sons (e.g. dn 3.1.5; mn 91.5; sā 264); the sacrificial horse is protected by 100 sons.¹¹⁴

3. The “Wheel Treasure” of the Cakkavatti plunges into the four seas (dn 17.1.10; mn 129.35); the horse is of cosmic stature, born in the western and eastern seas (Bṛhadāraṇyaka Upaniṣad 1.1.1–2).¹¹⁵

4. The Cakkavatti follows the Wheel across the land and all submit (dn 17.1.8–10; mn 129.35; mā 67); the horse sacrificer follows the horse across land and all must submit (Bṛhadāraṇyaka Upaniṣad 1.1.1–2).¹¹⁶

5. The horse of the Cakkavatti flies (dn 17.1.13; mn 129.37). The horse in the Vedas is identified with the Sun (wheel in sky), and is united with the chariot, the military vehicle of the Aryans.¹¹⁷

6. The duties of a Cakkavatti and the Aśvamedha are periodic astrological rites that renew the royal and solar powers (dn 26.3–9; Ṛg Veda 1.162, Yajurveda Vājasaneyi Saṁhitā 24.24–45, Śatapatha Brāhmaṇa 13.1.6.3).

1.5 Writing

Writing is unknown in the Suttas,¹¹⁸ while a few references to writing appear in the later strata of the Vinaya.¹¹⁹ This situation agrees with the archaeological evidence, which only attests to writing in the period after the Buddha, particularly the Asokan period and possibly a few earlier fragments. According to Megasthenes, writing was not known in India at the time of Candagutta, about 100 years after the Buddha’s death.¹²⁰ According to Nearchos, the admiral of Alexander the Great, however, writing was in use in India in 325 bce, slightly before the reign of Candagutta.¹²¹ These accounts may both be true, since Nearchos travelled no further east than the Indus river, whereas Megasthenes was residing in Pāṭaliputta in eastern India, about 2,000 km away.

Most scholars agree that there is no certain evidence for the use of writing in India before the time of Asoka, but they disagree as to whether it was known at all.^{122 123} Despite the lack of consensus,¹²⁴ the majority view is probably represented by Salomon who says:¹²⁵ “... there is every reason to think that Brāhmī [script] did not exist before the 3rd century bc, and that it was created then on the basis of a loose adaptation of one or more pre-existent Semitic scripts, with Kharoṣṭhī playing at least a partial role.”

The absence of any evidence of writing in the EBTs indicates their pre-Mauryan origin. In comparison, texts such as the Mahāyāna Sūtras frequently mention writing,¹²⁶ and this is one among many criteria for dating them to several centuries after the EBTs. It has in fact been suggested that the Mahāyāna Sūtras owe their very existence to the use of writing.¹²⁷

Chapter 2
Religious context

2.1 Contact with other religions

The EBTs frequently depict the Buddha and his disciples in dialogue with members of other religions and with sceptics.¹²⁸ This is radically different from all other Buddhist literature, which consists almost entirely of Buddhists speaking to other Buddhists. This difference makes sense if we consider that the EBTs largely stem from the life of the founder, one of whose tasks was to persuade others to his path.

2.2 Jainism

The Buddhist texts depict the Buddha as contemporary with Mahāvīra, the teacher of the Jains (e.g. mn 56/mā 133, mn 104.2/mā 196/T 85, dn 2). Jain tradition places Mahāvīra at 599–527 bce,¹²⁹ which agrees roughly with the traditional  Theravādin dating of the Buddha. While these dates are not historically precise, this does show that both Jain and Buddhist texts indicate that their teachers flourished in the same pre-Mauryan period.^{130 131}

Moreover, according to Norman,^{132 133} the EBTs and Jain literature show a number of similarities. And, although the language of the extant Jain literature stems from a later date,¹³⁴ this suggests that the two literary traditions are rooted roughly in the same period.¹³⁵ The similarities include the following:

1. Criticism of Brahmanical animal sacrifice;¹³⁶

2. Some ideas about karma, opposed to Brahmanical views;¹³⁷

3. The importance of generosity;¹³⁸

4. Social stratification: the khattiya caste is regarded as superior to the Brahmanical caste;¹³⁹

5. Paccekabuddhas: including some of the same names;^{140 141}

6. The idea of a sequence of past ‘ford-makers’ (titthakara), that is, founders of their respective religions;¹⁴²

7. Shared stories: the Pāyāsi Sutta (dn 23/dā 7/mā 71/T 45);¹⁴³

8. Shared similes;¹⁴⁴

9. Vocabulary: they have a large religious vocabulary in common, including epithets for their respective leaders,¹⁴⁵ and general religious terminology;^{146 147}

10. Common verses and verse lines;^{148 149}

11. Stylistic parallels: the use of some of the same metres in verse composition.¹⁵⁰

The earliest Jain texts often use the same name for Mahāvīra as the EBTs do, Nāyaputta/Nāyasuya, equivalent to the Pali Nātaputta.¹⁵¹ The doctrine of the four restraints is mentioned as being a Jain doctrine in both the EBTs and in early Jain Sūtras.¹⁵² The basic moral teachings of the EBTs and the early Jain texts are similar, especially the ideas of not killing, stealing, lying or indulging in sensory pleasures.¹⁵³

2.3 Brahmanism

A number of facts about the EBTs help us locate them in time vis-à-vis the Brahmanical texts. The EBTs (e.g. mn 91.2 and sā 255) refer only to three Vedas, not the four that became standard in later years,^{154 155} and they are unaware of the Mahābhārata and the Rāmāyana, which are of a later date.^{156 157 158} The EBTs depict uncertainty and questioning among the brahmans,¹⁵⁹ which is highly characteristic of the early, pre-Buddhist, Upaniṣads. And while the EBTs are aware of Upaniṣadic ideas,¹⁶⁰ they only name the Upaniṣads once, as Brahmanical lineages rather than as texts.¹⁶¹

The EBTs are unaware of Pāṇini, who can be dated no later than mid-4th century bce.¹⁶² This can be seen in the fact that the EBTs frequently refer to grammar, but as an ancient tradition, not as a modern innovation (e.g. mn 91.2).

The EBTs also depict many deities and practices that are found neither in the ancient Vedas, nor in the later Hinduism. In addition, they fail to mention many practices common to the later Brahmanical tradition, such as the worship of the Śiva-liṅgaṁ¹⁶³ and deities such as Krishna, Ganesh, Kali and Skanda.

2.4 The unity of Buddhism

Although the EBTs frequently mention disagreements within the Sangha, there is hardly any indication of Buddhism being split along sectarian lines. The only exceptions to this are the temporary split over the incident at Kosambī (Vin i 337–342 and Vin i 353–354; see also mn 48 and mn 128) and the more serious split in the Sangha caused by Devadatta (Vin ii 184–203, an 5.100). However, there is no connection between this schism and the later emergence of distinct Buddhist schools.

It is in post-Asokan literature, such as the Kathāvatthu, that true sectarianism is first mentioned. This fits with the general consensus among scholars that sect formation in Buddhism, with the possible exception of the Mahāsāṅghika schism, happened in the post-Asokan era.^{164 165} This again points to the EBTs largely being finalised before Asoka.¹⁶⁶

Chapter 3
Textual transmission

3.1 Comparative Studies

3.1.1 Overview

We possess EBTs from a substantial variety of ancient Indian Buddhist schools, including the Mahāvihāra (the ancestors of the contemporary Theravāda) of Sri Lanka, the Dharmaguptaka, Mahāsāṅghika, Mahīśāsaka, Mūlasarvāstivāda, Sarvāstivāda, and others of uncertain affiliation.¹⁶⁷ A century of detailed study has consistently shown that they are essentially identical in doctrine irrespective of transmission lineage.¹⁶⁸

The great Belgian scholar Étienne Lamotte says:¹⁶⁹ “However, with the exceptions of the Mahāyānist interpolations in the Ekottara, which are easily discernible, the variations in question [across the lines of transmission] affect hardly anything except the method of expression or the arrangement of the subjects. The doctrinal basis common to the Āgamas and Nikāyas is remarkably uniform.” This is in stark contrast to non-EBT texts.

The basic facts were discovered in the 19th century. In 1859 Samuel Beal published side by side translations of the Pali pātimokkha and the Dharmaguptaka prātimokṣa in Chinese, showing their virtually identical content. He noted that¹⁷⁰ “… the identity of the code in both cases, therefore seems to be established.” In 1882 Beal described detailed correspondences between Suttas in Chinese and Pali.¹⁷¹ He accurately predicted that¹⁷² “when the Vinaya and Āgama collections are thoroughly examined, I can have little doubt we shall find most if not all the Pali Suttas in Chinese form.”

This pioneering work was followed by a series of studies by Anesaki,¹⁷³ Akanuma,¹⁷⁴ Yin Shun,^{175 176} Thích Minh Châu,¹⁷⁷ and many other scholars. Beal’s findings regarding the pātimokkha were confirmed in 1928 by the Japanese scholar Ryūzan Nishimoto¹⁷⁸ and again, apparently independently, in 1955 by W. Pachow. Both of these studies showed that all existing recensions of the pātimokkha, which total around twenty texts from at least 7 different schools,¹⁷⁹ are very similar in content, with the exception of the most minor category of rules, known as the sekhiyas.¹⁸⁰

Recently, in his detailed and thorough Comparative Study of the Majjhima Nikāya, Anālayo shows that all significant aspects of early Buddhist doctrine are the same across all extant textual transmissions of the Suttas of the Majjhima Nikāya.¹⁸¹ Among the parallels to the Suttas of the Pali Majjhima Nikāya, the most important textual source, due to its completeness, is the Sarvāstivādin Madhyama Āgama preserved in Chinese. The Sarvāstivāda and Theravāda lineages must have separated approximately at the time of the Asokan missionary activities.¹⁸² This means that these texts have been transmitted separately for almost 2,300 years, including a period of separate oral transmission that lasted several centuries. And yet the doctrinal content is for all intents and purposes identical.¹⁸³ This shows how conservative and careful the individual schools were in preserving the EBTs.

Moreover, this conservatism must have been inherited from the more unified—both geographically and doctrinally—form of Buddhism that existed prior to Asoka. There is no reason to imagine that the separate schools would all be conservative in preserving their canonical texts unless they had been conservative prior to their separation. Since comparative studies show that the core doctrinal material of the EBTs has been reliably transmitted for almost 2,300 years, the reasonable inference is that it was reliably transmitted also in the first 150–200 years of Buddhist history.

In addition to the full scale study of the Majjhima Nikāya, there have been multiple smaller studies of various parts of the EBTs. These have confirmed that all the EBTs share a similar level of agreement to what we find between the Suttas of the Majjhima Nikāya and its parallels. Such studies have been carried out for much of the Saṁyutta Nikāya/Saṁyukta Āgamas,^{184 185} and to a lesser extent for the Dīgha Nikāya.¹⁸⁶

Caution needs to be exercised, however, regarding the Ekottara Āgama,¹⁸⁷ which is nominally the collection corresponding to the Pali Aṅguttara Nikāya. Although it shares some significant structural features with the Aṅguttara, the content is often very different.¹⁸⁸ The text is highly erratic and internally inconsistent, possibly being an unfinished draft. Scholars agree that it includes proto-Mahāyānist additions,^{189 190} thereby establishing its late date of completion compared to the rest of the EBTs.

This high degree of correspondence among the EBTs across different lines of transmission does not exist for any other texts of the vast Buddhist corpus.¹⁹¹ Even in the stylistically oldest part of the Khuddaka Nikāya, such as the Sutta Nipāta,^{192 193} the Udāna,^{194 195} and the Dhammapada,^{196 197 198} there is substantial divergence between the schools. This is despite the fact that these texts do have a common core, which is found across the different traditions. With texts such as the Abhidhamma, despite a small common core,¹⁹⁹ the divergence is even greater.^{200 201 202} But the vast majority of Buddhist texts are exclusive to the individual schools and do not have any parallels at all.

3.1.2 Mahāsāṅghika comparisons

The first doctrinal split in the Buddhist monastic community was that between the Sthaviras and the Mahāsāṅghikas. Because this was the first split, some scholars, such as Edward Conze and A. K. Warder, have suggested that material that is common to both these groups of schools be regarded as the most authentic.²⁰³

We do not agree with this criterion, since, as Frauwallner has argued, “the foundation of communities and the rise of dogmatic schools are two quite separate things”.²⁰⁴ The separation of textual lineages can happen for any number of reasons apart from sectarian identity, prominent among them geographical isolation. The texts of the Pali school were separated from the Kaśmīri Sarvāstivādins and the Gāndhārī Dharmaguptakas at least as long as the separation from the Mahāsāṅghikas.²⁰⁵ Moreover, the physical separation was even greater, at about 3,000 km.

Still, since the split between the Sthaviras and the Mahāsāṅghikas was the first schism motivated by doctrinal differences, it is worthwhile to consider the material in common between them. This research is hampered by the fact that there is little surviving early material from the Mahāsāṅghikas. It is therefore unrealistic to insist that only material common to the two schools be accepted as authentic. Nevertheless, judging from their account of the First Council²⁰⁶ and other indications, the Mahāsāṅghika did maintain a version of the EBTs, and specific Mahāsāṅghika literature emerged only later. Moreover, the Mahāsāṅghikas are known to have been particularly conservative in at least some respects, for example in rejecting as Canonical texts that were included in the Canon by other schools.²⁰⁷

Known similarities between the EBTs of the Mahāsāṅghikas and those of other schools include the following:

1. The Mahāsāṅghika pātimokkha is essentially the same as all other extant pātimokkhas.²⁰⁸

2. The canonical Vinaya material that falls outside of the pātimokkha and its explication is on the whole similar in content in all schools,²⁰⁹ although the Mahāsāṅghika text is structured very differently.²¹⁰

3. The Vinaya of the Lokottaravādins, a branch of the Mahāsāṅghikas, which is partially available in Hybrid Sanskrit, contains significant passages in common with other early collections, notably the first bhikkhunī ordination.²¹¹

4. The Mahāsāṅghika account of the First Council names four Āgama collections, which parallel the first four Pali Nikāyas, as well as indicating some of the content of its Saṁyukta Āgama. It names all the sections of the pātimokkha, as well as other Vinaya matters and the class of thullaccaya offences.^{212 213}

5. Fragments of the Mahāsāṅghika Mahāparinirvāṇa and Caṁgi (Pali: Caṅki) Sūtras have been identified in manuscripts dating to the 3rd–4th century ce.²¹⁴

6. The Mahāvastu is a Mahāsāṅghika text that narrates the life of the Buddha, largely in legendary form. Despite the elaborations not found in the EBTs, there are frequent references to early Buddhist doctrines. It also contains several dozens of Suttas, verses, and passages in a form little different from the EBTs.²¹⁵

7. The school of the Ekottara Āgama preserved in Chinese is problematic, although some scholars cautiously ascribe it to the Mahāsāṅghikas. In any case, while the text is erratic, inconsistent—both internally and with the EBTs as a whole—and contains much material that belongs to non-EBT literature, such as proto-Mahāyāna ideas,²¹⁶ most of the basic doctrine and many of the texts are similar to other EBTs.

8. The Śālistamba Sūtra is an early Mahāyāna Sūtra that is thought by Reat to derive from the Mahāsāṅghika school.²¹⁷ This Sutta contains many phrases and ideas on dependent origination that are found in identical terms in all the EBTs, while at the same time developing these ideas in new ways.²¹⁸ This too shows that the Mahāsāṅghikas had their own version of the EBTs, which was essentially the same as that of the other schools.

3.1.3 Later borrowing

Schopen suggests that the shared content of the EBTs may be due to later borrowing, levelling, and standardisation.²¹⁹ However, as recently shown by Anālayo, there is no evidence for this.²²⁰ At the same time, there is much evidence to the contrary. Moreover, the thesis of a shared origin is simpler, more natural, and more powerful than that of later borrowing.

Wynne has given some specific examples where later borrowing is very unlikely to have taken place, especially where incidental details are found shared between totally distinct types of texts or where the shared material, although irrelevant to any point of Buddhist doctrine, is counter to the usual Sutta presentation.²²¹ But the argument against later borrowing is not limited to a few specific examples.

Consider the Vinaya. Here we have a clearly demarcated text, where the pātimokkha is obviously older than the vibhaṅga material that surrounds it.^{222 223 224 225} The pātimokkha is almost identical across every school,²²⁶ while the vibhaṅga material is substantially different.^{227 228} If the later borrowing thesis were correct, one would not expect this difference in standardisation between the pātimokkha and the vibhaṅga. Thus the “shared origin” thesis is confirmed.

This is not just a characteristic of the Vinaya, but applies to every Buddhist text we have available for comparison. The Abhidhammas of various schools, to give another example, diverge far more than the EBTs, but what they have in common is especially those passages that quote the EBTs.²²⁹

Moreover, we have many examples of actual borrowing among non-EBTs and these compare very differently from parallel texts among the EBTs. For instance, the Buddha legend shows many signs of shared borrowings between the schools, but these are borrowings of ideas, episodes, and motifs, not the massive sharing of thousands of nearly identical texts, which would be required to explain the existence of the EBTs by means of borrowing.

Likewise, the Jātakas, and many other Buddhist stories, frequently appear to have been borrowed across the traditions, but the resulting similarities are quite different from those within the EBTs.

In the realm of philosophy, we also find borrowing. For example, key Abhidhammic ideas such as svabhāva/sabhāva, svalakṣaṇa/salakkhaṇa, and sāmānyalakṣaṇa/sāmaññalakkhaṇa are shared between the Pali, Sarvāstivādin, and other Abhidhamma systems, and are defined in similar ways.²³⁰ Yet the texts and philosophies in which the term appears differ widely depending on the school.

There are, therefore, many well-known examples of later borrowing between the schools, and the results of such borrowings are all quite different from the sort of uniformity we see between the EBTs.

3.2 The reliability of the oral tradition

For several hundred years, from the time that separate transmission lineages emerged in the Asokan period until the texts were written down, the EBTs were passed down orally in separate textual lineages. Comparative studies have shown that this oral transmission was highly reliable and that the core doctrinal material was essentially unchanged. How did this work, given what we know about the unreliability of memory? Indian culture provided the template for highly reliable oral preservation. It is known that the Ṛg Veda and other Vedic texts were transmitted orally—that is, by memory—with extreme accuracy for over two thousand years.^{231 232}

In his study of the Majjhima Nikāya, Anālayo considers the impact of oral transmission and concludes: “At the same time, rather than giving us a completely new picture of early Buddhism, what my comparative study of the parallels to the Majjhima Nikāya discourses yields is a reconfirmation of the essentials, with occasional divergence in details.”²³³

In the field of oral literature, there is a distinction between texts that are to be memorised verbatim, as in the Buddhist and Vedic traditions, and those which serve as springboards for storytelling, as in oral folk traditions. The latter are subject to natural evolution and variation; they are meant to adapt to the teller and the situation.²³⁴ Such, according to Oldenberg, are also found in Buddhist literature, specifically in the Jātaka collection of the Khuddaka Nikāya.²³⁵ But the former, which are codified and fixed texts, are meant to be preserved in exact form, as in the case of the EBTs.^{236 237}

The Indian oral culture developed various methods to ensure that this was achieved. Such methods pervade every aspect of the EBTs, and include:²³⁸

1. Repetitions of words, phrases, passages and whole Suttas;

2. Standardisation of words, phrases and passages;

3. The use of synonyms;

4. The use of the waxing syllable principle;²³⁹

5. Sound similarities;

6. Concatenation of Suttas or other textual units;²⁴⁰

7. Formal structures, especially aba;²⁴¹

8. “Summary” and “exposition”, which is a standard feature of Indian oral education;

9. Framing narratives to define the limits and give the context for the spoken material;

10. Verse summaries of prose teachings (especially in the Aṅguttara);

11. Similes (usually in an aba structure);

12. Numbered lists;

13. Group recitals.²⁴²

3.3 A democracy of conservatism

Unlike most religious organisations, the Sangha has traditionally not had a clear and established hierarchy, and after the Buddha there has never been a person of final authority, a sole arbiter of power.²⁴³ Instead, the Sangha has been decentralised and the texts were passed down through group recital (saṅgīti). This is the method recommended by the EBTs themselves for preserving the teachings.²⁴⁴

This means that no individual, or sub-Sangha group, would have been able to exert much influence on the content of the texts.^{245 246 247} It is thus improbable that the doctrinal core of the EBTs was deliberately changed.²⁴⁸

The differences between the recensions generally stem from editorial choices in the organisation and structure of the texts, as well as occasional inconsequential errors that arose during oral transmission.²⁴⁹ Doctrinal development, by contrast, happened by way of new texts and interpretations, and were not due to deliberate alterations of the EBTs.

3.4 How to make it up

There are a couple of passages in the Vinayas that give instructions on how to add missing material.²⁵⁰ For example, if the city where a discourse was spoken is unknown, it should be said that it took place in one of the six great cities mentioned in the EBTs.

This merely confirms what was known to Buddhist scholars already: that many of the incidental details in the texts have arisen from an artificial standardisation. This incidental detail, the framing narrative, must in all cases have been added after the Sutta was spoken.²⁵¹

This is confirmed by the tradition that at the First Council Mahākassapa asked Ānanda about the location, subject matter, protagonist and background story for each Sutta. These questions would have been superfluous if the Suttas already contained the narrative material. And the new material was not added to the substance of the text.²⁵²

The purpose of adding this material was to preserve the text by creating a “container” that gave the text a clear and separate identity.²⁵³ Schopen’s representation of this process as “If You Can’t Remember, How to Make It Up: Some Monastic Rules For Redacting Canonical Texts” is grossly misleading.²⁵⁴

3.5 The age of Pali

Early generations of Indologists accepted the traditional  Theravādin claim that Pali was the language of the Buddha. More recent linguistic studies, however, show that Pali is in part an artificial language, created some time after the Buddha.²⁵⁵ However, the differences in language between Pali and the historically datable language of the Asokan pillars are no more than minor phonetic changes that rarely, if ever, affect the meaning of the content. In all probability a similarly close relationship obtains between Pali and the historical Buddha’s own dialect.^{256 257 258}

Moreover, the Buddha himself may have used varying dialects depending on where he travelled,^{259 260} and certainly his disciples did: they were in fact encouraged to speak in their own dialect.²⁶¹ As Buddhism spread throughout Northern India, this diverging use of language must eventually have led to a need for standardisation, and this probably explains the introduction and development of Pali. Alternatively, or perhaps complementarily, Pali is related to a particular dialect in India from where the missions to Sri Lanka originated. A number of scholars are of the opinion that the West-Indian dialect associated with the Asokan rock edicts at Girnār and Bombay-Sopārā are closely related to Pali.^{262 263} The precise age of Pali, therefore, does not have any bearing on the age of the contents of the EBTs.

Finally, there are no certain traces of Sinhalese influence on the Pali EBTs.^{264 265} This suggests that the EBTs were in a standardised form when they arrived in Sri Lanka around the time of Asoka and that they are unlikely to have been changed after this. As Wynne says: “If the language of the Pali canon is north Indian in origin, and without substantial Sinhalese additions, it is likely that the canon was composed somewhere in north India before its introduction to Sri Lanka, and is therefore a source for the period of Buddhism in Northern India before this”,²⁶⁶ that is, before Asoka. In contrast, Pali texts actually composed in Sri Lanka do show influence from both Sinhalese and Dravidian.²⁶⁷

3.6 Manuscripts

The earliest Indian manuscripts are the Gāndhārī Buddhist texts from Afghanistan, the earliest of which date to the 1st century ce.^{268 269} These contain a Dhammapada, discourses in verse (for example the Rhinoceros Sutta) and discourses in prose (including several Saṁyutta and Aṅguttara style texts), as well as non-EBT texts, such as Avadānas and Pūrvayogas, commentaries and Abhidharma texts.

This pattern of textual distribution conforms to that expected of the period immediately after the beginning of the common era. Many of the texts belong to the EBTs, and those that have been studied in detail show a close affinity with other EBTs.²⁷⁰ The remainder of these texts are non-EBTs that form part of the textual development immediately based on the EBTs.

3.7 Scholarly opinion

Scholars thrive on disagreement. Nevertheless, among those who specialise in this field, there has been, since the beginnings of modern Buddhist studies, a rough working consensus that at least part of the EBTs are authentic. Academic sceptics of authenticity are almost always specialists in later Buddhism, making comments outside their own field. The following are some examples of the views of historical scholars of Buddhism.

1. Rupert Gethin: “Far from representing sectarian Buddhism, these Suttas [the four main Nikāyas] above all constitute the common ancient heritage of Buddhism”²⁷¹ and “it is extremely likely that at least some of these Suttas that come down to us are among the oldest surviving Buddhist texts and contain material that goes back directly to the Buddha.”²⁷²

2. Richard Gombrich says “the content of the main body of sermons, the four Nikāyas and of the main body of monastic rules … presents such originality, intelligence, grandeur and—most relevantly—coherence, that it is hard to see it as a composite work.” They are “the work of one genius”, the Buddha.²⁷³

3. Peter Harvey also affirms that “much” of the Pali Canon “must derive from his [the Buddha’s] teachings.”²⁷⁴

4. J. W. de Jong has said it would be “hypocritical” to assert that we can say nothing about the teachings of earliest Buddhism, arguing that “the basic ideas of Buddhism found in the canonical writings could very well have been proclaimed by him [the Buddha], transmitted and developed by his disciples and, finally, codified in fixed formulas.”²⁷⁵

5. Lamotte says that “Buddhism could not be explained unless we accept that it has its origin in the strong personality of its founder.”²⁷⁶

6. Hajime Nakamura writes that while nothing can be definitively attributed to Gotama as a historical figure, some sayings or phrases must derive from him.²⁷⁷

7. Noble Ross Reat: “Though the Pali Suttas obviously exaggerate and mythologize the qualities and deeds of the Buddha, with regard to his teachings they have every appearance of constituting for the most part a faithful and reasonably accurate record.”²⁷⁸

8. A. K. Warder has stated that “there is no evidence to suggest that it [the shared teaching of the early schools] was formulated by anyone other than the Buddha and his immediate followers.”²⁷⁹

9. Maurice Winternitz: “Speeches… which again and again recur word for word in the same form not only in many places in the Pali canon but also in Buddhist Sanskrit texts… [and] Tibetan and Chinese translation, we may perhaps consider as originating from Buddha…”²⁸⁰

10. Alexander Wynne: “I therefore agree with Rhys Davids, and disagree with sceptics such as Senart, Kern, and Schopen, that the internal evidence of the early Buddhist literature proves its historical authenticity.”²⁸¹

Examples of critics of authenticity, who are specialists in later Buddhism, include:

1. Edward Conze states that the confident attempts of European scholars to reconstruct the original teachings of the Buddha himself are “all mere guesswork”.²⁸²

2. Ronald Davidson has little confidence that much, if any, of surviving Buddhist scripture is actually the word of the historical Buddha.²⁸³

3. Karen Lang suggests that there is no way of knowing how closely the Canon written down in the last century bce resembles that of the present day.²⁸⁴

4. Kogen Mizuno suggests that the Suttas as we possess them may not contain the Buddha’s literal words, since they were not recorded as he spoke.²⁸⁵

5. Geoffrey Samuel says the Pali Canon largely derives from the work of Buddhaghosa and his colleagues in the 5th century ce.²⁸⁶

6. Gregory Schopen argues that it is not until the 5th to 6th centuries ce that we can know anything definite about the contents of the Canon.²⁸⁷

3.8 Stability of consensus

We cannot hope to represent the wide range of scholarly views over the past 100–150 years. Instead, we selectively quote some of the most distinguished early scholars in the field, and in the process hope to present a reasonable range of views of 19th century Indologists, showing that these tend to affirm the authenticity of the EBTs.

1. In 1882 Samuel Beal stated that the texts of the various schools were “founded on one and the same traditional record”.²⁸⁸

2. Beal said in 1884 that the pātimokkha “is the oldest, and in many respects the most important, material of the vinaya literature.”²⁸⁹ This position was the same as that taken by Rhys Davids and Oldenberg in 1879.²⁹⁰

3. Beal: “It is clear they [the Mahāyāna texts, specifically the Prajñā Pāramitā] cannot belong to the early school or have formed a part of the original collection of books.”²⁹¹

4. Dr. Hofrath Bühler said that the Pali Nikāyas are good evidence for the fifth or sixth century bce.²⁹²

5. Wilhelm Geiger: “But without a doubt it [the canon] contains a mass of utterances, speeches and teachings of the Master, as they were impressed on the memory of the disciples in their more or less accurate form.”²⁹³

6. Oldenberg argued that the pātimokkha and the “fundamental doctrines of the Dhamma” were from the same period, whereas the rest of the Vinaya was later,²⁹⁴ and that the Abhidhamma was later than the Sutta and the Vinaya Piṭakas.²⁹⁵

7. Oldenberg maintained that all “versions of the vinaya are based upon one foundation”,²⁹⁶ and said: “Hence our opinion of the early origin of the Buddhist texts [the EBTs], based as it is on external proofs, does not clash, but agrees with the internal possibility and probability.”²⁹⁷

8. Pachow, referring principally to Oldenberg and Kern, says: “It has been admitted by all eminent scholars that the Pmk [Prātimokṣa] is one of the oldest texts of the Buddhist Canon.”²⁹⁸ E.J. Thomas agrees.²⁹⁹

9. T.W. Rhys Davids: “Their composition, as to the Vinaya and the four Nikāyas (with the possible exception of the supplements),³⁰⁰ was complete within about a century of the Buddha’s death.”³⁰¹

10. T.W. Rhys Davids, referring to inscriptions on ancient stupas that mention suttantika, suttantakinī, pañcanekāyika, dhammakathika, and peṭakī, says these terms constitute “conclusive proof of the existence some considerable time before the date of the inscriptions, of a Buddhist literature called either a Piṭaka or the Piṭakas, containing suttantas, and divided into Five Nikāyas.”³⁰²

Having reviewed the conclusions of some of the most noted scholars, it seems clear that, despite difference in details, many of the basic insights of the first generation of 19th century Indologists remain valid. This is despite the notable advances and changes in understanding due to expanded sources and clearer analysis. The situation is comparable to that in biology. Just as the insights of Darwin, though often refined and modified, still provide the framework for modern biology, so this rough consensus extends beyond these general statements and provides the framework within which modern Buddhist scholars continue to work.

Finally, let us look at the more specific findings of one of the most noted pioneers of Buddhist studies. Here is a chronology of Pali Buddhist texts from T. W. Rhys Davids in the appendix to chapter 10 of Buddhist India.³⁰³

Chronological table of Buddhist literature
from the Buddha’s time to the time of Asoka

1. The simple statements of Buddhist doctrine now found, in identical words, in paragraphs or verses recurring in all the books.

2. Episodes found, in identical words, in two or more of the existing books.

3. The Sīlas, the Pārāyana, the Octades, the pātimokkha.

4. The Dīgha, Majjhima, Aṅguttara, and Saṁyutta Nikāyas.

5. The Sutta Nipāta, the Thera- and Therī-Gāthās, the Udānas, and the Khuddaka Pāṭha.

6. The Sutta Vibhaṅga and the Khandhakas.

7. The Jātakas and the Dhammapadas.

8. The Niddesa, the Itivuttakas, and the Paṭisambhidā.

9. The Peta- and Vimāna-Vatthus, the Apadānas, the Cariyā Piṭaka, and the Buddha Vaṁsa.

10. The Abhidhamma books; the last of which is the Kathā Vatthu, and the earliest probably the Puggala Paññatti.

This is obviously meant as a simple aid to orientation among the material, not as definitive or exhaustive. In an excellent example of the friendly and critical debates that characterised this period of early Indology, the list was swiftly criticised and nuanced by Bimala Law.³⁰⁴ Nevertheless, if we allow for the limitations of such a simple presentation of such a complex subject, his sequence holds up very well. The main changes we would suggest is to place the Itivuttaka with the Udāna and other texts at number 5, and to allow that the Abhidhamma was compiled alongside the other texts, probably from number 6 on, or at least from number 8.

Chapter 4
Character of the Early Buddhist Texts

Most of the EBTs state explicitly that the Buddha is their author. The character of the EBTs is most easily accounted for if this is, on the whole, true. There are obvious exceptions, such as where the speaker is specifically said to be someone else, or where there are indications that texts or passages have been added. These exceptions, however, should not be allowed to distract from the overall picture.³⁰⁵

The EBTs are stylistically different from other Buddhist texts, as well as internally consistent in both doctrine and phrasing. They are a separate genre of literature.³⁰⁶ This is most simply explained if they hark back to a common source in the earliest period. They have a different authorship from other Buddhist texts, and are the root texts from which the rest of Buddhist literary activity developed.

4.1 Vedic influence on the EBTs

The EBTs frequently bear the stamp of influence from Brahmanical literature in their literary style. The most obvious is the poetry, where we find that the metres are developed from Vedic precedent.^{307 308} Likewise, the characteristic feature of framing narratives is derived from the Vedas.³⁰⁹ In the Vedas we also find the models for such organising principles as the Saṁyutta principle of grouping texts by topic,³¹⁰ and the Aṅguttara principle of grouping them according to number.^{311 312} The EBTs frequently share metaphors and imagery with the Vedic literature. Indeed, we can point to several shared similes in just one Upaniṣadic passage, the dialogue between Yājñavalkya and his wife Maitreyī: the origin of the sound of the conch or the lute, (dn 23.19/dā 7/mā 71/T 45 vs. Bṛhadāraṇyaka Upaniṣad 2.4.7–9), the rivers that merge in the ocean (an 8:19 vs. Bṛhadāraṇyaka Upaniṣad 2.4.11), and the ocean that everywhere has one taste, the taste of salt (an 8.157 vs. Bṛhadāraṇyaka Upaniṣad 2.4.11). Finally, the EBTs borrowed a significant number of Brahmanical terms and ideas, often infusing them with a new meaning.³¹³ All of these ideas and forms were invariably adapted to a Buddhist use.

This shows that the EBTs adopted aspects of the Brahmanical model, which at that time was the only available example of a successful and lasting literary tradition. This is despite the enormous differences between the EBTs and the Vedas.^{314 315} All other forms of Buddhist texts, by contrast, are derived primarily from earlier forms of Buddhist literature.

4.2 Literary features

In his book Buddhist India, T.W. Rhys Davids says:

On the other hand, suppose a MS. were discovered containing, in the same handwriting, copies of Bacon’s Essays and of Hume’s Essay, with nothing to show when, or by whom, they were written; and that we knew nothing at all otherwise about the matter. Still we should know, with absolute certainty, which was relatively the older of the two; and should be able to determine, within a quite short period, the actual date of each of the two works. The evidence would be irresistible because it would consist of a very large number of minute points of language, of style, and, above all, of ideas expressed, all tending in the same direction.

This is the sort of internal evidence that we have before us in the Pali books. Any one who habitually reads Pali would know at once that the Nikāyas are older than the Dhamma Sangaṇi; that both are older than the Kathā Vatthu; that all three are older than the Milinda. And the Pali scholars most competent to judge are quite unanimous on the point, and on the general position of the Pali literature in the history of literature in India.

But this sort of evidence can appeal, of course, only to those familiar with the language and with the ideas.”³¹⁶

We will now have a look at some of the stylistic features that make the EBTs stand apart from other Buddhist texts.

4.2.1 Grammar

The geographical distribution of language features for the time shortly after the Buddha is known primarily from the Asokan edicts.³¹⁷ It is generally accepted that Pali has its roots in Western India in the region of Avanti,^{318 319} which is on the trade route to Sri Lanka.

At the same time, it has been shown that the vocabulary of the Pali EBTs preserves several early dialectical forms (Magadhisms)³²⁰ that are generally regarded as linguistic remnants from the time when Buddhism was geographically limited to the Eastern part of India.^{321 322}

These features are not found in other Buddhist literature, with the exception of the Kathāvatthu, which may have these forms due to its probable origin in Magadha.³²³

4.2.2 Vocabulary

The vocabulary of the EBTs is different from that of other Buddhist texts. The easiest way to see this difference is to compare passages in the EBTs with parallel passages in other Buddhist literature. Such comparison suggests that the other texts belong to a later chronological stratum than the EBTs. There are countless examples of this, one of which is the standard formula for satipaṭṭhāna as found in the EBTs compared with that found in the Vibhaṅga of the Abhidhamma.

1. The standard satipaṭṭhāna formula of the Suttas is as follows (e.g. mn 10.3):

   Idha, bhikkhave, bhikkhu kāye kāyānupassī viharati ātāpī sampajāno satimā, vineyya loke abhijjhādomanassaṁ.

2. The satipaṭṭhāna formula of the Suttanta-bhājanīya (“Analysis of the Sutta”) of the Vibhaṅga is as follows (Vibh 193,1):

   Idha bhikkhu ajjhattaṁ kāye kāyānupassī viharati bahiddhā kāye kāyānupassī viharati ajjhattabahiddhā kāye kāyānupassī viharati ātāpī sampajāno satimā vineyya loke abhijjhādomanassaṁ.

3. And the satipaṭṭhāna formula found in the Abhidhammabhājanīya (“Abhidhamma analysis”) of the Vibhaṅga is as follows (Vibh 203,1):

   Idha bhikkhu yasmiṁ samaye lokuttaraṁ jhānaṁ bhāveti niyyānikaṁ apacayagāmiṁ diṭṭhigatānaṁ pahānāya paṭhamāya bhūmiyā pattiyā vivicceva kāmehi …pe… paṭhamaṁ jhānaṁ upasampajja viharati dukkhapaṭipadaṁ dandhābhiññaṁ kāye kāyānupassī, yā tasmiṁ samaye sati anussati sammāsati satisambojjhaṅgo maggaṅgaṁ maggapariyāpannaṁ—idaṁ vuccati “satipaṭṭhānaṁ”. Avasesā dhammā satipaṭṭhānasampayuttā.

The first formula is the standard Sutta formula, whereas the second one constitutes the earliest development of the Abhidhammma.³²⁴ It is only at the third stage that we find the formula in fully fledged Abhidhamma style.

The formula in stage 2 is virtually identical with the Sutta formula, and in fact the formula is found in this form in some non-Pali EBTs.³²⁵ This close relationship is to be expected since this constitutes an analysis according to the Suttas. Yet further on in the same Sutta analysis typical Abhidhamma vocabulary, such as svāvatthitaṁ and vavatthāpeti, is introduced (Vibh. 193,24). We seem to be witnessing the beginning of the Abhidhamma method.

Stage 3 has the analysis according to the Abhidhamma. Here the whole satipaṭṭhāna formula is completely transformed and we see a range of vocabulary that is either unknown to the EBTs or used in a new sense: yasmiṁ samaye, lokuttaraṁ jhānaṁ, paṭhamāya bhūmiyā, avasesā dhammā satipaṭṭhānasampayuttā, and so on.

This is just one example of a pattern that plays out, not just here, but in every major doctrinal context or formula.

4.2.3 Metre

Metre—the rhythmic structure of verse—is constantly changing, and thus it is possible to identify the type of metre popular for any particular historical period. By careful metric analysis, it is possible to arrive at the approximate period in which a particular poem must have been composed.

It is sometimes suggested that certain verses found in the EBTs are particularly ancient due to archaic words and grammatical features.³²⁶ However, according to the late A. K. Warder, perhaps the world’s leading authority on Pali verse, archaic word forms in verse often are deliberate rather than showing the true chronological stratum to which the text belongs.³²⁷ Metrical analysis gets us around this problem and promises a more direct relative dating of Pali verse.

One of the earliest forms of the Pali metre, known as vatta, is close in structure to the vatta metre as used in the Upaniṣads, but chronologically slightly later. However, the same metre is earlier than that of the Bṛhaddevatā and still earlier than that of the Mahābhārata.³²⁸

According to Warder, “It seems certain that the verse parts of the Pali Canon represent quite a long period of composition”,³²⁹ and that “The differences of age thus suggested seem to agree with the general drift of subjective opinion on which sort of doctrine is earlier and which later.”³³⁰ But he also says that, “Although we cannot say whether any Pali verses in their present form date back to the time of the Buddha (i.e. circa 500 bce), on the one hand the changes in techniques within the Canon imply a considerable period of development prior to the 2nd century bce, and on the other hand the formal similarities between the Canon and those of the other early schools indicate a common origin of the original ‘kernels’ in a period before the sectarian divisions had separated them too far.”³³¹

Within the Pali canon, the metrically oldest verses are found in the Aṭṭhaka Vagga and the Pārāyana Vagga of the Sutta Nipāta.³³² Metrical analysis also shows that the Pārāyana Vagga can be further divided into a core of very ancient poems and the surrounding narrative framework.³³³ This fits with our general observation, and also with the results of comparative study, that the narratives are later than the actual words attributed to the Buddha.

4.2.4 Style

The EBTs are generally realistic and restrained in their portrayal of the Buddha and his environment, and the details do not seem unreasonable for what we know of the historical period and geographical area. For instance, the Buddha is rarely portrayed as displaying supernormal powers, and when he is they often have the hallmarks of being later insertions.³³⁴

Non-EBTs, by contrast, are often florid and full of supernormal phenomena. This is true for the entire Buddhist tradition, but especially so for the Mahāyāna Sūtras. The Mahāyāna Sūtras generally lack the realism of the EBTs.^{335 336}

This difference can be appreciated by considering the opening narrative of a typical text in the Majjhima Nikāya compared to the opening of a classical Mahāyāna text from the Prajñā-pāramitā corpus. A typical Sutta in the Majjhima Nikāya starts as follows:

Thus have I heard. On one occasion the Blessed One was living at Sāvatthī in Jeta’s Grove, Anāthapiṇḍika’s Park. There he addressed the bhikkhus thus: …”

A short extract (about 10%) of the opening of the the Large Prajñā-pāramitā Sūtra, according to Conze’s translation,³³⁷ reads as follows:

Thus have I heard. At one time the Lord dwelt at Rājagṛha, on the Vulture Peak, together with a large gathering of monks, with 1,250 monks, all of them Arhats— their outflows dried up, undefiled, fully controlled, quite freed in their hearts, well freed and wise, thoroughbreds, great Serpents, their work done, their task accomplished, their burden laid down, their own weal accomplished, with the fetters that bound them to becoming extinguished, their hearts well freed by right understanding, in perfect control of their whole minds—with 500 nuns, laymen, and laywomen, all of them liberated in this present life—and with hundreds of thousands of niyutas of koṭis of Bodhisattvas.[³³⁸] … Thereupon the Lord on that occasion put out his tongue. With it he covered the great trichiliocosm and many hundreds of thousands of niyutas of koṭis of rays issued from it. From each one of these rays there arose lotuses, made of the finest precious stones, of golden colour, and with thousands of petals; and on those lotuses there were, seated and standing, Buddha-frames demonstrating dharma, i.e. this very demonstration of dharma associated with the six perfections. They went in all the ten directions to countless world systems in each direction, and demonstrated dharma to beings, i.e. this very demonstration of dharma associated with the six perfections. And the beings who heard this demonstration of dharma, they became fixed on the utmost, right and perfect enlightenment.”

This is an entirely different kind of literature, one which is completely uninterested in history; so much so that it flaunts its creativity. The purpose of the introduction is no longer to identify time and place, but to lift the reader out of history to a timeless realm of Dharma.

Differences in style manifest in other ways too. Rhys Davids comments as follows on the stylistic difference between earlier and later works in the Theravāda tradition:

The philosophic ideas of the ancient Buddhism, and the psychological ideas on which they were based, were often curtly, naively, confusedly expressed. In Ceylon they had been much worked up, polished, elucidated, systematised. From several works now accessible we know fairly well the tone and manner of these later—and, as they must have seemed to Ceylon scholars, clearer, fuller—statements of the old ideas. In no single instance yet discovered has this later tone and manner found its way into the canonical books .”³³⁹

4.2.5 Oral vs. literary tradition

The EBTs depict the Buddha as being concerned about the long-term preservation of his message (dn 16.3.50/smps 19.7, dn 29.17), as saying that the EBTs would preserve the Dhamma (dn 16.3.50/smps 19.8–9),³⁴⁰ and as giving guidance on how to do this (dn 16.4.7–11/smps 24.2–52, mn 103.4–8). The mnemonic features of the EBTs (see section 3.2 above) serve precisely the purpose of ensuring the long-term reliable transmission of the texts.

These features are not incidental or added on, they are intrinsic to the EBTs. Let us consider one of these features, repetition. If repetition was a conscious device used by later redactors to ensure fidelity, but not part of the Buddha’s style, we would expect to find repetition only in those aspects of the texts that the redactors controlled. For example, we would find the same Sutta repeated in multiple collections. However, what we actually find is that repetition is prevalent at every level of the texts, whether the word, the phrase, the sentence, the paragraph, the passage, or the whole discourse. Repetition, like the other mnemonic features of the texts, is a characteristic of their composition, a characteristic that was further developed by the redactors. Since the EBTs were composed in the oral tradition, the mnemonic characteristics of the texts originated in the historical Buddha’s teaching style.

Most non-EBTs do not have such mnemonic features. This suggests that these texts were composed specifically for written rather than oral transmission. This would have been some time after the EBTs, when writing was in common use for Buddhist texts.³⁴¹

The exception to this is the later strata of canonical material, such as the Abhidhamma, much of the Vinaya, and some of the Suttas, especially in the Khuddaka Nikāya, but also some of the material in the main Nikāyas/Āgamas. This material retains the features of the oral tradition, yet on multiple other grounds it is clearly later than the EBTs. It would, of course, have taken considerable time for the literary style to emerge fully from the oral culture, just as the earliest builders in stone for a long time employed principles of wood work, or just as digital designers now employ skeuomorphism. Even the Mahāyāna Sūtras, which were composed in writing and which frequently refer to writing,^{342 343} retain some oral features such as repetition.³⁴⁴ This shows that the oral tradition was firmly established and must have flourished for a considerable period before the texts were put in writing.

4.3 Flavour of a single creator

This can be seen in a number of aspects of the EBTs, such as the large number of similes, analogies and metaphors that are vivid, precise in application, realistic and local, and formal in presentation;³⁴⁵ the analytical approach to language, which was unknown before the Buddha; use of irony and humour;³⁴⁶ and internal consistency and coherence.³⁴⁷ Moreover, many of the ideas presented in the EBTs are revolutionary for the time. This distinctive personal style is quite different from anything found in other Buddhist literature, or even in the Upaniṣads.

4.3.1 Consistency and coherence

The EBTs are characterised by a rigorous consistency in doctrinal teachings,³⁴⁸ together with a great variety in themes, contexts, settings and presentations. Whether there are any significant doctrinal contradictions is doubtful, and in any case there are far less than in any comparable ancient literature.

It is not easy to quantify this, but comparison with the contemporary hegemonic literature, the Upaniṣads, is instructive. Though the Upaniṣads compass a much smaller quantity of texts, they have far more diversity and contradiction in doctrines. Choosing two topics at random, here is a comparison between the EBTs and the earliest and largest Upaniṣad, the Bṛhadāraṇyaka. We believe that a more detailed study along similar lines would show that this kind of difference is systemic.³⁴⁹

1. Creation: The EBTs consistently assert that it is not possible to know a first point (agga) of creation.³⁵⁰ In the discourse “On Knowledge of Beginnings” (dn 27/dā 5/T 10/mā 154/eā 40.1) the Buddha starts his creation myth by saying, “There comes a time when the world ends …”. In the Bṛhadāraṇyaka there are six creation myths, which say that in the beginning there was either nothing (1.2.1), self (1.4.1, 1.4.17), Brahmā (1.4.10, 1.4.11), or water (5.5.1).³⁵¹

2. Senses: The EBTs always list six senses (the regular five plus the mind) in the same sequence, and they never mix the senses with other things. The Bṛhadāraṇyaka offers no less than thirteen different lists of the senses, usually mixed with such things as breath and voice. They almost always vary in both the contents of the list and the sequence.³⁵²

4.3.2 Internal cross-references

The EBTs frequently refer to doctrinal categories, verses, similes, specific teachings, and even to full discourses found elsewhere in the EBTs. When such references are made, the meaning is only transparent for someone who knows the passage referred to. Moreover, although there is extensive internal cross-referencing of this kind, there are no references to other Buddhist texts.

This shows that throughout Buddhist history, the EBTs have been regarded as a separate genre of texts, and that they were seen as an integrated body of teachings that constituted the Buddha’s message. Other Buddhist texts, by contrast, refer both to the EBTs and other texts.³⁵³

The amount of internal cross-referencing in the EBTs is truly staggering and only a thorough computer analysis might be able to uncover the full extent of it. Here we can only hope to give a taste of its extent:

1. Named collections of Suttas: Aṭṭhaka Vagga (Snp ch. 4, cited at sn 22:3 and Vin i 196,36³⁵⁴); Pārāyana Vagga (Snp ch. 5, cited at sn 12:31, an 3:32, an 3:33, an 4:41, an 6:61, and an 7:53).

2. Named Suttas: Ajitapañha/Ajitamāṇavapucchā (Snp 5:1, cited at sn 22:31); Brahmajāla (dn 1/dā 21/T 21/P 1021, cited at sn 41:3); Dhammacakka (sn 56:11/sā 379/T 109/T 110/eā 19.2/P 747/P 1003, cited at sn 8:7, mn 141.2, mn 26.25/mā 204, an 5:133, an 4:118, an 3:14); Kumāripañha (sn 4:25/sā 1092/sā² 31, cited at an 10:26); Mahāpañha (an 10:27/eā 46.8, cited at an 10:28 and sn 41:8); Māgaṇḍiyapañha (Snp 4:9, cited at sn 22:3); Metteyyapañha/
   Tissametteyyamāṇavapucchā (Snp 5:2, cited at an 6:61); Pañcālacaṇḍa (sn 2:7/sā 1305/sā² 304, cited at an 9:42); Puṇṇakapañha/Puṇṇakamāṇavapucchā (Snp 5:3, cited at an 3:32 and an 4:41); Sakkapañha (dn 21/dā 14/T 15/mā 134, cited at sn 22:4); Udayapañha/Udayamāṇavapucchā (Snp 5:13, cited at an 3:33).

3. Named doctrinal categories:³⁵⁵ four noble truths (dn 9.33, mn 77.14/mā 207, sn 48:8, an 5:15); the sets of dhammas that came to be known as the thirty-seven bodhipakkhiyā dhammā (dn 16.3.50/smps 19.9, dn 28.3, dn 29.17, mn 103.3, mn 104.5, mn 118.13, mn 149.10/sā 305/D 4094, mn 151.14, sn 22:81, sn 22:101, sn 43.5–11, sn 45:155, an 7:71, an 8:19, an 8:28, an 10:90); noble eightfold path (dn 16.5.27, mn 33.11, an 4:33, an 4:34, an 4:35, an 5:32, an 5:196, an 10:28); path (dn 3.2.21, dn 5.29, dn 16.6.5, mn 56.18, mn 91.36, mn 104.5, sn 8:7, an 3:125, an 4:25, an 4:76, an 8:13, an 8:21); seven factors of awakening (dn 28.2, an 1:74, an 6:57, an 10:28, an 10:61, an 10:95); five faculties (mn 70.13, an 6:55, an 7:56, an 10:28); four establishments of mindfulness (dn 28.2, mn 12.62, mn 33.12, mn 44.12, sn 22:80, sn 48:8, sn 48:11; an 5:15, an 5:136, an 6:57, an 10:28, an 10:61, an 10:95); four right efforts (mn 44.12, sn 4:4, sn 48.8, sn 48.11, an 5:15); four bases of psychic potency ( dn 16.3.3, sn 4:20, an 8:70); five hindrances (dn 3.2.21, dn 14.1.13, dn 16.1.17, dn 28.2, mn 56.18, mn 91.36, mn 151.10, sn 8:3, sn 46:54, sn 47:12, an 4:65, an 6:45, an 6:57, an 8:12, an 10:61, an 10:95); three kinds of good conduct (an 10:61); four factors of stream entry (sn 48:8, an 5:15); five aggregates (mn 9.15, mn 151.11, sn 4:16, sn 35:245, sn 56:11, an 3:61, an 4:254, an 5:30, an 6:63, an 10:27); dependent origination (mn 26.19, mn 98.13, sn 22:57); five lower fetters (sn 47:29, sn 47:30, an 7:53, an 8:21); pātimokkha (e.g. dn 2, mn 6.2/mā 105/eā 37.5, mn 53.7, mn 108.14/mā 145, sn 47:56, an 4:12, an 8:20, smps 41.2).³⁵⁶

4. Named specific teachings:³⁵⁷ craving being a dart (mn 105.18); whatever is felt is included in suffering (sn 36:11); sexual intercourse destroys the bridge (an 4:159); affliction being suffering (an 9:34); the samādhi that has knowledge as its fruit (an 9:37); five sense objects are a confinement (an 9:42).

5. Named similes: the simile of the saw is given at mn 21/mā 193 and named at mn 28.9/mā 30; the similes of the skeleton, the piece of meat, the grass torch, the pit of coals, the dream, the borrowed goods, the fruits on a tree are given at mn 54.15–21/mā 203 and named at mn 22.3/mā 200 and an 5:76; the simile of the butcher’s knife and block is given at mn 23.4 and named at mn 22.3 and an 5:76; the simile of the sword stake is given at sn 5:1 and named at mn 22.3 and an 5:76.

6. Verses referred to: Ud 7:5, cited at sn 41:5.

4.3.3 Vivid & realistic

The EBTs are full of vivid details of Indian society of the day, such as:

1. Everyday activities: Arrow making (mn 101.28/mā 19), boat building (an 4:196), chariot riding (mn 21.7/mā 193), conveying of greetings (mn 90.3–4/mā 212/P 1030), craftsmanship (mn 77.31), dying (mn 7.2/mā 93/T 51), debt and interest payment (an 6:45), dice playing (dn 23.27), fishing (dn 1.3.72), garland making (mn 56.30), herding cows (mn 19.7, mn 19.12, mn 33.15, mn 34.4), making jewelry (an 3:102), making raft and crossing water (mn 22.13), making sesame oil (mn 126.15), milking cows (mn 126.16), preparing bath powder (mn 39.15), producing fire (dn 23.21, mn 36.17–19, mn 93.11), refining gold (mn 140.20, an 3:101), spitting (mn 152.7/sā 282), swimming (mn 64.8/mā 205), taming elephants (mn 125.12/mā 198, mn 125.23/mā 198), taming horses (mn 65.33/mā 194), treating a wound (mn 105.24), warfare (an 5:75, an 5:139), wheel making (mn 5.31/mā 87/T 49/eā 25.6, an 3:15).

2. Scenes from everyday life: Acrobat (sn 47:19), caravan of merchants (dn 23.23), chariot (mn 27.2), chicks hatching (mn 16.27), clothes and accessories (mn 54.3), daughter-in-law (mn 28.10), domestic scene (mn 87.5), elephants playing (mn 35.5), giving travel directions (mn 107.14), leper (mn 75.13), monkey catching (sn 47:7), oil lamp burning (mn 140.24), polite gesture (mn 88.7/mā 214), poor man (mn 66.11), protocol when king meets Buddha (mn 89.8/mā 213/eā 38.10), relay chariots (mn 24.14), rescuing infant (mn 58.7), rich man (mn 66.12), royal elephant (mn 66.7), snake catching (mn 22.10–11), torture (mn 129.4), trumpeter (mn 99.24), weight loss program (sn 3:13).

3. Environment: Animals (mn 50.13/mā 131/T 66/T 67, mn 66.8/mā 192, an 4:33), autumn sun (mn 46.22), creeper destroying a tree (mn 45.4/mā 174), fortress (an 7:67), lotuses (mn 26.21, mn 152.6), mountain lake (mn 39.21), park (mn 89.4), plantain tree (mn 35.22), sal tree grove (mn 21.8/mā 193), towns (sn 55:21), weather (an 3:35).³⁵⁸

These descriptions are often given in detail and appear to be accurate, so far as we can assess. Even small details, such as the mañjeṭṭhika (Vin ii 256,26), a disease of sugar cane, has been positively identified as the disease ‘red rot’, still common in the region today.³⁵⁹ It would have been hard, perhaps impossible, to reconstruct such details a long time after the event and especially in a different society. Such reconstruction would in all likelihood have included errors or incongruities which would be easily detectable. Moreover, this realistic description is totally different from the past as imagined in other Buddhist literature, which is full of miracles and magic and lacks grounding in everyday realism.

4.3.4 Innovation

The EBTs contain a number of ideas that were revolutionary for their time. Most significant among these was the teaching of non-self (anattā).³⁶⁰ Other important innovations included the causal chain of dependent origination³⁶¹ and the Buddhist version of the law of karma.³⁶² Further, the EBTs contain a large number of new doctrinal structures, such as the four noble truths and the eightfold path, as well as new analyses of the contents of experience, such as the scheme of the five aggregates. Finally, the EBT pātimokkhas appear to be a novelty in the history of Indian law.³⁶³ Altogether the EBTs are vastly different from other contemporary literature, especially the Vedas.

Non-EBTs contain very little innovation and are mostly concerned with filling in any perceived gaps in the EBTs, working out their consequences, and systematising them. The EBTs stand out as the only truly innovative layer of Buddhist texts. A natural explanation for this is that the EBTs were the result of the insights of one exceptional person.³⁶⁴

4.3.5 Claims of incoherence

An important challenge to our contention that the EBTs are coherent comes from those who have argued that Buddhism contains fundamental teachings that are hard to reconcile. Probably the most important of these arguments is the claim that Buddhism, specifically the Buddhism of the Pali sources, gives contradictory accounts of the goal of the Buddhist practice, including contradictory accounts of the path of meditation that leads to these goals.³⁶⁵

This is not the place to assess these claims in detail, but a few general remarks seem called for. A major problem with these claims, here exemplified by those of Griffiths,³⁶⁶ is that they often do not distinguish between EBT and non-EBT material. Griffiths says, “The canonical and commentarial literature will be treated here as a unity … because the thrust of this paper is structural and philosophical rather than historical, and for such purposes differentiation between canon and commentary is of small importance.” This is assuming a point that needs to be proved. In the absence of such proof, it is not possible to ascertain the coherence of the EBTs, or the lack thereof, by relying on non-EBTs. The EBTs need to be considered on their own merits.

Another problem with Griffith’s proposition is his reliance on a very limited number of texts from the EBTs. His main reference is to the Satipaṭṭhāna Sutta. However, in establishing any point about the EBTs one needs to consider the literature as a whole.³⁶⁷ It is our contention that the problems identified by Griffiths and others fall away once this is done.³⁶⁸

4.4 Contradictions and oddities are not normalised

In many cases there is blatant disagreement between statements found in the EBTs and those found in other Buddhist literature. Despite the glaring inconsistencies, the EBTs were not changed to ensure greater harmony.

4.4.1 Odd details & incongruities

In a number of places the EBTs depict odd or incongruous behaviour, or behaviour that does not fit with how Buddhism or the Buddha is portrayed in other Buddhist texts. This contradicts any ideas of the EBTs being the product of a conscious, artificial propaganda, and is best seen as simply the ordinary realities of life. According to T. W. Rhys Davids:³⁶⁹ “It is a recognised rule of evidence in the courts of law that, if one entry be found in the books kept by a man in the ordinary course of his trade, which entry speaks against himself, then that entry is especially worthy of credence.” In other words, since the EBTs give details that contradict later ideas, they are especially trustworthy. It might be thought that this just makes those particular details credible. However, since the EBTs were edited and transmitted through many generations, and there would have been many opportunities to edit the oddities out, there must have been a general principle of conservatism among editors. This makes the entire corpus trustworthy. Examples of such oddities include:

1. It is Rāma, Uddaka’s father, who was spiritually attained, not Uddaka (mn 26.16/mā 204).³⁷⁰

2. Upaka hears the Dhamma from the Buddha but then walks off in the wrong direction (mn 26.25).

3. The Buddha lays down rules and later he modifies or even rescinds them (e.g. respectively at Vin iv 72–74 and Vin i 79/83).^{371 372}

4. The Buddha, in the middle of winter, sleeps on a pile of leaves on a cattle track (an 3:35).

5. The Buddha washes his own feet (mn 31.5).

6. The Buddha personally helps to tend an ill monk (Vin i 304).

7. The Buddha is not recognised as such, but is simply seen as a monk (mn 140.3–5/mā 162/T 511).

8. The Buddha is disparagingly called various names such as shaveling, recluse-like, menial, dark (dn 3.1.10) and outcast (sn 7:9).

9. A brahman verbally abuses the Buddha (sn 7.2).

10. The Buddha is reluctant to meet with a group of brahman householders who have come to offer him food because they are too noisy (an 5:30).

11. The Buddha dismisses a group of monks and says he does not want to live near them because they are too noisy, and a group of lay people then causes the Buddha to change his mind (mn 67.2–10/eā 45.2/T 137).

12. The Buddha says he is at ease in responding to the calls of nature when he is by himself (an 8:86).

13. The monk Meghiya not doing his duties for the Buddha, even when the Buddha asked him (an 9:3/mā 56).

14. The monks not delighting in a discourse given by the Buddha (mn 1/eā 44.6).³⁷³

15. Ānanda sometimes uses sneaky means to get the Buddha to give a talk (mn 26.3).

16. The monk Bhaddāli refuses to the Buddha’s face to keep the pātimokkha rule requiring monastics to not eat after midday (mn 65.2–4/mā 194/eā 49.7).

17. The out of context and seemingly unnecessary episode where king Pasenadi conveys a message to the Buddha from two otherwise unknown sisters (mn 90.3–4/mā 212/P 1030).³⁷⁴

18. The Buddha teaches king Pasenadi how to lose weight (sn 3:13).³⁷⁵

19. The Gandhabba Pañcasikha sings a love song to the Buddha, in which he compares his love for a particular female gandhabba to the love of arahants for the Dhamma, etc. (dn 21.1.5).

20. The Buddha needs to reflect at length in a vain attempt to help Devadatta (an 6:62).

21. When Sāriputta dies, his bowl and robe are taken to the Buddha, but there is no mention of relics (sn 47:13).

22. There is no mention in the EBTs of the circumstances of Moggallāna’s death.

23. The Buddha says the Sangha seems empty after Sāriputta and Moggallāna have passed away (sn 47:14).

24. King Ajātasattu not knowing where the Buddha is seated in the assembly (dn 2.11).

25. The Buddha complains of having a bad back, and then lies down in the middle of a Dhamma talk (mn 53.5).

26. The Buddha gets tired due to being asked excessively about the rebirth of various people (dn 16.2.8/smps 19.17).

27. The Buddha in his old age warms his back in the sun, his limbs are flaccid and wrinkled, and his body stooped (sn 48:41).

28. The Buddha says the rules he has laid down should be kept as they are (dn 16.1.6/smps 2.8), but later on he says the minor rules can be abolished (dn 16.6.3/smps 41.2).

29. Although the Buddha says the Sangha can abolish the lesser rules after his death (dn 16.6.3/smps 41.2), the Sangha does not know which rules he was referring to and therefore decides to keep them all (Vin ii 287–288).

30. The Buddha dies of bloody diarrhoea (dn 16.4.20).³⁷⁶

31. Despite spending so much time with the Buddha, Ānanda did not reach arahantship until after the Buddha passed away (Vin ii 286).

32. The Pali Vinaya says that only the Dhamma and the Vinaya were recited at the First Council, (Vin ii 286–287) but the Dīgha Nikāya commentary blatantly contradicts this by saying that the Abhidhamma was recited too (DN-a i 15).

33. Purāṇa says he will not remember the Dhamma as recited at the First Council, but according to what he himself has heard; this is recorded despite its implications for the diminished authority of the Council (Vin ii 289–290).

These episodes, and others, could easily have been edited out, but they were kept despite their awkwardness and their not fitting with later ideas. This indicates that the attitude towards preserving the EBTs was very conservative.

4.4.2 Later texts are obvious

In identifying later additions to the canon, it is usually possible to use multiple independent criteria. For example, in the case of mn 111, the Anupada Sutta, the following 5 criteria are relevant:³⁷⁷

1. No known parallels in non-Pali EBTs;

2. Extravagant praise of Sāriputta, not found elsewhere in the EBTs and akin to the flowery and exaggerated language of other Buddhist texts;³⁷⁸

3. Textual duplication and redundancy, such as upekkhā being mentioned twice as a jhāna factor for both the third and the fourth jhāna;

4. Abhidhamma type vocabulary, not found elsewhere in the EBTs, such as anupadavavaṭṭhita;

5. Juxtaposition of different literary styles, specifically one Sutta style list of jhāna factors connected with ca and one Abhidhamma style list of factors without ca.

Moreover, the redactors could easily have hidden their additions by careful editing, but they didn’t. Where they added new material, they did so without attempting to disguise it, and sometimes the additions are acknowledged as such in the commentaries, e.g., the final verses of the Mahāparinibbāna Sutta (DN-a ii 615),³⁷⁹ all the verses of the Lakkhaṇa Sutta (DN-a iii 922), and the entire Bakkula Sutta (MN-a iv 197).

Similarly, there are many Suttas that are set in the time following the Buddha’s demise, and this is acknowledged in the texts.³⁸⁰

Both the EBTs and the later tradition, then, quite readily acknowledge that some discourses were not spoken by the Buddha. In some cases, such as the final verses of the Mahāparinibbāna Sutta, this could not have been otherwise, as the context is after the Buddha’s passing away. In other cases, however, such as the verses of the Lakkhaṇa Sutta, even though the Sutta features the Buddha himself, the commentary is still happy to acknowledge that the verse were added later. The fact, therefore, that in the majority of cases there is no acknowledgement of lateness tends to authenticate these text as having the historical Buddha as their origin.

4.4.3 Life of the Buddha

The EBTs display little interest in the Buddha’s biography. This is in stark contrast to other Buddhist literature.³⁸¹ This is most naturally explained if the EBTs stem mainly from the historical Buddha himself. He was interested in teaching the Dhamma, not telling his life story. And where the Buddha does speak of his own life, it is always to give a teaching on how he practised to reach awakening, presumably to set an example to be emulated.³⁸² Those around him knew him personally and did not need a lengthy biography. Only after he died was there a need to develop a biographical literature, beginning with the Mahāparinibbāna Sutta, then the Khandhakas, then the full-fledged biographies, such as the Mahāvastu, the Lalitavistara, and the Jātaka Nidāna.

Moreover, the few details that the EBTs give of the Buddha’s life frequently contradict the later legends:

1. His practices in past lives “did not lead to Awakening” (dn 19.61, mn 83.21/mā 67/eā 50.4/P 1030), vs. practising as a bodhisatta for an incalculable period (jn 25–61).

2. In past lives he had no notion of being a bodhisatta (mn 81/mā 63/P 1030, mn 83/mā 67/eā 50.4/P 1030, an 3:15), vs. taking the bodhisatta vow under the mythological Buddha Dīpaṅkara an incalculable long time ago (jn 17–18).

3. He left home motivated by seeing the danger in human violence and social disturbance (Snp 4:15, verses 936–938) and to seek freedom from suffering and death (mn 26.12/mā 204), vs. becoming a bodhisatta out of compassion (jn 17–18).

4. He left home while his parents were crying (mn 26.14/mā 204), vs. leaving in the middle of the night while his family was asleep (jn 82–84).

5. Siddhattha is a bodhisatta only after leaving home to become an ascetic,³⁸³ vs. being a bodhisatta for an incalculable period of time (jn 25–61).³⁸⁴

6. Struggled in meditation (mn 128/mā 72, mn 19/mā 102, mn 4/eā 31.1), vs. smooth progress (jn 90).

7. Practised austerities due to wrong view (mn 85.10), vs. practised them to show the world his perseverance and endeavour (jn 89).

8. No mention of compassion as motivation until after Awakening (compare pre-Awakening mn 26.13/mā 204 with post-Awakening mn 26.21), vs. compassion being the motivation for becoming a bodhisatta (jn 17–18).³⁸⁵

9. The Buddha walked from Bodhgaya to Benares to start teaching (mn 26.25), vs. going to Benares in half a day, implying the use of psychic powers (jn 109).

10. Sakya being a small republic subject to Kosala (mn 89.19/mā 213/eā 38.10/T 1451/P 1035),³⁸⁶ vs. Sakya being a kingdom with the bodhisatta’s father as king and Kapilavatthu as the enormous and magnificent capital (jn 69, 76–77).

11. The Buddha attaining awakening (sammāsambodhi, sn 56:11), vs. the Buddha attaining omniscience (sabbaññutā) (jn 99).

Again, such details could easily have been removed from the EBTs. That they were kept despite the contradictions with other Buddhist texts, again shows the conservatism of the Sangha in keeping the EBTs as unchanged as possible.

4.5 Supernormal elements

From the perspective of a modern reader, the EBTs may seem to contain a number of supernatural elements that diminish the historical value of the whole genre. Steven Collins comments:

In the texts of the Pali Canon the Buddha is very frequently depicted as interacting with gods and other supernaturals, often giving them doctrinal talks. Many modern historians, who of course must be professionally at the very least agnostic about the existence of supernaturals, assume that one can ignore the nature of the Buddha’s interlocutors but still accept what he is depicted as saying as evidence of ‘his ideas’. But if later generations could invent gods for the Buddha to talk to they could also invent what he said to them. We have no way of distinguishing between transmission and invention.”³⁸⁷

The assertion on which Collins’ argument rests, that the Buddha is depicted in the Pali Canon as “very frequently giving them (the ‘supernaturals’) doctrinal talks,” is simply not true. Of all the 186 Suttas of the Dīgha and Majjhima Nikāyas, there is only a single one of which this is true, the Sakkapañha Sutta (dn 21).³⁸⁸

The truth is almost the exact reverse of what Collins would have us believe. Apart from the significant evidence from the Dīgha and Majjhima Nikāyas, an analysis of one of the main sections of the Canon that deals with supernormal beings, the Sagāthā Vagga of the Saṁyutta Nikāya, shows that they are almost exclusively confined to circumstantial material found in the narrative sections. The Sagāthā Vagga contains over 200 Suttas in which the Buddha is seen in conversation with divine beings. However, in only 21 of these does the actual conversation, as opposed to the surrounding material, suggest that one of the parties is non-human.³⁸⁹ Further, 15 of these 21 are conversations with Māra. But since Māra in the EBTs is often a name for a psychological state,³⁹⁰ it is likely that this is so in the majority (perhaps all) of these cases too. This leaves us with only six Suttas out of more than 200. But even this number does not give a fair representation of the state of affairs. All of these six Suttas consist of no more than the exchange of a few inspirational verses. They either lack doctrinal content completely or it is very limited. That is, we are probably not dealing with the sort of core doctrinal material that might be considered untouchable.

The notion that discussions between the Buddha and supernormal beings are rare or absent is also implied by a passage in the Mahāparinibbāna Sutta (dn 16.2.17). The Buddha is approached by the brightly dressed and flamboyant Licchavi princes, and, apparently amused, he says to the monks that for those who have never seen the gods of the Tāvatiṁsa realm, now is their chance, since they look just like these Licchavis! If the gods were visiting the Buddha regularly, filling the entire monastery with radiance, as the stock passage describes it, the monks could hardly have avoided noticing this.

This quick survey does not cover all the Suttas in the Pali Canon in which the Buddha is seen conversing with supernormal beings, but it probably comprises the vast majority of them. What we find, then, is that supernormal beings are no more than peripheral and mostly mentioned either in stories or in the narrative material that surrounds the core doctrinal content. In other words, whatever enhancement the EBTs underwent at the hands of redactors was limited to the circumstantial material and did not affect the core message of the Buddha’s teaching.³⁹¹

We do not mean to deny that the EBTs express a world-view in which supernormal phenomena are a part. Indeed, it is likely that this very world-view was partly responsible for the inclusion of such material in the narrative sections. That, and the prestige that this may have given the Suttas in the eyes of the intended audience, is sufficient to explain why it is there. There is no good reason for thinking that the existence of these elements shows that the transmission of the core doctrinal content has been unreliable.

Regardless of the actual status of supernormal beings, powers, or events in the life of the Buddha, the argument by Collins is of a peculiar sort. He moves from saying that later generations “could have” invented the Buddha’s statements—a straw man argument, since no-one disputes something so obviously true—to the general assertion that we have “no way” to distinguish transmission from invention. As we have shown throughout this essay, there are in fact many ways of distinguishing invention from transmission. Moreover, these are little different from the kinds of ordinary distinctions that we make in all forms of discourse. For example, consider the frequent evocations of “God” by American politicians. Do we assume that, if we are professionally agnostic towards the existence of a creator deity, we have no grounds for knowing whether any statements by that politician are true? Of course not. We take their religious beliefs as religious beliefs, and do not expect to find hard evidence for them. And we evaluate their factual claims by reference to knowable facts, just as we would do for anyone else.

Yet instead of making this sort of basic distinction, Collins sees the mention of supernormal elements as a reason for dismissing the EBTs outright, so that the very possibility of clearer understanding is denied. The proper course of a careful investigator, rather, would be to consider the nature and form of the supernormal elements, and to clarify the roles that they play within the literature.

4.6 Structure

4.6.1 The structure of the Saṃyutta

The EBTs emphasise the centrality of teachings such as the four noble truths, dependent origination, and the bodhipakkhiya dhammas. The early community recited and memorised such teachings, organising them into the ancestor of today’s Saṁyutta/Saṁyuktas. The Saṁyutta as a whole is shaped after the pattern of the broadest of all the teaching frameworks, the four noble truths (mn 28.2/mā 30).

The Saṁyutta Nikāya/Āgama collects Suttas (usually) by topic, and those topics neatly represent the four noble truths: the aggregates, senses, and elements under the truth of suffering; dependent origination under the truths of origination and cessation; and the subject of mental development under the truth of the path.^{392 393 394 395 396} If the Saṁyutta Nikāya/Āgama is the earliest collection of Suttas—as suggested by noted scholars such as Yin Shun—³⁹⁷from which the other Nikāyas/Āgamas evolved, then the Saṁyutta structure may be a literal implementation of Sāriputta’s statement that all good teachings are included within the four noble truths (mn 28.2/mā 30).

The Saṁyutta structure also closely relates to the 37 bodhipakkhiya dhammas, which make up most of the last section of this work, that dealing with the path. In several places the EBTs present the bodhipakkhiya dhammas as being a summary or the essence of the Buddha’s teachings (sn 22:81), and the Buddha exhorts the Sangha to keep an accurate memory of them.³⁹⁸

That the content of the collection in this way is reflected in its structure is another indication that the Saṁyutta/Saṁyuktas are an integrated whole stemming from a single source.

Finally, it has also been proposed that the first three of the 9 aṅgas should be identified with the proto-Saṁyutta, based on a statement by Asanga and on the pattern of distribution of texts within these collections.³⁹⁹ This would mean that the first three aṅgas formed a sub-structure within the proto-Saṁyutta. If this is correct, then we can discern the earliest form of a Buddhist “canon”, before the current organisation of texts into Nikāyas/Āgamas.⁴⁰⁰

4.6.2 Fractal structure

A fractal is a mathematical set that generates an object or pattern that is self-similar, that is, exactly or approximately similar at all scales. Unlike normal geometric patterns, fractals retain detail no matter how small the scale is.^{401 402}

Both the content and the form of the EBTs exhibit fractal characteristics. The best example in terms of content is the four noble truths. These may be expressed at the smallest level of linguistic meaning, the word: dukkha, samudaya, nirodha, magga. Each of these terms may be expanded indefinitely, thus displaying an ever increasing amount of detail (see e.g. mn 28/mā 30 and mn 141/mā 31/T 32/eā 27.1). At its largest scale, the four noble truths form the structure of the Saṁyutta Nikāya and, according to mn 28/mā 30, encompass all good teachings, that is, the Dhamma as a whole.

Likewise, the formal linguistic features of the EBTs exhibit fractal features. An example is the use of repetition. These occur at the level of the word, the sentence, the paragraph, the passage, the text, and the group of texts.

The resemblance to fractals is not coincidental. A fractal is formed through the recursive iteration of a simple operation. This is analogous to the manner in which complex Dhamma analyses are produced through repeated questioning of the previous statement (“And what, monks, is…?”; again see mn 28/mā 30). Moreover, a crucial benefit of such fractal patterns is that they are compressible. It is possible to learn the essence of the Dhamma in a few words, and all expansions exemplify the same pattern.

Such patterns are frequently observed in nature, and are an outstanding feature of organic as opposed to artificial growth. The fractal features of the EBTs, therefore, suggest that they also grew organically and holistically, not through the artificial bolting together of elements from different origins. The most obvious way that this would have happened is through a gradual expansion in details and contexts of the earliest teachings. The initial impetus for teaching may have been a particular insight, but an insight that allowed virtually endless variations and expansions in its exposition. This differs from non-EBT Buddhist literature, where we frequently see early elements juxtaposed with entirely different formulations.

Chapter 5
Archaeology

5.1 Overview

Towns, cities, and other sites described in the EBTs have for the most part been identified through archaeological excavations. Some of the findings go back to the time of Asoka or even before. Significantly, most of the sites were discovered and understood with the aid of the texts,⁴⁰³ but the sites themselves have generally provided independent evidence of their identity.⁴⁰⁴ The paucity of archaeological evidence for the period before Asoka is due to the fact that buildings were constructed of mud and timber,^{405 406 407} and because the Mauryan period, starting in about 300 bce, marks the beginning of the script and picture age in Indian Archaeology.⁴⁰⁸

According to Härtel, places of importance for Buddhist history have been “rebuilt again and again on the same ground” and “no archaeologist will dare to remove the mostly solid remains of the upper level”. He concludes that “the hope to recover the original structures and ruins of a town or habitation of the time of the Buddha, let us say Kapilavastu, is almost zero”.⁴⁰⁹ This, then, would seem to be the reason for the very limited pre-Mauryan archaeological finds.

5.2 Asokan edicts

The Asokan edicts are dated to about 150 years after the Buddha. They are the earliest datable epigraphic evidence in Indian history,⁴¹⁰ and the earliest significant archaeological finds in the middle Ganges area. They record the deeds and proclamations of the Mauryan emperor Asoka, who converted to Buddhism during his reign. Buddhism must therefore have been a major and growing force in Indian religion for a considerable period before Asoka. Elements in the inscriptions that agree with the EBTs, some of which are explicit references, include:

• RE 1 (Girnār version):⁴¹¹ No living beings are to be killed or offered in sacrifice.^{412 413}

• RE 2 (Girnār): Asoka made provisions “everywhere” (sarvata vijitamhi) for the medical treatment of both humans and animals, and he had wells dug and trees planted for their use.^{414 415}

• RE 3 (Girnār): “Respect for mother and father is good; generosity to friends, acquaintances, relatives, brahmans and ascetics is good; non-violence to animals is good; moderation in spending and possessions is good.”^{416 417}

• RE 4 (Girnār): Overlaps in part with re 3, adding “proper behaviour towards relatives, brahmans and ascetics, and respect for the aged.” Also “to instruct in the Dhamma is the highest work.”^{418 419}

• RE 5 (Kālsī): “Good deeds (kayāna [= kalyāṇa]) are difficult to perform” and “bad acts (pāpa) are easy to commit”, apparently quoting the EBTs (see Ud 5:8).^{420 421}

• RE 6 (Girnār): Asoka considers the welfare of the whole world (sarva-loka-hita) to be his duty.^{422 423}

• RE 8 (Girnār): Going on a Dhamma tour (dhaṁma-yātā), Asoka visited “sambodhi” (Bodhgaya). There was Dhamma instruction (dhaṁmānusasṭī) and Dhamma questioning (dhama-paripuchā) with the country people.^{424 425}

• RE 9 (Girnār): Asoka expresses his disapproval of excessive “auspicious” (maṁgala) ceremonies (which is comparable to what is truly auspicious according to the Maṅgala Sutta, Snp 2:4) and instead (along the same lines as re 3 and re 4) favours proper behaviour towards slaves and servants, respect to teachers, restraint in regard to animals, generosity towards brahmans and ascetics, all typical Buddhist virtues. This edict (as well as re 11) also says that “there is no gift like the gift of the Dhamma”, which is likely to be a quote from the EBTs (see an 9:5 or Dhp 354).^{426 427}

• RE 11 (Girnār): Many of the same points as mentioned above in re 9.^{428 429}

• RE 12 (Girnār): “One should listen to and respect the doctrine of others.”^{430 431}

• RE 13 (Kālsī): Asoka repents his conquering of the Kāliṅgas. He states that “conquest by Dhamma is the best conquest,” echoing the conquest of a wheel-turning monarch described at dn 17.1.8–9.^{432 433}

• Kaliṅga RE 1 (Dhaulī): Asoka tells his officials that they should practice the middle (majhaṁ paṭipādayema). This would seem to be a reference to the Buddhist middle way.⁴³⁴ Immediately following this he says that “one does not act right when acting from envy, anger, cruelty, hurry, indifference, laziness, or fatigue.”^{435 436}

• Kaliṅga RE 2 (Jaugaḍa): Asoka says his subjects should feel “the king acts towards us as a father would; he feels for us as he feels for himself; we are to the king as his own children.” This passage echoes the sentiment in the EBT description of the ideal king at dn 17.1.21/smps 34.25.^{437 438}

• MRE 1 (Gavīmaṭh): Asoka calls himself an upāsaka, a Buddhist lay disciple;⁴³⁹ a Śākya, one who belongs to the Buddha’s (spiritual) family (Rūpnāth version);⁴⁴⁰ and possibly a Budhaśake, a Sakyan of the Buddha (Maskī version);⁴⁴¹ and says he has visited the Sangha.^{442 443}

• MRE 2 (Yerrāguḍi): One should respect mother and father, as well as teachers (garu); one should have mercy for animals; one should speak the truth; pupils (aṁtevāsi) should respect their teachers (ācariya)—all typical Buddhist virtues.^{444 445}

• MRE 3 (Bairāt): Asoka honors (abhivādetūnaṁ (= abhivādetvā)) the Sangha and hopes it has little affliction (apābādhata) and is dwelling at ease (phāsu-vihālata), both expressions being from the EBTs. He expresses his respect (gālava) and faith (prasāda) in the “budha dhaṁma saṁgha”, and then says that all that was spoken by the Buddha was well-spoken (subhāsita). He speaks of “the long duration of the true dhamma (sadhaṁma)”, apparently quoting the EBTs (see e.g. sn 16:13). He then recommends a number of Buddhist texts (dhaṁma-paliyāya) for study: Vinaya-samukasa, Aliya-vasāni, Anāgata-bhayāni, Muni-gāthā, Moneya-sūta, Upatisa-pasina, Lāghulovāda. The last of these is specifically said to concern lying, which is also the subject matter of this Sutta in the EBTs.⁴⁴⁶ There is some disagreement as to the exact identity of some of these,^{447 448} since Buddhist texts frequently go by various titles. There is, however, no question of having to seek outside the EBTs to find them.⁴⁴⁹ This edict also uses the following EBT terminology: bhikhu, bhikhuni, upāsaka, and upāsikā, and the Buddha is called bhagavat.⁴⁵⁰

• PE 2:⁴⁵¹ “Dhamma consists of doing few bad deeds, many good deeds, being compassionate, generous, truthful, and pure.” Asoka further says that “many benefits have been conferred by him on bipeds, quadrupeds, birds and fish.”^{452 453}

• PE 3: “One should regard it thus: These are the ways to bad conduct, that is, fierceness, cruelty, anger, pride, envy. Let me not be ruined by them.”^{454 455}

• PE 4: Asoka abolishes death penalty.^{456 457}

• PE 5: Asoka prohibits violence against and the killing of various animals.^{458 459}

• PE 7 (Delhi-Toprā): Asoka summarises all his good deeds. He mentions the Jains (nigaṁṭha) and the Ājīvakas, both found in the EBTs.^{460 461} He says that, “People’s progress in Dhamma is achieved in two ways, by Dhamma rules and by conviction. Rules count for little; most is by conviction.”^{462 463} Otherwise this edict mostly repeats what he has already proclaimed elsewhere.^{464 465}

• MPE 1 (Lumbinī): “Here the Buddha, Sakya-muni, was born” (A: hida budhe jāte sakya-munī ti; B: hida bhagavaṁ jāte ti). Asoka then remits the taxation of the inhabitants of Lumbinī.^{466 467}

• MPE 2 (Sārnāth): Asoka condemns schism in the Sangha. He says that any monk or nun who causes such a schism is to be disrobed.⁴⁶⁸ This shows Asoka’s remarkable concern for the well-being of the Sangha. A variety of specialist EBT terms are employed: bhikhu-saṁgha, bhikhuni-saṁgha, sāsana, upāsaka, anuposatha, saṁgha bheta, saṁgha samaga (Sāñcī version), cila-thitīka (Sāñcī).^{469 470}

• PE at Nigālī-sāgar: Mentions the stupa of Buddha Konāgamana and that Asoka paid his respects there.⁴⁷¹

Gombrich makes the point that Asoka, like the Buddha, used current terminology in new ways, thereby turning worldly pursuits into Dhamma pursuits, all of which have strong Buddhist overtones. Perhaps Asoka was imitating the Buddha.⁴⁷²

Although much of the content of the Asokan edicts is not specifically Buddhist, the overall tenor is Buddhist. For example, many of the virtues listed above are similar in substance and detail to those of the EBTs.^{473 474} Moreover, and significantly, there is nothing in these edicts that contradicts the teachings in the EBTs.

5.3 Barabar caves

Asoka and his grandson Dasaratha made rock-cut caves for the ascetics of the Ājīvaka sect in the 3rd century bce.⁴⁷⁵ The Ājīvakas feature frequently in the EBTs.⁴⁷⁶

5.4 Bodh Gaya

According to Härtel, the vajrāsana (“the diamond seat”), meant to mark the spot where the Buddha attained awakening, is likely to be the oldest archaeological relic at Bodhgaya and is dated to the time of Asoka.⁴⁷⁷ The artwork on the railings at Bodhgaya seem to be from the 2nd century bce. It is more rudimentary than that in other locations.⁴⁷⁸ The scenes depicted are given in less detail, and thus they are often hard to identify.⁴⁷⁹ Nevertheless, some of the artwork can clearly be identified as Buddhist. For instance the lotus, an important early Buddhist symbol that often refers to the Buddha’s birth, features frequently.⁴⁸⁰ Other Buddhist motifs are as follows:

1. From the EBTs: first meditation (mn 36.31);⁴⁸¹ first teaching, symbolised by the dhammacakka (sn 56:11/sā 379/T 109/T 110/eā 19.2/P 747/P 1003);⁴⁸² Indra’s visit (dn 21/dā 14/T 15/MĀ 134);⁴⁸³ parinibbāna, symbolised by stupa (dn 16.6.7–9/dā 2/T 5/T 6/T 7/T 1451/smps 42.11–18).⁴⁸⁴

2. From non-EBTs: grass cutter’s gift (jn 93);⁴⁸⁵ wonder of parting the water and walking on dry ground (Vin i 32);⁴⁸⁶ return to Kapilavatthu (Vin i 82–83 and jn 118–119);⁴⁸⁷ gift of Jetavana (Vin ii 158–159);⁴⁸⁸ homage to Buddha by Pārileyyaka elephant (Vin i 353);⁴⁸⁹ a dozen Jātakas (7 unidentified).⁴⁹⁰

5.5 Deur Kothar

This Buddhist monument contains some of the earliest stupas, and it includes Asokan-era writing, rock shelters, simple art, an Asokan pillar, and Northern Black Polished Ware. It probably stems from the beginnings of Buddhist popular monuments (Mishra suggests the 3rd century bce),⁴⁹¹ which were greatly advanced by the time of Bhārhut, about a century later. The use of mainly abstract motifs is congruent with the EBTs, in which there is little concrete narrative, but much abstract teaching.^{492 493} Simple lotuses appear, lacking the sophisticated design of later times.

The monument was discovered by surveying the landscape of a large area that according to a number of sources had, in ancient times, been within the sphere of Buddhist influence and where a monument would thus be expected, but had not yet been discovered.⁴⁹⁴ That the location of monuments could be predicted in this way suggests that Buddhism was well established in this whole region by the time of Asoka.

There are two brahmī inscriptions at Deur Kothar that appear to connect donors (of the Bahuśrutīya school) directly to the Buddha through teacher/student lineages.⁴⁹⁵ These inscriptions include Buddhist terms such as bhagavat, Budha, upasaka, ācariya, and ātevāsi (= antevāsī).⁴⁹⁶ The inscriptions are damaged and full reconstruction is not possible; however, in so far as the reconstruction allows, they go back to the Buddha in 9 to 14 generations. This shows that the Buddha was regarded as a historical person who lived a couple of centuries prior to the inscriptions. Moreover, it shows the Buddha as the originator of the lineage and therefore as the source of Buddhism.

5.6 Bhārhut

This stupa was begun by Asoka, although most of what we can now see is dated to the Sunga dynasty in 2nd and early 1st century bce.^{497 498} It includes many Buddhist images and references. Representations of the Buddha are aniconic. Many details from the EBTs are depicted or mentioned in inscriptions, including:

1. Buddhas: Buddha Gotama and all five past Buddhas of the EBTs are mentioned, but none other: Sāka Muni;⁴⁹⁹ Kāsapa;⁵⁰⁰ Konigamena (= Konāgamana);⁵⁰¹ Vipasi;⁵⁰² Vesabhu;⁵⁰³ Kakusadha.⁵⁰⁴

2. People: Ajātasatru;⁵⁰⁵ Rāja Pasenaji Kosalo (= King Pasenadi of Kosala);⁵⁰⁶ Dighatapasi (= Dīgha Tapassī) (mn 56/mā 133).⁵⁰⁷

3. Places: Kosambī;⁵⁰⁸ Pāṭaliputa;⁵⁰⁹ Sudhamma deva sabhā (= Sudhamma Hall of the Gods) (mn 50.29);⁵¹⁰ Vijayanto Pāsāde (= Vejayanta Palace) (mn 37.8/sā 505/eā 19.3);⁵¹¹ Ida sāla guha (= Cave-hall of Indra) (dn 21.1.1).⁵¹²

4. Scenes from the Buddha’s life: Awakening (mn 26.18/mā 204, MN 36.43);⁵¹³ first discourse (sn 56:11/sā 379/T 109/T 110/eā 19.2/P 747/P 1003);⁵¹⁴ Indra’s visit (dn 21/dā 14/T 15/mā 134);⁵¹⁵ worship by King Ajātasattu (dn 2.99–101/dā 27/T 22/eā 43.7);⁵¹⁶ passing away (dn 16.6.7–9/dā 2/T 5/T 6/T 7/T 1451/smps 42.11–18).⁵¹⁷ Particularly interesting is a pillar that depicts and labels several scenes from the Mahāparinibbāna Sutta.⁵¹⁸ This suggests a familiarity with this discourse, which therefore must have existed.

5. Text reciters: sutantika;⁵¹⁹ pamcha-nekāyika (= pañcanekayika).^{520 521}

6. Technical terms: amtevāsi;⁵²² arāmika;⁵²³ avāsikā;⁵²⁴ bhatudesaka;⁵²⁵ navakamika;^{526 527} upadāna.⁵²⁸

7. Divine beings: Ajakālako (= Ajakalāpaka) (Ud 1:7);⁵²⁹ Erapato nāja rāja (= Erakapatta, king of nāgas) (an 4:67);^{530 531} Kupiro (= Kuvera; called Vessavana e.g. at an 7:53);⁵³² Muchalido nāgarāja (= Mucalinda, king of nāgas) (Ud 2:1);⁵³³ Suchiloma (sn 10:3/sā 1324/sā² 323);⁵³⁴ Sudhāvāsa devas;⁵³⁵ Vipachitta (= Vepacitti) (e.g. at sn 11:4/sā 1110/sā² 39/eā 34.8);⁵³⁶ Virudako (= Virūlhaka) (dn 18.11).⁵³⁷

8. Plants: Bodhi tree of Gotama Buddha;⁵³⁸ for Vesabhu this is specified as a Sal tree,⁵³⁹ for Konāgamana as an Udumbara tree, for Kakusandha as a Sirisha tree and for Kassapa as a Nyagrodha tree, all as in dn 14.1.8.⁵⁴⁰ Only the tree of Buddha Vipassī does not fit with the EBT description, which specifies the Pāṭalī tree. Instead it is given as the Asoka tree, as in the Mahāmāyūrī, a later Buddhist work.^{541 542}

9. Assemblies: Jātila Sabhā (Vin i 33);⁵⁴³ deva sabhā (an 3:37).⁵⁴⁴

10. Things: Dhamma wheel (an 4:36);⁵⁴⁵ Buddha’s footprint (an 4:36).⁵⁴⁶

11. Acts: Frequent mention of donations (dāna).⁵⁴⁷

12. Bhikkhunīs: Mentioned 16 times as donors.^{548 549}

Shortly after Asoka we thus find references to collections of scripture (see also section 5.7, “Sāñcī” below), to individual scriptures contained therein (see section 5.2, “Asokan edicts”), and to many people, places, ideas, and objects familiar from the EBTs. This is not just true of generalities, but also sometimes of details: according to Lüders “… the visit of Ajātasattu [to the Buddha] is depicted even in details exactly according to the Sāmaññaphala Sutta,” and “… the representation of the visit of Sakka [to the Buddha] follows the text of the Sakkapañha Sutta.”⁵⁵⁰ This suggests that the EBTs must have existed at this time.

There is further evidence to suggest that the EBTs were considerably earlier. Bhārhut’s early detailed Buddhist images and inscriptions depict a context later than that of the EBTs and familiar from Buddhist literature such as the developed Buddha legends and non-EBT Jātakas. Since such literature inherits and builds on the EBTs, this shows that the EBTs had been extant and influential for a considerable period of time prior to Bhārhut. Such non-EBT details include:

1. Buddha legend: Māyā’s dream, depicting the conception of the Bodhisatta as the descent of an elephant (jn 67);^{551 552} childhood contest (jn 78);⁵⁵³ Prince Siddhattha’s departure from Kapilavatthu on his horse Kanthaka, whose feet are upheld by four Yakkhas (jn 83);⁵⁵⁴ cutting of the Buddha-to-be’s hair (jn 86);⁵⁵⁵ worship by pārileyyaka elephants (Vin i 353);⁵⁵⁶ visit of nāga Erakapatta⁵⁵⁷ (Dhp-a iii 234).⁵⁵⁸

2. Jātakas: About 40 in all,⁵⁵⁹ of which 16 are mentioned by name, while others just appear in image. Some are still unidentified.⁵⁶⁰

3. Relic and stupa worship: (dn 16.6.27–28/smps 50–51).^{561 562}

4. Purchase of Jetavana: (Vin ii 158–159; jn 125).⁵⁶³

5. Saṅkika/Saṅkassa ladder: The Buddha’s descent from Tāvatiṁsa heaven (e.g. Dhp-a iii 224).⁵⁶⁴

6. Gadhakuti and Kosabakuti: Huts/houses of the Buddha’s according to non-EBT tradition (e.g. DN-a ii 1 and Jātakas).⁵⁶⁵

7. Patisamdhi: An Abhidhamma term that refers to rebirth.⁵⁶⁶

5.7 Sāñcī

This monastery/stupa complex is located near Vidisa, which is nearly 1,000 km from Pāṭaliputta. It originated in Asoka’s time, although, like Bhārhut, much of what we see today was added a century or so later.^{567 568 569 570}

At Sāñcī were discovered relics of the Buddha’s chief disciples, Sāriputta and Moggallāna, both of whom are prominent in the EBTs. There were also smaller stupas that contained relics and references to several later monks as “teachers of the Himalayas.” These monks are identical to several of the monks that according to the Sinhalese Vinaya Commentary (in Pali and Chinese) were missionaries in the time of King Asoka. This was one of the cardinal identifications that allowed scholars to conclude that the traditional accounts of Asoka contained some historical facts. Wynne has recently strengthened the case by making the point that the relics were found together in groups of five, the minimum size group required to confer ordination, thus pointing to their missionary purpose.⁵⁷¹ He also argues that some of the Asokan Edicts refer to the same missionary activities.⁵⁷²

The Eastern Gateway of the main stupa is decorated with bas-relief that depicts Asoka sending a branch of the Bodhi tree to Sri Lanka.⁵⁷³ This event is also mentioned in re 13.⁵⁷⁴ Both of these corroborate the story in the Mahāvaṁsa of the establishment of Buddhism in Sri Lanka at the time of Asoka (Mv 13.18–13.21, Mv 14.1–14.65).

5.7.1 Artwork

Although most of the artwork at Sāñcī is quite elaborate and reflects trends in the non-EBT Buddha biography, it is for the most part based on material found in the EBTs and ultimately refers back to this. As with Bhārhut, this suggests that the EBTs are the root from which this artwork sprang, but that there has been an extensive period of development and elaboration in the time after the EBTs. In the following we have referred both to the EBT root sources and the sources of the later elaboration, wherever applicable.

Most of the art at Sāñcī is concerned with the life of the Buddha. The following general development in the Buddha biography can be discerned in the scenes depicted at Sāñcī.

1. Straight from EBTs: Birth: lotus and standing Māyā (mn 123.15);^{575 576} austerities (mn 12/T 757, mn 36.20–30, eā 31.8);⁵⁷⁷ Awakening: Bodhi tree with empty seat (Vin i 1);^{578 579} first teaching: Dhammacakka and deer (mn 26.29–30/mā 204/eā 24.5);^{580 581 582} Bamboo Grove in Rājagaha;⁵⁸³ Indasālā: Sakka’s visit (dn 21/dā 14/T 15/mā 134);⁵⁸⁴ Ajātasattu’s visit (dn 2.12/dā 27/T 22/eā 43.7);⁵⁸⁵ Mahāparinibbāna: stupa (dn 16.6.7–9/dā 2/T 5/T 6/T 7/T 1451/smps 42.11–18);^{586 587} division of relics (dn 16.6.24–27/smps 51.8–21).⁵⁸⁸

2. EBTs, with elaboration: First meditation (mn 36.31/eā 31.8 and jn 77);⁵⁸⁹ divine messengers (dn 14.2–14);^{590 591} Brahmā and Sakka, with retinue, ask Buddha to teach (mn 26.20–21 and jn 108);^{592 593} strife brewing over the Buddha’s relics, “War of the Relics” (dn 16.6.25/smps 50).^{594 595}

3. Borderline EBT: Serpent Mucalinda (Ud 1:11/Vin i 3);⁵⁹⁶ two first disciples, Tapussa and Bhallika (Vin i 4);⁵⁹⁷ guardian devas give four bowls (Vin i 4);⁵⁹⁸ victory over nāga (Vin i 25);⁵⁹⁹ Indra and Brahmā visit Buddha at Uruvelā (Vin i 26–27);^{600 601} the Buddha stops splitting of wood and lighting of fire (Vin i 31);⁶⁰² parts water and walks on dry ground (Vin i 32);^{603 604} visits Kapilavatthu (Vin i 82–83 and jn 118–119);⁶⁰⁵ converts Jaṭilas;⁶⁰⁶ Bimbisāra’s visit (Vin i 35–39 and jn 111–112);^{607 608 609} Buddha’s footprint with dhammacakka (an 4:36).⁶¹⁰

4. Mostly non-EBT, but with EBT kernel: Māyā’s dream, conception (mn 123.6–7/mā 32 and jn 67);^{611 612} temptation and assault by Māra, his daughters and host of demons (Snp 3:2 and jn 95–98, 105–106);^{613 614} Buddha’s residences at Jetavana, and the spread of coins for buying the Jetavana (jn 125);^{615 616 617} gift of Nigrodhārāma to Sangha (jn 119);⁶¹⁸ teaching the Sakyas (jn 119–120);^{619 620} offering of honey by monkey in Pārileyyaka forest⁶²¹ (Dhp-a i 59–60).⁶²²

5. Non-EBT: Buddha-to-be leaves on the horse Kanthaka (jn 83–86);^{623 624} Kanthaka’s legs supported by devas (jn 83);⁶²⁵ Sujātā offers milk rice (jn 92);⁶²⁶ Sotthiya (Svastika) offers grass to Buddha at Bodhi tree (jn 93);⁶²⁷ Buddha walks meditating after Awakening (jn 104);⁶²⁸ jewel house where Buddha contemplates Abhidhamma after Awakening (jn 104);⁶²⁹ twin wonder under mango tree at Sāvatthī (Jā no. 483);⁶³⁰ Buddha goes to Tāvatiṁsa heaven to teach Abhidhamma to mother (Dhp-a iii 216);^{631 632} Saṅkassa: descent from Tāvatiṁsa (e.g. Dhp-a iii 224).⁶³³

The art at Sāñcī also includes some scenes that are not from the Buddha’s biography. An evolution similar to that in the Buddha biography can be seen in this art too:

1. Straight from EBTs: Six Buddhas and their trees (dn 14.1.8): Vipassī with pāṭalī, the trumpet flower tree;⁶³⁴ Sikhī with puṇḍarīka, the white mango tree; Vessabhū with sālā tree; Kakusandha with sirīsa, the acacia sirissa tree; Konāgamana with udumbara, fig tree; Kassapa with nigrodha, Banyan tree; Gotama with assattha, the Ficus Religiosa.⁶³⁵ Cosmography: Four Great Kings, Kubera (Vessavaṇa), Virūḷhaka, Virūpakkha, and Dhataraṭṭha (e.g. dn 18.12).⁶³⁶ The six kāmāvacara heavens (an 3:70).⁶³⁷

2. Mostly non-EBT, but with EBT kernel: Maitreya with his tree of awakening.^{638 639}

3. Non-EBT: Relic worship (dn 16.6.28/smps 50–51);⁶⁴⁰ Asoka visits stupa at Rāmagāma (Aśokāvadāna);^{641 642} Asoka visits Bodhi tree (re 5 and Aśokāvadāna);^{643 644 645} five Jātaka stories.⁶⁴⁶

5.7.2 Inscriptions

There are few inscriptions at Sāñcī compared to Bhārhut. Still, the few inscriptions that are found there include a number of terms that either refer to the EBTs or include vocabulary which has the EBTs as its source.

1. Reciters of scriptural collections: sutātika,⁶⁴⁷ sutātikinī,⁶⁴⁸ sūtātikinī,^{649 650} dhamakathika,⁶⁵¹ pacenekayika (= pañcanekāyika),⁶⁵² sutatikiniyā.⁶⁵³

2. Technical terms: āchariya,⁶⁵⁴ anantarya,^{655 656} antevāsi,⁶⁵⁷ araha (= arahant),⁶⁵⁸ ārya-Sangha,⁶⁵⁹ bhagavat,⁶⁶⁰ bhikhu,⁶⁶¹ bhikhuni,⁶⁶² budha,⁶⁶³ chilathitika,⁶⁶⁴ dhama,^{665 666} nikāya,⁶⁶⁷ sadhivihārī,⁶⁶⁸ samaga,⁶⁶⁹ samanera,⁶⁷⁰ sagha (= sangha),⁶⁷¹ saghadana (= sanghadāna),⁶⁷² sapurisa,^{673 674} thera,^{675 676} upāsikā,^{677 678} upāsaka,⁶⁷⁹ vināyaka,^{680 681} vinayakāna.⁶⁸²

3. Names: Sāriputa (relics),⁶⁸³ Mahā Mogalāna (relics),⁶⁸⁴ Idadeva (= Sakka).⁶⁸⁵

5.8 Piprahwa/Ganwaria

A soapstone casket (the “Peppé casket”) with Asokan Brahmi inscriptions, apparently containing relics of the Buddha, was discovered in a large stupa at Piprahwa towards the end of the nineteenth century.⁶⁸⁶ The inscription, as well as the location of the find, implies that the Sakyas received some of the relics of the Buddha after his cremation and that the Sakyans belonged to this area.^{687 688} This echoes the distribution of relics described at the end of the Mahāparinibbāna Sutta (dn 16.6.24–26/smps 51.9–21). The exact interpretation of the inscription is disputed,⁶⁸⁹ but it seems likely it should be understood as follows:

This receptacle for the relics of the Blessed Buddha is that of the Sakyan brothers called Suki, together with their wives and children.”

Sukiti bhatinaṁ sa-puta-dalanaṁ iyam salila-nidhane Budhasa bhagavate sakiyānaṁ.⁶⁹⁰

Further excavations revealed another relic casket at a lower level than the Peppé casket, once again with bone relics, but without inscription, and dated to the Mauryan period.⁶⁹¹ Excavations a few hundred metres from the stupa yielded a number of sealings with the inscription Kapilavastu.⁶⁹² This discovery substantially strengthens the case that the Piprahwa find is linked to the relic distribution described in the Mahāparinibbāna Sutta.

The earliest archaeological finds of Piprahwa and the nearby site of Ganwaria, such as the mud walls closest to the natural soil level, probably date to around 500 bce,⁶⁹³ again consonant with the dates of the Buddha and descriptions in the EBTs.

5.9 Rājagaha

The hills around Rājagaha feature in the EBTs, and the striking geography inspired several discourses, similes, and stories. The hot springs, which were the occasion for a Vinaya rule, are still present. Many ruins of stupas and monasteries are found, although it is difficult to know what period they may be traced back to.

There is a cave that has been identified with the Sattapaṇṇi Cave, in which the First Council was held; if this is correct, the Council must have been a humble affair, for the cave is small.

The hills feature a dry-stone wall some 40 kilometres long, the remnant of ancient fortifications. These are probably pre-Mauryan, from the time when Rājagaha was still the main city of Magadha. The EBTs mention that king Ajātasattu was fortifying Rājagaha soon after the Buddha had died (mn 108/mā 145).⁶⁹⁴ This may be equivalent to the wall from New Rājagaha, which has been firmly dated to the 4th century bce, thus matching closely with the information given in the EBTs.⁶⁹⁵

5.10 Mathurā

Mathurā appears in the archaeological record as a major centre of Buddhism from around 150 bce.⁶⁹⁶ The stone artwork has a characteristic “Mathuran” style, which suggests a period of development in perishable materials during an earlier unattested artistic culture.⁶⁹⁷ Prior to this we have some traces of mud brick walls from the 3rd century bce, and earlier than that only some Northern Black Polished Ware, terracotta, and metal items.⁶⁹⁸ In the time of the Buddha, therefore, Mathurā was probably a minor trading town.⁶⁹⁹

This agrees with the depiction of Mathurā in the EBTs, where Mathurā is described as having five drawbacks: the ground is uneven, there is a lot of dust, the dogs are fierce, there are evil yakkhas and it is difficult to get alms food (an 5:220).

The growth of Mathurā from a dusty town to a prosperous city agrees with mn 84 and an 2:38, which are set after the time of the Buddha. The mention of a king, and the concern with winning over the ruling class of Mathurā, suggest the growing importance of Mathurā as Buddhism spread to the north-west.

5.11 Other Mauryan excavations

In addition to the above, Asokan era Buddhist remains have been discovered at all of the most important locations mentioned in the EBTs, and also at many lesser locations, such as: Kapilavatthu⁷⁰⁰ (mn 14.1/mā 100/T 54/T 55);^{701 702} Kosambī (sn 22:81/sā 973/sā² 207);^{703 704} Kusinārā (dn 16.5.1/smps 32.4);⁷⁰⁵ Lumbinī;⁷⁰⁶ Nāḷandā (dn 1.1.1);⁷⁰⁷ Pāṭaligāma/Pāṭaliputta/Patna (dn 16.1.19/smps 4.2);⁷⁰⁸ Pāvā (dn 29.1);⁷⁰⁹ Rāmagāma (dn 16.6.24/smps 51.13);⁷¹⁰ Sārnāth (sn 56:11);^{711 712 713} Sāvatthī, including the Jetavana (e.g. mn 4.1/eā 31.1);^{714 715} and Vesālī (e.g. mn 12.1/T 757).^{716 717}

At Sāvatthī archaeological excavations have established that the ancient town goes back at least to the middle of the 6th century bce.⁷¹⁸ Regarding Vesālī, Härtel dates the earliest habitation to around 500 bce.⁷¹⁹

At Kosambī a stone slab from the Kuṣāṇa period (1st–3rd century ce) with the inscription Ghositārāma Vihāra links this find to the EBTs.⁷²⁰ Archaeological excavations at Kosambī have unearthed fortifications dating to as early as the 6th century bce.⁷²¹ Again, this fits with the dating of the Buddha to the 5th century bce and the mention of Kosambī in the EBTs.

In general, there is a remarkable correspondence between the archaeological record and the main towns of the middle Ganges plain as described in the EBTs. Bronkhorst says:⁷²² “Erdosy recalls that Buddhist tradition recognizes six cities of outstanding importance which would have been fit to receive the mortal remains of the Buddha—Campā, Kāśī, Śrāvastī, Kauśāmbī, Rājagṛha and Sāketa—and points out that the first five of these correspond to the earliest urban centres reconstructed from archaeological evidence, omitting only Ujjain.”

5.12 South Indian art and inscriptions

Amarāvatī, a major Buddhist site in Andhra Pradesh, attests to the spread of Buddhism to the south shortly after Asoka. It has been estimated that the earliest Buddhist art at Amarāvatī stems from the second or even third century bce.⁷²³ It shares the aniconic features and workmanship of early North Indian Buddhist art.⁷²⁴

According to the Chinese pilgrim Xuan Zang, who visited South India in the 7th century, there was a monastery built by Mahinda, Asokas’s son, in the Pandyan country, which is close to Sri Lanka.⁷²⁵ Thus there is an ancient tradition that Buddhism arrived in South India in the Asokan period.

The fact that Buddhism was well-established in the south of India at such an early period attests to a movement that had already been in existence for a considerable period of time.

5.13 Northern Black Polished Ware culture

The sparse archaeological finds in the period from around 500 bce to the Mauryan era are characterised as the Northern Black Polished Ware (nbpw) culture, named after the glazed pottery common to this era.⁷²⁶ The cultural complex of this time also includes iron and the gradual introduction of burnt brick and coinage.

It is likely that Magadha’s rise to power was connected to iron ore mines south of Rājagaha, the iron from which could be used for weapons and tools. The furnaces used to prepare the iron may have been instrumental in the production of nbpw;⁷²⁷ at the very least the simultaneous appearance of iron and new forms of pottery attest to a development in the control of fire. The use of iron is widely attested in the EBTs and, to a lesser extent, also the use of coins. Burnt bricks, however, are only attested in the vibhaṅga and Khandhakas of the Vinaya,⁷²⁸ which are generally later than the EBTs.⁷²⁹

The geographical distribution of the nbpw culture is, apart from the outlier of Amarāvatī, similar to the geographical area known to the EBTs, although the nbpw culture is found over a slightly wider area due to its prevalence until the Asokan period.⁷³⁰ This confirms that, despite the political divisions, this region was a relatively unified culture in this period, which again is consonant with the EBTs.

The small scale and low level of urbanisation found in the nbpw culture also agree with the EBTs, which do not contain descriptions of large cities.

Chapter 6
Development of Buddhism

Earlier and later texts can be distinguished because the later ones contain developments not found in the earlier texts. At the same time the later texts refer to the earlier texts and assume their existence. Without the earlier texts the later ones do not make sense, whereas the EBTs are independent of later developments.

6.1 First Council

The First Council is recorded in the Vinaya of all schools whose texts are available.^{731 732} While they differ on details, each implying that it was their own version of the canon that was recited,⁷³³ they all agree that the business of the council was to recite the EBTs. These Vinaya accounts of the first Council all contain details about which Suttas and which parts of the Vinaya were recited, showing that the EBTs existed in some form in all these schools.⁷³⁴

Some Vinayas—the Dharmaguptaka, Sarvāstivāda, Mūlasarvāstivāda, and the Vinayamātṛkā Sūtra—also add the Abhidhamma, although the Pali, Mahāsāṅghika, and Mahīśāsaka Vinayas notably omit any mention of the Abhidhamma.^{735 736}

This is significant, as the Abhidhamma was a vital part of  Theravādin self-identity, and, despite this omission, later generations came to accept that it was spoken by the Buddha. Rather than make a simple addition to their canonical text, the  Theravādins preferred to authenticate the Abhidhamma through a legend of the Buddha visiting his mother in heaven and teaching her there.^{737 738} They also used the simple expedient of including the Abhidhamma in the flexible and expanding content of the Khuddaka Nikāya.⁷³⁹ Thus they developed and expanded their understanding of the Buddha’s teachings, but through interpretation and later legends, not by altering the EBTs.⁷⁴⁰

6.2 Second Council

The second council, 100 years after the parinibbāna, was convened to decide whether certain practices taken up by certain bhikkhus were in line with the Vinaya. Most of these practices were of minor or dubious importance, while at least one—the use of money by monks—had major ramifications. The Sangha was in crisis, and monks gathered from a large part of Northern India to ensure the new practices were rejected.^{741 742}

This illustrates the conservative nature of the early Sangha. Their duty, as expressed countless times, was to ensure that “the True Dhamma last long” (dn 29.17). Unlike modern academia, where innovation is rewarded, the monks were expected to hand down the teachings intact, and even minor changes or variant interpretations were considered unacceptable.⁷⁴³ Moreover, there was no leader who had the authority to make such changes.⁷⁴⁴

A common understanding of the teachings, at least while Buddhism was still confined to a relatively small area, was promoted by monks travelling and mixing with monks from different regions.

6.3 Literary developments

Two trends in Buddhism in the Mauryan period were particularly influential on Buddhist literature. On the one hand, the Sangha became much larger, better endowed, and more specialised, with detailed systems of Buddhist exegesis. On the other hand, the Dhamma reached out to a much broader spectrum of the lay community.

These trends correspond to the developments of the content in the doctrinally secondary books of the early canons. On the one hand, Buddhist literature saw the evolution of the Abhidhamma texts,⁷⁴⁵ a highly abstruse and specialised literature. On the other hand, there was the development of the popular stories of the Jātakas, Vimānavatthu, Buddha biography, and the like. This literature is thus situated naturally within the Mauryan period, and is quite unlike the literature of the EBTs, on which it depends.

Within the Pali school, the literary style of texts composed in Sri Lanka, such as the commentaries and the chronicles, is substantially different from the style of texts inherited from India.⁷⁴⁶ And since the new style that emerged in Sri Lanka is not encountered in the EBTs, it would appear that the Pali EBTs were considered closed upon their arrival on the island, that is, at the time of Asoka.⁷⁴⁷

Also the absence of any mention in the EBTs of the Buddha visiting Sri Lanka is significant. This legend, found in several early non-EBT sources,⁷⁴⁸ is an important part of the Sri Lankan sense of identity and it would have been tempting to slip it into the Pali EBTs. That this was not done again testifies to the conservatism of the tradition.

As for Mahāyāna texts, although the texts themselves tell us that they were spoken by the Buddha, they clearly belong to a later literary period than the EBTs. This is clear for both stylistic and linguistic reasons, and is agreed on by most Buddhist academics.^{749 750 751}

6.3.1 Jātakas

Jātakas are depicted in several of the earliest Buddhist monuments, such as Bhārhut and Sāñcī, but they are clearly later than the EBTs.⁷⁵² This is despite the fact that they often depict social conditions that are earlier than the time of the Buddha (e.g. the king in Benares) and many of them must have their roots in that period.⁷⁵³ The large number of Jātakas originated over a long period, some being pre-Buddhist, and were adopted into Buddhism as it became a popular religion in the Mauryan and following eras. Important developments include:

1. The Jātakas, unlike the EBTs, are almost entirely narrative.

2. Even the earliest layer of the Jātakas, the verses, generally lack the distinct Buddhist terminology and doctrinal terms found in the EBTs.

3. Whereas in the EBTs the Buddha or one of his chief disciples invariably is the protagonist, in the Jātakas the protagonist is always the bodhisatta.

4. The structure of the Jātaka tales is different from Suttas of the EBTs.⁷⁵⁴

5. The Jātakas mention things not found in the EBTs such as bricks (iṭṭhaka) and trade routes to countries outside of India (e.g. Jā no. 339).

6.4 Doctrinal developments

There are numerous doctrinal developments between the EBTs and non-EBTs. Such developments are not occasional or arbitrary, but follow broad-based patterns that are characteristic of philosophical and doctrinal development. This development is thoroughgoing and it is possible to pick virtually any significant teaching found in the EBTs and show how it gradually develops and evolves through the later literature. Whether such changes are merely rewordings and clarifications, or whether they indicate a genuine shift in the doctrine, is irrelevant for this point. The wording of the teachings definitely changed, and this is enough to establish the relative chronology. Some specific examples are:

1. Karma as destiny (Vimāna- and Peta-vatthu);⁷⁵⁵

2. Transference of merit (Petavatthu);^{756 757}

3. Elaboration on and fascination with good and bad destinations of rebirth (Vimāna- and Peta-vatthu);

4. Shift in attitude to jhāna, especially the Abhidhammic concept of lokuttarajjhāna (Vibh);

5. Dependent origination happening on an ‘occasion’ (Vibh 145);

6. Emphasis on and detailed explanation of obscure doctrines in the EBTs, e.g. the relay coaches of mn 24/mā 9/eā 39.10 and the kasiṇas (Vism);

7. Increased detail and expansion in the exposition of core teachings of the EBTs, such as the description of insight (Vism);

8. Abstract teachings as opposed to applied (Abhidhamma, Paṭisambhidāmagga);

9. Systematisation of “dhammas” (Dhammasaṅgaṇī);

10. Explanation of dhammas as “bearing their own essence”, sabhāva (As 39,11);

11. Khaṇikavāda, the “doctrine of momentariness” (e.g. Vism 268,14; Vibh-a 27,3);

12. Systematisation of “conditions” (Paṭṭhāna);

13. Omniscience of the Buddha (jn 99);

14. Lineage of past Buddhas (Buddhavaṁsa);

15. Pāramīs (Cariyāpiṭaka);

16. Bodhisatta path (Cariyāpiṭaka and commentaries);

17. Books on history (Aśokāvadāna, Dīpavaṁsa, Mahāvaṁsa).

Texts such as the Vibhaṅga of the Abhidhamma make it particularly clear what was the direction of these developments. The Vibhaṅga is divided into two parts: an analysis according to the Suttas and one according to the Abhidhamma. The Sutta analysis is similar to the EBTs, whereas the Abhidhamma analysis is a new development.⁷⁵⁸ Since the Sutta material forms part of the Abhidhamma but the Abhidhamma material is not, or very rarely, found in the Suttas, the direction of development is unambiguous.

In general, the direction of doctrinal development is obvious. The texts containing new doctrines always look back to the EBTs as their starting point. This is very clear with the commentarial literature, but also with such works as the Visuddhimagga, which quotes extensively from the Suttas. The same is true of the Abhidhamma. For most non-EBT texts it can easily be shown that they base their philosophy on the EBT material but take it further, sometimes in new directions. The EBTs, by contrast, do not refer to ideas that are developed in other Buddhist literature.

6.4.1 Absence of sectarian views

There are extremely few examples of definitively sectarian views in the EBTs. One rare exception is the occasional interpolation of the three times (past, present, and future) in the Sarvāstivāda Saṁyukta Āgama.⁷⁵⁹ This means that the schools generally did not insert their own views into the EBTs, but developed them in other literature, such as the Abhidhamma.^{760 761}

6.4.2 Early Abhidhamma

The earliest stratum of Abhidhamma literature has been identified by Frauwallner, Warder, and others, on the grounds of doctrine, style, and agreement between schools.^{762 763} This literature consists of texts that quote from the Suttas and then offer analysis or elaboration of those texts.⁷⁶⁴ In some places they actually say they are analysing in accordance with the Suttas, as contrasted with the analysis according to the Abhidhamma.⁷⁶⁵ But even in cases where it follows the Sutta method, the analysis involves a more developed, systematic mode of analysis compared with the EBTs, though little overt change in the doctrine.⁷⁶⁶

The fact that these texts contain both EBT material (or very close) and new Abhidhamma material, whereas the EBTs have very few traces of Abhidhamma material, shows the direction of the development: the Abhidhamma must have come into existence after the EBTs.

Further, since some of the Abhidhamma material is shared between schools and some is not, the Abhidhamma project must have begun in the nascent sectarian period (around 300 bce). The origin of the shared material would be even earlier, which means the EBTs must have existed before then. But even in the shared basis of these texts, the level of agreement is far less than between parallel versions of the EBTs.

6.4.3 Kathāvatthu

The Kathāvatthu of the Pali Abhidhamma was begun in the time of Asoka, or not long thereafter, which places its beginnings around 150–200 years after the Buddha. This dating is based on multiple lines of evidence:

1. The traditional account, which states that it was composed under Asoka.⁷⁶⁷

2. The traces of Māgadhan dialect, which are more numerous in the early sections.⁷⁶⁸

3. The agreement with the Vijñānakāya, a similar but much briefer analysis in Sanskrit.⁷⁶⁹ This mentions the Kathāvatthu’s author by name.⁷⁷⁰ Thus the Kathāvatthu must have been started before the Sangha’s dispersal to Sri Lanka and Northern India.

4. The focus on the fundamental schismatic issues, which were the hot topics of the time. Later additions to the Kathāvatthu address later issues. The Mahāyāna is never mentioned.⁷⁷¹

The Kathāvatthu is in the form of a dialogue where the texts and ideas of the EBTs are presented and opposing interpretations given. For the Kathāvatthu, the EBTs, or more precisely the four main Nikāyas/Āgamas plus a small part of the Khuddaka Nikāya, are always the final authority in settling opposing views, not the Abhidhamma or any other Buddhist text.⁷⁷² In fact it argues about practically everything except the actual letter of the text of the EBTs.

It must have taken considerable time to develop such systematic doctrinal readings of the EBTs, and even longer to develop the Kathāvatthu’s formal logic, which is the first example of rigorously applied formal logic in India. So the EBTs must be much earlier than this.

In several cases the Kathāvatthu quotes from EBT passages that contradict or appear to contradict the teachings of the  Theravādin school itself.⁷⁷³ Nevertheless, there is never any question of whether these passages might not be the word of the Buddha. In other cases, passages from the EBTs that were used by non-Theravādins in support of their special interpretations, but were awkward from a Theravādin point of view, were left untouched.⁷⁷⁴ The disagreements are on the interpretation, not on the contents of the EBTs.

6.5 Texts rejected by some

Authenticity is not a modern notion. It was central to the early community’s understanding of its own religion, and the authenticity of certain texts is questioned in the traditions. These texts are among those considered by modern authorities as late additions. Texts rejected by various schools include:

1. According to the Sinhalese chronicle the Dīpavaṁsa (5.37 of Oldenberg’s translation), the Mahāsāṅghika rejected the Parivāra (the last book of the Pali Vinaya Piṭaka), the six books of the Abhidhamma (six, because this is said to have occurred before the composition of the Kathāvatthu), the Paṭisambhidāmagga, the Niddesa, and some of the Jātakas.

2. According to the Pali Abhidhamma commentary, a student was of the opinion that the Abhidhamma was not spoken by the Buddha.⁷⁷⁵ This was refuted based on the occurrences of the word Abhidhamma in the Vinaya. However, this shows that these parts of the Vinaya are late rather than showing that the Abhidhamma is early.⁷⁷⁶

3. Several Chinese monks around the 5th century rejected some or all of the Mahāyāna Sūtras: Hui-tao doubted the Pañcaviṁśatisāhasrikā Prajñāpāramitā Sūtra; T’an-le disparaged the Lotus Sūtra; Seng-yuan belittled the Mahāyāna Mahāparinirvāna Sūtra.⁷⁷⁷

4. The Sautrāntikas rejected the Abhidhamma.⁷⁷⁸

Chapter 7
Theoretical considerations

7.1 Scientific characteristics

There is no complete agreement on what constitutes a scientific theory, but there are, however a number of well-recognised characteristics.^{779 780} The theory that the EBTs are authentic (toa—Theory of Authenticity) fulfils these criteria, while the theory that the EBTs are inauthentic (toi—Theory of Inauthenticity) does not.

1. Falsifiability: The toa may be easily falsified; that is, it is a “risky” theory in Popperian terms. For example, a single epigraphic reference to a Mahāyāna Sūtra dated to the 4th or 5th century bce would conclusively refute the toa. On the other hand, it is not easy to see how the toi could be falsified, even in principle, still less in practice. Such theorists argue that “we do not know”, or that “the evidence may be incomplete”, or that there may have been a “systematic removal” of texts. Since all empirical knowledge is uncertain, incomplete, or subject to distortion, such objections can never be finally refuted.

2. Plasticity: A good theory is refined and improved when new evidence or arguments come to light. The toa has repeatedly done this. Examples include the date of the historical Buddha, whether Pali was his language, and whether the original texts were the same as the Pali, an unknown Ur-Canon, or a more loosely defined body of texts.

   By contrast, proponents of the toi, despite repeated criticism by experts in the field, have not substantially changed or adapted their arguments.

3. Predictive power:⁷⁸¹ The toa has accurately predicted many subsequent developments in the field. A dramatic example is the statement by Samuel Beal in 1882: “… when the Vinaya and Āgama collections are thoroughly examined, I can have little doubt we shall find most if not all the Pali Suttas in a Chinese form.”⁷⁸² This has been thoroughly substantiated by over a hundred years of comparative studies.

   Another example is the archaeological discoveries by Alexander Cunningham. He did not have the Pali or other EBTs available to him, but using the Greek sources, the Hindu Purāṇas, and especially the work of the Chinese pilgrim Xuanzang as his guide, he was able to locate many of the sites referred to in the EBTs. While more detailed work has necessitated subsequent revision of some of his findings, he was able to successfully map out the overall geography of ancient India, and most of his relevant identifications have been confirmed by positive archaeological identifications.

   A further example is the relative dating of the books of the Pali canon by Rhys Davids, a list that has been mostly corroborated by subsequent studies.⁷⁸³ This is not just true of the overall picture, but of many details. One example is the question of the minor class of Vinaya rules, known as dukkaṭa. Oldenberg, based on his study of the Pali Vinaya, firmly said that this category arose only after the Buddha, because the monks would not make changes to the original classes of offences.⁷⁸⁴ At the time, comparative study of the Vinaya was still in its infancy, and Oldenberg would not have known that the Mahāsāṅghika Vinaya entirely lacks the category of dukkaṭa, but has the term vinayātikrama in a similar role.

   The toi, on the other hand, has never predicted any major discoveries.

4. Fecundity: The toa has been a highly fecund theory, that is, it has resulted in multiple further developments and discoveries. These include the reconstruction of ancient Indian history, the establishment of Indian archaeology, and the historical understanding of the relations between and development of the Buddhist texts of the various schools and languages. In addition to these purely academic results, the theory has inspired a spiritual resurgence in traditional and new Buddhist cultures.

   By contrast, the toi has not resulted in any significant practical or theoretical developments. Its main impact seems to have been the disparagement of any study of Early Buddhism as useless because “we don’t know”. This has resulted in the neglect of the field, despite the very large amount of basic work that still needs to be done, as very many early texts have never been translated or adequately studied. This sterility has impacted on a variety of related specialities, such as archaeology, philosophy, and psychology, where non-specialists have been led to believe that the authenticity of the EBTs has been conclusively refuted, and therefore that there is no value in even trying to study them.

5. Simplicity: A good theory is able to account for a large number of facts with a small number of assumptions. The toa is based on the simple, rational assumption that the EBTs attributed to the Buddha were largely spoken by him, unless there are good reasons to believe otherwise. With this simple starting point we are able to trace in a meaningful way the development of Buddhism.

   By way of comparison, it is difficult to work out exactly how the toi proposes to explain the existence of the Buddhist texts. Sometimes the toi theorists seem to imply that the texts were composed by later generations of monks, but mostly they claim the question is uncertain, which is, of course, a non-theory.

6. Extensiveness: The toa encompasses a very wide range of facts known about Early Buddhism, as we have demonstrated throughout. By contrast, the toi scholars have only considered a tiny fraction of the relevant evidence.

7. Coherence: The toa is internally consistent, and does not violate any normal scientific or common sense principles. The toi, on the other hand, sometimes asks us to believe that ancient monks invented a vast doctrinally and historically stratified literature for their own selfish interests.

8. Responsiveness: A good theory is not an isolated set of concepts; it lives in a constant climate of give and take, listening to critiques seriously and responding to them meaningfully. We have cited some examples of this above, such as the article by B. C. Law that responds to, corrects, and improves on Rhys Davids’ chronological list of Pali texts. Another example is the study by Noble Ross Reat of the Śālistamba Sūtra. Earlier, the sceptical scholar Edward Conze had argued that we can only accept as authentic teachings that are shared between the Mahāsāṅghika and Sthavira groups of schools. Reat took up the challenge and was able to meet Conze’s criterion.⁷⁸⁵

   While the toi theorists have engaged in debate, this remains essentially superficial, as the theoretical mood of the toi precludes acceptance of any genuine knowledge of the period.

7.2 The character of inductive theories

The toa is similar to inductive theories in the classical sense, that is, it uses observation of particulars to test, correct, and improve general hypotheses, which in turn suggest avenues for further investigation and research.^{786 787} It attempts to account for the entire range of what is known about the period. The arguments of the toi theorists, on the other hand, are based on theoretical positions, with only occasional attempts at providing corroborating evidence.⁷⁸⁸

Examples of inductive theories include the theories of evolution and global warming. One of the characteristics of such theories is that they are probabilistic, and hence much better at establishing generalities than specifics.⁷⁸⁹ This problem is well known in the case of global warming: the theory cannot predict whether any specific day will be hot or cold, but it can say with a high degree of probability that there will be more and more hot days in coming years.

Similarly, while we cannot say with any certainty that any specific text was literally spoken by the historical Buddha, we can say with a high degree of probability that the texts as a whole stem from him. This distinction is one of the rhetorical fault lines in the discussion of authenticity.⁷⁹⁰ Sceptics assert that we can’t know for certain whether any specific phrase was spoken by the Buddha,⁷⁹¹ while advocates affirm that the doctrinal teachings as a whole must come from him. While the sceptical assertion is true, it is trivially so, and has no real effect on our understanding of the status of the texts.

7.3 The problem of specifics

Here is one example of this problem of specifics. The Mahāparinibbāna Sutta says that when faithful devotees visit the site of the Buddha’s birth, they will reflect, “Here the Buddha was born.” In Pali this is idha tathāgato jāto ti (dn 16.5.8). The corresponding Sanskrit text uses bhagavā instead of tathāgata: iha bhagavāñ jātaḥ⁷⁹²—different words with the same meaning. This passage seems to be quoted in the Lumbinī Asokan pillar, which however uses two distinct phrases: A: hida budhe jāte sakyamunī ti; B: hida bhagavaṁ jāte ti. The ti indicates “end quote”. Thus one of the forms on the pillar corresponds with the extant Sanskrit text, while one is different from both the Pali and the Sanskrit.

On a word level these are all equivalent; they are common terms that refer to the same individual. We might distinguish tathāgata, which is normally used by the Buddha when he refers to himself, whereas bhagavā and buddha are used by others. This would correspond with a situation where the Pali form is that spoken by the Buddha, while the other forms are those spoken by others about the Buddha.

Still, the meaning is the same in all cases. Thus focussing on individual differences obscures the larger point, which is that several diverse sources, from Sri Lanka, Nepal, and Kashmir (and doubtless other sources not covered here) all contain the same phrase in the same context. The general inference, then, is that the early Buddhist tradition acknowledged the Buddha’s birthplace, that early Buddhists were encouraged to reflect on this at the place itself (reflection being implied by the “ti”), and that the textual and epigraphic sources confirm each other. From this, and the reference to Asoka visiting sambodhi (Bodh Gaya), it is reasonable to infer that the other sacred sites mentioned in the Mahāparinibbāna Sutta were also places of pilgrimage from an early date, probably from immediately after the Buddha’s parinibbāna.

These are not isolated or uninteresting details. They raise questions as to the nature of the worship of the Buddha, the extent to which he was regarded as a historical figure, located in time and place, the emotional attitude of the early Buddhist community towards the Buddha, as well as possible economic and doctrinal implications of an early practice of pilgrimage. If we simply dismiss the evidence on the grounds that it is uncertain, we lose the chance to learn meaningful things about the Buddha and his early community.

7.4 Denialist Buddhism

Critics of Early Buddhism have adopted a rhetoric of scepticism in order to dismiss the notion of authenticity. Their arguments are apparently intended to be hard-nosed and unsentimental, but when examined closely they are reminiscent of arguments by denialists of various types, such as those relating to the harmful effects of tobacco, creationism, or the reality of man-made climate change. Just as sceptics characterise the search for authenticity as “Protestant Buddhism”, it seems appropriate to describe this form of scepticism as “Denialist Buddhism”.

The unifying characteristic of the various forms of denialism is their insistence on extreme, unreasonable scepticism regarding any truth claims they oppose. The following quote is from a tobacco industry executive, but it might just as well describe the fundamental principle of the sceptics of Early Buddhism.

… we are committed to an ill-defined middle ground which is articulated by variations on the theme that ‘the case is not proved.’ ”⁷⁹³

Denialist movements share a common intellectual ancestry, in that they appropriate elements of postmodern thought in order to marginalise or denigrate science. And they share an unmistakable “tell”, by which they can be distinguished from genuine scepticism. Denialists purport to be about a certain field, but they make no contributions to that field. Creationism has made no contribution to biology; climate change denial has made no contribution to climate science; and Denialist Buddhism has made no contribution to understanding the Buddha or his teachings. Its effect has been to dissuade people from even trying.

There are genuine epistemological problems in the study of Early Buddhism; in that we can all agree. But to argue, as the denialists do, from epistemological uncertainty to dismiss the entire field, is to get everything backwards. To study Early Buddhism is to study the origins of one of the great spiritual movements of humanity. It is a topic of intrinsic interest and worth. Of course there are difficulties, as there are in any field. But genuine enquirers take the problems as a challenge, and figure out ways to solve the problems. That’s how science works. You make the unknown knowable. You don’t run away from the problem and dismiss those who are trying to learn.

Diethelm and McKee have identified various characteristics of denialists.⁷⁹⁴ Below we show how these characteristics apply to sceptics of early Buddhism.

For examples of the sceptical point of view we use Gregory Schopen’s Bones, Stones, and Buddhist Monks, which has probably been the single most influential book in Buddhist studies in the last generation, in some countries at least. Schopen advocates an essentially negative attitude, saying, “It is however, very likely that [the canonical texts] will not tell us very much, and this, perhaps, gives rise to the broadest generalisation that we can make.”⁷⁹⁵ With this as its starting point, there is little wonder that this movement has produced little to help us understand the EBTs. The main product of this form of scepticism is doubt, and the main outcome of such doubt is to give up trying to learn.⁷⁹⁶

1. The identification of perceived conspiracies (including belief in corrupted peer review and inversionism).

   Schopen attributes the absence of rules regarding stupas in the Pali Vinaya to their “systematic removal” by the  Theravādin monks. He argues that a couple of references in much later texts, and some in the Pali Vinaya itself,⁷⁹⁷ indicate that there were rules regarding these matters in the Vinaya, and that their loss could not be explained by accidental omission.⁷⁹⁸ He fails, however, to give any convincing account as to how or why such a change would be made. Moreover, his evidence is easily accounted for.^{799 800 801} Thus on the basis of evidence that is to start with extremely thin, and which is easily explained by more simple means, Schopen postulates a vast hidden conspiracy that was able to make major systematic changes to an ancient scripture without anyone noticing. Elsewhere he does not even try to gather any evidence, but merely throws out the suggestion that the texts have been “intentionally altered.”⁸⁰²

2. The use of fake experts (often with the smearing of real experts).

   Climate change denialists often cite “scientists” who disagree with the consensus on global warming. On examination, the scientists almost invariably turn out to be experts in fields other than climate change. Similarly, most of the hard sceptics come from a background of later Buddhism or Tibetan studies, while those who specialise in Early Buddhism or Indian studies generally tend to affirm authenticity. The supposed “protestant” bias of early Indologists is criticised,⁸⁰³ but the denialists do not acknowledge their own biases as secularists, materialists, postmodernists, or academics. Everyone has biases. Appreciating a person’s perspective helps understand their approach to a subject, but this needs to be done without dismissing reasonable arguments and conclusions.

3. Demanding impossible standards for research.

   Critics point to the absence of early manuscripts,⁸⁰⁴ or the absence of specific details in the archaeological record, such as early evidence of monasteries. This parallels the practice in climate change denialism or creationism of claiming to refute the science by pointing to the absence of certain concrete pieces of evidence. These theories, however, like the toa, do not depend on specific, singular proofs, but on the convergence of large amounts of supporting data for which there is no better explanation.

   Inference, when properly used, does not rely on a chain of uncertain claims, one after another, diminishing in probability when compounded; but on the convergence of multiple lines of reasoning and evidence that independently confirm the result. The cautious use of such empirical method is described by Rhys Davids: “… before drawing the conclusion that, therefore, the Nikāyas, as we have them, are older than the existing text of the Mahābhārata, we should want a very much larger number of such cases, all tending the same way, and also the certainty that there were no cases of an opposite tendency that could not otherwise be explained.”⁸⁰⁵

4. Use of fallacy, including misrepresentation and false analogy.

   Schopen rarely considers seriously the methods used by scholars of early Buddhism, and when he does he misrepresents them. For example, he says the cardinal tenet of the higher criticism is that if all versions of a text agree they must stem from a pre-sectarian tradition.⁸⁰⁶ This ignores the fact, which is a very basic one, that the main picture of Buddhist history in India was developed in the 19th century, before any substantial comparative work was done. Scholars used a variety of methods, some of which we have outlined here, to arrive at conclusions that have, by and large, been confirmed by later comparative studies.

We live in an era of unprecedented potential for the study of Early Buddhism. Internationally, more people than ever are interested in Buddhism. Emerging technologies have opened up entirely new ways of making texts available and analysing them. Much of the potential of the field remains virtually untouched. Most of the EBTs remain untranslated into English.⁸⁰⁷ There has been little or no application of digital or statistical analysis to the texts, which surely must be a field that offers promising insights. When so much basic work is left undone, it is absurd to claim that we cannot know anything. We simply don’t know what we can know. Early Buddhism is a field of study in its infancy, with a huge scope of texts and other evidence, and a tiny population of serious scholars. We need to be supporting innovation and opening up frontiers, not prematurely dismissing the very possibility of knowledge.

Conclusion

We know when the Buddha lived, where he lived, who he associated with, how he lived, and what he taught. We know these things with greater certainty than for almost any other historical figure from a comparable period. And we know this because of the EBTs. All other historical and archaeological information about the period depends on the EBTs to make sense.

One of the distinctive features of the EBTs is their careful and conservative attitude to knowledge. “Buddha”, after all, means “Awakened”, and the development of wisdom is right at the heart of the Dhamma. The Buddha did not just teach the truth as he found it, but carefully detailed the way to discover the truth, and warned against leaping to rash conclusions. When we hear words of caution against accepting anything on mere faith, we are hearing the unmistakable inflection of the historical Buddha’s own voice.

The Buddha’s caution in what we can know is, however, balanced by his optimism, his belief, confirmed by his own experience, in the human mind. We can know. Not perfectly, perhaps, but well enough. There is no point in seeking for some ultimate truth in the messiness of the world we live in. What we need is a practical truth, one that is good enough.

In the Sandaka Sutta, the Buddha points out the dangers in living the spiritual life based on belief in a scriptural tradition. He says, with his characteristic reasonableness, that what is passed down in tradition may be well learnt or badly learnt, it may be true or it may be otherwise (mn 76.24). Only if a spiritual life leads to the ending of suffering is it of true value.

The point of establishing the authenticity of the EBTs, then, is not to prove that they are all true. It is to show that they are useful. Within the corpus we indeed find that some things are well learnt, others badly learnt; some true, and some otherwise. With some reflection and effort, we are able to discern these things. What remains is a powerful, clear, balanced, and profound approach to the spiritual life. This approach has been of benefit to countless people, and we believe that it remains so today. By encouraging the study of the EBTs, we believe more people will be moved to apply these teachings and test them in the only way that the Buddha cared about: to reach the end of suffering.

-- END OF BOOK --
Wisdom & Wonders