Pali Canon

• 1. Introduction

  ›

  The Pali Canon is the most complete surviving collection of early Buddhist scriptures, preserved in a language called Pali and holding a place in the Theravada tradition that no other text can match. It was not printed until 1900, when the first complete edition appeared in Burma in 38 volumes. Yet long before ink touched paper, monks were expected to carry its entire contents inside their heads. A Burmese monk named Vicittasara memorized the whole Canon for the Sixth Buddhist Council, a feat that gestures at just how differently the ancient world thought about what it meant to "keep" a text. Where did this immense body of teaching come from? Who actually wrote it? And how did words spoken in northeastern India survive the centuries to become a living scripture for millions of people today?

• 2. Three Baskets

  ›

  Pitaka means "basket" in Pali, a word that refers to the actual palm-leaf receptacles in which manuscripts were stored. The Canon divides into three of these baskets, and that division shapes everything about how Theravada Buddhism understands its own teachings. The first basket, the Vinaya Pitaka, deals with the rules of the monastic community. The rules were not issued all at once; according to the stories embedded in the text, the Buddha laid them down one by one as specific behavioral problems or disputes arose among his followers. The Vinaya breaks into three internal parts: a commentary on the Patimokkha, a code of rules for monks and nuns; a second section of 22 chapters grouping other rules by topic; and a third section that analyzes the rules from multiple angles.

  The second basket, the Sutta Pitaka, is the largest. It holds the discourses and sermons attributed to the Buddha, organized into five collections called nikayas. The Digha Nikaya gathers 34 long discourses; scholar Joy Manné argues that its high proportion of debates and devotional material suggests it was designed to make converts. The Majjhima Nikaya holds 152 medium-length discourses, which Manné reads as aimed at giving new converts a solid grounding in the teaching. The Samyutta Nikaya collects thousands of short discourses in roughly fifty groups arranged by subject or person; Bhikkhu Bodhi, in his translation, describes it as the nikaya with the most detailed doctrinal explanations. The Anguttara Nikaya contains thousands of short discourses arranged numerically, from ones to elevens, and carries more elementary teaching than its three sister collections.

  The third basket, the Abhidhamma Pitaka, stands apart. Where the suttas address the hearer in a flexible, adapted way, the abhidhamma pursues systematic analysis of mind and doctrine across seven distinct books. Cousins summarized the distinction well: where the suttas think in terms of sequences or processes, the abhidhamma thinks in terms of specific events or occasions. The Abhidhamma is strictly a Theravada collection. Its seven books have little in common with the Abhidharma works recognized by other Buddhist schools, and the first of them, the Dhammasangani, runs to enumeration, definition, and classification of dhammas at a level of technical precision the suttas rarely attempt.

• 3. From Voice to Palm Leaf

  ›

  Buddhist tradition places the Canon's beginnings in Rajgir, three months after the death of Gautama Buddha, at the First Buddhist Council. There, a monk named Ananda recited the Sutta Pitaka and a monk named Upali recited the Vinaya Pitaka. The assembled arhats accepted the recitations, and from that point the teachings traveled not on paper but inside living human beings. The Theravada tradition holds that the Canon was recited orally from the 5th century to the 1st century BC, kept alive by regular communal recitations that reinforced memorization.

  The transition to writing happened in Sri Lanka. The Tipitaka transmitted there during the reign of King Asoka was first committed to palm leaves at the Fourth Buddhist Council in 29 BC, roughly 454 years after the Buddha's death. The Sri Lankan Mahavamsa places this event in the reign of King Vattagamani. The move was not uncontested: more conservative monks likely opposed the innovation, and it took time before written transmission was broadly accepted. The Bhanakas, a tradition of specialist reciters, continued to exist alongside written scriptures for many centuries afterward, which shows that the act of writing the Canon down was not the end of oral tradition but the beginning of a new layer alongside it.

  The oldest surviving manuscripts of the Pali Canon are from late in the 15th century. The climate of Theravada countries is harsh on organic materials, and apart from brief quotations in inscriptions and a two-page fragment from the 8th or 9th century found in Nepal, nothing older survives. The earliest textual fragments of canonical Pali outside Sri Lanka were found in the Pyu city-states in Burma, dating to the mid-5th to mid-6th century CE.

• 4. Who Wrote These Words

  ›

  Richard Gombrich argues that the main preachings in the Vinaya and Sutta Pitaka are coherent and cogent enough to be the work of a single mind, namely the Buddha himself rather than a committee assembled after his death. Peter Harvey takes a more cautious line, saying that while "much" of the Canon must derive from the Buddha's teaching, "parts of the Pali Canon clearly originated after the time of the Buddha." J.W. de Jong put it plainly: it would be "hypocritical" to assert that nothing can be known about earliest Buddhism, and the basic ideas found in the canonical writings could very well have been proclaimed by the Buddha, transmitted by disciples, and finally set into fixed formulas.

  Prayudh Payutto argues that the Canon represents the Buddha's teachings essentially unchanged apart from minor modifications. His case rests partly on the processes of the first council and partly on the methods for memorization that monks used during the Buddha's own lifetime. Bhikkhu Sujato and Bhikkhu Brahmali support early dating on several grounds: the technology described in the Canon matches the rapid technological development of the Buddha's era; the texts contain no back-written prophecies concerning King Ashoka, which Mahayana texts typically include; and the political geography depicted in the Canon reflects India at the time of the Buddha rather than the changed landscape that followed his death.

  Some scholars lean toward agnosticism. Ronald Davidson says he has little confidence that much, if any, surviving Buddhist scripture is actually the word of the historical Buddha. Geoffrey Samuel attributes much of what the Pali Canon represents to the work of Buddhaghosa and his colleagues in the 5th century AD. Gregory Schopen argues that definite evidence about the Canon's contents does not appear until the 5th-6th centuries AD. That claim drew criticism from Alexander Wynne, who points out that the Calcutta-Bairat edict of Ashoka lists several texts from the Canon by name, suggesting some of them were already fixed during Ashoka's reign from 304-232 BC.

• 5. Pali, Language and Place

  ›

  Theravada tradition has long identified Pali as the language of Magadhi, the dialect spoken by the Buddha in the kingdom of Magadha. Linguists disagree. Careful analysis shows that Pali is more closely related to the Prakrit languages of western India, and the preserved examples of Magadhi reveal substantial incompatibilities with Pali. The geographical fingerprints inside the Canon itself complicate the picture further: identifiable texts cluster around the Ganges region of northeastern India, touching the kingdoms of Kosala, Kasi, Vajji, and Magadha, which is consistent with Magadhan origins. But linguistic research suggests the Buddha's teachings may have been recorded first in an eastern Indian language and then transposed into a western Indian precursor of Pali sometime before the Asokan era.

  The Canon was then redacted extensively in the 5th or 6th century CE in Sri Lanka, nearly a thousand years after the Buddha's death. That redaction is the version most scholars work from today. The Sri Lankan Pali Canon itself acknowledges an earlier redaction toward the end of the 1st century BC. Early Buddhism scholar Lars Fogelin concluded that the Sri Lankan Canon is a modified Canon, and that there is no good reason to assume that Sri Lankan Buddhism resembles Early Buddhism as it existed on the Indian mainland.

  Dr. Peter Masefield's research on Indochinese Pali, sometimes called Kham Pali, adds another strand. Long dismissed as a degraded form, Masefield found it to be an internally consistent dialect. Records in Thailand indicate that when monastic ordination died out in Sri Lanka, many texts were lost; surviving texts had traveled with the third re-introduction of Theravada Buddhism into Sri Lanka from Siam, and portions had been translated into Indochinese Pali and then, at least in part, back into Pali again.

• 6. Printing, Translation, and Access

  ›

  The Pali Text Society published the first scholarly edition of the Pali text in Roman script between 1877 and 1927, eventually filling 57 volumes including indexes. That project made the Canon legible to Western readers for the first time, but the translations that accompanied it drew sharp criticism from within the Society itself. The then-President stated in 1994 that most of the translations were unsatisfactory. A former President said in 2003 that most had been done very badly. Paul Griffiths coined the term "Buddhist Hybrid English" for the resulting style, calling it "deplorable" and "comprehensible only to the initiate, written by and for Buddhologists."

  Translation work continued through the 20th and into the 21st century. Bhikkhu Nanamoli and Bhikkhu Bodhi's translation of the Majjhima Nikaya appeared from Wisdom Publications in 1995. Bhikkhu Bodhi's translation of the Samyutta Nikaya came from the same publisher in 2003, and his Anguttara Nikaya followed in 2012. In 2018, Bhikkhu Sujato made new translations of all five Nikayas freely available on the website SuttaCentral, releasing them into the public domain.

  A Japanese translation edited by Takakusu Junjiro appeared in 65 volumes between 1935 and 1941 as The Mahatripitaka of the Southern Tradition. A Chinese translation of that Japanese edition was undertaken between 1990 and 1998 and then printed under the patronage of Yuan Heng Temple in Kaohsiung. Digital access accelerated in 1996, when the Dhammakaya Foundation and the Pali Text Society cooperated to produce PALITEXT version 1.0, a CD-ROM database of the entire Pali Canon. A searchable transcript of over 16,000 pages later became available through BudhgayaNews, allowing users to find any word across the whole Canon at once.

• 7. Memory as Practice

  ›

  Even after two millennia in written form, the Canon's oral roots shape how it is used. Memorization and recitation remain active practices in Theravada communities, and the texts most frequently recited are the Paritta. Lay people typically know at least a few short texts by heart and recite them regularly; within the tradition, this is considered a form of meditation, provided the practitioner understands the meaning. Monks are expected to hold a considerably larger portion of the Canon in memory.

  Buddhaghosa, working in the 4th-5th century AD, compiled the authoritative traditional interpretation of nearly the entire Canon, drawing mainly on earlier materials that have since been lost. His Visuddhimagga summarizes the traditional Theravadin understanding and remains the single most important secondary text in the tradition. Subcommentaries written after Buddhaghosa pushed the interpretive project further, commenting on both the Canon and on Buddhaghosa's own commentaries.

  A spokesman for the Burma Sasana Council described the Canon as containing everything needed to show the path to nirvana, while noting that the commentaries and subcommentaries sometimes include speculative matter alongside illuminating illustrations. Rupert Gethin offered a wider frame: he suggests that the whole of Buddhist history may be read as a working out of the implications of the early scriptures, a claim that positions the Canon not as a closed archive but as a generative source whose meanings are still being drawn out.