Writings: Buddhism

How were Buddhist texts handed down?

The early teachings were memorized first, then written down, and each school or tradition kept its own canon, its set of authoritative texts.¹ Theravada is the branch of Buddhism followed in Sri Lanka and Southeast Asia. It holds that its canon, in the Pali language, was first written down in Sri Lanka in the first century BCE. Lewis Lancaster gives the tradition's date as 29 BCE, Charles Prebish gives 25 BCE, and the Access to Insight timeline gives about 100 BCE.²³⁴

Korea's nomination of its woodblock canon to UNESCO, a government heritage document, states that the teachings were first written down 200 years after the Buddha's death. That date differs from Theravada's.⁵

Mahayana, the "Great Vehicle" of Buddhism, holds that the Buddha himself taught its Perfection of Wisdom sutras, discourses on emptiness, in his second turning of the wheel of teaching. The Tibetan canon project 84000, which reports this, adds that traditional accounts say a longer version, hidden in a non-human realm, was retrieved by Nagarjuna.⁶⁷ Lancaster dates the emergence of Mahayana to around the beginning of the common era, a date that Gregory Schopen holds was fixed mostly by convention. Paul Harrison holds that Buddhists judged a teaching by whether it was meaningful, accorded with the dharma, and led toward liberation. In this way, he argues, some of them came to accept the Mahayana sutras as the word of the Buddha.²⁷¹

How much survives, and how old are the oldest copies?

Harrison writes that most texts written down in India from about the first century BCE were lost. The only canon to survive entire, he writes, is the Theravada one, in Pali. Parts of other schools' canons survive in Chinese or Tibetan translation, in vast canons with many editions.¹ By Stefan Baums's 2014 count, more than 89 Gandharan scrolls with about 115 texts, mostly in Gandhari, had been described in print. That counts one early regional group, and no source consulted counts all Buddhist manuscripts.⁸

• The tradition's account. Theravada tradition places the Buddha's death in 544 or 543 BCE and holds that his teaching was recited soon after. Heinz Bechert writes that traditional dates for the death range from 2420 to 290 BCE.⁹⁴¹⁰
• The critical dating. No date is agreed. Bechert holds that no date for the Buddha's life has been unanimously accepted by tradition and scholars. Contributors to a 1988 symposium proposed 486 to 261 BCE for the death, and Bechert himself dates it between 420 and 350 BCE. Harrison holds that Mahayana sutras were produced in India from about the first century CE onward.⁹¹
• The oldest copies. Baums dates the Gandharan scrolls to the first and second centuries CE, some perhaps as old as the first century BCE. The Library of Congress gives a radiocarbon range from the first century BCE to the first century CE for its scroll. Bhikkhu Sujato and Bhikkhu Brahmali, authors who argue a single thesis in a book-length work, write that the earliest Indian manuscripts are Gandhari texts from Afghanistan, the earliest dating to the first century CE.⁸¹¹¹² Harry Falk and Ingo Strauch report a central radiocarbon date around 74 CE (81 percent probability for 47 to 127 CE) for what they call the earliest known Perfection of Wisdom manuscript. It is in the Split Collection. A test on another roll gave 184 to 46 BCE, which they say does not concur with what they believe they know of the script's stages.¹³ The earliest entry is Ashoka's Bairat edict, cut in stone and naming seven teachings by title.¹⁴
• The oldest Pali manuscript. Oskar von Hinuber calls four leaves of the Pali monastic code in Kathmandu the oldest surviving Pali manuscript and a singular exception among much later ones. Cecil Bendall dated them to the end of the 8th or the beginning of the 9th century CE. Von Hinuber dates them to the 9th century in his 1991 book and to the eighth or ninth century in the Encyclopedia of Buddhism.¹⁵¹⁶
• The gap, computed here. The oldest Pali manuscript is taken here as copied between 775 and 900 CE. The 775 is this site's reading of Bendall's end of the 8th century; the Encyclopedia of Buddhism's eighth or ninth century would allow a start about 75 years earlier, so each gap could be up to about 75 years shorter at its low end. From the Buddha's death, the gap is about 1,320 to 1,445 years on the Theravada reckoning (544 BCE) and about 1,035 to 1,385 years on the symposium's range. From the Theravada writing-down date it is about 800 to 930 years on Lancaster's and Prebish's dates and about 875 to 1,000 on Access to Insight's. To Gandhari scrolls of the first century CE it is about 545 to 645 years on the Theravada reckoning and about 260 to 585 on the symposium's range. It is less if the first-century BCE ranges hold. On the critical dating, Mahayana differs. Harrison's first century CE for the production of its sutras and Falk and Strauch's 74 CE for a Perfection of Wisdom manuscript fall in the same period. Harrison's date covers the sutras as a group. The Mahayana accounts consulted give no year, so no gap can be counted on that view.

How should the entries be read?

Equal slots do not mean equal evidence. Five of the ten entries are Gandhari or Sanskrit manuscripts from northwest South Asia and Central Asia. One entry is in Pali, the language of the Theravada canon, and two are East Asian. One is Tibetan, the Kangyur printed in 1410, and one is Ashoka's Bairat edict, cut in stone.¹⁷ Baums writes that the exact find-spots of most Gandharan scrolls are unknown, because they were traded on the art market.⁸ See Buddhist Scriptures for the three canons and how ancient texts are dated for the methods.