The Historical Buddha and the Early Texts

"He came himself and worshipped (this spot), because the Buddha Śakyamuni was born here." — Ashoka's pillar at Lumbini, about 250 BC, translated by E. Hultzsch¹

"We know, and have known for some time, that the Pali canon as we have it … cannot be taken back further than the last quarter of the first century B.C.E." — Gregory Schopen, 1985²

"This gives a strong impression that we are indeed dealing with the authentic remains of an oral transmission, not with material that is for the most part the product of later harmonization and levelling." — Bhikkhu Anālayo, 2012³

Summary

Christian apologists often say that the Gospels were written within living memory of Jesus, while the scriptures of other religions were written down centuries after their founders.  For Buddhism the claim is roughly right, but it is usually made without the numbers, and the numbers cut in more than one direction.  This article sets out what is known about the Buddha from stones, inscriptions and manuscripts, how Buddhist scholars themselves argue for the antiquity of their texts, and how the result compares with the evidence for Jesus when the same yardstick is used on both.

Conclusions, stated first.

1. A teacher called Gotama, the "sage of the Sakyas", almost certainly lived.  The best single piece of evidence is a royal inscription.  In the twentieth year of his reign, about 250 BC, the emperor Ashoka set up a pillar at Lumbini in southern Nepal and recorded that he came to worship there "because the Buddha Śakyamuni was born here."¹  That is about 150 years after the Buddha's death on the date most specialists now prefer, or about 235 on the older date.  The same king enlarged the stupa of an earlier "Buddha Konakamana," whom no historian thinks was real,¹ so the pillar proves a belief held in the third century BC, not a fact.  The case for a real man rests on that belief together with the texts' realistic, pre-imperial setting.  Almost every specialist accepts it;  one recent sceptic, David Drewes, argues that there are no grounds for speaking of a historical Buddha at all.⁴

2. His dates are not settled.  The Sri Lankan chronicles put his death 218 years before Ashoka's consecration, which gives about 483 BC.  Northern Buddhist sources say about 100 years, which gives the late 300s BC.  Since a symposium held near Göttingen in 1988, most Western specialists have put the death within about twenty years of 400 BC.⁵⁶  The 2013 excavation under the Maya Devi temple at Lumbini was announced as support for the earlier date.  Its own radiocarbon dates cannot do that:  read at face value they make the timber structure older than the Buddha on every chronology, and read as maximum ages they fit every chronology.⁷

3. The teaching was oral for three to four centuries.  Ashoka named seven Buddhist texts that he wanted monks and lay people to hear,¹ and donor inscriptions of about the second century BC describe monks as one "who knows the Pitakas" and one "who knows the five nikayas."⁸  The Sri Lankan chronicles say the canon was first written down under King Vaṭṭagāmaṇi in the first century BC, but those chronicles were themselves composed four to five centuries later.  The oldest Buddhist manuscripts that survive are birch-bark scrolls from Gandhāra, in what is now Pakistan and Afghanistan, from about the first century BC to the second century AD.⁹  The earliest of them come roughly 300 to 485 years after the Buddha, depending on his date.  The oldest surviving Pāli text is a gold book of about the fourth or fifth century AD.

4. Buddhist scholars make a strong case for an early common core, and their own best defenders state its limits.  Discourses preserved separately in Pāli, in Chinese translation and in Gāndhārī agree on their essential teachings, which is best explained by a shared source older than the split between the schools in the third century BC.  Bhikkhu Anālayo, the leading comparative scholar, concludes that the Pāli discourses go back to a time before Ashoka, "allowing us to catch a glimpse of Buddhist thought predominantly from the pre-Asokan period."³  He also concedes that Buddhist oral transmission "could not arrive at a level of precision comparable to the Vedic reciters" and changed gradually.³  The core reaches back to perhaps a century after the Buddha.  That the core is the Buddha's own words is an inference from it, not something any manuscript shows.

5. By the same yardstick, every interval for Jesus is shorter, except one.  On mainstream dates Paul wrote about 20 years after the crucifixion, Mark about 40, and John 50 to 80.¹⁰¹⁰¹⁰  Outsiders name Jesus within about 60 to 85 years;  the first Western writer to name the Buddha writes about 600 years after him.¹¹  The oldest Gospel fragment is about a century or more after Jesus, the oldest Buddhist manuscripts three to five centuries after the Buddha.  But no inscription names Jesus for about two centuries, while a royal inscription names the Buddha within 150 to 235 years.  Apologists who date P52 to "AD 125" overstate the Gospel side.  On the Buddhist side, the claim that the bulk of the discourses are the Buddha's own words goes, as its defenders Sujato and Brahmali say themselves, beyond what most specialists affirm.¹²

6. History can show that Gotama lived and roughly what he taught.  It cannot show the miracles or the cosmology.  The earliest layer of the Buddhist texts already contains the miracles:  at his birth the infant takes seven steps and announces that this is his last life, while the "ten-thousandfold galaxy shakes and rocks and trembles."¹³¹⁴  The Gospels' earliest layer also contains miracles.  A historian who brackets the one set of miracles must give reasons before accepting the other, and the reasons for the resurrection are argued in The Historical Case for the Resurrection, not here.  What the comparison does show is that the Christian claim rests on a much shorter chain of transmission.

Part 1: When did the Buddha live?

No ancient source gives the Buddha's death in a calendar that can be converted to the modern one.  Every date is a count of years from a later, datable event, and the only such event available is the reign of the emperor Ashoka.  So two questions have to be answered:  when Ashoka reigned, and how many years separated him from the Buddha.

Fixing Ashoka

Ashoka's thirteenth rock edict boasts that his conquest by moral teaching had reached the lands of five Greek kings, whom it names:  Antiyoka, Turamaya, Antikini, Maka and Alikasudara.¹  Greek and Roman sources date these men independently.  They are Antiochus II of Syria, Ptolemy II of Egypt, Antigonus Gonatas of Macedonia, Magas of Cyrene, and an Alexander of Epirus or of Corinth.  The historian Romila Thapar gives their reigns as 261–246, 285–247, 276–239 BC, Magas dying about 252–250, and Alexander of Epirus 272–255 or of Corinth 252–244 BC.¹⁵  All five were on their thrones together only between 261 BC, when Antiochus II succeeded, and 255 BC if the Alexander is of Epirus, or about 250 BC if he is of Corinth.  Thapar put the edict at about 256–255 BC.¹⁵  Ashoka began issuing his major edicts in his thirteenth regnal year, so his consecration falls about twelve years earlier than the edict, in the late 270s or the 260s BC.¹⁶  Most writers use 268 BC.⁶  Rupert Gethin of Bristol notes that "recent research suggests that Asoka's accession may be plausibly placed anywhere between 280 and 267 bce."⁶  This is the one external anchor in the whole chronology, and it is good to within about a decade.

Counting back from Ashoka

The Sri Lankan chronicles give the gap.  The Mahāvaṃsa says:  "Be it known, that two hundred and eighteen years had passed from the nibbana of the Master unto Asoka's consecration."¹⁷  The older Dīpavaṃsa says the same.¹⁸  Added to a consecration about 268 BC, 218 years gives a death about 486 BC, and Wilhelm Geiger, translating the Mahāvaṃsa in 1912, worked out 483 BC.¹⁷  This "corrected long chronology," with a life of 563–483 BC, is the date most readers have met.

Buddhist sources from north India and Central Asia, preserved in Sanskrit, Chinese and Tibetan, put Ashoka only about 100 years after the Buddha.  That "short chronology" gives a death in the late fourth century BC.  Theravāda countries keep a third reckoning, an era starting in 544/543 BC, but Geiger already judged it "a relatively late fabrication,"¹⁷ and Thapar calls it "a later fabrication" that "does not tally with any other evidence."¹⁵  A fourth figure, the "dotted record" said to have been kept at Canton by adding one dot each year, had 975 dots in AD 489 and so implies 486 BC.  Geiger himself doubted that it ran "nearly a thousand years" without "a single error."¹⁷

Why most specialists now prefer about 400 BC

In 1988 Heinz Bechert of Göttingen gathered the field at a symposium whose papers fill three volumes.  The question of whether the Buddha existed "was not even raised" there.¹⁹  What changed was the date.  Three lines of argument pulled it later.

1. Round numbers. "218" and "100" may not be counts at all. Gethin: "such figures as 218 and 100 should properly be seen as ideal round numbers."⁶
2. The teachers' lineage. The Dīpavaṃsa lists the monks who led the monastic discipline from the Buddha's death to Ashoka: Upāli, Dāsaka, Sonaka, Siggava and Moggaliputta, with the years each had been ordained.¹⁸ Richard Gombrich of Oxford worked through these figures and found they span much less than 218 years. His table puts the death in 404 BC and Ashoka's consecration 136 years later.²⁰ Charles Prebish, who doubts parts of Gombrich's argument, still found the 404 BC date "quite compelling."²⁰
3. Archaeology and culture. On the archaeology, the towns and trade the discourses describe fit the fifth and fourth centuries better than the sixth. Lance Cousins, reviewing the conference volumes, judged the archaeological evidence "the major factor tending to support a later date," while adding that "It is not however clear whether it is as yet overwhelming."⁵

The results cluster.  Cousins reported that Hajime Nakamura, K. R. Norman and Gombrich "all propose dates within the range suggested by André Bareau:  around 400 B.C. with a margin of twenty years on either side."⁵  Bechert himself argued for the fourth century.²⁰²¹  Prebish counted "the vast majority of the speculations offered by the scholars attending the 1988 Göttingen symposium" as falling "between 368 and 404 B.C.E."²⁰  Gethin summed up in 1998:  "While there is no scholarly consensus on the precise dates of the Buddha, a detailed examination of all the available data and arguments by scholars in recent years has resulted in a general tendency to bring the date of the Buddha considerably forward and place his death much nearer 400 bce than 500 bce."⁶

How much the uncertainty matters.  The spread between the long and short dates is about 115 years, and the favoured median sits in between.  For the comparisons later in this article, every interval is given for both a death about 400 BC and one in 483 BC.  None of the conclusions depends on which is right.

Part 2: What the stones say

Ashoka names the Buddha

Three of Ashoka's inscriptions mention the Buddha.  They are the oldest datable Buddhist documents of any kind.

The Lumbini pillar.  In 1896 a sandstone pillar was dug out at Rummindei, in the Nepalese lowlands just north of the Indian border.  Its five-line inscription reads, in E. Hultzsch's standard edition:  "When king Devanampriya Priyadarśin had been anointed twenty years, he came himself and worshipped (this spot), because the Buddha Śakyamuni was born here."  It goes on to say he set up a stone pillar "(in order to show) that the Blessed one was born here," and made the village of Lummini free of taxes.¹  "Devanampriya Priyadarśin," Beloved of the Gods, is the title Ashoka uses for himself throughout his edicts.  The twentieth year of his reign is dated 244 BC by Hultzsch, about 252 BC by Richard Salomon, and about 248 BC by Harry Falk, depending on the date each uses for his consecration.¹²²²³  This article uses "about 250 BC."

The Nigali Sagar pillar, a few miles away, records that in his fourteenth year Ashoka "enlarged the Stupa of the Buddha Konakamana to the double (of its original size)."¹  Konakamana is one of the Buddhas said to have lived before Gotama.  Hultzsch's own note calls him "This mythical Buddha."¹  So the same king, in the same district, honoured the shrine of a Buddha no historian believes in.  The Lumbini pillar therefore proves what people believed about Gotama around 250 BC.  By itself it does not prove that he lived.  What makes the difference is everything else:  the Konakamana cult left no texts, no named disciples and no rival schools that remembered his teaching.

The Calcutta-Bairat rock edict, addressed to the monks, opens with Ashoka's "reverence and faith in the Buddha, the Dharma (and) the Samgha" and says that "Whatsoever, Sirs, has been spoken by the blessed Buddha, all that is quite well spoken."  It then lists seven "expositions of the Dharma" that Ashoka wants monks, nuns and lay people to hear often, among them "the Muni-gathas," "the Moneya-suta," "the Upatisa-pasina," and "the Laghulovada which was spoken by the blessed Buddha concerning falsehood."¹  This is the earliest evidence of named Buddhist texts.  Several of the titles match discourses in the Pāli canon, though Gregory Schopen points out that only three have been identified "with anything approaching unanimity … and even these are not certain."²

Was the Lumbini pillar faked?

The question is fair, because the archaeologist who publicised the find was a fraud.  Alois Anton Führer, an assistant surveyor for the Archaeological Survey of India, was by Falk's account "making up excavation reports, forging at least one inscription and distributing fake relics free of charge, including a supposed tooth of the Buddha to a Burmese monk."  Vincent Smith exposed his claims about Kapilavastu in 1901.²³  The writer Terence Phelps has since argued that the Lumbini and Piprahwa finds were part of the same deception.²⁴

The evidence keeps the Lumbini pillar apart from Führer's frauds.  The pillar was found by the local governor, General Khadga Shamsher, and Führer only publicised it.²³  J. F. Fleet, who rejected the common reading of the Piprahwa vase, saw "no reason for supposing that the place where the inscribed portion of the column was found, standing and partly buried, is not the place where the column was originally set up."²⁵  The Chinese pilgrim Xuanzang, who visited in the seventh century AD, saw at the birthplace "a great stone pillar, on the top of which is the figure of a horse, which was built by Asoka-raja," by then broken and fallen.²⁶  So an Ashokan pillar stood at Lumbini long before any European arrived.  Führer's known fakes were reports and finds that later surveyors could not confirm, which is how he was caught.  This article treats the Lumbini inscription as genuine.

The Piprahwa reliquary

In January 1898 William Peppé, the manager of an estate at Piprahwa, about eight miles from Lumbini,²⁵ dug into a brick mound.  Inside it he found a large stone coffer holding soapstone and crystal vases, pieces of bone and hundreds of small jewels.  "The only inscription of any kind is scratched on the cover of the smaller urn," he reported.²⁷  The words run, in one transliteration:  sukitibhatinaṃ sabhagiṇikanaṃ saputadalanaṃ iyaṃ salilanidhane budhasa bhagavate sakiyanaṃ.⁴

What it says is disputed.  Georg Bühler, in 1898, read it as "This relic-shrine of divine Buddha (is the donation) of the Sakya Sukiti-brothers … associated with their sisters, sons, and wives."²⁸  Vincent Smith took the bones to be "part of the sacred body … of Gautama Buddha himself," but added:  "Whether or not the depositors' belief was actually well-founded no man can say."²⁷  Fleet argued in 1906 that the vase was "a memorial, not of Buddha, but of the kinsmen of Buddha," the Sakyas massacred in a later legend.²⁵  Harry Falk, the leading modern authority on Ashokan sites, reads the last word sakiyanaṃ as attaching both to the donors and to the Buddha, and judges that "The double meaning was probably intended."²³

Its date is disputed too.  Bühler thought the script older than Ashoka.  Salomon reports that this view is "no longer generally accepted, mainly on archaeological grounds," and that Herbert Härtel put the vase no earlier than the second century BC.²²  Falk argues that the text "should be older than Bharhut, and a time of origin in the later part of the third century b.c. is not at all impossible."²³

And its authenticity has been challenged.  Phelps argues that Führer, who was in contact with Peppé while the excavation went on, engraved the text himself:  "Is the Piprahwa inscription simply another Fuhrer forgery? Fuhrer certainly had the knowledge to do this".  He concludes that Peppé's own notes show "that it was simply a modern forgery."²⁴  Falk replied that Phelps had compared the wrong object, since the vase on display in Calcutta is "just one of several soapstone copies made at the end of the 1890s," and that Phelps's evidence produces "not a single reasonable argument against the genuineness of the Piprahwa inscription."²³  Apart from Andrew Huxley, whose "Siam Conspiracy" papers Falk also rejects,²³ the forgery theory has found no support among epigraphists in the work read for this article, and Salomon's standard handbook treats the vase as a genuine record of a relic deposit.²²  But the theory has not been tested by any scientific study of the object, and that gap should be stated.

Excavations at Piprahwa in the 1970s found seals of a much later age naming the "Kapilavastu" monastic community, which ties the site to the Buddha's home town of the texts.⁴  The jewels from Peppé's find were offered for auction by Sotheby's in Hong Kong in May 2025.  India's government objected, the sale was cancelled, and the gems were sold privately to the Indian Godrej group for public display.²⁹

What Piprahwa shows.  By the third or second century BC, people who called themselves Sakyas, in the region where the texts place the Buddha's clan, enshrined bones they believed were his.  That supports the texts' geography and the reality of the clan.  It does not show whose bones they were.

The 2013 dig under the Maya Devi temple

In 2011 and 2012 a team led by Robin Coningham of Durham University dug beneath the Ashokan brick temple at Lumbini.  Under it they found an older brick kerb and, under that, a line of postholes around an open centre with traces of large roots.  They proposed that this was a timber shrine built around a tree, older than Ashoka, and called it "the earliest Buddhist shrine."³⁰  National Geographic's headline said it was "dating to around 550 B.C."³¹

What the dates are.  Charcoal from two posthole fills gave radiocarbon ages of 2540 ± 30 and 2548 ± 30 years before present.  The team calibrated these to 799–546 BC and 801–548 BC.  An optically stimulated luminescence date on an early land surface gave 545 ± 235 BC.³⁰

What the dates show.  Human activity at the spot began before Ashoka.  That is a real result.  The authors went further:  "if the posthole alignment is related to the earliest veneration of the Buddha, shortly after his mahaparanirvana, we may also have the first archaeological evidence regarding the date of the life of Buddha.  As noted above, dates of the sixth century BC lend support to the longer chronology".³⁰  The dates cannot carry that.  Recalibrated against the current radiocarbon curve, both posthole samples fall wholly before 546 BC.  The chance that either charcoal sample is younger than 483 BC, the long chronology's date for the Buddha's death, is about one in a thousand.⁷  So if the charcoal dates the posts, the posts are older than the Buddha's death on every chronology, and the structure is not a shrine to him.  If the charcoal is older than the posts, as charcoal often is, the dates are only upper limits and fit any chronology.  The luminescence date's error of ± 235 years is more than twice the whole gap between the long and short chronologies.

Specialists said the same.  Richard Gombrich:  "In nothing that has been unearthed is there a single trace to suggest anything to do with Buddhism."  He noted that the luminescence span "is of course wide enough to include every date for the Buddha's death under possible consideration."²¹  Oskar von Hinüber read the finds as "vestiges of pre-Buddhist tree worship."¹⁹  Julia Shaw of University College London called the claims "convincing but speculative," since an "older tree shrine quite disconnected from the worship of the historical Buddha" was also possible.³¹  Coningham's team also misreported the debate, describing the short chronology as "in the vicinity of 480 BC" and attributing it to Heinz Bechert, who in fact favoured the fourth century.³⁰²¹

The Lumbini dig is a useful case for apologists of every kind.  A real excavation produced real dates, and a press release turned them into a confirmation that the dates do not give.

Part 3: From recitation to writing

The tradition of the First Council

The monastic law code of the Theravāda school ends with an account of a council held at Rājagaha in the rainy season after the Buddha's death.  The senior monk Mahā Kassapa proposes:  "Come, Sirs, let us chant together the Dhamma and the Vinaya before what is not Dhamma is spread abroad, and what is Dhamma is put aside".  He chooses "five hundred Arahats less one."  Upāli is questioned on the monastic rules, and Ānanda, the Buddha's attendant, on the discourses.³²  The account is candid about friction.  Ānanda admits that he never asked the Buddha which were the "lesser and minor precepts" that the monks might drop, and the elders cannot agree on what they were.³²

Few critical scholars take the council as described for history.  Gombrich, with André Bareau and others, regards it as "essentially legendary."²⁰  Étienne Lamotte, the great Belgian historian of early Buddhism, took a middle view:  "in the first century after the Nirvana, one or more groups of specialists, whether assembled in council or not, attempted to codify the word of the Buddha" and produced a body of doctrine and rules "which constituted the common heritage of the Buddhist sects which were subsequently to develop."³³

How the texts were kept without writing

As Wynne notes, "there is no evidence for writing before Aśoka."⁴  Von Hinüber:  "at the time of the Buddha there was no script, so nothing written can be traced back to his time."¹⁹  The discourses were memorised and chanted by groups of monks.  Their style shows it:  stock passages repeated word for word, numbered lists, refrains.  One discourse has the Buddha tell his followers to meet "again and again" and recite together, comparing "meaning with meaning" and "letter with letter," so that the holy life will endure.⁴

How accurate was this?  The Buddhist community included Brahmins who knew that the Vedas had been passed on orally, word for word, for many centuries.⁴  But Anālayo points out a difference that cuts against the Buddhist texts.  Vedic boys memorised sounds before they understood them.  Buddhist monks began after ordination, as adults, and were taught that the meaning mattered most.  "This makes it only natural that the early Buddhist oral transmission could not arrive at a level of precision comparable to the Vedic reciters."  The result, he concludes, "inevitably resulted in a gradual change of the transmitted material."³

What the inscriptions show about the oral canon

The railing of the Bharhut stūpa in central India, carved in about the second century BC, carries the names of its donors.  One rail-bar is "the gift of the venerable (arya) Jāta, who knows the Pitakas"; another is "The gift of Budharakhita … who knows the five nikayas."⁸  The Pāli canon is the only canon known to divide its discourses into five nikāyas.  Heinrich Lüders, who edited the inscriptions, concluded that "the expression pamchanekāyika confirms that the Pali canon was in existence in the 2nd cent. B.C. in Western India."⁸  Schopen accepts that the titles show "at least some Buddhist literature had been classified as sutta and vinaya, and that some idea of a Pitaka had already emerged.  But we have no idea what these classifications included."²  Both readings are fair.  The inscriptions show a recited body of texts with the same divisions as the Pāli canon, some 200 to 300 years after the Buddha.  They do not show its contents.

When the canon was written down

The Sri Lankan chronicles say it happened under King Vaṭṭagāmaṇi Abhaya.  The Dīpavaṃsa:  "Before this time, the wise Bhikkhus had orally handed down the text of the three Pitakas and also the Atthakatha.  At this time, the Bhikkhus who perceived the decay of created beings, assembled and in order that the Religion might endure for a long time, they recorded (the above-mentioned texts) in written books."¹⁸  The Mahāvaṃsa says the same, adding that the monks acted because "the people were falling away (from religion)."¹⁷  Wilhelm Geiger's chronology puts Vaṭṭagāmaṇi's reign at 29–17 BC, and Schopen accepts "29 to 17 B.C.E." as the date of the earliest redaction anyone can know about.²  Von Hinüber gives both 29–17 and 89–77 BC.³⁴  On either reckoning, the written canon begins roughly 310 to 470 years after the Buddha's death.

The report is late too.  Von Hinüber judges that the Dīpavaṃsa was composed "not long after 350 AD," and the Mahāvaṃsa about a century later.³⁴  The first statement that the canon was written down therefore comes about four centuries after the event it describes.  The claim is plausible, because it fits the age of the oldest manuscripts from Gandhāra, but it is not a contemporary record.

The oldest surviving manuscripts

Gandhāra.  Since the 1990s, collections of birch-bark scrolls written in the Gāndhārī language and the Kharoṣṭhī script have surfaced in Afghanistan and Pakistan.  Richard Salomon of the University of Washington, who has led their study, calls them "the oldest surviving specimens of original Buddhist texts in the world, as well as the oldest Indian manuscripts of any type."⁹  He writes that "The earliest Buddhist texts in Gandhari can now be securely dated to the first century BCE, or possibly even earlier, while most of the others date from the first and second centuries CE."⁹  The British Library scrolls mention two Indo-Scythian rulers known from coins and inscriptions of the early first century AD.⁹  The pot that held the Senior scrolls is inscribed "year 12," almost certainly of the Kushan emperor Kaniṣka's era, which works out to about AD 140, a date Salomon calls "consistent with radiocarbon tests."⁹  One scroll about many Buddhas has been radiocarbon-dated to around the first century BC.⁹

What they contain matters.  Twenty-nine of the thirty-five discourses in the Senior collection correspond to texts in the Saṃyukta collections of other canons, and their selection shows that they were drawn from a complete collection "whose structure resembled that of the corresponding collections in the Pali and Chinese canons."  Salomon concludes that "This finding supports the traditional view that the central sections of the canons were common to all branches of early Indian Buddhism and that they represent the oldest and most fundamental teachings."⁹  But the same finds also include a Mahāyāna Perfection of Wisdom sūtra, radiocarbon-dated to between AD 47 and 147.⁹  No scholar attributes that text to the historical Buddha.  The oldest manuscripts therefore show early discourses and later compositions circulating side by side, which is why their age alone settles nothing about authorship.

Pāli.  The oldest Pāli writing is a book of twenty gold leaves found in a relic chamber at Śrī Kṣetra in Burma.  Janice Stargardt of Cambridge dated its script to "around the mid-or late fifth century."  She found that it and the related finds preserve canonical texts "in a language and in recensions that do not differ significantly from those recorded in the Pali Text Society editions," which she took as proof of "how scrupulously such texts were transmitted" through the first and second millennia AD.³⁵  Wynne, citing Stargardt and Falk, puts them a little earlier, in the late third or early fourth century AD.⁴  The oldest Pāli manuscript proper, a fragment of the monastic rules kept in Kathmandu, was copied in the eighth or ninth century.  Most Pāli manuscripts are far younger.  Von Hinüber:  "The continuous manuscript tradition with complete texts begins only during the late 15th century.  Thus the sources immediately available for Theravada literature are separated from the Buddha by almost 2000 years.  It should be kept in mind, however, that the age of the manuscripts has little to do with the age of the texts they contain."³⁴

Chinese.  The four main collections of discourses that parallel the Pāli Nikāyas were translated into Chinese between AD 384 and the 430s, from Indian originals belonging to other schools:  the Ekottarika in 384, the Madhyama in 397–398, the Dīrgha in 413, and the Saṃyukta from 435.^3637383

Part 4: How old is the common core?

Inscriptions show that Buddhist texts existed and were sorted into collections before the canon was written down.  They do not show what those texts said.  For that, scholars compare the versions of each discourse that different Buddhist schools preserved.  This part states the Buddhist case first, in its defenders' words, then the sceptical case, then weighs them.

The case for an early core, as its defenders make it

The fullest statement comes from two Western-born Theravāda monks, Bhikkhu Sujato and Bhikkhu Brahmali, in The Authenticity of the Early Buddhist Texts (2014).  By "Early Buddhist Texts" they mean the four main collections of discourses, parts of a fifth, and the monastic rules, not the later commentaries, biographies or Mahāyāna sūtras.  Their 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."¹²  They say plainly that this goes further than most of their colleagues:  "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."¹²  They also concede that the texts "were edited and arranged over a few centuries following the Buddha's demise" and are "not a verbatim record of the Buddha's utterances."¹²

Their argument has three main legs.

Separate lines agree.  The Theravāda school of Sri Lanka and the Sarvāstivāda school of north-west India went their separate ways at about the time of Ashoka's missions.  Their collections of middle-length discourses survive, one in Pāli and one in a Chinese translation of AD 397–398,³⁶ and they teach the same doctrines.  Sujato and Brahmali argue that conservatism this strong must have been inherited from before the split, so "the reasonable inference is that it was reliably transmitted also in the first 150–200 years of Buddhist history."¹²  Bhikkhu Anālayo, a German-born monk who has taught at the University of Hamburg, has done the detailed comparison discourse by discourse.  His verdict on the 152 discourses of the Majjhima-nikāya and their parallels:  "numerous differences in details alongside considerable agreement in regard to essential teachings," the marks of "a genuine oral transmission, with all its vicissitudes, but also with all its strengths."¹³

The texts do not know the empire.  Chandragupta and Ashoka, who built the first empire across north India in the late fourth and third centuries BC, appear nowhere in the early discourses, though later Buddhist books are full of Ashoka and even have the Buddha predict him.  Sujato and Brahmali:  "Candagutta and Asoka are two of India's greatest rulers, yet they are not mentioned anywhere in the EBTs, which indicates that they were closed by the time of Candagutta."¹²  The discourses describe small kingdoms and modest towns, not the large cities of the Mauryan age.  The latest king they seem to mention, Muṇḍa of Magadha, ruled about 350 BC.¹²

Outside evidence keeps agreeing with the texts.  Alexander Wynne, an Oxford-trained Indologist, lists the cases.  Reliquaries at Sanchi name the "Himalayan" monks whom the Pāli chronicles say were sent north in Ashoka's reign.⁴  Two inscriptions found at Deorkothar in central India, from the early second century BC, record a line of teachers descended from Anuruddha, a disciple whom the texts place at the Buddha's death.⁴  Richard Salomon and Joseph Marino, who published them, called them "strong and unexpected support for the historical veracity of traditional voices."⁴  Wynne's conclusion:  "Beyond any reasonable doubt, we can conclude that early Buddhist texts are ancient and sufficiently objective for us to 'know' the Buddha.  Details can be doubted – whether this or that text was really spoken by the Buddha."⁴

The sceptical case

The strongest sceptic is not a critic of Buddhism.  Gregory Schopen, an American Buddhologist who has worked mainly from inscriptions and monastic law codes, set out in 1985 what he thought was actually known.  "We know, and have known for some time, that the Pali canon as we have it … cannot be taken back further than the last quarter of the first century B.C.E., the date of the Alu-vihara redaction, the earliest redaction that we can have some knowledge of".  He went further:  "it is not until the time of the commentaries of Buddhaghosa, Dhammapala, and others – that is to say, the fifth to sixth centuries C.E. – that we can know anything definite about the actual contents of this canon."²

He adds three points.  Ashoka names seven texts but never mentions a canon or its divisions, and "only three of these have been identified with anything approaching unanimity … and even these are not certain."  The texts that are demonstrably old are "all short texts in verse."²  And the comparison argument can be turned around.  Every version being compared survives in late copies, so "If all known versions of a text or passage agree, that text or passage is probably late; that is, it probably represents the results of the conflation and gradual leveling and harmonization of earlier existing traditions."²

The Buddhologist David Drewes goes furthest.  Surveying two centuries of scholarship, he argued in 2017 that "nothing about him has ever been established as fact," and proposed "accepting that we do not have grounds for speaking of a historical Buddha at all."  He did not say that no such person lived:  "it is possible that there was some single, actual person behind the nebulous 'śramaṇa Gautama' of the early texts".⁴  Oskar von Hinüber, the senior German Pāli philologist, replied that the question of historicity "was not even raised" at the 1988 Göttingen conference because "historians of ancient India are very confident about the historicity of the Buddha."¹⁹

Weighing them

Schopen's inversion has been tested on his own example, and the example does not show it.  Schopen illustrated levelling with a story about the stūpa of an earlier Buddha, Kassapa, which he said four schools' monastic law codes share.  Anālayo went back to the texts and found that the Theravāda law code has no stūpa in the story at all:  "the case chosen by Schopen does not conflict with the tenet that regards agreement between parallel versions as a criterion for earliness, simply because in the present case the parallel versions do not agree."³  Surveying discourse after discourse, he concluded that "we are indeed dealing with the authentic remains of an oral transmission, not with material that is for the most part the product of later harmonization and levelling."³  Harmonization across schools that were rivals, spread from Sri Lanka to Central Asia, and preserved the same discourses with different wording, numbering and order is harder to picture than a common source.

Schopen wrote before the oldest manuscripts were found.  His 1985 claim that "probably none of the Chinese sources go back beyond the second century C.E." was right about the Chinese, and his manuscript evidence was "uniformly late."²  The Gāndhārī scrolls described in Part 3 changed that.  They include discourses with Pāli parallels, written in another language, in another region, by another school, by about the first century AD.  That does not reach back to the Buddha, but it does show the same discourses circulating separately at the turn of the era.

The argument from silence about Ashoka is as strong as its Christian twin, and no stronger.  A text that never mentions the greatest king of its world was probably fixed before him, unless he was simply irrelevant to it.  New Testament scholars make the same argument about the Gospels' silence on the fall of Jerusalem in AD 70.  Raymond Brown, a mainstream Catholic scholar who dated Mark to about 68–73, still wrote that this silence "is very hard to explain … but we should not pretend that we have a satisfactory answer."¹⁰  The same standard applies to both.  An argument from silence counts for about as much as the expectation that the event would have been mentioned.  That expectation is high for a Buddhist literature that later made Ashoka its hero, and moderate for Gospels written in a Jewish-Christian setting where the temple's fall mattered greatly.  See When Were the Gospels Written? for the Gospel side.

A prediction that came true counts the same way in both traditions.  On his last journey the Buddha passes the village of Pāṭaligāma and predicts that it will become the great city Pāṭaliputta, as it did under the Mauryas.  Wynne allows that "Perhaps this story was invented to relate Pāṭaliputta to the Buddha."⁴  Critical New Testament scholars read Jesus' prediction of the temple's destruction the same way, as a sign that the passage was written after 70.  A reader who accepts one inference should accept the other.  A reader who resists it for the Gospels should resist it for the discourses too.

Where this leaves the core.  The case for a body of discourses older than the split between the schools, so older than about 250 BC, is strong.  It rests on independent lines agreeing, on a pre-Mauryan setting, and on inscriptions that keep confirming small details.  The case that the core goes back to the Buddha himself, rather than to the first few generations of his followers, is an inference from coherence and from the absence of any rival founder.  Wynne and Sujato make that inference with confidence.  Anālayo makes it more cautiously, and concedes that oral recitation "inevitably resulted in a gradual change of the transmitted material."³  Schopen's narrower claim stands:  no manuscript or inscription shows the wording of any discourse before the first century AD.

Part 5: The Buddha and Jesus, by the same yardstick

The numbers side by side

The table uses mainstream dates for both men:  a death about 400 BC, or 483 BC on the older reckoning, for the Buddha, and about AD 30 for Jesus.  The Gospel dates are those of Raymond Brown's standard Introduction to the New Testament, and the papyrus dates are the palaeographers' current ranges, set out in New Testament Manuscripts & Variants and When Were the Gospels Written?.

On every measure but one, the evidence for Jesus stands closer to the man.  Paul's letters come about twenty years after the crucifixion,¹⁰ and he says he stayed fifteen days with Peter and met "James, the Lord's brother."³⁹  The first Buddhist text certainly written down comes three to four centuries after the Buddha.  Two non-Christian historians name Jesus within about 60 and 85 years.⁴⁰⁴¹  The Greek envoy Megasthenes, at the Mauryan court about 300 BC, described Indian ascetics he called "Garmanes" without naming any Buddha, as far as his account survives in Strabo.⁴²  What Philip Almond calls "unarguably the first reference to the Buddha in the West"⁴³ comes from the Christian writer Clement of Alexandria, about six centuries after the Buddha:  "Some, too, of the Indians obey the precepts of Buddha; whom, on account of his extraordinary sanctity, they have raised to divine honours."¹¹  The oldest Indian texts outside Buddhism are silent too.  Rupert Gethin, a Pāli scholar, writes that in "the earliest texts of the Jain and brahmanical traditions, there is no explicit mention of the Buddha at all."⁶  Later Jain and brahmanical texts were not searched for this article.

The exception is inscriptions.  The oldest inscription naming Jesus that this research found is a floor mosaic in a Christian prayer room uncovered in 2005 at Kefar 'Othnay, beside Megiddo in northern Israel.  It records that a woman "has offered the table to God Jesus Christ as a memorial."⁴⁴  Its excavator, Yotam Tepper, dated the building to about AD 230;  Vassilios Tzaferis argued for the second half of the third century;  and Edward Adams, surveying the earliest Christian meeting places, concludes that "the site needs further investigation for the excavator's interpretation and dating to be established securely."⁴⁵  So the first inscription naming Jesus comes about two centuries or more after his death, further from him than Ashoka's pillar stands from the Buddha on the date most specialists prefer.  The reason is not hard to find:  Christians had no royal patron before Constantine in the fourth century, and a small and often persecuted minority left few public inscriptions of any kind.  But an explanation of the silence is not evidence, and on this one measure the Buddha is better attested.

Where Christian apologists have overstated their case

The comparison is often made carelessly, and a skeptic who checks the sources will notice.

1. "Not put into writing until after the Christian era." In Lee Strobel's The Case for Christ, the historian Edwin Yamauchi says that "The scriptures of Buddha, who lived in the sixth century B.C., were not put into writing until after the Christian era, and the first biography of Buddha was written in the first century A.D."⁴⁶ The Sri Lankan chronicles put the writing in the first century BC, and some Gāndhārī manuscripts are that old too.⁹ The sixth-century date is the older chronology, which most specialists have left.
2. "600 to 900 years." Gary Habermas and Benjamin Shaw write that "Many of Buddha's major writings date from 600 to 900 years after his death, with oral teachings being the norm for the first 500 years," citing Edward Conze, and add that "Conze concedes that that is why Buddhists cannot compete with Christians regarding the reliability of their traditions."⁴⁷ Conze's figure describes his own anthology: "The bulk of the selections in this book was written down between A.D. 100 and 400."⁴⁸ Most of those selections are later Mahāyāna scriptures, not the early discourses. Conze's introduction contains no such concession. What he does say cuts in a different direction: "Buddhists possess nothing that corresponds to the 'New Testament'," but "some of it must represent the direct and actual sayings of the Buddha himself. At present we have, however, no objective criterion which would allow us to isolate the original gospel."⁴⁸
3. "More than two hundred years" to the first mention. Sean McDowell quotes Justin Bass: "That is a gap of more than two hundred years between when the Buddha died and our earliest source referencing him."⁴⁹ That is right only on the older date of 483 BC. On the date most specialists prefer the gap is about 150 years.
4. "Gaps of hundreds to well over a thousand years." A Christian Research Institute programme summary says founders' teachings were recorded after "usual gaps of hundreds to well over a thousand years," and that the Buddha's dates involve "differences of a thousand years or more."⁵⁰ For the Buddha neither is true of modern scholarship: the dates in serious dispute differ by about a century, and the oldest manuscripts are three to five centuries after him.
5. The Gospel side, too. The claims that P52 dates to AD 125 and that a fragment of Mark from the first century had been found are retired in this site's New Testament Manuscripts & Variants. P52's possible dates run from about 125 into the early third century,⁵¹⁵² and Pasquale Orsini and Willy Clarysse conclude that "There are no first-century New Testament papyri and only very few papyri can be attributed to the (second half of the) second century."⁵³

None of these errors changes the direction of the comparison.  They do make the Christian case look stronger than it is, and a reader who finds one will distrust the rest.

Where the Buddhist case is stronger than apologists allow

1. A royal inscription within one to two centuries. Nothing of the kind exists for Jesus. An emperor's own words, cut in stone at the birthplace within about 150 years of the death, are better evidence of an early cult than anything physical that survives for Christianity before the third century.
2. Separate lines of transmission. The Pāli, Gāndhārī and Chinese collections were kept by rival schools for centuries apart, yet share a core. The first three Gospels are not independent in that way: on the usual view Matthew and Luke copied Mark. Christians argue instead from several early sources inside the New Testament, such as Paul's letters, Mark and material shared by Matthew and Luke. That is a different and in some ways weaker kind of corroboration than independent recensions.
3. Small details confirmed in stone. The Deorkothar lineage of Anuruddha, the Sanchi reliquaries of the Himalayan missionaries, and the Sakyas of Piprahwa all match details in the texts.⁴

Where it is weaker than its popular presentation

1. The oral period is long. Three to four centuries passed before the first writing, against twenty to seventy years for the New Testament. Anālayo, the tradition's most careful defender, concedes that the reciters could not match Vedic precision and that the texts changed.³
2. The Pāli manuscripts are late. The oldest Pāli text is about a thousand years after the Buddha, and complete Pāli manuscripts survive only from the late fifteenth century. Von Hinüber: "the sources immediately available for Theravada literature are separated from the Buddha by almost 2000 years."³⁴ New Testament manuscripts, by contrast, "become more common" in the third and fourth centuries.⁵³
3. The report of the writing is itself late. The claim that the canon was written in the first century BC comes from chronicles of the fourth and fifth centuries AD.³⁴
4. The strongest claim goes beyond the field. Sujato and Brahmali hold that "the bulk of the sayings in the EBTs that are attributed to the Buddha were actually spoken by him." They say plainly that most academic specialists affirm only that the texts may "contain some authentic sayings of the Buddha."¹² The comparative evidence supports an early core. It does not by itself show whose words the core preserves.

The same tests, applied both ways

The age of a manuscript is not the age of a text.  Von Hinüber's warning that "the age of the manuscripts has little to do with the age of the texts they contain" is true, and it applies to the Gospels too.³⁴  Late copies can carry early texts.  A Buddhist may fairly use that point against an apologist who dismisses the Pāli canon because its manuscripts are medieval.  A Christian may use it the same way against a skeptic who dismisses the Gospels because P52 is a century late.

Oral tradition.  If monastic recitation could carry a core of teaching across three centuries, as Anālayo, Sujato and Wynne argue, a skeptic cannot also hold that the followers of Jesus could not carry his sayings across twenty to forty years.  And if a skeptic doubts that forty years of oral tradition can be trusted for Jesus, consistency demands far more doubt about three hundred years for the Buddha.  Either way the Gospels come out ahead on this test.

Silence and prophecy.  Part 4 showed that the argument from silence and the argument from fulfilled prediction count the same way in both traditions.

Part 6: What the evidence can and cannot establish

"Did the Buddha exist?", "What did he teach?" and "Did the wonders happen?" are three different questions.  The evidence answers them to very different degrees.

That Gotama lived: strong

A teacher of the Sakya clan, remembered as the Buddha, almost certainly lived in the Ganges plain in the fifth or early fourth century BC.  The evidence is a converging set, not one proof.  Within about 150 to 235 years a king marked his birthplace.  An early reliquary was dedicated by Sakyas in the area where the texts place them.  Inscriptions of the second century BC record teaching lineages from his named disciples.  And the discourses show a realistic, pre-imperial world that later writers did not know how to fake.  Drewes's objection is that the Buddha "has not been linked to any historical facts."¹⁹  It is fair to note that the same objection would erase most founders of ancient movements, and that it treats a belief attested in a royal inscription within two centuries as worth nothing.  By the standard this site applies to Jesus, Muhammad and others, the Buddha passes.  See Did Muhammad Exist? for the same test on another founder.

What he taught: probable in outline, not in wording

The shared core of the discourses gives the outline of an early teaching, the doctrine that schools separated for centuries still held in common.¹³  That this outline was taught by the Buddha himself is the most economical explanation of its spread across rival schools, and most specialists accept it.  But even the defenders do not claim the wording.  Sujato and Brahmali call the texts "not a verbatim record of the Buddha's utterances."¹²  Anālayo describes reciters who preserved meaning rather than exact sound, so that change was inevitable.³  Whether rebirth and non-self fit together is a separate question, taken up in Non-Self, Rebirth and Nirvana.

The miracles and the cosmology: in the earliest layer, and beyond the reach of history

The early discourses are not a sober record that later legend decorated.  Gods and wonders are already in the shared core.  The Acchariyabbhuta Sutta of the Majjhima collection describes the birth:  the newborn "stands firm with his own feet on the ground.  Facing north, he takes seven strides with a white parasol held above him, surveys all quarters, and makes this dramatic proclamation:  'I am the foremost in the world! … This is my last rebirth'."  At the same moment "this ten-thousandfold galaxy shakes and rocks and trembles."¹⁴  The Chinese parallel from the Sarvāstivāda school also has the great light, the earthquake and the seven steps, though each version lists wonders the other lacks.¹³  Anālayo, after tracing the Buddha's fire miracles, judged most of them "later developments," yet found that other "supernormal abilities" such as "Celestial travels by the Buddha and his disciples … form a common heritage among the early discourses."³  In the same discourses the Buddha says that he knows gods exist.³

History can establish that his early followers believed these things.  It cannot establish that they happened, because a historian has no way to test a cosmic earthquake reported centuries later by the people who revered him.  The same is true of the Gospel miracles in so far as they are reported only by believers.

The same standard, applied to the Gospels

Fairness cuts both ways here, and the Christian reader should feel the force of it.  The earliest layer of the Gospels also contains miracles.  Mark, the earliest Gospel on the usual view, is full of healings and exorcisms.  A historian who sets aside the Buddha's seven steps because they are wonders told by devotees cannot accept Jesus' walking on water simply because Mark tells it.

But the two cases are not alike in every respect, and the differences are matters of evidence, not of faith.

1. Distance. The Buddhist birth wonders sit in texts that reached writing about four centuries after the Buddha. Paul's list of resurrection witnesses in 1 Corinthians 15 was written about 25 years after the crucifixion,¹⁰ and names living people a reader could question.
2. What the religion stakes on history. Paul writes: "If Christ has not been raised, your faith is vain; you are still in your sins."³⁹ The Christian claim stands or falls on one event at one place and time. Buddhism stakes far less on history. In a discourse often quoted by modern Buddhists, the Buddha tells the Kālāma people: "don't go by oral transmission, don't go by lineage, don't go by testament, don't go by canonical authority … But when you know for yourselves …"⁵⁴ On that view a gap of four centuries hurts the history of Buddhism more than its practice.
3. What is tested. The resurrection is argued on the historical evidence in The Historical Case for the Resurrection. That case has to be made there. This article shows only that its witnesses stand far nearer the events than the Buddhist texts stand to the Buddha.

Objections and responses

1.  "The Gospels were written by believers too, so comparing gaps proves nothing."  The Gospels are believers' books, and that is a fair reason to read them critically.  The comparison here is about something else:  how far the earliest evidence stands from the person, and how many separate lines carry it.  On those measures the Gospels stand much closer.  Jesus also has outsiders, Josephus and Tacitus, within a century.  No Western writer names the Buddha for about six centuries.

2.  "Indian oral tradition was far more exact than Greek or Jewish tradition.  The Vedas prove it."  The Vedas do show that an oral culture can keep a text word for word for many centuries.  But the Buddhist texts were not kept the Vedic way.  Anālayo, a monk and the leading comparative scholar of these texts, explains that Vedic boys learnt sounds before meaning, while Buddhist monks learnt as adults and were taught to care for meaning.  The result was a tradition that kept doctrine well and wording less well.³  The Buddhist texts' own variety across schools shows this.¹³

3.  "History does not matter to Buddhism.  The teaching is tested by practice, not by records."  Many modern Buddhists say this, and the Kālāma discourse supports them.⁵⁴  It is a real difference from Christianity, whose central claim is a datable event.  But traditional Buddhism also rests much on the Buddha's own testimony about things no practice can check, for instance his assertion that gods exist.³  How reliably that testimony was passed on is a historical question.

4.  "David Drewes is right:  nothing about the Buddha has been established."  Drewes is right that there is no record from the Buddha's own lifetime.  But a king marked his birthplace within about 150 years.  Sakyas enshrined his supposed relics near the place the texts name.  Teachers two centuries later traced their lineage to his named disciple.  The texts describe a pre-imperial world their later editors no longer knew.  Von Hinüber and Wynne answer Drewes on these grounds, and almost every specialist agrees with them.¹⁹⁴  By the standards used for other ancient figures, the Buddha existed.

5.  "The 2013 excavation proved the Buddha lived in the sixth century BC."  It did not.  The posthole charcoal is older than the Buddha's death on every chronology, so either the structure was pre-Buddhist, or the dates are only upper limits that fit any chronology.⁷  Richard Gombrich found nothing in the finds "to suggest anything to do with Buddhism."²¹

6.  "Lumbini and Piprahwa are colonial forgeries."  Führer was a forger, but the Lumbini pillar was found by the Nepalese governor, and an Ashokan pillar stood at the birthplace when Xuanzang visited in the seventh century.²⁶²³  The Piprahwa forgery case has been argued in detail by Terence Phelps and rejected in detail by Harry Falk.²³²⁴  No specialist in Indian epigraphy has accepted it.  The honest gap is that no laboratory study of the Piprahwa vase has been published.

7.  "Apologists inflate the New Testament evidence, so the whole comparison is propaganda."  Several apologists have overstated it, as Part 5 shows.  This article uses the skeptics' own dates for the papyri and the mainstream dates for the Gospels.  The comparison holds on those dates.

8.  "If the miracles in the earliest Buddhist texts are not history, neither are the miracles in Mark."  The rule is right, and this article accepts it.  Early attestation alone does not make a miracle report true, in either tradition.  The Christian case for the resurrection does not rest on early attestation alone.  It rests on named witnesses within a few years, on hostile converts such as Paul and James, and on the failure of the alternative explanations.  Those arguments belong to The Historical Case for the Resurrection and must be judged there.

9.  "Legends grow quickly.  Forty years is long enough for Jesus' miracles to be invented."  Legends can grow in forty years.  The Buddhist evidence shows what growth over centuries looks like:  fire miracles that Anālayo judges "for the most part identifiable as later developments," and birth wonders that differ from school to school.¹³³  The Gospel interval is a tenth as long, and Paul's earlier letters already report the resurrection as the founding claim.  How much legend could form in the shorter interval is argued in the resurrection article, not settled by this comparison.

Issues and unanswered questions

1. The radiocarbon reports for the Gāndhārī scrolls were not read. This article relies on Richard Salomon's summaries of them. The primary reports, Allon, Salomon, Jacobsen and Zoppi (2006) and Falk's papers on the "Split" collection (2011) and the Gāndhārī Perfection of Wisdom (2012), would give the exact ranges.
2. Bechert's own conclusions and Gombrich's 1992 paper were not read. They are known here through Cousins, Prebish, Gethin and Gombrich's 2013 note. The 404 BC figure and the Vinaya-lineage argument should be checked in Gombrich's paper.
3. Drewes 2017 was not obtained. His words are quoted as Wynne and von Hinüber quote them. Von Hinüber says Drewes "claimed that the Buddha never existed"; Drewes's quoted words are narrower.
4. The date of the oldest Pāli text is disputed. Stargardt put the Śrī Kṣetra gold leaves in the mid or late fifth century AD; Wynne, citing Stargardt and Falk, puts them in the late third or early fourth. Falk's paper was not read.
5. Vaṭṭagāmaṇi's reign is 29–17 BC on Geiger's chronology and 89–77 BC on another. Von Hinüber gives both. No argument for the earlier date was found.
6. The forgery debate rests partly on unread sources. Charles Allen's The Buddha and Dr Führer (2008), Andrew Huxley's papers on Führer, and Phelps's separate case against Lumbini were not obtained. A scientific study of the Piprahwa vase would settle the question.
7. The Lumbini excavators never explained a natural-soil sample dated 1681–1521 BC beneath the structure, and their later work at Tilaurakot and elsewhere was not read for this article.
8. The Bharhut evidence is read two ways. Lüders took the donor title pañcanekāyika, one "who knows the five nikayas," to prove the Pāli canon in the second century BC; Schopen, after Lamotte, doubts that it refers to the five collections at all.

Verdict

A Buddhist teacher called Gotama, of the Sakya clan, very probably lived in the Ganges plain and died about 400 BC, perhaps as early as 483 BC.  The best evidence is a royal inscription at his birthplace about a century and a half after his death, joined by a reliquary of his clan, lineage inscriptions naming his disciples, and a body of discourses that rival schools preserved separately and that describe the world before the Mauryan empire.  That body of discourses was oral for three to four centuries.  Its oldest manuscripts are Gāndhārī scrolls of the first centuries BC and AD.  Its shared core is best explained by a source older than Ashoka.  Buddhist scholars such as Anālayo, Sujato and Wynne make that case well and state its limits honestly.

Measured the same way, the evidence for Jesus stands much nearer the man:  letters within about twenty years, narratives within forty to seventy, outsiders within a century, and fragments of the texts within about one to two centuries.  The Buddhist case is stronger than apologists usually allow in one respect, an early royal inscription.  Apologists have also overstated the Buddhist gaps and their own papyri.  Neither tradition's early texts can, by their age alone, establish the miracles they report.  What the comparison does establish is that the Christian claims come down a much shorter chain.

Sources