Trusting our sources: manuscripts, archaeology, and what we “cannot know”

I am fresh back from the “Buddhism and Social Justice” conference hosted by Leiden University, The Netherlands.

This will be the first in what I hope will be a number of posts in the coming weeks about individual papers and ideas flowing from the conference, posted both here and/or at my own blog, American Buddhist Perspective(s). This post has to do with methodology and how we approach our sources, so I think it is something everyone here can appreciate and, I hope, offer feedback on. At the conference Prof. Steven Collins made the very interesting plea:

Don’t touch early Buddhism. It doesn’t exist. We don’t know anything about it. Anything you say about it is a complete fantasy. But, if you start out a lecture course or a book with early Buddhism, necessarily, whatever your fantasy, the rest of the book is going to look like a degeneration, or an accommodation, or something else.

So then we get this strange thing in Buddhist Studies, which you don’t get in Catholic Studies: theory vs practice. Nobody would think of giving a lecture course on Catholicism with: ‘here’s the theory, but here’s the practice.’

And it seems so natural for us in Buddhism, because of the essentially 19th century Romantic ideas that both Buddhists and Western scholars had about ‘the early period’. So it is not only that it is difficult to fantasize about early Buddhism, it is positively harmful for your health.

Prof. Jonathan Silk, who spoke earlier on the same day, made the similar observation that many of our common assumptions about Buddhist ideologies and communities are challenged by the reality of Buddhist history and of contemporary Buddhist societies (e.g. records of slave ownership by Buddhist institutions in various times and places – specific instances in Tibet, China, Burma, and Korea were all discussed separately in the conference).

After outlining some other instances of Buddhists behaving badly, Silk asked the very important question: “Just what is the relation between our expectations and the materials we study?”

And it is this question I hope we might ponder as a collection of students and scholars of Indian Philosophy.

Silk went on to note that much work on Buddhism begins with an examination of the ideas of the Buddha himself; notably Richard Gombrich’s 2009 book “What the Buddha Thought.” He then said, “Speaking for myself, I am entirely convinced that we have no idea, and never will have any idea what the Buddha thought, or believed or did. But I am equally convinced that, from the point of view of disinterested scholarship, this of absolutely no importance whatsoever. What we do or can know is what Buddhists, or at least some Buddhists, thought the Buddha thought.”

This leads me to the natural question, “What are your thoughts on what Silk thinks about Buddhists thinking about what the Buddha thought?” (insert grimace)

Some of you, I hope, will have done some comparative work on thinkers roughly contemporaneous to the Buddha such as Plato and Aristotle. I didn’t get too far with either of these, but I do seem to remember that Aristotle’s work disappeared from the Western (Latin) world for some time and only survived through Arabic sources which then had to be translated into Latin and eventually European vernaculars.* Apparently, at least some Greek sources also survived to be transmitted directly into Latin, etc. As the Stanford Encyclopedia of Philosophy entry on Aristotle’s Biology reports:

The surviving corpus of Aristotle derives from medieval manuscripts based on a 1st century BC edition. There were no commentaries on the biological works written until they were collectively translated into Arabic. The first appearance of Aristotle’s biological writings in the West are Latin translations of an Arabic edition by Michael Scot, which forms the basis of Albertus Magnus’ De animalibus. In the 13th century William of Moerbeke produced a Latin translation directly from the Greek. The first printed editions and translations date to the late 15th century, the most widely circulated being that of Theodorus Gaza. In addition to the three works traditionally referred to as History of AnimalsParts of Animals and Generation of Animals, there are a number of briefer ‘essays’ on more specialized topics: On animal motionOn animal locomotionOn respiration,On life and deathOn youth and old ageOn length and shortness of lifeOn sleeping and wakingOn the senses and their objects (the last six being included in the so-called Parva naturalia). Whether one should consider De Anima (On the soul) part of this project or not is a difficult question. What is certainly clear, however, is that there are important connections between the theoretical approach to the relationship between body and soul defended in that work and the distinctive way that Aristotle approaches the investigation of animals.

Our foundational edition dates to the 1st century B.C.E., some 200 to 300 years after Aristotle’s death. How can we have any faith that this edition accurately maintain’s Aristotle’s ideas? Prof. Martha Nussbaum, in her work on De Motu Animalium (MA), gives a compendium written by Nicolaus of Damascus (b. ~64 B.C.E.), paraphrases in the works of Themistius (317-390 C.E.) and Simplicius (c. 490- c. 560 C.E.), and a reference by Philoponus (490 – 570 C.E.) as being the evidence that there is “as much early external evidence for the authenticity of the MA as there is for any of the major works with whose tradition it is associated” (p.5).

It is worth noting that, according to wikipedia at least, our manuscript evidence of the first of these itself only goes back to the 14th century C.E. It seems then, that hard evidence about anything that Aristotle thought is completely lacking given the strict criteria that Silk and Collins seem to espouse.

Perhaps we could attribute greater fidelity to them because the Greeks used writing rather than oral tradition, but there is at least some scholarship that shows that texts preserved in writing fair no better than teachings that are orally preserved (I think here of various books by Bart Ehrman on the evolution of the Bible, but suggestions closer to our purposes, either pro or con, would be greatly welcomed).

That brings us to the next problem, the connection between those 1st century B.C.E. foundations and what existed in the 9th century. We know that the Bible was altered over the centuries as manuscripts were copied, and for skeptically minded scholars, it will seem almost certain that at least some elements of the Pāli Canon and Aristotle’s works were as well.

So can we talk meaningfully about “Aristotle’s ideas” or is this merely a projection of 19th century Romantics, as Collins suggests was the case with the lingering interest in “early Buddhism”? How about other Indian philosophers? How much of what we “know” about their ideas comes down only through now-long-lost manuscripts and only very late extant materials which may or may not accurately reflect their thought?

I must admit that when I studied ancient philosophy, there was no skepticism of this sort in my professors. Certainly, when it came to the Pre-Socratics for whom we have merely fragmentary evidence to begin with, our understanding is also fragmentary, but we were never told that, for instance “the fragments of Heraclitus tell us more about what philosphers, or at least some philosophers, thought Heraclitus thought.”

Those of you familiar with Gregory Schopen’s “Bones, Stones, and Buddhist Monks” (1997) will know these arguments well enough. In the first chapter of that book, entitled “Archaeology and Protestant Presuppositions in the Study of Indian Buddhism” he writes:

When Europeans first began to study Indian Buddhism systematically there were already two bodies of data available to them, and the same is true today. There was, and is, a large body of archaeological and epigraphical material, material that can be reasonably well located in time and space, and material that is largely unedited and much of which was never intended to be “read.” This material records or reflects at least a part of what Buddhists-both lay people and monks-actually practiced and believed. There was, and is, an equally large body of literary material that in most cases cannot actually be dated and that survives only in very recent manuscript traditions.) It has been heavily edited, it is considered canonical or sacred, and it was intended–at the very least–to inculcate an ideal. This material records what a small, atypical part of the Buddhist community wanted that community co believe or practice. Both bodies of material, it is important to note, became available to Western scholars more or less simultaneously. The choice of sources for the scholar interested in knowing what Indian Buddhism had been would seem obvious.

The choice, of course, is the archaeological material, which turned out to be the road not taken, by and large to this day.

To me this marks an at times rather sharp division between historical and philosophical (sometimes called phenomenological) approaches to the study of religion. I’d like to be charitable and say that both approaches are valid and are capable of yielding new and complimentary understanding of ancient thought and ways of life. We can debate the thought of Aristotle based on quite late evidence, holding out that perhaps a discovery will add to his body of works or, contrarily, somehow show that some of what we have taken to be “his” work is in fact an invention by later authors. However, this last week I was reminded once again that for at least some, this charitable approach will not do, and only one way of approaching philosophy – or intellectual history – can produce results worth pursuing.

So, my questions for philosophers is: how do you work with historians, especially those who are sceptical of the possibility of understanding what early philosophers meant or thought? Is the analogy with Ancient Greek thinkers fatally flawed for some reason? Or are we just decades behind our counterparts in this area when it comes to carefully debating the merits of the authenticity of texts?

Aristotle’s De Motu Animalium: Text with Translation, Commentary, and Interpretive Essays. Princeton: Princeton University Press (1978) by Martha Nussbaum offers insight into how a contemporary philosopher approaches these questions.

De Anima by Nussbaum and Amelie Rorty presents much the same picture.

About Justin Whitaker

Justin Whitaker is a Ph.D. Candidate at Goldsmiths, University of London, focusing on a comparison of early Buddhist ethics (primarily from the Nikāyas) and the ethics of Immanuel Kant (1724-1804). He has a B.A. in Philosophy from the University of Montana and an M.A. in Buddhist Studies from Bristol University, UK.

29 thoughts on “Trusting our sources: manuscripts, archaeology, and what we “cannot know”

  1. Thanks, this is a really fantastic article and certainly echoes things I myself have been thinking about lately, about what seems to me to be a pervasive skepticism in Buddhist studies. Obviously, a degree of doubt and skepticism is probably necessary to avoid the pitfall of romanticism, but I, for one, think the pervasive and perduring attitude for studying Buddhism (indeed, anything) should be one of curiosity rather than skepticism. In other words, I think it’s important to focus on answering the question “What can we know?” Rather than ” Can we really know anything?” That skeptical attitude has some social capital, perhaps especially in the academy, where it’s considered good scholarship to reign greater understanding by saying “well most benighted schmucks think X (i.e. With regard to the past, the historicity of the Buddha), but if you look at the evidence, there’s really no way to say that;” which is also a much easier position to maintain than the positive assertion “well most benighted schmucks think X, but really, if you look carefully, it’s more like Y, Z, and Q,” which is a much more difficult position to maintain, as this can be further criticized more easily, but I think represents a greater understanding. Ok, I’m making some pretty broad and abstract accusations, but basically all I’m trying to say is the common observation that it’s easier to criticize than contribute. In that sense, I would agree with the last question you posed (although of course, I don’t have much expertise in this regard) that we are behind our peers in understanding, but that I think the remedy is to foster a deeper sense of curiosity towards our subject rather than skepticism. Sorry for the long rambling comment.
    fwiw, TN

    • Many thanks, Tom.

      I think you have summed up things very well and I particularly agree with your stress on the importance of curiosity and your statement that “it’s easier to criticize than contribute.”

      I’m sure that Silk, Collins, and Schopen would agree with this, and each does have his own body of contributions to the field. Here I focused on their criticisms, and perhaps some would say I did so too much. But in terms of what you note is a “pervasive skepticism in Buddhist studies” I think these examples are illustrative, if nothing else.

      I think Schopen in particular is dismissive of what we can really know about the Buddha, but I am not sure if Silk and Collins share his level of skepticism. As a cursory reading of either of their works will suggest, we can know quite a bit about early Buddhism, if not the Buddha himself, based on close examination of the texts.

  2. I won’t say much, other than to say read Vens. Sujato and Brahmali’s “Authenticity of the Ancient Buddhist Texts”‎

    The Buddha taught his Dhamma to a group of his monks, some of whom were charged with memorizing the teachings verbatim. After the Buddha’s death, the monks met to determine what was needed to be remembered and passed on. The monks met again at the Second Council, and the oral tradition was passed on. King Ashoka revived Buddhism and supported the transmission of the teachings until they arrived in Sri Lanka and were first written down. A good, disciplined oral tradition is nearly as good as a written one. The writings that passed through Sri Lanka and China largely echo each other. The historical events recited in these ancient texts match up with the dates attributed to the Buddha. There’s no scientific reason to doubt that the experiences of the original Sangha ( the Dhamma and Vinayay) have not been largely maintained through this process.

  3. Alas, I work constantly with manuscripts, and I look at them both as as archaeological artifacts as well as bearer of texts. Basically, I work for you guys, since I try to reconstruct slowly, but very slowly, how, why, when, by whom etc. manuscripts were produced, I catalog them and enable you to access them in the easiest way. Should I stop doing this? I mean, if it’s pointless to try and reconstruct the past, just tell me and I’ll change my line of work.

    Instead, I would like to ask a question to all those Buddhologists, so to say, who stress the fact that one should look at contemporary Buddhism (i.e. ritual studies): do you think that a historian of the Christian church would do the same, just stroll into a contemporary Catholich church in a village in the middle of nowhere in Spain or Italy, and pretend to reconstruct the history of early Christian communities based on his observations of present day Catholic rituals? Of course not, you would answer. But it is different in South Asia, you would argue, the tradition is still a living one etc. We all know the argument, right? But do you truly think it’s that different?

    So, we are left with nothing. We cannot try and reconstruct, let’s say, early Buddhism only by means of archaeological, textual, epigraphical evidence, and maybe also with a little help of field research in contemporary Buddhist societies. Even less can we try and reconstruct a fair approximation of what the historical Buddha thought. So, what is all the fuss for? Why did you write this post, and why am I posting a comment?

    Well, I think I’ll just go on using all the sources we have at our disposal for historical research, and keep on writing and saying something like: “well most benighted schmucks think X, but really, if you look carefully, it’s more like Y, Z, and Q,”

    • First, thank you for your work with manuscripts. I definitely think it’s valuable and should be continued. As I said toward the end, I think historical and philosophical (and I should have said “and other”) approaches all can produce complimentary understandings. Reconstructing the past, as you put it, is also what we try to do in philosophy, so -again- I think we can happily work together on this.

      What I would worry about is if you were to determine that a manuscript came from 14th century Bhutan and then said that this manuscript can *only* tell us anything meaningful about 13th century Bhutan, especially if the text claims to be a commentary by a direct disciple of Atiśa. If you accept that something from the 14th century in one region could tell us, with at least approximate accuracy, about someone/something/some mode of thought in the 11th century in quite possibly a different region, then we’re on the same page here.

      The fuss, for me, is simply that I *do* think we can know important things about early Buddhism and the Buddha based on textual, archaeological, and other sources (just as philosophers take for granted that we know a great deal about what Aristotle thought). So I am curious/worried about the growing number of scholars, most of whom I really respect, saying we cannot know X, Y, or Z.

      As for your refutations, as long as you’re kind to the poor schmucks and clear in your refutation and evidence for different understandings, I’m 100% on your side.

  4. Talking specifically of Early Buddhism, it is definitely possible to know something about it from the literary evidence itself (this doesn’t however mean we should ignore the archaeological evidence).

    The theravada canon is the earliest surviving Buddhist canon, much of which linguistically dates to the pre-classical i.e. the late-Vedic period Specifically the Sutta Pitaka of this canon is common to all the Buddhist schools (even the non-Theravada schools). These contain the inspired utterances (suttas i.e sūktas) attributed by tradition to the Buddha, but these sūktas are regurgitations of what the Buddha was remembered (by other bhikṣus) to have uttered, hence we can’t be sure that they are not ‘imagined utterances’. If we ignore this outer core, the canon has an inner core, where a different picture of Buddhism emerges. This inner-core consists of predominantly verse sūktas, which are very likely authentic since not only do they display inconsistencies with the content and language of the prose sūktas but also accord with the medium of literal oral transmission which verses are more suited to – à la the Vedic oral tradition.

    • :) Yep.

      I’m not sure if Prof. Collins researched the history of debates about theoria and praxis in Western thought, but yes, they do seem to be all over the place. So indeed, I imagine that scholars of Catholicism would make this very distinction a starting point for a course or book.

      I can quickly cite/note from Arabic/Aristotelian sources:

      Decades after Al-Farabi, Ibn Sina (Avicenna) compiled the ideas of many Muslim philosophers of the previous centuries and established a new school which is known as Avicennism.[21][27] After this period, Greek philosophy went into a decline in the Islamic world. Theologians such as Al-Ghazali argued that many realms of logic only worked in theory, not in reality. (via

      And Immanuel Kant’s interest and worry about the distinction in his 1793 essay on this topic:

      So those are at least two empirical data-points for concerns about ‘theory vs practice (or reality)’ that have nothing to do with 19th century Romantic ideas.

      • I can’t help but wonder if some of the theory/practice talk came from the nature of that specific conference, social justice, and it was perhaps critiquing the way people tend to put Buddha and Buddhism on the “good guy” side of the tally fairly quickly.

        Also, If there really is a difference between how people treat Aristotle (etc.) vs. Buddha regarding a concern with what they “really thought”, then perhaps some of it has to do with the nature of their contributions. Aristotle left us arguments, mainly. Buddha indeed gave arguments, but legitimated by claims about his own contemplative discoveries, and hence personal authority. There is a reason that some Buddhist philosophers (Dignaga, etc.) were so concerned to argue that his utterances were a pramana and he was an apta (indeed, pramana-bhuta, the very form of a source of knowledge) of the highest rank. This way of approaching him does stress more on Buddha the person.

        • Excellent points, Matthew. And sorry for such a late response.

          Yes, it’s quite possible that the concern in mind had a lot to do with the particular topic of the conference. But then I would rather it be stated that there is often a world of difference between the Buddhism of the texts and what we observe Buddhists doing in historical settings. We don’t need, imho, to so strongly deny the reality of the Buddhism of the texts, especially as the texts and practices are appealed to time and again by later Buddhists.

          Regarding the nature of Aristotle’s thought vs the Buddha’s, I think you’re right. However, this is an area I’d like to see explored more. We know that thinkers like Kant and Descartes were keenly interested in religion and even had religion built in to certain aspects of their philosophy. And Aristotle too had arguments for the existence of God and, if I remember correctly, thought that through theorizing we humans could participate in the activity of God, or something along those lines.

          So it’s not unreasonable to wonder if a certain ‘religious’ nature could be ascribed to his philosophy and perhaps there were even cults developed around him after his death.

          If so, we might further say that the fact that such cults died out and our current appreciation of Aristotle is relatively irreligious in content, while the Buddha’s cult survived and thus our view of him as a religious figure is just a matter of historical contingency.

          But, lest we fall too far into speculation, I’ll say that these are all empirical questions that we *can* find evidence for one way or the other.

          • I don’t think there’s much to be gained by debating whether Aristotle’s thought falls into that misleading and overused category “religious”. The more salient point here might be: to what extent did Aristotle have a following comparable to the Buddha’s, and how might their respective followings affect the way we read their thought?

            As I recall, Julia Annas in The Morality of Happiness speaks a lot about later Aristotelians (Peripatetics). That book is well worth a read for scholars of Buddhism – I’ve long thought Hellenistic philosophy is a much closer comparisand for Buddhist thought (and for other practically oriented Indian schools) than are the usual contemporary philosophical suspects (utilitarianism, Kant, even Aristotle).

  5. Angarika Dharmapala says (sentence numbering mine):

    1. The Buddha taught his Dhamma to a group of his monks, some of whom were charged with memorizing the teachings verbatim.
    2. After the Buddha’s death, the monks met to determine what was needed to be remembered and passed on.
    3. The monks met again at the Second Council, and the oral tradition was passed on.
    4. King Ashoka revived Buddhism and supported the transmission of the teachings until they arrived in Sri Lanka and were first written down.
    5. A good, disciplined oral tradition is nearly as good as a written one. 6.The writings that passed through Sri Lanka and China largely echo each other.
    7. The historical events recited in these ancient texts match up with the dates attributed to the Buddha.
    8. There’s no scientific reason to doubt that the experiences of the original Sangha ( the Dhamma and Vinayay) have not been largely maintained through this process.

    My response:
    1. The Buddha (or anyone for that matter) never charged any monk with memorizing anything (certainly not any ‘prose’ sutta that exists in the present sutta pitaka), therefore this is evidently blind belief, unless you can present any evidence that this was the case. If at all anyone voluntarily memorized anything during the Buddha’s lifetime, they were not prose suttas, but mainly/only verses like the ones in the sutta nipata. The very name sutta nipāta means “suttas falling down” (from the Buddha’s time?).

    2. The vast majority of the sutta pitaka (much/all of the prose suttas) were composed after the Buddha’s passing away. However there is no evidence for any oral tradition for these prose suttas, all of the suttas were evidently written down. You can have a pitaka (basket) of written texts, but you can’t have a basket of texts in your mind. Canonization of extant suttas (and standardizing their written language) evidently happened later.

    3. These councils that you keep referring to – nobody today knows what happened in them, how do you know what they did? Why do you feel the need to invent your own history of Early Buddhism?

    4. The idea that the Buddhist canon was first written down in Sri Lanka defies all rationality. There was evidently no writing in Sri Lanka until the Buddhists reached there, so where and how did these Sri Lankan Buddhists find the art of writing in Sri Lanka? Writing entered the sub-continent in the 4th century BCE from the north-western parts (now Afghanistan/Pakistan) and much if not all of these early writings are Buddhist. Therefore Buddhism had a reason to adopt writing in north-western India, 4th century BCE onwards – to preserve their canon.

    5 & 6 – no comments.

    7. No they dont match up at all. The dates of the Buddha’s life are as much a matter of conjencture and controversy today as they have been during the last 200 years. We only know that he may have lived sometime during the 5th or 4th century BCE, that too is not certain, he may have lived in the 6th century.

    8. There is every reason to believe that Pali is not a language at all but rather an epigraphically standardized form of the canonical texts, and moreover that there is no independent evidence for the existence of any middle-Indic language in the time of the Buddha. So what is there in the Pali canon is a linguistically distorted copy of Buddhavacana, and the prose suttas are entirely unreliable as a record since they were first composed after the Buddha’s death based on the recollections and interpolations of some Bhikṣus.

    • Ramakrishnan, read Ven. Sujato and Brahmali’s book and let me know what you think. You are making certain assumptions in your notes that may have been refuted by others. As I am not a EBT scholar (my letter above is a friendly conversation piece), I’ll leave the proofs to those doing the heavy scholarly lifting. I have a large degree of confidence in the work of Vens. Sujato and Brahmali…they’re both serious, scholarly, and while some might suggest they have a bias I feel they were really interested in scientific proofs. Whether the early Nikayas are Buddhavacana in total, or partially, matters not that much to me, as the Dhamma itself seems to me a strong enough foundation to build my spiritual house upon. Namaste.

  6. But more generally, I love the comparison you make with Aristotle’s writings. I had no idea that our earliest surviving Aristotle manuscripts were from the 1st century CE. I wonder whether Collins and Silk and Schopen have any idea either. Though I also wonder whether they would care. (If one isn’t a philosopher, one can simply wield one’s destructive hammer and laugh at all those horrible pre-21th-century people who were so much dumber than one’s own brilliant philological self.)

    To the extent that their criticisms ring true at all, I think Gombrich has the right approach to them. Rather than a historical person, “the Buddha” becomes a character to whom we can attribute a certain consistency even if that consistency was only in the mind of his redactors. He is a hypothesis we use in interpreting the texts (Alexander Nehamas’s “postulated author”); we start out assuming a coherent worldview worth learning from and the burden of proof is on those who want to say that that isn’t there.

    Why? Because we’re philosophers. The whole point is to read constructively and learn from a real set of ideas that fits together.

  7. Let me start out by saying that I teach comparative religions. So I frequently deal with this same question in many different contexts and in regard to several different figures/teachings/texts.

    My general approach is to quote the tradition…….as in the following: “what the Buddhist tradition takes to be the Buddha’s earliest teachings blah blah” or “the tradition has established the following story to be the authoritative version of events following the Buddha’s death even though we have no historic evidence…” And I do the same in regard to Christianity, Judaism, etc (the first quote in the above article notwithstanding).

    My understanding is that the first few centuries of every religious tradition is pretty much the same in this regard: lacking in solid historic data. Founders do not tend to leave good notes. All we have is myth, legend, hagiography. And in every religious tradition there is a great distance between theory and practice.

    • Thanks for this, Lyone, and my apologies for the late response. My worry with this approach: “he first few centuries of every religious tradition is pretty much the same in this regard: lacking in solid historic data” is that there are sometimes vastly different amounts of solid historical data among religions. So I couldn’t say “All we have is myth, legend, hagiography.” We do have those, but we also have archaeology, historians such as Josephus, linguistic analysis, etc. So I think we can and should do our best to separate out what is strictly claimed within the tradition (especially when the claims just appear much later in the tradition) and ideas and events which can be attested to from multiple sources.

  8. Hi Justin,

    Some interesting comments already. Sujato & Brahmali’s book on the authenticity of the texts has been mentioned. Elsewhere I responded to this book: “shock news, bhikkhus believe texts are authentic!”. Along with them we can group Bhikkhus Anālayo and Bodhi. This is one end of a spectrum. Religieux with a vested interest in tallying received tradition with the evidence such as it is. The result is an increasing sophisticated apologetic literature, spurred no doubt by criticisms from the other end of the spectrum.

    Collins, Silk and Schopen represent the other end of a spectrum of (mostly USA) scholarly opinion on the texts. They don’t accept any deductions. They are saying that the literature we do have has no *discernible* history. It may well have a history, but we have no direct evidence and all deductions are mere conjecture that cannot pass muster in a post-Popperian world where conjectures must be a priori refutable in order to be taken seriously. Schopen in particular has pointed out that were we do have archaeological evidence it contradicts textual evidence – particularly with respect to how monks live and conducted themselves. One has to respect their individual and collective brilliance, but the position is based on a method which sets the evidentiary bar very high indeed. And the question is: is that a reasonable stance to take? In Schopen’s case in particular we wonder whether he is simply being contrarian.

    In something like a middle-ground we find scholars such as Alexander Wynn and Richard Gombrich and dozen’s of others who write about early Buddhism with more or fewer qualifications. This grouping accept that valid deductions can be made about the authors of the texts from the texts and in the absence of other kinds of evidence. As Gombrich says, we have no good reason to disbelieve the idea of that the Buddha lived (unless we hold the Buddhist tradition to be wholly unreliable). If anyone is a Popperian it is Gombrich, he proof-read Popper’s book “Conjecture and Refutation” as an undergraduate and Popper was a close friend of his illustrious father.

    Clearly *someone* composed the texts. Comparisons with Chinese, Gāndhārī and Sanskrit versions show only relatively minor differences in form and organisation, but almost no differences in intellectual content. A well developed and apparently long evolved (and thus diverse and conflicting). Buddhist literature clearly existed by the 1st Century AD (as datable Gāndhārī texts demonstrate). The question then becomes how long it pre-dates the first manuscripts and Chinese translations? Gombrich and his teacher K R Norman showed that Buddhist texts were cognizant of Brahmanical ideas in the early Upaniṣads. One avenue I’d like to follow up is to outline the view of Brahmins in the Canon and see how that lines up with the views of Brahmins themselves.

    This difference of opinion (and this is what it is) is based around questions of evidence. What constitutes evidence? Do the Buddhist Canonical texts constitute evidence? And evidence of what? What objective criteria exist to allow us to decide this question?

    Calling into question the validity of the evidence of texts and the uses to which the evidence is put is a helpful function of a well functioning scholarly ecology. Schopen has raised the game of the entire discipline. But it’s worth reading Alexander Wynn’s reposte to Schopen. It’s also worth reading Sujato’s resposte (with one idea on his bias towards the Theravāda version of events).

    I’m at the Schopen end of the middle ground. Gombrich just (very cordially) turned down an article of mine for the JOCBS. I argued that the stories about the Buddha were all made up to suit a Brahmanised (post-Asoka) audience long after his putative death; that the Buddha is at best a legendary figure whose qualities represent the values of the day. This is based on a study of the names of the Buddha and his family – Brahmin names almost without fail. He thought I was rather too sceptical about the sources, and not well enough informed on Indian cultural custom (I still think the article is worthwhile so I’ll try another journal).

    On the other hand I have published an article proposing (following the suggestion of Michael Witzel) to connect the history of Buddhism to an influx of people from Iran ca. 1000 BC, bringing with them Zoroastrian ideas. In other words I seek to anchor the idea of early Buddhism in deeper history by constructing a narrative that predates their arrival on the margins of Kosala. By connecting the literature to a much earlier time I hope to show that Buddhism certainly does have a history before Asoka. My next published article will be looking at how Zoroastrian ideas might have shaped the early Buddhist view of karma. This will help to reinforce the idea. A third article on the emergence of the idea of Hell in India will complete the trilogy.

    I think the textual description of the stage of development of the second urbanisation must place the accounts of the early Buddhist texts before the Empire building efforts of the Mauryas. The milieu of the early Buddhist texts is largely one of self-contained city states ruled by warring petite kings. There is no sign of the Mauryans and they made such a huge impact on the heartland of Buddhism that the absence of evidence counts as evidence of absence in this case. However note the absence of Mauryans in Mahāyāna Buddhist texts also – we have to see this as a deliberate archaism. So if the Mahāyanist were deliberately archaic, why weren’t the early Buddhists? It seems to me that the early Buddhist account is too detailed and specific to be an archaism.

    However I also think we see clear signs of post-Mauryan Brahmanical influence on the Canon. I think the conclusion it is that the texts were composed over a lengthy period, beginning in the pre-Mauryan era, but edited and collated later. The level of fidelity is therefore low. And the telling of the history is a complex task. The conflicts and contradictions are insufficiently acknowledged to date.

    Another issue with the field of Buddhist studies is the influx of professing Buddhists – many of them ordained (I count myself in this group). Buddhist studies has long struggled with emic/etic issues. Scholars have been blinded by their sympathies for Buddhism. An increase of emic scholars writing in the style of academics but with fundamentalist agendas is a problem for us. Those of us who are Buddhists have preconceptions it is almost impossible to overcome. Our underlying narratives all too often involve absolute truths, unreasoning faith, and an uncritical eye.

    How many Buddhist scholars have asked themselves why our founder has high status Brahmanical given and family names? Why do his mother and aunt also have high status Brahmin names? Why is his father never referred to as Gautama – as the head of the family it ought to have virtually been like a title (and his son ought to have been Gautamya). My literature review showed that the last person to address this issue was D D Kosambi in 1944, and he sought to smooth over the cracks rather than dig deeper. How does this relate to the process of Brahmanisation that occurred only after the collapse of the Mauryan Empire? And so on.

    In the back ground to this is an issue the present generation of scholars may only be dimly aware of (or perhaps they are better informed than I am?). The study of texts was for a time thoroughly discredited through the influence of post-modernism. The post-modern approach to texts destroyed the very notion of philology: learning about another culture through reading their texts. Pomo argued that nothing could be learned about an author from their text because what we read is a negotiated construction between text and reader. In reading a text we only discover our own preconceptions – we tell the story that we always tell or that we want to tell. A century ago people like us who study texts would have happily called ourselves philologists. K R Norman is a philologist. Perhaps the last of a generation. Now we are Indologists, Sanskritists, Historians, etc., anything but Philologists. Philology was also caught up in the scandal of orientalism – since many of the great orientalists were philologists. The situation is gradually changing. Doing philology is not so gauche and rude as it was in the 60s and 70s. Still I bet no one reading this calls themselves a philologist even though what most of us do is study ancient cultures through their texts – which is philology.

    So Schopen et al are doing us a service by issuing these strongly worded and decisive challenges. Thinking critically about what they say requires the rest of us to work much harder. But are they ultimately correct in their assertions? I think it’s too early to say. The archaeology of India is so far patchy and superficial in most areas. Who knows what else lurks under the soil? The practice of philology is gradually being rehabilitated and who knows it may become pukka again.

    • Dear Jayarava – many thanks for all of this. It’s a bit overwhelming in terms of what to try to respond do…

      First, I’m not sure claims about “what the Buddha thought” aren’t a priori refutable. The body of texts we have can be used to argue for different interpretations and new texts may still be found to upset the whole thing. As you later stated, Gombrich is definitely a Popperian, so the (also later) point about “what counts as evidence” is really the key problem here. And as you also note, aside from Schopen’s possible contrarianism, the main arguments for distrusting texts come from Postmodernism.

      Next, “Schopen in particular has pointed out that were we do have archaeological evidence it contradicts textual evidence –particularly with respect to how monks live and conducted themselves.” I think it’s fairer to say that *some* archaeological evidence contradicts textual evidence, and/or gives us new information. I think we have to be much more circumspect on what evidence Schopen presents and what conclusions he draws from it.

      Schopen and others may have upped the game, so to speak, but I hope that in the process they have not also discouraged many great minds from looking at texts and thinking hard about what they mean and meant to Buddhists, past and present.

  9. Pingback: 163th Philosophers’ CarnivalThe Indian Philosophy Blog | The Indian Philosophy Blog

  10. Pingback: 163th Philosophers’ Carnival | elisa freschi

  11. Dear Philosophers, I really like the blog and this post. I hope in a later comment to state my views on the philosophical and historiographical issues raised in the original post, but I first wanted to offer some information and corrections about the evidential basis of Aristotle Studies. In this post I will address this part of Justin’s post:

    “Aristotle’s work disappeared from the Western (Latin) world for some time and only survived through Arabic sources which then had to be translated into Latin and eventually European vernaculars. Apparently, at least some Greek sources also survived to be transmitted directly into Latin, etc. … Our foundational edition dates to the 1st century B.C.E., some 200 to 300 years after Aristotle’s death.”

    First, at the risk of pedantry, allow me to define some terms. By “lost works” we mean works attributed to Aristotle, and that are included on at least one of the three ancient lists of Aristotle or are referred to by later writers, but that do not survive in their own manuscript tradition. Papyri can be an important source for such works. For example, there is a substantial papyrus fragment of a History of the Constitution of Athens usually attributed to Aristotle (see P. J. Rhodes, A Commentary on the Aristotelian Athenaion Politeia, Oxford 1981). Most lost works survive primarily in what we loosely speaking call “fragments”, that is, testimony, reports, paraphrases, quotations, etc. contained in later authors. These fragments can be used to reconstruct lost works. Recent examples of this include Richard Janko’s recent reconstruction of Aristotle’s lost dialogue On Poets (Philodemus, On Poems, Books 3-4, with the Fragments of Aristotle, On Poets, Oxford 2010); and work that I am doing with D. S. Hutchinson on a reconstruction of Aristotle’s lost dialogue the Protrepticus or Exhortation to Philosophy (see

    By “Aristotle Corpus” we mean the works included in the canonical Berlin Academy edition published by Immanuel Bekker in 1835, which edition remains the standard pagination and system of reference by book and chapter numbers (with certain exceptions). The number of papyrus fragments of works in the Aristotle Corpus is negligible (see: I. Privitera, “Aristotle and the Papyri: the Direct Tradition”, Quaestio 11 (2011), 115-140). The basis of the works in the Aristotle Corpus are the manuscript traditions of these works (and, often, groups of these works), but NONE OF THE MANUSCRIPTS OF THE WORKS OF THE ARISTOTLE CORPUS ARE EARLIER THAN THE NINTH CENTURY (see the authoritative article by L. Minio-Paluello (henceforth M.-P.): “Aristotle: tradition and influence” in C. C. Gillispie (ed.), Dictionary of scientific biography (Vol. 1, pp. 250-281), New York, 1970: “only minor fragments of papyri containing Aristotle’s texts from the corpus and no manuscript older than the ninth century exist”, p.268). Thus the situation with manuscripts is much worse than Justin imagines. Some key works do not exist in manuscripts until much later, e.g. the Politics, which does not appear in any manuscript older than the thirteenth century (see M.-P., p. 268). What Justin refers to as “our foundational edition dates to the 1st century B.C.E.” is the HYPOTHETICAL edition, sometimes referred to as the “Andronican edition”, attributed to “Andronicus of Rhodes”. Recent scholars have expressed skepticism about the importance of the “Andronican” edition (see, e.g. J. Barnes, “Roman Aristotle” in J. Barnes and M. Griffin (eds.), Philosophia Togata II: Plato and Aristotle at Rome (Oxford 1997), p. 1-69.). But whether or not the Andronican edition was a significant influence on our manuscript traditions, we should not call it “our foundational edition” for the simple reason that it does not exist any more.

    Now among the works of the Aristotle Corpus, some are authentic, others are dubious, and still others are spurious. By “authentic” we mean that there is a consensus of scholars that the work is attributable, at least in large part (excluding, e.g., intrusive glosses), to Aristotle of Stagira (384-322 B.C.). By “dubious” we mean that there is no consensus of scholars on the authenticity of the work (such as the Magna Moralia or the Problems); by “spurious” we mean that there is a consensus of scholars on the inauthenticity of the work (such as On the Universe or Physiognomics). (The standard classification can be seen at a glance in the table of contents to J. Barnes edition “The Complete Works of Aristotle: the revised Oxford translation”, 2 vols., Princeton 1984). It is because On the Movement of Animals was by some considered a dubious work (but is not in, e.g. the Revised Oxford translation) that Nussbaum felt obliged to address the issue in the passages of her edition and commentary cited by Justin. The way she addresses the issue—by citing other ancient scholars that refer to it—indicates an important way to compensate from the relative lateness of our manuscript tradition. References to works of the Aristotle Corpus by authors in the Greek, Latin, and Semitic traditions are key to corroborating the authenticity of the works of the Aristotle Corpus.

    By “the Greek tradition” I mean the tradition of transcribing and commenting on Aristotle’s works in Greek, a tradition that spans from the mid fourth century B.C. (with Aristotle himself) to the present day. By “the Latin tradition” I mean the tradition of translating and commenting on Aristotle’s works in Latin, a tradition that spans from the fourth century A.D. to the seventeenth century. By “the Semitic tradition” I mean this with respect to Syriac, Arabic, and Hebrew (fifth century A.D. to sixteenth century). I set aside the Armenian, Neo-Latin, German, English, and other traditions that are not significant for the establishment of Aristotle’s text. (These categories and their definitions are taken from M.-P. For the Armenian tradition and its importance in establishing the textual evidence for Aristotle, see J. Barnes in J. Barnes and V. Calzolari (eds.), L’oeuvre de David l’Invincible).

    It is worth pointing out here that the chief basis of authenticity of the Aristotle Corpus, other than the prima facie attribution in the manuscript traditions themselves and their internal coherence, is the continuity of the Greek manuscript tradition. Although the texts were not always available in the Latin West, there seems to have been a continuous tradition of copying Aristotle’s works in Greek, from the beginnings by his own students (especially Theophrastus and Eudemus) but by memebers associated with his school over a period of about 250 years, down to the hypothetical Andronican edition of circa 70 B.C., through the subsequent Greek Commentators and the Latin, Semitic, and Armenian traditions. Although there must have been reduction of this activity with the closure of Pagan schools in 529, the revival of the process of writing and copying manuscripts of the works in Aristotle Corpus in the ninth century seems to have been based on texts that extend back to the sixth century (this is the position of M.-P.). Thus although the Latin and Semitic traditions are often the source of variant readings (and occasionally very important in this respect), they are not absolutely essential to the establishment of the Aristotle Corpus. The corroboration of the Greek commentators, especially Alexander of Aphrodisias in circa 200 A.D. is, however, extremely important. As for the claim, restated by Justin, about the dependence of the tradition on Arabic texts, “it cannot be emphasized too strongly that Aristotle was Latinized from the Greek much more than from the Arabic and, with very few exceptions, earlier from the Greek than from the Arabic. Although competent scholars have tried to make this fact known, the commonly held view of historians of ideas and of people in general is the wrong view: that the Latin Middle Ages owed their knowledge of Aristotle first and foremost to the translations from the Arabic.” (M.-P., p. 270).

    I apologize for the length of this post, and the fact that I have not managed to comment on the substantive issues raised in Justin’s post. Hopefully what I have said will, however, underscore his overarching point, that these issues deserve serious consideration not only by scholars working in the Indian but also in the Greek tradition, and in particular in the Aristotelian tradition.

  12. Many thanks for this, Monte. It is tremendously helpful. If I may recap a few of your key points:

    The “canon” for Aristotelian works dates to Bekker in 1835. There were earlier (partial?) catalogs and lists but this is our current academic standard. In terms of *manuscripts* nothing we have predates the 9th century and much is later than that, e.g. the Politics, 13th century. The 1st century B.C.E. “Andronican edition” thought to be a key source for the later tradition does not exist and my not have been that important after all.

    In terms of what we have today, texts are classified as *authentic, dubious, or spurious* (note: as far as I know we don’t have any such accepted system in discussions of texts attributed to the Buddha). These are necessarily somewhat fluid categories, but the criteria of prima facie attribution, internal coherence, and the existence of a continual (trustworthy?) manuscript tradition all help ground arguments for and against authenticity. And lastly, it’s time once and for all to drop the over-emphasis of our debt to Arabic sources for preserving Aristotle.

    I hope that is a fair summary of your key points.

    Again I will have to do some further research -or hope someone jumps in here- on the relative *trustworthiness* of the Pali oral -> manuscript tradition. There has been work to show the more-or-less obvious redactions, though again we don’t have anything like a scholarly consensus on anything related to authentic utterances of the Buddha. My sense is, echoing Gombrich, that we’re something like 100 years behind Western scholarship on Aristotle (he made the case for us being 100 years behind Biblical scholarship in this regard).

    Granted that we’re behind, I wonder what sense we can make of claims to just ignore questions of authenticity and just get on with what I’ve termed “historical’ work. As Mathew suggested above, perhaps the idea of just getting on with other work is limited to the context of social justice, as in “we should look at what they’re doing and not get caught up in the red herring of textual ideals.” On the other hand, if we try to approach the Buddha as a philosopher on par with Aristotle, we ought to emulate the progress of the last 100 years of Aristotelian studies.

    In that vein, I wonder if there have been historians who have said similar things about Aristotle – that we cannot really know what *he* wrote about anything because all we have are *very* late manuscripts (and what manuscripts preserve is *not* what a great thinker has said or written, but instead what the *scribe* wants us to think the great thinker thought…), and if so, how have other Aristotle scholars responded to such skepticism?

  13. I’m not sure that Aristotle is really the parallel to consider. Textually, it may be similar–a tradition that traveled over time from one region to another through translation. But Aristotle is a known historical author.

    I would think the Buddha would be more akin to Socrates, Confucius, or Christ: Oral teachers whose disciples gave testimony to their teachings. It’s only natural to be doubtful of whether those testimonies are accurate or not. That’s a different situation from wondering if a text was copied or translated accurately. Even the original copy is suspect of containing misrepresentations, additions, or omissions. Add to that the usual textual history problems that they sometimes turn into literary open source projects over the millenia and–yes, I agree: We don’t know what exactly the Buddha taught.

    But I would say we do know what Buddhists have taught. That’s enough for me to chew on, personally.

  14. Pingback: That all-too-human urge to understand the heart of things

  15. Greetings from Department of Archaeology and Museums ,Telangana, India!
    The Department of Archaeology and Museums Telangana, India, is the premier organization for archaeological research and protection of the cultural heritage of Telangana state.
    We feel privileged to connect with people from all over the world, involved in preservation and conservation of heritage, maintenance of museums and other similar activities.
    By means of this strategy, we intend to collaborate with the global fraternity of historians,curators and others, aiding each others efforts towards preserving the worldwide heritage.You can connect with us to get regular information regarding the Archaeology and Museums department and its efforts towards maintenance of centuries old sites and monuments in this region which had witnessed the rule of several dynasties
    Want to know more about us ? Please visit our official website and Facebook page.

  16. I wonder if we might revive this discussion, possibly with a new post, because of some recent and very striking claims that pertain to the issue.

    I am referring to Christopher I. Beckwith’s discussion of the value of Megasthenes (350-290 BC) as a source of information on early Buddhism. Megasthenes’ Indica is a lost work, some fragments of which are preserved in the Geography of Strabo (64 BC-AD 24). Beckwith explains that Strabo’s version of Indica “had been interpolated and expanded by others”, that Strabo’s method of selection was compromised by the period’s preference for “light, chatty, titillating stories”, and that the process of transmission through medieval scribes was imperfect, being especially corrupt when it comes to foreign names. Nevertheless, “it preserves part of the earliest dated eyewitness account of Indian philosophical-religious practices and ideas by far. It is therefore incalculably more important than any of the other texts traditionally considered to represent or reflect Early Buddhism” (Greek Buddha: Pyrrho’s encounter with early Buddhism in central Asia, Princeton, 2015, p.68, emphasis in original).

    Beckwith remarks in a footnote to the above passage: “It must be noted that Indologists in particular are wont to discount Greek sources on early India, referring to all the above-noted problems. However, the same problems, and much worse ones, affect Indian sources, which are mostly a millennium or more younger, have never been properly edited, and consist largely of fantasy” (Greek Buddha, n23).

    I would like to hear from Indologist’s about these claims. I am less interested at present in hearing about the perceived value of Megasthenes as a source, or about the relative value of the Indian sources to the Greek ones. What I am eager to hear what readers think of Beckwith’s low estimation of the value of the Indian sources themselves, which seems to be based on three issues: (1) the remoteness and lateness of the Indian sources; (2) the lack of properly edited critical editions of the texts; and (3) the later contamination and corruption of the ideas expressed in the texts.

    Since it seems like the most tangible issue, perhaps we can begin by focusing on the claim about the lack of properly edited critical editions. Is it true that the Indian sources on early Buddhism have never been properly edited?

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>