Steven Collin’s article On the Very Idea of the Pali Canon is an influential monograph that recontextualizes the notion of the “Pali Canon”, aiming to show it in a “properly focused and realistic historical perspective”. This response consists of some brief notes to assess whether it achieves its aim.
If there is one thing we have learned in the past year, it is this: beware “realists”. Too often, the idea of “realism” is a mask for cynicism and reductionism. Thus Collins proposes that the Pali Canon, rather than being passively maintained by the Theravada school, “should be seen as a product of that school, as part of the strategy of legitimation by the monks of the Mahavihara lineage.”
In the same way, Collins’ article should be seen as a strategy of legitimation by Collins himself, as he jockeys for kudos and prestige in the cutthroat academic world.
Collins claims that “we have no evidence of any kind which can be securely dated before Asoka”. To make this kind of claim, we have to implicitly define what kinds of evidence can be accepted as “secure”. Collins himself merely assumes that we all know this, and that we share his assumption that “secure” means “material”, notwithstanding the fact that archeological evidence is absolutely as subject to exactly the same kinds of uncertainties as any other form of information.
He says we must “rely on inferences”, as if this were a problem, and not merely a bland description of literally all forms of knowledge in every field of knowledge that has ever or will ever exist.
I personally have no “secure” knowledge of Steven Collins’ existence, and must rely on the admittedly partial and uncertain evidence of certain articles and writings which, among the unreliable morass of information on the internet, have been attributed (by who?) to him. Given that there have been many notable academic frauds, given that there is clearly an institutional need for radical theses to drive popularity and enrollments, and given that many of the so-called “memories” of his life have the unmistakable tinge of hagiography, it is clear we must be “more fully aware” regarding the figure whose personhood “we denote by the short-hand term”, “Steven Collins”.
“Collins” points out that the word tipiṭaka is attested from around the 1st century CE, in the Milindapanha and some inscriptions. He says, “I think it should be taken to denote not three closed lists of documents, but rather three different genres within the tradition”.
“Collins” presents precisely zero evidence to support his contention, only what he thinks. It is in any case a distinction without a difference. Of course they are three genres; they are also (fairly) fixed lists of texts. That the word is only mentioned in the very latest of such texts merely shows that it was formalized by that time.
Obviously the “Tipitaka” did not exist in the Buddha’s day. It began with the nine angas, which were reorganized into the nikayas, which were then incorporated into the Tipitaka we have today. The entire Abhidhamma was composed after the time of the Buddha, as was much of the Vinaya, as well as portions of the Suttas. It is an intellectual failure to assert that everything in the Tipitaka was taught by the Buddha. It is equally an intellectual failure to assert that the lack of 100% clarity and precision means that we know nothing. There is a genuine historical question to be asked as to the details of this process, but “Collins” is not really interested in clarifying it.
“Collins” goes on to try to undermine the authority of the historical sense of the phrase buddhavacana. This is used in early texts in the sense, “whatever is spoken by the Buddha is well-spoken”, but later texts, chiefly Mahayana, turned it to become “whatever is well-spoken is spoken by the Buddha”, thus allowing more flexibility in what is considered to be buddhavacana.
He points out that there is a Pali text spoken by the monk Uttara where “grammatically there would be nothing wrong with interpreting his remark in the Mahayanist sense”, i.e. that “whatever is well-spoken is spoken by the Buddha.” However, he acknowledges that in context, “Uttara is saying that whatever he teaches comes from the Buddha”. In other words, contextually it clearly does not mean that Uttara feels he can just make stuff up and say it came from the Buddha, but if we remove it from context and assign it an arbitrarily possible grammatical construal, we can imagine that it might have said that. So says the realist.
I have written in more detail on this text here. “Collins” nowhere discusses the highly unusual and late character of this text.
“Collins” asks, as if he has just posed a great mystery, “Why then did what has become Theravada ‘orthodoxy’ choose to emphasize an historicist and exclusivist idea of its ‘Canon’, ‘the Buddha’s Word(s)’?” Why indeed? But any answers he provides, such as they are, are either baseless or obvious. Of course the formation and maintenance of the Tipitaka has a political dimension and is a part of the legitimation of religious and political authority. Most of the Tipitaka editions available today are either royal editions or are government-sponsored.
But a tradition consists of more than just the mindless instantiation of an orthodoxy. The First Council tells us of the 500 arahants, but also of Purāṇa who remembered things his own way. The Second Council account lays bare the political maneuvering among the Sangha to resolve differences. The Third Council resulted in the Kathavatthu, with literally hundreds of discussions of points of difference. The Dipavamsa points to competing notions of canonicity as the reason for the first schism. The commentaries regularly feature competing interpretations. The later Theravada tradition—especially post-12th century—is full of contested interpretations of Abhidhamma.
A tradition is both static and dynamic, it looks both forward and back, and it moves to the center even as it pulls to the periphery. The idea that there are contesting notions of what the teaching really is is no radical thesis, it is central to the stories that the tradition tell about itself. It has been there all along.
“Collins” goes on to “assert”—without supporting “argumentation”, but with an “abundance” of annoying “scare quotes”—that “we simply have no idea which texts in fact pre-date Asoka”. On the contrary, we have a fairly clear idea of this. But to understand this requires knowing the difference between “not 100% certain” and “having no idea”.
As usual with skeptical “realists” of “Collins’” sort, they feel no need to actually get in the trenches and make an argument that engages with the evidence. They simply assert a set of epistemological requirements that they already know cannot be met, without justifying why they must be met, then build sweeping theories as if they have just proven something.
“Collins” quotes Rahula and Malalasekera as saying we cannot know the exact relation between the Pali Canon as it was first written down and today. There are two fundamental problems here.
Firstly, he takes the most skeptical comments by these scholars and ignores their broader contribution. Clearly neither of them believed that the Tipitaka was just made up and has no relation to what the Buddha taught. “Collins” is not interested in understanding what these scholars believed and how they negotiated the encounter between tradition and modernity, only to use them as points to score in an argument. He characterizes them as “staunchly orthodox modern Theravadins”, a reductive and othering description that elides the dynamic, reformist nature of the work of these two scholars.
The second problem is, he cites from articles in 1928 and 1956, but it never occurs to him to ask, “If at that time there was a lacuna in understanding, has that now changed? Or if not, is there any way I could contribute to clarifying this knowledge.” This kind of postmodern nihilism is only interested in undermining the work of those who actually get things done.
“Collins” goes on to point out that the actual texts in practical use in a tradition are not the same thing as what is enshrined as a “Canon”. He asks whether “these and only these texts function as ‘scripture’, with no others having canonical authority?” He rather boldly answers “No” to his own question, a boldness that is untempered by the complete lack of any evidence.
The only reference he cites towards the idea that there has ever been a confusion in the tradition about what is a “Canon” is a passage in the Culavamsa, where Buddhaghosa’s work is acclaimed as being “like the Pali" (pāliṁ viya). But the word “like” can have been used for no other reason than to emphasize that, while they are not actually the thing, they are in some respects comparable to the thing. Something that is like the Canon cannot be the Canon.
Continuing in his “just asking questions” style, he speculates, “It might well be that the content of most smaller monastic libraries is in effect a ‘ritual canon’”. And further, “A monastic library … might perhaps be compared to a modern academic library”. With this, who can but concur? It certainly might.
On the other hand, if he had ever visited an actual monastery, perhaps the distinction between the “canon”, which is typically contained in a separate, locked cabinet in a sacred position next to the altar, and the other books, including various chanting books and the like, which are on ordinary shelves, might have helped clear things up for him.
Whenever I read skeptical scholars of that era, whether “Collins” or the equally dubious figure “Schopen”—whose debunked work “Collins” cites approvingly—I am struck by how they prefigure the internet troll. Seeing a flicker of uncertainty as proof of falsity; treating authority or expertise as inherently corrupted; reaching for ad hominem dismissal as a first line of argumentation; applying skeptical standards to others but not themselves; “just asking questions”; an overweening sense of confidence in the compelling rationality of their own thought; and most of all, an ideological commitment to undermining any form of received knowledge, while avoiding any responsibility for creating and building new knowledge. I can’t help feeling that there is more than a coincidence here; that they represent a manner of teaching in US colleges that birthed a generation of graduates who mistook this sort of thing for a genuine intellectual pursuit.
“Collins” ends up by saying:
we need empirical research into each individual case, not a simple deduction from the existence of the closed tipiṭaka produced by the Mahāvihāra. We need more research, for example, historical and enthographic, on the possession and use of texts …
I have not been able to uncover any evidence that Collins himself ever made any effort to do this “needed” empirical research.
If we cannot be certain of such a basic fact, how much more do we not know of his existence?
