European scholars did not at first realise that the many religious sects of East and Southeast Asia were closely related. So different were the extant forms of Buddhism from each other that they seemed like totally different religions. However, it soon became apparent that they all claimed to have been founded by a figure called "Buddha" (A story now told several times: Almond 1988, Bluck 2006, Franklin 2008).
It was apparent that all the different forms of Buddhism must have some common history. In the early decades of the 1800s, scholars assumed that the history of Buddhism would follow the same course that all (true) religions were believed to follow. They believed that the history of European Christianity was an ideal to which the history of other religions would conform. European imperialists saw their own nations, as well as their own self-views and values, as reflecting universal ideals. And they set about fitting Buddhism into this mould. There was, at that time, no critique of his monolithic, Eurocentric view. European imperialism was at its peak. From initial speculations about primitive Buddhism, eventually emerged the idea that we now call "early Buddhism" (I drop the quotation marks from this point, but in my mind they are always applicable to this phrase). The idea of early Buddhism is that we can and have rediscovered the origins of Buddhism within the Pāli suttas.
Since taking those early faltering steps, recovering primitive Buddhism or early Buddhism has become a Holy Grail for Buddhist Studies. Buddhist modernists even seem to believe in the possibility of re-enacting early Buddhism.
And yet we can say with some confidence that, before the 19th century, no Buddhist in the world was concerned with early Buddhism. Until their encounters with European imperialists, Buddhists around the world simply did their thing -- in the many and various ways that Buddhists do. As far as Buddhists were concerned, they were already following the ancient ways of their ancestors, which could be traced back to the Buddha. Even now, most Buddhists take this view of their own traditions. The mythology of the Buddha's life was always part of this tradition, but the story was steeped in magic and superstition. The Buddha was known, for example, to have been born from his mother's side (rather than her vagina), and immediately after this he took seven steps and spoke some lines. And this is the story that Buddhists told until very recently.
This is not to say that Buddhists did not adopt reform movements from time to time. But these reform movements were seldom if ever based on literal readings of texts. Reform mostly emerged from personal experiences of awakening and this usually led to entirely novel formulations of Buddhist doctrines.
Assuming, for the sake of argument, that the Pāli suttas reflect real life to some extent, the simple fact is that over time, gradually perhaps but inexorably, all Buddhists abandoned those teachings we find in Pāli suttas. Significantly, these are the very same teachings that supposedly constitute early Buddhism. Buddhists themselves, all Buddhists everywhere, abandoned the lifestyles, attitudes, and practices that we find described in the suttas. All Buddhists everywhere chose to believe other, newer, ideas; chose to perform new types of rituals; to practice a new (communal) form of monasticism. Buddhists stopped memorising and copying those texts and they memorised and copied other, newer, texts. They stopped doing this kind of practice and they started doing that kind. And so on, right across the board. Later Buddhists abandoned early Buddhism. Presumably for what seemed like good reasons at the time.
What this means is that the various versions of early Buddhism identified by scholars, and more recently by religieux, have almost no relationship with any modern form of Buddhism. For example, Sujato (who now describes himself as "not Theravādin" but somehow still a "bhikkhu"), has self-published a (relatively conservative) list of differences between early Buddhism and Theravāda to emphasise the discontinuity.
Buddhists did not just adopt new teachings across the board, they also repeatedly disagreed about which of the new teachings were better and splintered into competing sects that began to change things in different ways. This process was already well underway by the time Buddhist beliefs began to be documented (and thus became a historical phenomenon for the first time).
Early Buddhism is an answer to a question that no Buddhist in the world before about 1840 would have thought to ask.
Jonathan S. Walters (1998) has contested terms like "Buddhism" or "early Buddhism". The problem is that such terms hypostatise and take agency away from what Buddhists think and do. This causes us to think of "Buddhism" as a transcendental essence. We start to see phrases like "Buddhism says..." or worse "Buddhism teaches...", giving the impression that Buddhism is monolithic and has its own agency. In the final analysis, "Buddhism" is an abstraction. Abstractions don't have agency; agency is a product of sentience. For the purposes of this essay, agency is human. When we focus on, for example, what actual Buddhists teach (as opposed to what Buddhism teaches), we see that they teach a huge range of ideas, attitudes, and practices that don't form a single, coherent whole. In many cases what one group of Buddhists does and says qua Buddhists may seem totally unrelated to what another group does. Different Buddhist groups are not necessarily comprehensible to each other (though this is more and more downplayed as we head into the 21st century).
That said, Walters notes that Buddhists tend to present a united front when interacting with non-Buddhists, even while they argue vehemently when dealing with other Buddhists who accept a different orthodoxy. Scholars have played along, presenting Buddhist history and doctrine in terms of "What Buddhism says..."
As noted, once some more coherent views on early Buddhism began to emerge from the mining of the Pāli Canon, it became clear that Buddhists in Pāli suttas behaved in ways that were unlike anything being practised by living Buddhists (as already noted, Buddhists moved on from early Buddhism and adopted late Buddhism). When early Buddhism emerged as an idea, no Buddhists anywhere still practised anything like what we now think of as early Buddhism. Notably, this included all of the various Theravādin sects (some of whom don't even recognise each other's ordination lineages).
Joseph Walser (2022) has emphasised this kind of disconnect. An example of a monolithic "Buddhism teaches X" scenario, found in virtually every scholarly account of Buddhism, is the idea that Buddhism teaches the anātman doctrine (for brevity's sake I will set aside the ongoing dissensus on what this doctrine actually means). Walser notes that if you go to traditionally Buddhist countries and ask about their beliefs, almost no one has heard of the anātman doctrine that features so prominently in textbook accounts of Buddhist belief. And, on the contrary, almost everyone in traditional Buddhist countries accepts some kind of ātmavāda.
Such inconvenient facts are usually absent from textbooks and encyclopaedias on Buddhism because they don't fit the idealised account of hypostatised "Buddhism". The hypostatized account of early Buddhism, which is widely promoted in academic studies of Buddhism, is based on reading certain suttas literally while excluding those suttas that resist a literalist reading. What can be read literally, is taken as a factual, and can be used without any further caveats to reconstruct early Buddhism, while what cannot be read literally, is not read.
If Buddhists were not asking these questions, then who was? Jonathan Walters (1998) has argued that it was Protestant European imperialists. That is to say, people who believed that Europe (not to say England) represented humanity in the ideal, and who sought to impose European values wherever they went, and—without any irony at all—at the same time, expropriating indigenous peoples' lands and murdering those who resisted. Moreover, they saw Protestant Christianity as the ideal religion. So when they met Buddhism in, say, Sri Lanka, British scholars interpreted the history of Buddhism on the model of European Christianity (and ultimately on the history of Rome). It was this parallel that motivated them to try to identify and valorise the origins of Buddhism as portrayed in Buddhist normative texts and hagiographies.
Lacking modern historical training, the Victorians simply ignored questions about the provenance and dating of the Pāli texts. Rather, they proceeded as though these religious documents were exactly what they purported to be: the very words of the Buddha.
This innocence regarding historical theories and methods of inquiry has remained a feature of Buddhist Studies down to the present, much to the annoyance of those with historical training. I also picked up these bad habits because I taught myself Buddhist Studies by reading publications. It wasn't until I decided to educate myself in some basic approaches to historical inquiry that I realised the problems with the standard accounts of what Buddhism says. Now my thinking is undergoing a paradigm shift.
Outside of the academe, we regularly find that Buddhists have a religious conception of Buddhist history. Ironically, the Buddhism that many modern Buddhists believe in and assert is an idealised account of ancient Buddhism that emerged from the European imperialist project in the 19th century. Many Buddhists believe that early Buddhism was a more authentic form of the Buddhist religion and, importantly, that it can be again. Some even believe that we should try to re-enact early Buddhism. Indeed, my impression is that some Theravādins believe that they are re-enacting early Buddhism.
The Buddha's historicity—that is, the question of whether the Buddha was a historical figure—is particularly vexed because common sense and folklore are not aligned with the academic field History. Those with no education in History routinely take statements like "the Buddha is not a historical figure" or "the Buddha is a figure of Buddhist mythology", to mean that "the Buddha didn't exist". To be historical, in this view, is to be real. And to be mythological is to be unreal.
The resulting confusion is projected so that people confused in this way see the confusion as a fault of historians writing in counter-intuitive ways. In trying to discuss these issues in a Buddhist forum, for example, promoting the historian's approach attracted waves of hostile criticism. Users denounced the view that the Buddha was not historical based on an uneducated layperson's understanding of "historical". Some also vehemently denounced what they saw as the whole corrupt enterprise of academia which produced this view.
To try to bridge the gap between historians and Buddhists, I want to go over in brief how historians approach history. What follows is an outline of some basic historical methods based on about a dozen modern accounts of what History is and how historians approach learning.
—History—
The theory and methods of History are hotly contested, especially following the emergence of the postmodernist critique in the 1960s and the publication of Edward Said's (1978) critique of "Orientalism". Some lay people are aware of these criticisms and like to use them as a club to beat academics with. Lay people often forget that the post-modernists and Said were themselves academics. Said was professor of literature at Columbia University. These criticisms came from within academia.
It's not the case, for example, that uneducated lay people are driving some intellectual movement away from the benighted past towards a more enlightened view. With respect to academia, even well-read laypeople are spectators at best. So lay people writing vicious critiques of academia that no academic ever reads make no difference whatever. As I have discovered, even publishing in academic journals is no guarantee that academics will take one seriously.
Said's target was mostly 18th and 19th-century historians. And, by the way, Said had no problem with the term "oriental" (he uses it throughout the book), he just wanted European historians to address their evident biases. This they did and continue to do. One sometimes sees the suggestion that Orientalism is still prevalent in academic writing, but the story of Orientalism in academia did not stop in 1978. Edward Said was taken very seriously by European academics, as were the critiques emerging from mid-20th century France which undermined many aspects of historical methods at the time. Historians wrestled with these new ideas and incorporated what was useful into their approach.
The irony here is that "early Buddhism" is itself an Orientalist construct. It's part of the attempt to describe ancient Indian religions by shoehorning them into idealised European moulds. It relies on assumptions introduced by those early 19th century Indologists, such as the idea that Pāli suttas can be mined for history by ignoring everything that cannot be read literally. If the Buddha meets a king, it can be read literally and taken as historical, despite the absence of any corroborating historical facts related to that king; whereas, if the Buddha meets a god such as the Vedic creator, Brahmā, that cannot be read literally and can be considered mythological. Those involved in early Buddhism re-enactment societies tend to see "mythological" as synonymous with "irrelevant".
This historical/mythical distinction is not a distinction that the authors of the Pāli suttas made or would even recognise. This distinction belongs to the European Enlightenment, not the Buddha's enlightenment. For the authors of the Pāli canon, all these stories are on the same level. Traditionally, all suttas are buddhavācana. Suttas are not traditionally divided into "real" and "imaginary".
There is some use of allegory in Pāli. There are extended similes. Some suttas are obviously not meant to be taken literally. But they are mixed in with all the other texts and they are not marked in any way. When the topical collection of the Saṃyutta Nikāya was composed, sometimes more obviously mythological suttas were clumped together (such as the Devatā Saṃyutta SN 1). Still, at other times they were simply mixed in with the rest. The process of curating the canon of Buddhist writings never seems to have made a distinction between texts we can take literally and those we cannot. Nor, to the best of my knowledge is this distinction found in modern-day traditional Buddhist countries. This is a modernist, European distinction.
Despite ongoing debates about historical theories and methods, there are some ideas and practices that are commonly taught and applied by all historians (though not by all people who have written histories of Buddhism).
Historians study documented events. And thus ancient history effectively begins when writing begins. In India, evidence of the use of writing slightly precedes the appearance of coherent documents (Asoka's edicts). See Strauch (2024) for a recent overview of this issue. But the first documents in writing in India are precisely the Asoka edicts. These do mention the Buddha, but they provide only very scant information and it is mixed in with a lot of propaganda.
Archaeology allows us to see what people did in antiquity but we get little or no information about why they did that (and not something else). Lay people seem to think of History as simply telling the story of the past. Historians, by contrast, see History as an attempt to explain the past. Historians prefer causal explanations that tell us why people did what they did. Such explanations require that we have access to the thought processes involved. It is not until people start writing things down that we begin to get clues about why events unfolded the way they did.
In the sense that historians seek causal explanations, History is similar to science. However, in scientific discourse, causal explanations generalise, and can often be expressed in mathematical form, such as Newton's laws of motion, or the various gas laws. Scientific laws enable us to make statements like: F = ma. What this means is that if a force of 1 N is applied to a mass of 1 kg, that mass will accelerate at 1 ms-2 in the same direction as the force operates. Such equations can be used to predict the future.
Causal explanations in history don't generalise in this way, since they concern the motivations of people, and people are vastly more complex that a simple 1 kg mass. We can explain history only retrospectively, and even then such explanations don't easily extrapolate for the deeper past or the future.
So a historical figure (a figure who possesses historicity) is someone who is described in writing by someone who met them and wrote about it at the time. Such first-hand accounts are referred to as "primary sources". Without primary sources, there is no history.
Popular or general accounts of history may simply narrate the past and rely on secondary sources; that is to say documents written later and/or by people who did not personally witness the events. A good example of this approach is Sally Wriggins' (2004) popular book of Xuanzang's journey to China. Wriggins summarises historical research, sometimes making use of relatively obscure secondary sources, but she does not contribute anything new to our understanding of Xuanzang and does not use any primary sources. This is not the same as historical research which is how we produce new explanations of the past. Wriggins is merely describing a journey, and provides us with no causal explanations for the events she narrates. Wriggins is referred to as "a writer and and lecturer", rather than "a historian". And this is fine. Her book is well-written, very entertaining, and conveys such facts as were considered "known" at the time. But if you want to read critically about Xuanzang, then you need to turn to, for example, the oeuvre of Max Deeg and especially his forthcoming multi-volume commentary on Xuanzang's Records of the West (Dà Táng Xīyùjì 大唐西域記; T 2087).
Historians take documents as reflecting views at the time of writing, since, even when someone is ostensibly writing about the past, they do so from their own point of view (note that this orientation is at least partly a result of the postmodern critique of History). This is especially true in the ancient Indian world when historical awareness was often lacking. On the other hand, in China, for example, writing and historical awareness both arose much earlier than in India. So I can say with confidence that the oldest Heart Sutra artefact was commissioned on 16 March 661 CE, by Yang Shesheng 楊社生, a minor aristocrat serving in the military with the rank of Guoyi duwei 果毅都尉 (Courageous Commander). Because such things were recorded for posterity in China.
Another basic point is that historians require a fact to be corroborated. A single source does not suffice to write a history. Ideally, in writing a history we have access to multiple primary sources written from different points of view. That said, ancient written documents inevitably reflect the concerns of a literate elite. When Asoka wrote his first edict, there cannot be been more than a handful of literate people within 5000 miles of his capital, Pataliputta. So, while the edicts reflect his views, they are not corroborated by other documentary sources and they are not a general account of the history of India. Rather, they are primary sources that have to be interpreted with care.
In explaining these simple facts to Buddhists, I have often seen the complaint that this "places far too many restrictions on us". Historians are concerned that their work has integrity. The limitations help to ensure this. That such limitations are routinely ignored by Buddhist Studies scholars is an indictment of the intellectual integrity of the field.
I want to now return to the topic of Pāli and the Buddha.
—Buddha and Pāli—
No documents were written at the time of the Buddha for the simple reason that writing was not in use at that time, in that place. Writing came into use slightly before Asoka's edicts in the mid-3rd century BCE, but the earliest documents with written texts are the edicts themselves. And the Buddha is supposed to have died ca 400 BCE, if not substantially earlier. For this reason, the Buddha is not a historical figure. This is not at all controversial. Amongst historians, there is no dispute over this. Rather the disputation comes from academic philologers and Buddhist apologists (with considerable overlap) who don't understand the theories and methods of historians.
Presenting this material to Buddhists—especially to (Pāli-worshipping) Theravādins—inevitably raises a storm of protests. Though they don't come right out and say it, the main message seems to be: "How dare you say the Buddha didn't exist?"
But existence and being a historical figure are distinct issues. No one, I think, is denying that the Buddha existed. Given the way that religions work, it is certainly possible that Buddhism had a founder. Indeed, to most historians, a founding figure seems both plausible and likely. Of course, we cannot be certain because there is no documentary evidence of the Buddha during the period in question and no corroboration from other sources. All that we know was part of a religious tradition that was passed down orally for many centuries before being written down, and that by the mid-3rd century, Buddhists attributed their religion to Buddha.
Nor is there any archaeological support for the Buddha. While archaeology confirms that cities such as Sāvatthī and Rājagaha did exist, and when they began to exist, there is no physical sign of Buddhism before Asoka, let alone a person called Buddha.
Much of the dissent is focused on the Pāli suttas. On one side we have a group of positivist-historicists who claim to be able to prove that the Pāli suttas are "authentic" records of the Buddha's teaching from that time. Some on this side are happy to reassert the old Theravāda conceit that the Buddha spoke Pāli (whereas I have argued that we will never know what language the Buddha spoke).
On the other side are historians who point out that the oldest extant Pāli documents are from the 5th century CE. The earliest corroborating evidence for a canon of Pāli texts is also from the 5th century CE. There are much older manuscripts in Gāndhārī and much older texts in China, there is no Indian canon in either place. The Chinese had to curate a canon of their own. So did the Tibetans.
Buddhist myth tells us that the Pāli canon was written down in the first century BCE, but again, this is from sources composed in Sri Lanka in the fifth century CE. As documents, the Pāli suttas are primary sources for 5th-century Sri Lankan Buddhism.
The corollary of this is that, despite the elaborate special pleading, the Pāli texts can tell us little or nothing about "early Buddhism", using the methods that have been adopted. From the historian's point of view, we can date the written Pāli texts to the 5th century CE. And as such the Pāli texts are primary texts for Sri Lankan beliefs in the 5th century, and they are not primary sources for Indian beliefs in the 5th century BCE.
This is not to say that nothing happened in India in the 5th century, or that Indian beliefs were not encoded in this way. Sri Lankan Buddhist converts obviously adopted and developed a form of the Indian religion. But Sri Lankan Buddhism is far from simple or straightforward. Antagonism and competition between the great monasteries is a defining feature of Sri Lankan Buddhism. And in the end the most conservative monastery won out and got to define Sri Lankan Buddhism down to the present. In practice, this involved purging Mahāyāna and Tantric forms of Buddhism which were popular at the time, as well as purging sects of Theravāda that did not conform to the Mahāvihāra ideal. The Pāli texts are not simply Sri Lankan in origin, they are the records of the Mahāvihāra.
Indian Buddhists were already diversifying by this point, but Pāli obscures this on the whole. The Kathavatthu is an exception and shows how much the Theravādins were at odds with other Buddhists about the correct interpretation of the Buddha's doctrines; the disagreements were fundamental and vehement. Buddhists were not unified even at this early stage of Buddhist history (and likely this reflects trends from Buddhist pre-history, though here we enter the realm of speculation).
Historians are bound to respect the epistemic horizon represented by the advent of writing and to remain silent about the history of places and time for which there are no primary sources (such as pre-Asoka India). While historians may indulge in speculation, as we all like to do, historians are bound to make a distinction between what happened, which is unknown, and their speculations about what might have happened.
—Summary—
Buddhists emerge onto the stage of history as members of several related and competing religious movements and alongside some other religious movements with similar methods and concerns. We get a sense of these new religious movements being partly a response to the emergence of relatively belligerent city-states along the middle Ganges Valley, and partly to the influx of Brahmin migrants from the west. Why did Brahmins migrate into lands that they had previously described, e.g. in the Śatapatha Brahmaṇa, in disparaging terms? No one seems to have an explanation for this.
In addition to the two groups of Indic-speaking peoples in the region, North Indian people spoke a range of languages from at least two other families, Dravidian and Munda, as well as many "tribal" languages, which in modern times include a number of language isolates. And Buddhism clearly shows influence from a range of cultures. Yakkhas, for example, are definitely not Indo-European.
As Walters (1998: 247) has said:
Buddhologists, anthropologists, and historians of religion have raised serious doubts about the naïve use of the suttas as sources for reconstructing Theravāda Buddhist history.
The whole idea of early Buddhism emerged from Orientalist, imperialist European scholarship. That it has been taken up by Buddhist Modernists should not distract us from the orientalist nature of the project.
Buddhist Modernists have taken the re-enactment of early Buddhism as the ultimate solution to Buddhists' long-standing over the authenticity of their teachings. In this view, ultimate authority is vested in the person of the Buddha, and in the "records" of his sermons. A living Buddhist authority is thus an expert in Pāli (with a sideline in Chinese and/or Gāndhārī). One can see this being played out in, for example, the remaining Buddhist online forums.
It is interesting that this move to co-opt the scholarly conceit of early Buddhism is paralleled by the widespread emergence of openly enlightened teachers. When I started blogging it was still widely considered a major faux pas to "claim" that one had any kind of attainment (including even first dhyāna). There was a lot of discussion of "claims". Fast forward to the present and many Buddhists now openly talk about enlightenment from their personal experiences. Such Buddhists may or may not refer to Pāli texts, but the point is that they don't need to refer to any text because they know from experience. A sentiment, incidentally, that we find peppered throughout the Pāli suttas but which is mostly downplayed by Modernist Buddhists.
The focus on Early Buddhism has led to a hypostatisation of the idea. This gives agency to an idea and removes it from Buddhists themselves.
There is no "early Buddhism" apart from what actual Buddhists did and said in that period. Knowledge of this type requires writing. And there was no writing at that time. And of this period we can know little and do know less.
The reconstruction of early Buddhism relies on privileging Pāli suttas that can be read literally and rejecting those that cannot be read literally. A sutta that can be read literally is assumed to reflect history. However, the authors of the suttas did not think this way.
The Buddhism that is reconstructed by this method does not resemble any extant Buddhist tradition. This is partly because Buddhists abandoned early Buddhism in favour of newer doctrines. This is no surprise to anthropologists, but it rather undermines the claim to authenticity that accompanies this reconstruction.
Discussing this with Buddhists is more or less impossible because they are not aware of how historical enquiry works nor what words like "historical" mean in this context. This is reinforced by Buddhist apologists and allies in academia who are also (apparently) ignorant of the methods of the field of History.
There are numerous caveats on using Pāli suttas for historical research. The oldest extant Pāli documents are from the 5th century CE. The idea that they were written down earlier is based on 5th-century Sri Lanka texts. There is no corroboration for a canon of Pāli texts until the 5th century. The impression is that no such canon was known in Gandhāra (where the surviving texts are far more eclectic, and largely standalone or part of smaller collections). The Chinese and Tibetans had to invent their own canons, which they did in very different ways.
There is no archaeological support for the existence of Buddhism until Asoka. And the evidence we get from Asoka is thin at best. Substantial evidence for Buddhism emerges a little later in the form of elaborate stupas with narrative friezes illustrating the Buddha's life story (including Jataka stories). No figure from the Pāli canon has ever been linked to a historical fact or event. Not even the kings are historical figures.
The early Buddhist narrative is largely focussed on valorising Pāli texts, and other texts only when they reinforce the monolithic, hypostatised view of early Buddhism.
Early Buddhism (as currently presented) is a fantasy that is only plausible in ignorance of the theories and methods of history. Having belatedly studied these methods, I find my attitudes have substantially changed. In fact, we know nothing and can know nothing about early Buddhism because there are no primary sources relating to those times and places. Adopting the practice of treating suttas that can be read literally as though they are factual documents, is not consistent with the best practice adopted by historians.
Early Buddhism probably never existed in the form that it is presented, and even if it did exist, we'd have no way of knowing how accurate the portrayal is. What the modern accounts of early Buddhism seem intended to do is to salve the constant anxieties that Buddhists feel about the authenticity and authority of the copies of copies of copies of 5th century Buddhist documents that they have inherited. When one has no personal experience of awakening to fall back on, religious literature takes on a much greater significance as an arbiter of orthodoxy.
—Bibliography—
Almond, Philip C. (1988). The British Discovery of Buddhism. Cambridge University Press.
Bluck, Robert. (2006). British Buddhism: Teachings, Practice, and Development. Routledge.
Carter, John Ross. (1977). "A History of Early Buddhism". Religious Studies 13(3): 263-287. https://www.jstor.org/stable/20005420
Franklin, Jeffrey J. (2008). The Lotus and the Lion: Buddhism and the British Empire. Cornell University Press
Lillie, Arthur. (1881) Buddha and Early Buddhism. London: Trübner & Co.
Strauch, Ingo. (2024). "Aśoka and the Use of Writing in Ancient India". In The Ancient World Revisited: Material Dimensions of Written Artefacts. Edited by Marilina Betrò, Michael Friedrich, and Cécile Michel. De Gruyter.
Walser, Joseph. (2022). "Buddhism without Buddhists? Academia & Learning to See Buddhism Like a State". Pacific World. Series 4, Vol. 4: 103-170. https://www.shin-ibs.edu/documents/pwj4/3/4-3-4-Walser.pdf
Walters, Jonathan S. (1998). Finding Buddhists in Global History. American Historical Association.
Wriggins, Salley Hovey. (2004). The Silk Road Journey with Xuanzang. (Rev Ed.) Icon Editions, Westview Press.
