> There’s yet another problem: the sheer volume of texts. When it comes to Indian and Buddhist traditions, for example, the number of ancient manuscripts that have survived but are yet to be studied has been estimated at around 10m, though Friedrich says he has seen estimates as high as 30m. There simply aren’t enough scholars with the right expertise, including the necessary language skills, to do the work.
I was worried a lot about this during my PhD. Well-cited papers were just the "tip of the iceberg". I did some fairly thorough searches and came across quite a few papers in my field that were important in my view and unfortunately overlooked.
The problem doesn't seem intractable to me. More researchers should spend some time searching the literature where they think no one else did. If a certain percentage of people did this, then I think it would benefit everyone including those who don't do particularly deep searches.
First off i don't understand how this 10-30k number is physically or culturally possible. ( i admit my tired brain thought million first - confusing british/roman systems! )
Also having read sporadically through various dzogchen texts on "dream yoga" which is an older mapping and superset of what we today call lucid dreaming i'm positive that there could be an incredible amount of "useful knowledge" of various "weirder" aspects and concepts of consciousness and dreams waiting for us in some of those texts.
Note that it's the number of manuscripts, not the number of distinct works. Many of the manuscripts would be copies of the same popular works (people preserve whatever's most useful), but there is a long tail of works each preserved in very few manuscripts.
The 30-million estimate has been arrived at independently by several people AFAIK, but one of them is David Pingree, and look at e.g. https://www.google.com/search?q=pingree+estimate+million+man... for several sources. In particular, this paper by M. D. Srinivas [1] gives several independent estimates of the number of manuscripts, all in millions. There have been 4.2 million manuscripts actually documented already (as of 2019 when the paper was written), and this paper's estimate is that "10 million is indeed a reasonable estimate for the number of Indian manuscripts that are extant […] the number of manuscripts in Sanskrit can be expected to be of the order of 8 million."
And although this post is about literature, that paper is about Indian astronomy&mathematics, which is perhaps representative:
> We find that of the estimated 9,000 source-works of Indian Astronomy and Mathematics (which are preserved in around 30,000 manuscripts), only about 150 texts were edited, and just 30 texts translated during 1800-1947. During 1948-2019, there has been significant progress and another about 300 texts have been edited and 66 texts translated, many of them with detailed explanatory notes. Thus, only about 450 (or 5% of the estimated 9000 source-texts available) have been edited and published so far; even among the published works, only 96 texts have been seriously studied via translations and explanations with a view to bring out their technical (mathematical-astronomical) content.
Do you have any sense of how long these works are?
It would be challenging for a single human to read 10,000 books in a lifetime, say, of 85 years, beginning at age 5. That would be 125 books per year, or nearly 2.5 per week. Every week, every year, for life.
If these were shorter works --- the equivalent of pamphlets or essays (or more likely: epic poems), of a few pages each, that might be more viable.
That said, 10 million works might represent a cultural accumulation, but would be difficult to describe as a cultural tradition, simply on the basis that at any one time, any person could know at best a small fraction of the works.
The linked paper quotes "2.96 lakhs manuscripts (2.61 Crore Pages)" which works out to an average of about 88 pages per manuscript. I imagine one "page" (folio) of a (usually) palm-leaf manuscript is probably about half a printed page's worth of material, so maybe each manuscript on average is the equivalent of a 50-page printed book.
(Again, note that 10 or 30 million manuscripts is probably fewer works: the quoted 30000 manuscripts <-> 9000 works relationship for astronomy/mathematics may not hold across the entire manuscript corpus. Also, for comparison, both the Library of Congress and the British Library seem to have on the order of tens of millions of books, and Harvard's massive Widener Library has 3.5 million books.)
I'm not sure I understand the remark about "a cultural tradition": surely it's just as much the case for "Western culture", or even say "French culture", that a given person could have read only a small fraction of the works. It does not mean there isn't a French tradition or whatever. The intended meaning is just that from the (pre-modern) Indian "world" or "sphere" or whatever, there are those many manuscripts, many of them uncatalogued, let alone studied/digitized/edited/published/translated.
By "cultural tradition", I'm talking of a literature which would be generally familiar to a population, or perhaps to its literate class.
Books and stories contribute not only knowledge or entertainment, but a shared knowledge and common metaphorical or conceptual language. And these need not be written traditions. I could reference, say, Psy's "Gangnam Style", Harry Potter, the Marvel Cinematic Universe, the Black/Blue / Yellow/White dress, Michelangelo's David, or Pikachu's face, and ... at least a large portion of readers would share that understanding.
If I were to reference instead, say, Sutro's "Affected", Hamilton Holt's "Commercialism and Journalism", Dersu Uzala, my uni flatmate's bronze torso, or the Vuck I'd spotted on a roadtrip some years back, there'd be a far smaller awareness of each.
There's a difference between information that's preserved, and information that remains culturally significant. Moreover, if the knowledge is considered to have both an explicit (book-transmissible) and tacit (experiential) component (think the difference between your chem book & lecture, and the lab section), what happens when the experiential knowledge dies?
What's the minimum requirement for knowledge to be considered live, in the sense that there's a community of practice which can sustain itself generationally?
I'm fairly familiar with counts of books and other types of records:
- US Library of Congress: ~40 million books (includes 15.5 m unclassified), 131 million unclassified records. Records > 400k new copyrights/year. Of these, there were > 800k research requests from Congress and other agencies, and ~360,000 items were circulated. Note that that last is > 0.1% of the total books, and 0.2% of the total holdings of the library. Data from 2020: https://loc.gov/about/general-information/#year-at-a-glance
- University of California Library System has a total of 40.8 million print volumes across 100+ libraries system-wide.
- Bowker, the US ISBN registrar, was issuing about 300k "traditional" and on the order of 1 million self-published ISBNs annually through most of the 2000--2020 period best I can tell. (These data used to be more readily available.)
- Data on sales by specific title are ... hard to obtain. But I'd suspect that the top-10 title account for a large fraction, and the top-100 probably a majority of total sales. (Anyone inside Amazon know numbers?)
- Of copies of individual books ever printed, The Bible appears to be first, with an estimated 5 billion copies. It's trailed by the Quran (~800m) and Quotations from Chairman Mao Tse-tung (estimates vary widely from 800m -- 6.5 billion. It's interesting to note that all three of these are literally propaganda, two in the original religious sense, the third in the more modern ideological one. Six novels are thought to have sold over 100 million copies: A Tale of Two Cities, The Little Prince, Harry Potter and the Philosopher's Stone, And Then There Were None, Dream of the Red Chamber, and The Hobbit. https://en.wikipedia.org/wiki/List_of_best-selling_books
A friend who'd dropped out of a maths PhD programme once described the culture as "studying some specialised field in which there might be five or six other people in the world who could understand what you were doing." And finding this somewhat less than satisfying.
And I've had the exerience myself of studying lesser-known works and topics, or even some well-known ones, and finding that, variously, there was considerable insight to be found, or that what was widely "known" about a work ... didn't adhere closely to the work itself. To what extent then can the cultural awareness be said to be accurate.
Not sure if this makes my meaning clearer, but again, with an archive of millions to tens of millions of works, or even multiple copies of some smaller count, and much smaller pre-industrial populations of whom a small fraction were literate ... the living knowledge of those works must have been much smaller. At 7.5 books per year, an adult might read some 450 books in a lifetime. If half of those are popular, then the popular common canon is on the order of 200 works. What this figure might have been in ancient India I don't know, though for some cultures (Greece and Rome, and even 19th century Britain), knowlege of the Illiad and Odessey might have sufficed to be considered literate. Two books from a much larger accumulation.
>Do you have any sense of how long these works are?
While I have no clue on these specific works, I tend to use things like books of the Bible as a benchmark. Other works have similar length "books," like Caesar's "Gallic War" is split into 8 books of 5000 to 15000 words each or Aristotle wrote 400 books of about a thousand lines each.
Maybe some are significantly longer, but I'd assume the majority are that length. Keeping longer works in one piece for centuries is incredibly challenging.
Are you sure it's thousand? "m" is the common British abbreviation for million, while "M" would be the Roman abbreviation for a thousand. Plus, frankly only 10,000 Indian manuscripts seems way too low. India is huge, and has been literate for thousands of years.
There's a lot that I would say is troubling about this number.
Psychotherapists have been prescribing mindfulness-based interventions at increasing rates in the past few decades. You can't escape mindfulness, you'll hear about it everywhere you go. Yet, most of the literature about mindfulness or meditation and its impact on health is written in Sanskrit. Some of it is translated, and some of the translations make their way into English-language papers in psychotherapy or psychology journals, but it's such a small amount.
Rather that dig through Buddhist texts, we have created our own, new set of practices called "mindfulness meditation", prescribing it left and right, going on vipassana retreats, reading books on transcendental meditation, etc.
Mindfulness meditation is a western invention assembled out of pieces, taken out of context, of Buddhist and other traditions. Transcendental meditation was invented by a yogi in the 1950s. To be clear, I'm not talking about how recent these are in order to imply that older traditions are better--but older practices are better studied and we don't read about the older practices.
You can find a ton of studies in western medical journals about the benefits of various MBIs, but these studies, taken as a whole, are somewhat troubling. One troubling aspect is that many of the benefits are based on data which is self-reported, and the nature of questions in a self-reported study is limited. This is normal and expected in these kinds of studies, but it gives us a very narrow selection of observations about the effects of MBIs, and these questions / observations are often selected in order to prove positive effects--researchers, of course, want to prove positive effects of MBIs.
Meanwhile, there's two millenia of scholarly work, including descriptions of negative outcomes from meditation--what those negative outcomes were and how to structure the practice of meditation to prevent those negative outcomes--but these scholarly works are, again, not written in English and you hardly ever see literature reviews of these works in western medical journals.
Anyway. As a metaphor, it seems like we're exploring a continent, and there are people already living here, but we're ignoring them because they speak Sanskrit, and some of us are getting hurt. It's not even really the ancient texts, but modern texts, even modern texts with English translations.
I'm not sure how much scientific value can be found in texts written before the scientific method was developed. Aristotle's writings on biology and cosmology are almost entirely nonsense, for example.
MBIs are based on pre-scientific practices to begin with. Wouldn't it be a bit weird to say that MBIs are effective, but then say that all of the associated literature from 3,000 years is probably nonsense and not worth investigating?
We like to talk about how smart the ancient Greeks are. Greeks invented geometry, Greeks theorized that the Earth is round and measured its diameter, Greeks theorized that matter was made of atoms. Dalton gets credit for proving atomism experimentally, but he started out by gathering information from previous thinkers and experimenters.
The same is true of other cultures besides the Greeks and other theories besides atomism. It's easy to pick on Aristotle. He said some really stupid things; he said that men have more teeth than women. We can find plenty of quacks from the 20th century like Freud and Jung, plenty of barbarism like when Egas Moniz who cut up people's brains to change their behavior. Meanwhile, we keep discovering practices performed by various "primitive" peoples that are effective, and with investigation, we can understand the reason why it is effective. You don't need to read a 2,000-year-old Sanskrit text and take it at face value, but if you are studying the effects of meditation, it seems completely natural that you might want to do a literature review!
This sounds like a fascinating unexplored area. Are there any English texts (books/journals/theses) exploring this? I’d love to look into what limited analysis of this large reserve of literature does exist in English.
It's a weird area. There's a continuum between research in Buddhism and research in psychology & psychotherapy, and in the middle you get journals like Psychology of Religion and Spirituality or Mental Health, Religion, and Culture. Interdisciplinary research, for various reasons, just seems so difficult, journals that cover interdisciplinary topics are not as prestigious, and most researchers are loathe to give up their specialty even though they agree that interdisciplinary research is important.
I've seen some articles in these interdisciplinary journals that talk about MBIs and have citations that point towards western clinical practices as well as citations that point towards Buddhist texts... and I find it difficult to follow the citations in both directions.
It reminds me a bit of the divides between machine learning and statistics, or biostatistics and clinical research, etc. Someone who does clinical research will say, "Yes, biostatistics is very important," and then turn around and run another clinical study without talking to their colleagues at the same university in the biostatistics department.
I was worried a lot about this during my PhD. Well-cited papers were just the "tip of the iceberg". I did some fairly thorough searches and came across quite a few papers in my field that were important in my view and unfortunately overlooked.
The problem doesn't seem intractable to me. More researchers should spend some time searching the literature where they think no one else did. If a certain percentage of people did this, then I think it would benefit everyone including those who don't do particularly deep searches.