Someone who struck me as authoritative -- I do not remember who -- wrote -- I do not remember where. I should write these sorts of things down more often. It may have been in the Bryn Mawr Classical Review, which I read often and recommend heartily -- that the manuscripts of Aristotle are literally myriad. I then consulted the Oxford English Dictionary, and saw that "myriad" literally means "10,000."
[PS, 19 February 2018: My memory was faulty here. I finally tracked down the mention of literally myriad manuscripts. It was a reference to manuscripts, not of Aristotle, but of Augustine. Nevermind.]
Attempting to verify that there really are as many as 10,000 manuscripts of the works of Aristotle, I found that, as of the writing of the article on Aristotle in the 1972 Encyclopaedia Britannica, there were 47 surviving philosophical works attributed to Aristotle, and that he actually wrote many more. Not from the encyclopaedia, I learned that these 47 works were often copied individually, as opposed to huge volumes each containing many of the works. I learned that several of these works survive in Latin translations in several hundred manuscripts each (Aristotle wrote in Greek, and was very popular among Medieval scholars of Western Europe who could read Latin but not Greek.). If several hundred Latin copies is typical for each of those 47 works, then perhaps there really are over 10,000 manuscripts of Aristotle surviving in our time, and the vast majority of them are Latin translations. (Several hundred X 47 = more than 10,000.) I'm assuming that untranslated Greek manuscripts of Aristotle are not nearly so numerous, but perhaps I'm wrong about that.
I have absolutely no ideas how many manuscripts of Aristotle in Arabic translation have survived to our day, or in other languages, for that matter.
Some time ago, I read in Rackham's Loeb edition and translation of Aristotle's Politics
that the manuscripts of that work "are not very good nor very old. The oldest evidence for the text is a translation in barbarous Latin by a Dominican monk of the thirteenth century, William of Moerbeke[...]The five best extant Greek copies are of the fifteenth century[...]" That was the first time that I had read anything about the transmission of Aristotle's texts. And so I mistakenly assumed that there were not many manuscripts of anything written by Aristotle. It turns out that Moerbeke is one of the Latin translators of Aristotle who has been copied into hundreds of surviving manuscripts, per work, having translated other works by Aristorle besides the Politics, and that not everyone has shared Rackham's low opinion of his Latin prose.
So, is Aristotle in 2nd place among ancient authors, behind only the Bible, in terms of numbers of surviving manuscripts? I don't know. One reason I don't know is because the experts on ancient Greek and Latin literature themselves don't know how many surviving manuscripts there are of the authors in which they specialize. And the reason they often don't know is because they don't much care. How can this be? Well, you see, the most important aspect of their jobs is get a version of those ancient texts as close as possible to what the ancient authors originally wrote. And for the purpose of determining those texts, the great majority of the manuscripts can be dismissed, if it has been determined that they are all copies, or copies of copies, or copies of copies of copies, etc, of some other surviving manuscripts. There is often a very great difference between the number of manuscripts which scholars use to determine the text, and all of the surviving manuscripts of that text. Oh, so there are X number of manuscript copies of Moerbeke's translation of the Politics? Hey, that's great. But because I have the actual copy which Moerbeke made (or high-res photos of that copy), I don't need all those hundreds of others. Is how those scholars will often react, if they see their job as editing the text.
There are other reasons for looking at all of the other copies. For example, someone has to determine where they came from, whether manuscript J was a copy of a copy of manuscript R, or what exactly. Or maybe Professor Y thinks that Professor X made a mistake when he or she said that J was a copy of a copy of manuscript R, and wants to check for him- or herself.
Another reason is if we want to get a general idea of how popular that ancient author was in a certain time and place. We can only get a very general idea of this, because we know that a lot of manuscripts have disappeared, and we don't know how many. Just because there are hundreds of manuscripts today of Ovid, and none at all of Pompeius Trogus, doesn't mean that Ovid was read by more people in the 2nd century AD than Trogus. But the great number of 12th-century manuscripts of Ovid (compared to surviving 12th-century manuscripts in general), combined with other things such as frequent mentions of him by 12th-century writers, mean that we're probably pretty safe in saying that Ovid was widely-read in the 12th century. Probably.
It seems to me that typically, there are more 15th-century manuscripts of a given Classical Latin author than manuscripts of any other one century, and sometimes more than all the other centuries put together. It seems that way. But I don't know for sure, because I only have those century-by-century numbers in the case of a few Classical Latin authors. Maybe they're pretty typical of the rest, maybe they're not. After the 15th century, the numbers of manuscripts of Classical Latin authors drops away to almost nothing, because of the invention of the printing press. One notable exception to that is the text of the 1st-century novel Satyricon by Petronius,
the inspiration for Fellini's film of the same name, liked by Fellini fans, less well-liked by Classicists who feel that Fellini missed much of Petronius' message. The text of Satyricon has been patched together like Frankenstein's monster from various manuscripts each containing just a part of the whole. 4 of those manuscripts were written in the late 16th century, and just recently, Maria Salanitro has found what she believes are still more parts of the novel, contained in a 17th-century manuscript.
How much of the preponderance of 15th-century manuscripts -- assuming I'm correct in assuming it exists -- is due to an actual rise in the reading of ancient Latin Classics in the 15th-century, and how much is due to people being suddently much more careful to preserve manuscripts? I have no idea.
It was nice of Martin Wohlrab to list and comment on all 147 of the manuscripts of Plato which he could find, late in the 19th century, and it was also nice of the University of California to re-print his list
in the 21st century. Did Wohlrab include manuscripts of Latin translations of Plato (or translations into still other languages) in his list? I'm going to have to examine this list a little more closely and get back to you on that one. Were there ever very many manuscripts of Latin translations of Plato? Hey, that's another really swell question. I know that Latin translations of Plato were made after the invention of printing.
Are the numbers of manuscripts of Cicero or Vergil comparable to those of Aristotle? Another thing I really wish I knew.
Why do I care so much about it? Am I about to help these professors in their task of sorting out which manuscripts derive from which, by the process they call collation? No. Am I interested in the numbers of readers these authors have had? To be honest: only slightly. I think I care about these numbers of manuscripts because autism. (It would also be great if I could demonstrate that there are more manuscripts of one Classical author or another than of the Bible, but I suspect that the Bible-thumpers out there who're saying that there are only 20 manuscripts of Livy [There are hundreds. How many hundreds? I wish I knew. Hey, there might be thousands for all I know.], and so forth, have also drastically under-counted the total number of Biblical manuscripts.)