Beginning in this Wrong Monkey blog post and then in several others, I've had some things to say about how many manuscripts there are of this or that ancient text -- manuscripts of the Bible, for example, or of ab urbe condita, Livy's history of Rome. I wrote that first post back in 2009 because I'd seen some figures which I suspected, rightly, as it turned out, were way off.
And all along I've realized that the number of manuscripts, by itself, is far from a comprehensive statement about how well the text has survived from ancient times down to our own time. So why have I become so fascinated with learning numbers of this or that sort of manuscript? Maybe because I'm autistic and have an autistic relationship to numbers. However, it has occurred to me that I may have been misleading my readers by giving them such numbers without other information which is very important to understanding the significance of those numbers.
For one thing, sometimes one manuscript of a text is much more significant than many other manuscripts of parts of that text, simply by virtue of length. I was thinking for example of the 4th-century Codex Sinaiticus, which contains the Greek New Testament plus about half of the Greek Old Testament. Still, it's counted as one Biblical manuscript, since it all originally belonged to one copy of the Bible between one set of covers. Over the centuries, many Biblical manuscripts have been made which never contained the entire Bible: some contained the Old Testament, some contained the New Testament, some contained only the Old Testament Book of Psalms, some contained only the four canonical Net Testament Gospels, some contained some other book or a few other books, still others contained just passages from this book and that. Yet, each one is counted as one Biblical manuscript, because each one originally was one bookmaking project, 1 volume which stood alone. When we say "a manuscript of an ancient text," we are referring to a manuscript which contains the entire text, or a tiny fragment of the text, or anything in between.
Actually, the Codex Sinaiticus was not discovered all at once, but in several pieces. But those pieces are all counted together as one manuscript, because originally they were all one huge volume. If any pieces of the rest of that original volume are found, they plus what we now have will still be counted as just 1 manuscript.
When the Codex Sinaiticus was discovered in the 19th century, it was by far the oldest manuscript of any part of the Bible then known. Since its discovery, more Biblical manuscripts have been found which are as old and in some cases even older. But many of these manuscripts are just scraps of papyrus or parchment with only a few words on them. Sometimes it the writing is so brief and faint that it has only been with difficulty that someone has determined that it contains a text from the Bible. But that little scrap, if it can't be shown to have originally been part of the same book as some other little scrap, is counted as 1 manuscript. The Codex Sinaiticus, containing most of the Bible; a 12th-century Psalter (a volume containing just the Psalms is called a Psalter); and a little 4th-century piece of papyrus containing about a dozen words from the Bible: each one is counted as one Biblical manuscript.
But if two or more such little scraps can be shown to have originally been part of the same manuscript, then, just the same as with the pages of the Codex Sinaiticus found separately, those little scraps will now be counted together as 1 manuscript. The same way, if it is proven that a book containing the Psalms and another containing the Gospels were originally made as 1 book, then what used to be counted as 2 manuscripts is now counted as 1. The same way if different pieces of parchment or some other material with writing on them are demonstrated to have originally all been parts of a one-volume Bible.
Another consideration, when we talk about Biblical manuscripts, is that not everyone agrees what is or isn't a part of the Bible. From ancient times down to the present, different groups have included different books in the Bible. And then in the past couple of centuries, manuscripts of books which were rejected by those who eventually became the dominant churches and have been missing since ancient times have been found by archaeologists and others: the Gospel of Thomas, the Gospel of Judas, the Gospel of Mary Magdalene, etc.
Now let's move from the Bible to ab urbe condita, the history of Rome written in the first century BC by Livy. In the case of the Bible, many manuscripts contain an entire Bible, both Old and New Testament, and the Codex Sinaiticus and some other very old manuscripts contain most of it. Of the 142 books of the ab urbe condita, all of the manuscripts currently known, all put together, add up to 35 of those 142 books, books 1-10 and 21-45, plus a couple pages from book 91 and a couple of sentences from book 11. And as far as I know, no single manuscript contains more than 10 books. So, although the total number of manuscripts of Livy is impressively large, the number of manuscripts which contain ALL of his work is 0 -- as compared to however many manuscripts contain the entire Bible, dozens or hundreds or however many it may be.
Another thing: often the greatest specialists in a certain ancient text not only don't know the total number of manuscripts of that text, or even a close guess about how many there are -- oftentimes they don't particularly care how many there are. And they're being more sensible about this than I am, with my hunger to know exactly how many known manuscripts there are of Caesar's Gallic War or Lucan's Civil War. Why? Because every single manuscript doesn't always matter that much when it comes to editing the texts: coming up with the most accurate possible version of the text along with a reasonable number of guesses about variations, given in the footnotes. And editing texts is what a lot of these experts do all day long every day, while I flutter around the fringes of their profession being a weirdo.
Why doesn't every single manuscript always matter all that much? Well, for instance, let's take Ammianus Marcellinus, who in the late 4th century AD wrote a history which he may have considered to be a continuation of the history of the 1st-century-AD history of Tacitus, who may have considered his work to be a continuation of Livy's. Ammianus' history was 31 books long; today we have books 14 through 31 on 2 9th-century manuscripts and 14 15th-century manuscripts. However, it has been shown that all 14 of those 15th-century manuscripts come from 1 of the 9th-century manuscripts, that 4 of them are copied directly from it, and that all 10 of the remaining manuscripts are copied directly or indirectly from 1 of those 4. One page from that 9th-century manuscript is now missing, giving the 15th-century manuscripts most of the scholarly value they now have.
Sometimes an ancient text is known to us from only 1 manuscript. Sometimes an ancient text is known to us from no manuscripts at all. How can this be? It happens if early printed copies of the work survive, but all of its manuscripts have gone missing since they were first printed. That has happened a couple of times. Somewhat more common is that manuscripts survive, but an early printed version still contains some passages which are now missing from all known manuscripts.
And let's not forget Phillip Patterson, who recently spent 4 years' worth of his spare time copying out the King James Bible by hand. That means there's at least one more manuscript of the entire Bible than there were before Patterson started, because a text written with a pen on paper is a manuscript.
The numbers of manuscripts of ancient texts such as the Bible and Livy and Marcellinus tend to drop off sharply after the 15th century, because of the spread of printing, but occasionally a more recent manuscript plays a large role in establishing an ancient text.