Given a book coming to us from ancient Greece/Rome (no reason to necessarily limit just to those two, if there's a wider answer), is there a standard way for me to find out how the text came down to us?

As a motivating example, I was reading about Aristotle's History of Animals, and on its Wikipedia page I couldn't find any mention of which manuscripts were the earliest, how they were preserved, have there been controversies over corrections or non-canonical parts or widely varying variants and such. In fact, the Wikipedia article mentions it had been translated into Arabic and from Arabic into Latin, as was often the case, but it doesn't mention anything about whether the Greek original was preserved in Europe and where. There's a Loeb edition so apparently it survived. Where should I go to learn the details? Not just for this particular book, but as a rule?

  • 2
    I hope not - authority is anathema to science. I prefer my history to rely on methodology than authority.
    – MCW
    May 11, 2023 at 17:17
  • 5
    I think that you'll need to look on a per-document basis. For example, I happen to know that Beowulf's lineage is well-known, if not beyond all disputes. It stems from a single manuscript, but comes down to us by a complex web of copying and translating. I saw a paper in Science IIRC, some years back (a decade?) where scholars applied the sw developed to trace evolutionary trees using DNA sequences to trace Beowulf's document history, with typos (scribos?) being the equivalent of mutations. I've also seen references to publications tracing various Roman works, but don't recall any details.
    – Mark Olson
    May 11, 2023 at 19:32
  • 2
    Serious (academic) editions of these texts usually have an accompanying article which describes the transmission history and the authority of the current edition. But there seems to be no single book which collects these data. For most Greek and Roman texts, the earliest sources are of medieval origin.
    – Alex
    May 12, 2023 at 11:50

2 Answers 2


I don't believe there is, no. The only work I'm aware of that pretends to have anything like the scope you're talking about is Wikipedia, and you have to work at it to get that kind of information even for The Bible (the most copied work of literature in human history).

Of course it wouldn't be a simple matter to organize such thing, because we often have only parts of manuscripts of ancient documents, so it would practically have to be done on a Bible-esque "chapter and verse" basis. Even worse, sometimes documents or bits of documents survive only as quotations found in copies of other works. And of course copies made by humans don't always match. Sometimes one can even use these differences to establish a lineage of various strains of copies. I'm not entirely sure what the interface to a database that tried to encapsulate all that kind of information would look like.

If someone reading this feels like creating "TextualCriticismDB.com" (or it already exists), feel free to let me know.

  • 3
    "interface to database": tei-c.org/release/doc/tei-p5-doc/en/html/TC.html ?
    – Tomas By
    May 11, 2023 at 18:25
  • 2
    @TomasBy - Cool! So it looks like someone has thought about what the interface to the database would look like at least. Of course that appears to be an XML schema that could be used to do such a thing. I don't see anything there or on their Github about there being one central location for book encodings (using that XML schema of course) that one could go to find the info for one particular book. Perhaps I'm missing it.
    – T.E.D.
    May 11, 2023 at 18:31
  • 1
    Some links here en.wikipedia.org/wiki/Text_Encoding_Initiative (bottom half of page).
    – Tomas By
    May 11, 2023 at 18:38
  • 1
    @TomasBy - It looks like in this case the Perseus Digital Library would be the most likely place to start looking. However, they don't appear to have encoded that particular work of Aristotle (yet?).
    – T.E.D.
    May 11, 2023 at 18:44
  • No idea, I think it is mostly used for research.
    – Tomas By
    May 11, 2023 at 18:46

While there is no single resource for the textual history of how every classical work has come down to us, there are several books and databases which contain the known raw facts as to which manuscripts exist and where they are. The relationships between these copies is almost always a matter of modern reconstruction, debated by scholars and therefore not always suitable for a fixed-form database summary. For certain authors, works, topics and periods, there may be specific resources that present a more comprehensive account; results will vary depending on how motivated scholars have been over the last several centuries to investigate a given topic. Aristotle is among the authors who are very intensely studied.

In the case of the History of Animals, the Loeb edition you cite does include an account of the manuscript sources on which its version of the text was based. These are the same as in the edition by Leonhard Dittmeyer, Aristotelis de Animalibus Historia (Teubner, 1907). They are cross-cited to Inventaire des manuscrits grecs d'Aristote by André Wartelle (1963), a catalogue of 2283 Aristotelian manuscripts. The book also covers important printed versions, whether based on these manuscripts or on other sources now lost to us. For example, one of the most important sources is given in Loeb as:

Ca Laurentianus LXXXVII. 4. 14th century. Written by Joanicius. Contains Books I–IX in the following order (which is the order normally found in the mss.): I–VI, VIII, IX, VII. Collated by Bandini against Du Val's 1619 edition for Camus; this was apparently the first time it had been used for establishing the text. Collated also by Bekker and Dittmeyer.

(References like "Camus" are explained elsewhere - for example, that means Armand-Gaston Camus who published his version of the History in Paris in 1783. So this gives some more hints about which texts were used by which authors of post-mediaeval editions as well.)

The citation "Laurentianus LXXXVII. 4" is not completely obvious; it refers to a book in the collection of the Biblioteca Medicea Laurenziana in Florence, which includes several different works by Aristotle. Its version of the History of Animals starts on the verso of f70. Much more information on this particular codex is available at its comprehensive bibliography entry of Commentaria in Aristotelem Graeca et Byzantina, which as its name suggests includes a manuscript database of the Greek and Byzantine side of Aristotle's transmission. Many more versions of the text are in its summary page.

Information there shows not only the physical structure of the manuscript and which bits of the text it contains, but also its provenance, and citations of scholars who have written about it.

Just looking through these pages, we can make some meta-observations:

  1. Sources for the raw data are pretty accessible, in the sense that a great deal of information is out there on the Internet, either in modern databases, or digitizations of earlier works (such as 19th-century critical editions).
  2. The data are pretty inaccessible, in the sense that fully appreciating any particular database entry might demand a knowledge of Greek, Italian and German, even aside from interpreting what is meant by "167 (κ[β´])".
  3. We often do not have all that many different copies of a text, and new discoveries are rare. There may not be very much we can reasonably say about different versions and how they came about.
  4. Still, a lot of authors have written a lot of words about their different theories of what is going on. There is not necessarily a single accepted version of a text's lineage.
  5. Due to historical divisions within scholarship, sources will tend to focus on just one segment of the transmission. For example, it's common to find accounts which end their story at the advent of print, while different authors take up the tale of how various print editions relate to one another. There are databases for papyrus fragments, which are distinct resources from the databases for mediaeval manuscripts.

Moreover, even the apparently comprehensive information above actually omits an important part of the transmission, namely the Arabic tradition. We know that Camus' print edition, mentioned above, took account of the Latin translation by Michael Scot from about 1215, which was made from Arabic sources. It was this Latin version which tended to be used by mediaeval scholars such as Albert the Great, so that is an important part of the history of the work's reception. Additionally, the Latin has been used to inform modern versions of the Greek text. Despite all this, the first printed edition of Scot was made as recently as 2020 (by Aafke M.I. van Oppenraay). Her introduction mentions many more manuscript sources from the Arabic-Latin side, not found in CAGB. This type of prose account is probably the best way to get an understanding of what we know, and what we can infer.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.