Most Greek scholars are aware that sigma has a few different forms. In most current printed editions, it has a medial (σ) and final (ς) form, even though for a large part of antiquity up to the Byzantine era, it was usually written as a so-called "lunate sigma" (ϲ).

My question is a little more specific: it seems like the scholarly consensus has been to distinguish between medial and final forms for lowercase sigma, even if the underlying manuscripts are uncials (=all caps) or use lunate sigmas. This is the case for most scholarly editions I'm aware of (e.g. Nestle-Aland and most Oxford Classical Texts), and all Greek textbooks seem to follow suit.

I was confused, then, when I recently was looking through the OCT version of Sophoclis Fabulae. Not only does it use lunate sigmas throughout the whole edition, but (as far as I can tell) the editors didn't even think that this choice was worthy of mention in their preface. What's even stranger is that this is a relatively recent OCT from 1990/92.

I have a few related questions: Is this choice as uncommon as I'm making it out to be? Is this just a matter of taste or preference, or is there some principled reason for adopting the lunate sigma in this particular edition?

1 Answer 1


I don't have an exact chronology down, but it is indeed something that Oxford has been doing with their newer OCTs. I first noticed it in Merkelbach/West's edition of the Fragmenta Hesiodea, which was published in 1967 and Solmsen's edition of Hesiod from 1970. Some other editions I've found which use the lunate sigma include Hollis' edition of Callimachus' Hecale from 1990 and Diggle's edition of Fragmenta Selecta Tragicorum Graecorum from 1998. It's also not exclusively the domain of fragmentary works: Bude's edition of Herodotus (first published 1906) uses the traditional sigma, but Wilson's new edition from 2015 uses the lunate.

Michael Hendry (whose blog really takes me back! miss those days) found it as early as 1970 in a Cambridge edition of Euripides' Phaethon, so it's not just Oxford. (Interestingly, it's also a fragmentary work.)

It's also not all of Oxford. West doesn't use the lunate sigma for his commentaries on the Theogony or Works and Days (published in 1966 and 1978, respectively), nor does Carlos Steel in his editions of Proclus' Commentaries (2007–2009). Given the lack of uniformity, my guess is that it's up to editorial discretion.

Earliest discussion online I've seen so far dates to 1995, with Carl Conrad calling it a recent trend "among Classicists." So no, it's not just your text; no one discusses it because it's just not a conversation point anymore, and hasn't been for at least 20 years. While it might have started off as the choice sigma for fragmentary texts (which often come from papyri that use the lunate sigma), it's now used in complete texts as well. Given the length of time it can be found in modern texts and its adoption by a variety of editors, any discussion of editorial decisions on its use wouldn't be found in the introduction.

  • 1
    Really helpful (+1). It does strike me as jarring--do you know of any good articles that lay out the debate? Clearly enough scholars thought there were good enough reasons to break the prior consensus.
    – brianpck
    Jun 6, 2022 at 18:00
  • 1
    @brianpck I didn't see any immediately, but I can keep looking for it. I should expect there to be something somewhere!
    – cmw
    Jun 6, 2022 at 21:03

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.