I recently visited Jordan and noticed that many mosaic are commented with included text. The text seems mostly ancient Greek alphabet, but it also contains non Greek characters such as C, obviously standing for Σ, at least for some words. Given I forgot most I ever knew of Greek, I mostly looked at city names.

For example yous have the cities NϵAΠOΛIC and ΔIOCΠωΛIC on two mosaics from the 8th century in the Church of St. Stephen at Umm al Rasas. Note also the different spellings for city (POLIS).

Similarly. on the 6th century mosaic map of the Holy Land in the Church of St George at Madaba, there is the city NIKOΠOΛIC. Look below Bethleem (number 2) on the large this map:h century map of Holy Land

This cannot be the cyrillic alphabet, which does not contain ω and was created in the ninth century anyway, its authors (saint Cyril and Methodius being born between 815 and 827).

This seems to indicate and evolution in the use of Greek characters in that part of the world. Unless the use of C was justified because the ending sigma ς is not easily rendered in mosaic. This does not seem too likely as mosaics are pretty good at that level of details (look at the λ for Bethleem).

Can someone give some light on this. Was there an evolution in writing. How much of it was already done when the two brothers created the Cyrillic alphabet, and would it explain some of the new characters in it?

Also worth noting on the third mosaic (map of the Holy Land) is the Greek spelling of Bethleem, with two different epsilons: BHϑλϵEM. And there is also the frequent mixing of lower and upper case (if I may called them that way, given the time). Was that common?

  • 1
    The Cyrillic alphabet does contain ω, it is still used in Church Slavonic.
    – Yellow Sky
    Commented Apr 28, 2014 at 13:18

1 Answer 1


These are all normal Greek characters. C is a form of sigma: it's called lunate sigma, and is a variant that's sometimes used in printed texts these days too. Lunate sigma is a Hellenistic development which occurred in handwritten Greek (not specific to mosaics) for speed of writing. (It's also the origin of the Cyrillic C for [s].)

In ΔIOCΠωΛIC the omega must be a spelling mistake: at this period Greek no longer distinguished the sound of omega (originally [ɔː]) from that of omicron ([o]), and such confusions are common. The two epsilons in "Bethlehem" look almost the same to me; one's a bit more rounded, but that's well within the usual range of shape variation in pre-print era texts. The lowercase-uppercase distinction would not arise for another several centuries.

  • Thank you. I am a bit surprised by the answer as I never had any hint in my ancient Greek textbooks. As I recall, event ancient texts were written (in the books I had) following the modern use of uppercase and lowercase. But then, when did the two character sets appear, and why? How old are the most ancient original handwritten documents we have? Could the lunate sigma replace both sigmas or only the ending one ς ?
    – babou
    Commented Apr 28, 2014 at 8:05
  • 2
    Lunate sigma is actually older than σ ς, which are derived from it by closing the loop and by adding a tail, respectively. Modern conventions of printing Greek are different from ancient ones in several ways: not just uppercase and lowercase but the use of spaces and punctuation, and of breathings and accents (another Hellenistic invention, which wasn't consistently used until Byzantine times). The lowercase/uppercase distinction basically arose with printing, which imported an existing manuscript convention of making some initial letters larger and fancier.
    – TKR
    Commented Apr 28, 2014 at 17:20
  • @babou No, neither Romans nor Greeks used upper/lower case the way it's used nowadays. Capital letters were used for inscriptions in monuments as well as books on stuff deemed of special importance (such as religious books or major works of poetry/history). Small letters were used always for mostly everything else when writing in papyrus, patch or the like. Have a look on photos of ancient hexagaplas to see how they're written in all minusculae, not capitals.
    – Joe Pineda
    Commented Apr 29, 2014 at 2:37
  • @babou If I recall it correctly, Lunate Sigma is simply a cursive form of capital sigma, so it can appear anywhere in the word.
    – Joe Pineda
    Commented Apr 29, 2014 at 2:40
  • 2
    @JoePineda Minuscule letters didn't exist in antiquity -- they're a Byzantine invention. The first known Greek minuscules are from the ninth century. Ancient papyri used basically the same letter forms as inscriptions, though with some adaptations for handwriting, like lunate sigma, lunate epsilon, a somewhat different alpha, and what is now the lowercase form of omega. (What are hexagaplas?)
    – TKR
    Commented Apr 29, 2014 at 4:20

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.