Concepts

Before the Athenians adopted what we think of as the “Greek” alphabet in the archonship of Eukleides (403 BCE), Athens, like other Greek city states, used a local (epichoric) alphabet for public documents. The “universal” standard of Unicode does not recognize this, and instead imagines that there is only a single “Greek” alphabet used between the end of the Bronze Age and the magic date of 1453 CE. If we wish to work with digital editions of Athenian documents before 403 BCE, we therefore are compelled to devise an encoding system for this orthography.

In print publications, the practice has been to print archaic Attic texts using visually similar glyphs. An upper case epsilon, Ε, for example, is used for the fifth letter of the Attic alphabet, which can represent sounds corresponding to ε, η or (in some circumstances) ει in later literary orthography. Some glyphs are borrowed from Latin alphabets. The Attic rough breathing character is expressed with “h”.

Epigraphists normally add accents corresponding to traditional accentuation of literary texts, although written Greek accents are only invented long after the Attic alphabet has ceased to be used, and never appear in our epigraphic texts.

The encoding defined in AtticGreek.jl mimics modern publications. Whenever possible, it encodes an Attic character with a visually similar Unicode codepoint from the Greek range in form :NFKC. It supplements these with Latin alphabet code points. Accent combinations of omega or eta with a circumflex cannot be encoded using the visually similar characters for omicron and epsilon (since those short vowels can never have a circumflex), so we use the accented Roman characters ô for omega with circumflex and ê for eta with circumflex.