About
What this is, and where the readings came from.
I built this while taking GRK30, an advanced Ancient Greek course at Dartmouth with Professor Rachel Dubit. What I wanted while studying a passage was to have all of it visible at once, on the text itself: the translation, the parse of each word, what the word was doing in its sentence, and the vocabulary worth keeping. So I made it.
The readings collected here are mostly descriptions of landscape, which was the theme of the course: Calypso's island, the plane tree on the Ilissus, Theocritus' herdsmen and their countryside.
The unit here is a reading. Each one is a chapter of its text, self-contained, with its own page and its own tools, and nothing groups them above that but the book they belong to. Adding another is a matter of adding one more chapter.
The Latin
There is also a Latin reading here, Vergil's first Eclogue, from the Latin version of the same site that I used for LAT30. It is one reading, not a collection: Tityrus keeping his farm while Meliboeus loses his, which suits the landscape theme well enough to sit beside the Greek. The format is identical, so many more could be added, and the instructions cover Latin as well as Greek.
What I'm reading now
Since the courses I have been using it for my own reading, and those texts sit on their own shelf on the contents page: Xenophon's Anabasis in Greek and Seneca's letters to Lucilius (the first sixty-five) in Latin. They go up a chapter or a letter at a time, as I get through them, so they are always unfinished.
How it works
The text sits on the left, a deliberately literal English translation on the right, aligned clause by clause. Every word is clickable and opens its lemma, part of speech, full parse, gloss, and a note on what it is doing in the sentence.
Everything else is derived from that same word-level data rather than written separately, which means it cannot drift out of step with the text:
- the parse trainer, which draws forms from the readings themselves and cites where each one stands;
- form drills and grammar review, sorted into twelve buckets;
- practice, which cuts a roughly 20-line window out of a book and generates translation, parsing and vocabulary questions over it;
- an Anki vocabulary deck, nested so you can drill a whole book at once.
The twenty-line window is measured, not counted: a verse segment weighs its actual line span and a prose segment weighs about eight Greek words to the line, so “twenty lines” means the same quantity of reading in Homer as it does in Plato.
About the analysis
I used AI to do the first pass of parsing on the Greek and to prepare each new batch of readings. The course readings are marked human-verified in the word panel. The texts I am reading now start out as morpheus+model: every inflected form carries a parse that the Morpheus analyser accepts for it, but the choice among those parses, and the glosses, notes and translation, are generated, and they stay marked that way until I have read the chapter back myself. Names Morpheus does not know are marked model-only. The reader carries that mark on every token precisely so that generated analysis can never quietly present itself as checked analysis.
Texts and credits
The Greek and Latin of the Anabasis (Marchant's Oxford text) and of Seneca's letters (Gummere's Loeb text) come from the Perseus Digital Library, used under CC BY-SA 4.0; the chapter files made from them are shared under the same licence. Morphological analysis is checked against Morpheus, via the Perseids Project.
Adding your own readings
The courses are over, but the tool is not specific to them. Any Greek or Latin text can be put through it, and I have written up the process I used so you can add readings of your own: Adding readings.
Who it is for
Anyone reading Greek or Latin. It was built for one course, but what it actually does is put the whole apparatus on the page so you can read without breaking stride to look something up. That is what I wanted at every stage of learning the language, so I have left it up.