Next winter at SCS (Society for Classical Studies) 2023, there will be a panel on what it means to teach students to read Latin. Reading Latin. It seems so obvious what it means, right? But no. What does it mean to read Latin? Of all the approaches to take, looking at data is a good starting point. Let’s start with what reading Latin has meant in the first year Latin classroom for decades…

What better place than the self-described “reading method” of textbooks such as Cambridge and Ecce, Romani? The latter’s first chapter begins with a cold-open paragraph of Latin. Here are the details:

• 70 total words in length (i.e., tokens, see below)
• 29 unique words

Text Coverage
Text coverage is measured by tokens, or total words. There are five tokens in the sentence “the bird sees the cat.” Two of the tokens in that sentence happen to be the same word. Therefore, “the” represents 40% text coverage. If the reader doesn’t know “the,” they have a text coverage of 60%. The reader who knows everything except “cat” would have a text coverage of 80%. It’s a simple example, but not hard to see what can happen at that 80% level comprehension-wise. The reader understands “the cat sees the ____,” so the unknown word is a big piece of missing information. Imagine reading a whole paragraph about the cat and ____ without knowing what ____ is and then being asked about ____. That’s not a very fun experience. And now imagine grading some kind of assessment on that experience! Don’t do it!

In that Ecce textbook example above, est appears 7 times and isn’t glossed. You gotta guess what it means from context. Luckily, most kids do. Those who don’t, though, miss out on 10% of the text coverage. A text coverage of 90% isn’t good enough for comprehension to have a solid chance, either (Laufer 1989, Laufer 1992, Hu & Nation 2000, Laufer 2010, Schitt, Jiang & Grabe 2011, Herman & Leeser 2022), but est isn’t the best example. Let’s look a little more into what “reading” means in this first textbook paragraph…

Glosses
There are 22 glossed words of the 29. The proper nouns are presumed to be recognizable, as is pictūra, a clear cognate. We know est doesn’t make the cut, and in also is left out, presumably both understood fully with ease. Best case? The reader fills in all these meanings. So, what’s the rest of the so-called “reading” like for the beginner? Here’s the likely process:

1. Read ecce!
2. Check glossary, it means look!
3. Read in pictūrā
4. Check glossary, not there, guess meaning
5. Read est
6. Check glossary, not there, guess meaning
7. Read puella
8. Check glossary, it means girl
9. Reread look, in picture is
10. Forget puella
11. Check glossary, it means girl
12. Reread look, in picture is girl
13. Think “that sounds wierd”
14. Read nōmine
15. Check glossary, it means by name, named
16. Read Cornēlia
17. Check glossary, not there, guess meaning ‘cuz it kinda looks like a name you’ve heard of
18. Reread look, in picture is girl
19. Forget nomine
20. Check glossary, it means by name, named
21. Reread look, in picture is girl, named Cornēlia.

Alrighty. That was 21 steps, more or less, to read “look, in the picture is a girl named Cornelia.” High five. It turns out that “reading” is more like a guess & check process, constantly referring to glosses and/or a dictionary. It’s a process that involves piecing together. It’s matching words with their meanings, and coming up with something that makes sense.

Limitless Vocab
Most of the time, the texts used in a first year Latin class also have many one-off words, or very few words that repeat. To get a sense of the exposure to input (I), here’s how many times a student reads Latin words more than once in the whole paragraph of Latin beyond the first sentence:

• 8 = in
• 7 = est
• 5 = habitō
• 5 = villa
• 4 = puella
• 3 = iam
• 3 = pictura
• 2 = etiam
• 2 = laetus
• 2 = legō
• 2 = nōmen
• 2 = quī
• 2 = rusticus

The other 10 words (34%) occur just once. For Latin students given these kinds of texts, then, “reading” doesn’t provide a lot of exposure to input (I) that’s comprehensible (C), which is no surprise why it’s mostly the students with good memories who succeed at this kind of “reading.” You can’t really do much reading with these texts, either, certainly nothing close to extensive reading. If you could, though, a recent study by Herman & Leeser confirms what we know about extensive reading, which is that students should know at least 98% of the words (i.e., tokens) to have a chance at comprehension. That would be reading. Like…reading reading…and not some “sort of reading” or “other” kind of reading. Circling back to text coverage…the process outlined above exposes how the student’s first—FIRST—text presumes something like 45% text coverage (i.e., all words except glosses). That’s nowhere close to 98%+ needed in order to just…read.

When you look at it from this point of view, it’s hardly considered “reading” by anyone’s definition, except of course the textbook company. Did you know that “the reading method” is just a proprietary description for what’s in the book? It’s true. No one in the literature refers to “the reading method” outside of citing the few Latin textbooks that created the term. Perhaps it’s best not to go down the road of what constitutes a “method,” though. We’ll leave things there for now.