Eric Herman released memos 45 and 46 mid-March, and wow do they take “authentic texts”…to task! I’m not gonna post a bunch of juicy quotes from the 16 different authors cited in the articles. There’s plenty of convincing evidence to back up Eric’s claims, so go read the original if you’re a skeptic. Instead, here’s just one to get things started:

“They do not, however, provide any empirical evidence that this approach is more effective than adapting the texts themselves” (Gilmore, 2007, p. 109).

Gilmore is referring to advocates of the famous mantra that has led second language teachers astray for decades: “edit the task, not the text.” This push for “all-things-authentic,” a term for which there’s almost no consensus, has resulted in teachers justifying the use of unbelievably out-of-range texts given to language students to “read.” This by no means is unique to Latin teachers, but the kind of Latin that’s been given to students for centuries is probably the most extreme example of texts that no learner should have any business attempting to read. As a result, language development has suffered to the point of degree-holding teachers themselves being unable to read the very texts they’re giving their own students. Now that’s a trip. Eric goes on to offer some commentary on the situation:

“The motto is a sweet-sounding way of saying that the task does not need to involve reading for general comprehension nor any more demanding task such as reading to learn or reading critically. It’s obviously true that any text can be used if the task is not to comprehend the text. We can come up with something for learners to do with any text, but are these tasks the best use of class time? If you believe that NS (Native Speaker) texts have “inherent value” (Zyzik & Polio, 2017, p. 127), then you’re willing to “edit the task” when learners can’t comprehend the texts.”

That’s a pretty good summary of Latin teaching and learning: not comprehending texts, making use of English translations to derive meaning, exploring Greek and Roman culture—in English—and reducing the learning of Latin to SAT word building and intellectualized literary analysis. I shared some snippets of this research and Eric’s 1-page preview with the Latin Best Practices group. There was very little engagement. I’m guessing the ideas were dismissed as something like “well this couldn’t possibly apply to Latin” given the profession’s focus on obsession with ancient NS texts, especially since they’re regarded as having the “inherent value” that Eric mentioned. In fact, Day & Bamford referred to the “cult of authenticity” back in…wait for it…1998. Perhaps reading about the implications was just too much to be acknowledge by teachers who have been using these absurdly above-level NS texts their entire careers. For more on how it’s not their fault, see this 2017 post. Whatever the reason, this really should become part of the discussion. In fact, how could it not?

In this follow-up to what reading Latin has meant in the first year Latin classroom for decades, it’s worth listing some of the common tasks—instead of reading—that have been edited by teachers when they give students ancient NS texts far above their level:

• translating
  • guessing
  • checking a glossary
  • looking up words in a dictionary
  • stringing together English words (from that glossary or dictionary)
• identifying parts of speech
• searching for certain grammatical structures

Things get murky when we bring ACTFL into the mix, too. The performance descriptors and proficiency guidelines give us a sense of what the beginner (Novice level) Latin student is expected to do with an NS text. Let me restate that. These describe the kind of task a learner is expected to do when given above-level NS texts. Theses do NOT describe what learners can do when given below-, or at-level texts, which is…reading and understanding! Also, note the simplicity of the NS text expected, knowing that the ancient texts Latin teachers have been giving students are written far, far higher at a Distinguished or Superior level. Not only do we have no NS texts written for the beginner, the tasks, processes, and results that you see below are hardly considered “reading” by any definition:

• skim & scan
• recognize key words and formulaic phrases
• derive meaning when texts are supported by visuals (i.e., use pictures mostly for comprehension)
• comprehend some limited vocab
• recognize roots, prefixes, and suffixes
• get a limited amount of information
• recognition of key words, cognates, and formulaic phrases makes comprehension possible

Here’s what the Intermediate level states about the “reading” process as well as text features:

• texts are not complex and have a predictable pattern of presentation
• minimally connected and primarily organized in individual sentences and strings of sentences containing predominantly high frequency vocabulary
• Intermediate-level readers are most accurate when getting meaning from simple, straightforward texts
• can understand information conveyed in simple, predictable, loosely connected texts

And for the record, most students won’t reach Intermediate Mid unless they have six+ years of language. It’s nothing short of amazing how ACTFL still pushes the “edit the task, not the text” narrative, right?! Given that push along with the modest (at best) language development expectations below, no wonder many comprehension-based language teachers are seeing their students able to comprehend far more with learner-centered and adapted texts (vs. unadapted NS texts)!

In fact, we need a completely new chart and some updated descriptions. What can students do when they’re given appropriate level texts right from the start? How long does it take to be able to read more advanced and those high level NS texts when a learner’s foundation is comprised of TONS of comprehension-based and communicative input, not just some random tasks?

For Latin, I wonder if this all is a classic Classicist move of confusing the means with the ends. If the goal is to eventually read ancient NS texts, the way to get there is with adapted texts at a level written for the language learner not struggling with something way out of range, or succeeding in some kind of task that has almost nothing to do with comprehension and reading. Have Latin teachers taken themselves and the literature too seriously, expecting to spend their time “reading” these texts with Novice and Intermediate students? Surely, the “skim & scan” type tasks can be put to rest by now, no?

So, forget editing the task, and instead edit the text.

Don’t take my word for it. This is what the researchers are saying.