In August of 2020, I wrote 0 To 70: Five Years Of Latin Novellas. Two years later, the number of Latin novellas nearly doubled. Well, another two years have gone by and we’ve seen 40 more, which is an increase of 31%! Above all, there are now 38 authors out there writing different kinds of Latin. In this post, I want to celebrate that different kind of Latin while revisiting some findings…
Continue readingSupport
Latin & CI Criticism: Challenging Work & Higher-Order Thinking
A frequent complaint of comprehension-based Latin teaching is that students aren’t challenged enough. Critics point to Bloom’s “understanding” level, assuming that everything stops there, imagining students hanging out in the lower-order thinking levels. Given that assumption, the critic’s takeaway is that “CI Latin” is unchallenging, and lacks higher-order thinking.
Untrue.
What is true is that comprehension-based Latin teaching prioritizes understanding as step zero. Under such an approach, nothing is done without first understanding the Latin in front of all students. One immediate implication is that Latin texts must be level-appropriate for beginners in the first years of a new language. And that level is quite low. Therefore, it is true that these texts are at a much, much, much lower level than the kinds of texts traditionally used in Latin classrooms, but that’s just text level. This says nothing about thinking level. Do these truths mean that students do NOT engage in higher-order thinking with these lower level texts? Are students NOT challenged when reading and discussing in class?
Continue readingRevised AP Latin 2025: Good News…Still Bad News…And Maybe Even Worse News
**Update 7.20.23 – This post was SOMEHOW deleted and nowhere in my trash folder?! I’m guessing it occurred when WordPress updated to the JetPack app. I know, you don’t care, but I Just wanted to say that I found the post in its ENTIRETY using the Wayback Machine. Here it is if you wanna see!**
When the draft of the new revised AP Latin course was released, I simultaneously couldn’t care less yet was also amazingly intrigued. So, I did another analysis like the one a few years ago. In short, the good news is that 1) the total amount of Latin has been reduced by about 25%, 2) quite a bit of “OTHER” Latin has been suggested as comprising a third of the overall syllabus alongside Virgil & Pliny, and 3) the revised course has been officially recognized as a second year college equivalent. Now, the bad news…
The Same
The revised course is fundamentally the same. While the draft name-drops the trending “language acquisition” phrase, the core texts are still very, very far above the reading level of nearly every student who will take AP Latin. This, in no way, is a tenet of language acquisition, whatsoever. Even methods and theories encouraging a “productive struggle” would draw the line well before anything like Vergil & Pliny after a few years of a new language. Beyond that, most high school students have no business taking a second year college course, anyway. BuT Ap Is AdVaNcEd, RiGhT?! Sure. The problem, though, is all the schools that choose AP as their program capstone without any other option, weeding out kids who “can’t cut it,” and/or backwards designing the whole program down to freshmen year (or lower for “feeder” middle school programs). When AP is used this way—which is very common—most students are denied a senior year language experience, or are unnecessarily forced to struggle through the content.
Core Text Vocab
The revised Vergil & Pliny selections vocab comes to:
- 5,500 total words in length
- 3,500 forms (i.e. aberant + abest = 2)
- 1,800 meanings/lemmas (i.e. aberant + abest = 1 meaning of “awayness”)*
This particular Vergil & Pliny content not as bad as the current Vergil & Caesar content (i.e., down from 11,700 total words in length, 5,700 forms, and 2,600 total meanings/lemmas). Progress, right? But still nowhere close to what typical students have acquired by senior year. Using the same generous figure of seniors knowing 750 words by the time they begin AP coursework, there’s still more unknown than known vocab. Like, way more (i.e., 1050 unknown words of Vergil & Pliny remain!!). The AP draft also states that “the focus of the course is continued Latin language acquisition with the inclusion of some textual analysis and contextualization skills.” Uh…some…analysis and contextualization skills?! If the focus were actually on acquisition, students would be reading a LOT. To be clear, students still won’t be reading this kind of Latin. To truly “read” these texts, students would need to understand 98% of the Vergil & Pliny selections to have a chance at comprehending what’s going on (see Text Coverage). That ain’t gonna happen. But there’s more, and it might be worse…
The stats so far represent just 65% of the AP syllabus content!
Teacher Choice
The AP draft outlines setting aside about 1350 total words of poetry and 1650 total words of prose, all of the teacher’s choosing. This constitutes a third of the revised course content aside from Vergil & Pliny. Now, for this analysis, I got lucky with randomly choosing the recommended Ovid passages from Metamorphoses (i.e., 1450 words), and the Ciceronian letters (i.e., EXACTLY 1650 words). When we add these texts to the core Virgil & Pliny, unfortunately, we get a chart that basically mirrors the current AP one:
While the 2025 AP draft shows the total amount of Latin being reduced by about 25%, it turns out that the vocab difference is only about 4% less (i.e., 2600 total words currently vs. 2300). Progress…right? Here are a few different charts to help visualize all that:
Wildcard
Yet I chose Ovid & Cicero. The big wildcard is Teacher’s Choice portion of the AP syllabus. Surely, the impact of choosing text Y or author Z will vary, yet by how much? I was curious about that impact, but I do have a day job, other hobbies, and Ph.D. coursework to do. Rather than look at multiple combinations of poetry and prose, perhaps looking for the most extreme outcomes, I kept Ovid and replaced the Cicero letters with different prose selections amounting to ~1650 words of Latin to see what came up.
Starting with Sallust (Bellum Catilinae,, 5-15), the results were pretty close with just 10 more words added given all the vocab in common with Vergil, Pliny, and Ovid. Next, I pieced together an assortment of Gellius’ Vestal passage in Noctes Atticae (1.12), Eutropius’ founding of Rome in Breviarium Historiae Romanae (1.1-8), and the first five characters of Hyginus’ Fabulae. After all, the stated point of including this “OTHER” Latin is for encouraging teachers “to explore Latin texts from different time periods written by a variety of authors.” The difference between using this assortment of selections instead of Cicero wasn’t staggeringly worse. Then again, it did add about 50 more unique words, increasing those vocab demands, now coming within 200 words of the current AP syllabus!. The point? The Teacher’s Choice portion—a third of the AP syllabus—has the potential to meet (surpass?) the vocab demands of the current AP!
In sum, the revised AP course seems to be an all-too-common case in education: making progress in some areas yet ultimately falling short of addressing a fundamental problem. The 2025 AP is basically the same old thing (i.e., vocab demands are very close to the current AP with just ~4% fewer words) wrapped up in new packaging. I’m sure it will appeal to more teachers, but it doesn’t seem to deliver much else in terms of pedagogical improvements besides reduced workload and a tip of a hat to everyone reading “OTHER” Latin. So, maybe 2035 draft will be promising? I’ll be long gone from the classroom by then. Best of luck to everyone out there!
*Once again, for this analysis mostly done via Voyant Tools & Google Sheets, I rounded to the nearest 100, in the AP’s favor. Errors are unlikely to change things remarkably, and in all likelihood the situation is a little worse than what you see.
ALIRA: “All Your Datum Are Belong To Us…Plz?”
In the original draft of this post, I compared two data sets of students taking the ALIRA. However, I’m not really comfortable publishing that. I really don’t need anyone trying to play the victim when it’s been me going on a decade now defending my teaching practices and the kind of Latin that I read (and write) with students. It’s too bad, too, because the data are quite compelling. Some day, I’ll share the charts. Until then, you’ll have to take my word on it. You probably already know that I don’t fuck around, either, so my word is solid.
In short, the charts will contradict the claim that reading non-Classical Latin leaves students unprepared for reading Classical Latin. They will suggest that reading non-Classical Latin texts, such as those rife with Cognates & Latinglish via class texts and novellas, is of no disadvantage. They will also suggest that reading Classical texts is of no advantage. That’s all I’m prepared to share, for now.
Once a lot more data like these will be presented, though, the jury will start to come in on the matter of what kind of Latin prepares students for any other Latin. From what I’ve seen so far, it looks like A LOT of any Latin can prepare students to read other Latin, and that’s a good thing. These emerging data show that concerns and claims over certain kinds of Latin don’t play out in reality. Still, it’d be good to have more scores, not just the 532 ones currently submitted to that ALIRA form. If this all seems mysterious, it kind of has been. I haven’t shared the spreadsheet yet for viewing. That changes today!
Continue reading90%, 95%, 96%, 98%, 100%—Misinterpreted Numbers Demystified
Here are some common numbers floating around the language teaching world, their misinterpretations, and some clarifications to go with each one:
90%
This figure comes from an ACTFL position statement on target language use. The biggest misinterpretation I’ve seen is thinking something like students should be speaking the target language 90% of the time. The 90% figure is actually about percentage OF language during class that IS the target language. For example, if you’re reading a 100% Latin text during class and ask SOME questions about it 100% in English, it’s quite possible that target language use is still above 90% (e.g., students read 1,000 total words of Latin and hear 100 words of English questions = 90% target language use). Students need input. This 90% figure is an attempt to prioritize high quantities of language. The alternative to avoid would be something like doing a close reading on a short paragraph of Latin with the discussion entirely in English, perhaps resulting in the opposite 10% target language use, which was quite common in the teaching of languages prior to ACTFL’s statement. Somehow, you’ll still find such low levels of target language use today, despite ACTFL’s efforts back in 2010. N.B. if you do a general search, some ACTFL page might come up that contradicts their original 2010 statement, referring explicitly to class time. It also name-drops Krashen, pointlessly throws in the i+1 concept that itself became misunderstood, cites Vygotsky in some kind of weird sociocultural nod, and ends with an obligatory reference to Long & Swain’s work on output theory. This hodgepodge of ideas is startling, but probably represents how ACTFL tends to remain as NEUTRAL as possible, taking bits and pieces from way too many popular ideas and approaches as if it makes sense to combine them all. Don’t fall into that trap. Follow principles that you know align with each other and don’t conflict!
95%
See 98%, below.
96%
Perhaps more commonly known as “4%,” the figure of 96% represents the students who typically do NOT continue beyond the second year of language study, here in the US, with the dominant paradigm having been grammar-based teaching. Although programs have been moving to comprehension-based, and/or communicative approaches, it’s still possible that the 4% figure represents some programs today, though the origin dates back to a 1969 speech, detailed in this post. A common misinterpretation has been to claim that individual “4%ers” are the only students capable of learning a language via grammar-translation (GT). This is untrue. The “4%ers” are simply those who tend to stay enrolled. Recently, I’ve argued they represent the pedagogically immune.
98%
I’ve seen this (as well as 95%, and even 90% in some cases) shared as “minimum comprehension level” that students should have when reading. This figure actually represents recommended minimum text coverage while reading. It’s from various studies, detailed in this post. In short, a text coverage as high as 98% has been shown to produce comprehension scores as low as 70%. That’s not incredibly good. Also from the studies, a text coverage of 95% produced scores as low as 55%! You don’t want to know what the scores were for 80%!
100%
I’ve heard people say that knowing 100% of words means a text will be 100% understandable. That’s not true (see above). In an age of promoting “productive struggle” and “grit,” the accusation is that a 100% understandable text is undesirable because students will somehow not get their +1 level to their i (again, see this post on that whole concept demystified). This is most definitely untrue. Even when comprehension is 100% there are aspects of language (e.g., word order, forms, etc.) that can still be learned. Given what is known about text coverage, there’s very little justification for using target language students don’t know, like above-level texts. Therefore, the 100% figure should be a goal for text coverage or known vocab.
A Year Of Grading Research: 30 Articles, 8 Books, 1 Pilot Study
You’re looking at my school desk. There’s some wormwood lotion for our desert-like winter classroom conditions here in New England, some peacock feathers (why not?), one of the deck prisms my great grandfather made in his line of work, the growing collection of my ancient wisdom series obsession, and what remains of this year’s unread novella order. What’s not there is the stack of articles and research reports that had been piling up since last spring. I’ve finally read them all during my planning periods. Of course, each report itself produced at least another to read, and often two or three more, making the review process more like attacking a hydra, but those are now tucked away in a “To Read/Review” folder in Drive. My desk is clear, and that’s enough of an accomplishment for me while teaching full-time. Aside from the reports, I’ve read 8 books, too:
- Hacking Assessment 1.0 & 2.0 (Sackstein, 2015 & 2022)
- Ungrading (Blum, 2020)
- Point-less: An English Teacher’s Guide to More Meaningful Grading (Zerwin, 2020)
- Proficiency-Based Instruction: Rethinking Lesson Design and Delivery (Twadell, et al. 2019)
- Embedded Formative Assessment (Wiliam, 2018)
- Assessment 3.0 (Barnes, 2015)
- Grading and Reporting Student Progress in an Age of Standards (Trumbull & Farr, 2000)
- Punished By Rewards (Kohn, 1993)
In case you’re wondering and were to ask for my current top five, which includes Grading for Equity (Feldman, 2018) that I read a couple years ago, it’d have to be Ungrading, Pointless, Punished by Rewards, and Hacking Assessment. Beyond the books, this year I also completed a small-scale pilot study, which I’ll be presenting at the CANE Annual Meeting. While not specific to Latin teaching, a case could easily be made that *any* grading research can apply to *every* content area. In fact, it’s somewhat remarkable what researchers have found, yet the profession just doesn’t seem to know. And there’s consensus. I’m not prepared to make sweeping claims and cite anything specific, but my impression of the consensus so far is:
- Grading does more harm than most people think. It’s one of the few relics of antiquated education still practiced today en masse, in pretty much the same way, too. Considering everything that’s changed for educators in the past two, five, 10, 20, and 50 years even, now realize that the current dominant grading paradigm predates all of that. The fact that most grading systems are still based on the 0-100 scale with a “hodgepodge” of assessment products that are averaged together to arrive at a course grade is nothing short of astonishing.
- Schools with a more contemporary (i.e., 30-year old) approach that claim to have standards-based learning (SBL) and grading (SBG) systems are actually still in their infancy, with some not really implementing the systems with much fidelity at all, thus, giving a lot of SBG-derived or SBG-adjacent practices a bad name. It’s mostly teacher/school misinterpretation and poor rollouts of these practices that render the efforts ineffective, not the practices themselves.
- Gradelessly ungrading is probably the only sure bet for fixing the mess that grades have gotten us into. If you’re putting all your time and effort into SBG, I recommend that the second you understand the basics, see if you can skip right on over to a) using portfolios, b) getting rid of all those points, and c) having students self-assess & self-grade just once at the end of the term. You’re gonna need to provide a bit of feedback with this kind of system, too, so maybe try Barnes’ SE2R model.
Cognates & Latinglish: I’m Not Finished.
Gotta love a trilogy, right? This is my final Winter 2023 post on the whole cognate kerfuffle.
One reason Chinese takes native English speakers a LOT longer to acquire than something like Spanish is because Chinese has approximately zero cognates. Terry Waltz has reported that Spanish has about 3,000 words most English speakers can understand without being taught any of them. Languages with more cognates are acquired faster because cognates play a role in the first developmental stage of second language learning. Check out what ACTFL has to say about the novice language learner and cognates:
Cognates & Unranked Caesar/Virgil
Cognates. Whether you love ’em or hate ’em, how much Latin are we talking about? When they’re used, do cognates end up comprising most of a book’s Latin? Are we talking half? A quarter? Less? And then what does that mean about the rest of a book? How much of a book’s Latin is being dismissed amongst the cognate concerns? One of the concerns is that cognates weren’t used by Classical authors, the claim being that using infrequent words is a problem, and the conclusion being that cognates are unhelpful for today’s Latin learner. I don’t have this concern or have found evidence to support the claim. Nevertheless, I have been wondering just how much—or little—the concerns constitute, percentage-wise, of a book. I also have been wondering if there were many words Classical authors themselves used that other Classical authors didn’t really use. N.B., I don’t mean hapax, just words that were rarely used by others. Thus, if Classical authors preferred to use rare words, too, that would help illustrate how the current cognate fuss is more about preference than anything else. Let’s see…
Continue readingPunished By Rewards & Advantages To The Single-Point Rubric
As if researching how to eliminate grading and reduce assessment couldn’t get much better, I’ve now got something else. Alfie Kohn’s 1993 masterpiece really ought to be required reading for every educator. Coming up on its 30 year anniversary, the author at the time reviewed studies dating just as far back to the 1960s. This post is gonna focus on self-assessments. When it comes to students self-assessing, evidence suggests that the more students think about HOW WELL they’re doing (vs. WHAT they’re doing), they do it poorly.
That’s crazy-unintuitive, right?!
Continue readingWriting Challenge #3: Description
As we’re winding down the month’s writing challenges, let’s recognize that over just a couple weeks, contributors produced nearly 3,000 words of Latin for the beginner. These short stories share some themes and common vocab. Not bad at all! While sheltering vocab is not everything, it’s most things, but let’s add something onto keeping word count low, shall we? Descriptions. Among other uses of description, a character’s quality or how they do some action becomes an instant question for students: “are you also like the character?” or “would you do things the same way?”
So, Challenge #3 is to write a highly descriptive short story using as few of the following core verbs and function words as possible in order to focus on description:
- esse, habēre, velle, īre, placēre
- et, quoque, quia, sed
- ā/ab, ad, cum, ē/ex, in
- ergō, iam, nōn, subitō, valdē
For Challenge #3, there will be an overall unique word limit (excluding names, and different forms of words). Also, don’t forget about referring to the cognate list for adjectives, and don’t forget to make adverbs from them!
BOSS level sheltering: no more than 15 words
CONFIDENT level sheltering: no more than 25 words
NOOB level sheltering: no more than 35 words
Here’s the link for Challenge #3. And here’s where I’ll put the stories once they start rolling in.
Magister P. 