RSS Feed

Author Archives: Carla Hurt

A complete guide to Classical Latin pronunciation: the sounds of Golden Age Latin

Posted on

I have now made a Classical Latin pronunciation series on Youtube! Check out the first three parts here:

These videos are suitable for complete beginners and advanced students alike. I speak 100% in Latin, with no English in these videos. This means I make no verbal descriptions of the sounds or comparisons to any variety of English. Instead, I focus completely on demonstrating the target sounds out loud and allowing for cycles of listen-and-repeat-after-me. There is no preamble, just straight-to-the-point practice. “Show, don’t tell.”

The three videos above teach the letters, vowels, diphthongs, consonants, and various combinations of letters. I make a quantitative distinction between long and short vowels. I follow the updated Calabrese 5-vowel system, which supercedes the vowel system from Sidney Allen’s Vox Latina as a more historically likely representation of Golden Age Latin vowels. I have not yet covered syllable length, accentuation or elision in these videos, but a follow-up that explains these syllabic features could be my video on Pronunciation Tips for Scansion.

Why I created this series

I had always assumed that there were already some very good “basic” pronunciation guides to Latin, but when I searched for one, all I came up with was:

  • People reading aloud inaccurate instructions from a textbook
  • Advanced, very accurate deep-dives into single phonemes

I have no complaints about the second category, the advanced deep dives, but they are not very beginner friendly because it is hard to get a systematic overview when all phonemes are split into separate videos.

I do, however, have serious complaints about the first category!

The instructions for pronouncing Latin commonly printed in textbooks are not completely accurate representations of Latin. Because textbooks are print resources and not audio resources, they rely on “close fit” comparisons to English sounds that don’t necessarily convey the true nature of Latin sounds. The result is that the beginner will inevitably speak with their native English accent while speaking Latin – because the instructions, delivered in English, can only reference English phonemes.

There are several key weaknesses to textbook written descriptions of Latin sounds:

  • For one thing, there are multiple varieties of English pronunciation, so instructions like “o as in pot” will be interpreted differently by a standard American and a Received Pronunciation (RP) English speaker.
  • Some textbooks reference particular dialects of English, saying for example that the “r is trilled, as in Scottish”. This instruction is totally useless for me because where I live (Australia) the Scottish accent is vanishingly rare. Almost everyone I have met on this continent who identifies as Scottish has an Aussie accent.
  • Textbooks are often outright misleading when explaining aspirated vs. unaspirated consonants. For example, they instruct students to pronounce “p as in pot” and “c as in cat”. Actually, both of those consonants correspond to the aspirated versions “ph” and “ch”, as English requires all initial p, t, c to be aspirated. You can test this yourself by holding a piece of paper to your lips and feeling it flutter with the puff of air that escapes as you say “pot” or “cat”, but not “spot” or “scatter”.
  • The instruction “ph as in uphill” and “th as in anthill” lead to students and teachers making bizarre “puh huh” and “tuh huh” sounds that don’t represent aspiration.
  • Very few beginner pronunciation tutorials even hint at the nasal final -m, and those that do tend to use verbal descriptions that are very hard to replicate independently (what does it mean to “nasalise” a vowel?). This phoneme is extremely common in Latin (think of all the -am and -um endings!). Knowing that the -m is barely pronounced but is mostly a marker for nasalisation helps explain why -m is ignored for elision; I think it is a shame that students only learn that sound exists in their final years when they are scanning poetry, and not throughout the course.
  • Many Latin teachers in these videos don’t know how to pronounce the Greek “y” and say “u” instead.
  • Many Latin teachers in these videos make “ū” into something that sounds like the Greek “y” – saying “ūnus” like “oonus”.

I hope that these videos fill the gap and provide a more Latin-through-Latin approach to explaining Latin pronunciation.

Key choices

Some people may be surprised that I don’t group “ui”, “eu” and “ei” on an equal level with the other diphthongs “ae, au, oe”. I did this because there are also words in which the combinations “ui”, “eu” and “ei” are treated as separate syllables. There are only about four words with “ui” as a diphthong, and three words with “ei” as a diphthong, so I deal with these diphthongs separately and show the specific words in which they do apply.

I also decided on the Koine Greek “zz” pronunciation of “z”, which may be suprising as others have chosen Attic “zd”. The reason is simple: the Attic pronunciation of zeta as “zd” had already evolved into Koine “zz” centuries before Golden Age Latin. It is completely anachronistic to mix 500 B.C. Attic pronunciation with 100 B.C.-100 A.D. Classical Latin – the time difference is huge. The Romans of the Golden Age would have been exposed to varieties of Koine pronunciation, not Attic. To recommend “zd” for zeta in the time of Augustus would be as weird as recommending 1500 A.D. medieval French phonology for French quotations inside a 20th century English novel. At least, in my opinion. I just don’t think it is likely a contemporary of Cicero ever heard “zd”, as it wouldn’t have been normal for the Greek speakers of the period.

My vices

On recording and editing these videos, I noticed that I have a tendency to shorten final -ī into a short -i, as in when I said “vēnī, vīdī, vīcī” or “Annō Dominī”. mea culpa! Those final -ī sounds should have been long.

Let me know if you hear any other vices in my pronunciation, and I will endeavour to work on those in my future videos.

Grammar Analysis scores don’t correlate with unseen translation ability

Posted on

Some of the types of assessments Latin teachers have been traditionally setting are quite weird, the kind of things that aren’t normally done in any other languages. Consequently, it is hard to find recent research on whether what we are testing really matters.

Perhaps one of the strangest things we do is Grammar Analysis, where a student is asked to remember the “ablative of this” and the “genitive of that”.

A year or so ago, there was an alarm among Victorian Latin teachers when (due to an unfortunate miscommunication from the writers of the curriculum) we thought that Grammar Analysis would be completely removed from the new Year 12 Latin study design. At the time, we reacted quite defensively. We attended a consultation meeting with the curriculum committee where we went around the table repeating all the same things: that we needed Grammar Analysis in some form in the year 12 study design, because otherwise we would lose the rigour of the subject, students wouldn’t bother to learn grammar at earlier year levels, and the quality of unseen translation would drop.

It was a drama and half, and I shouldn’t say more about the episode, except that we got what we asked for and Grammar Analysis (“questions on accidence and syntax”) was prominently reinstated into the Year 12 study design.

But, was Grammar Analysis really the right hill to die on?

The new Victorian Latin study design gives leeway for how deeply teachers want to teach Grammar Analysis, as it is only internally assessed. How much should we emphasise Grammar Analysis among our students? How useful is it to be able to identify a subjective vs. objective genitive, or a dative of person judging, or a circumstantial participle?

For better or worse, the skill we care most about in Victorian Latin is the unseen translation. It remains just under half the marks of the final year 12 exam and is one of the best discriminators of student performance in the subject.

I decided to look into the assessment data of my year 10-12 students in the years 2018-2021 to see what metrics are the closest predictors of success in unseen translation.

Limitations

Some limitations with my study are that correlation does not equal causation – I expect at least some of these metrics will align to unseen translation without actually helping translation, because relative levels of diligence and motivation would cause the same students to do well in all hard assessments. If I plotted how the students did in Maths, it would probably correlate somewhat to how they did in unseen translation. That wouldn’t necessarily prove that skills in Maths help students translate Latin.

What this study might show, however, is a relative lack of correlation. If we expect a type of test to legitimately test student’s language skills, then we expect it to correlate with their ability to translate unseens. If we see a scattershot, perhaps we are not really measuring language skills that students use in the unseens.

A further limitation is the size of my sample. The number of data points in comparing performance within the same year was 41 (eg. John in year 10 vs. John in year 10, John in Year 11 vs. John in year 11), and the number of data points for comparing performance from a previous year to a following year was 24 (eg. John in year 10 vs. John in year 11, John in year 11 vs. John in year 12). Only 17 unique students made up the three cohorts graduating year 12 in 2019, 2020, and 2021, and there was some attrition as students moved schools during this time.

However, a positive of this data set is that all the students had the same Latin teacher the whole time (myself), and the way the assessments were taught and marked and their relative difficulty was consistent year on year.

With all this, let’s see what test scores appear to correlate most strongly to high scores in unseens.

Previous performance on unseen translation

Multiple R = 0.7243

In the graph above, I plotted the average unseen translation scores for the same student in two consecutive years. The horizontal axis represents average unseen scores in one year, and the vertical axis represents how well they did the following year.

This is a bit of an obvious point, but being previously good at unseen translations is a really good predictor for how good you’ll be at unseen translation in the future.

The Multiple R value for this correlation is 0.7243 – the tightest correlation of all the data in my study. (a value of 0 would indicate no correlation, a value of 1 would indicate strong correlation.)

Other types of assessment

The main other types of assessment (excluding seen translation) we typically do in high school Latin include:

  • Grammar Analysis
  • Vocabulary tests
  • Oral assessments
  • English to Latin translation

Grammar Analysis

In Grammar Analysis, students identify accidence and syntax in seen or prepared texts. Questions can include:

  • What case is this noun, and why was this case used? (eg. Posessive Genitive)
  • What is the person, number, tense, voice, and mood of this verb?
  • What use of the subjunctive is exemplified by this word?
  • What part of speech is this word?
  • What word does this adjective agree with?
Multiple R = 0.4039

Multiple R = 0.3646

In the first graph, I plotted how well students did in Grammar Analysis (GA) versus their unseen translation scores in the same year. In the second, I plotted their previous GA score versus their performance in unseens in the following year.

Both graphs exhibit a very wide scatter. A student in the same year could get 100% in GA but less than 85% in unseens, or conversely, they could get 90% in unseens but 60% in GA. (Multiple R = 0.4039)

Moreover, the correlation between doing well in GA one year and then doing well in unseens the next year is even weaker. (Multiple R = 0.3646)

GA does not appear to be a good predictor of success in unseen translation, especially in future years.

Anecdotally, I think all Latin teachers have experienced students who show conflicting results in GA and unseen translation. When a student comes along with great instincts at reading Latin in context but does rather woefully at remembering the Ablative of Something, we worry that all the missing information about grammar categories is going to trip them up in translation in the future, so we might advise them to work on that weakness. Conversely, when a student is doing brilliantly at GA but can’t apply the rules in translation, we might have advised them to re-read their grammar notes on such-and-such that they missed in the unseen, because they seem to respond well to explicit grammar instruction when they do GA. I’m not entirely convinced that these approaches are helpful.

If Grammar Analysis is so vital to good unseen skills, it is not clearly indicated as such by my student data.

Vocabulary

In this study, I only included vocabulary testing where it was in the traditional format of “provide one definition for this word”. The direction is always Latin-to-English. At year 10 level, I use the 20-word chapter vocab lists that come with my textbook series, but my vocabulary tests at year 11-12 level draw from very long lists of Latin words in order of frequency. For example, 100 word test is drawing from the top 100 words in Latin, the 200 word test is from the top 200, etc.

Students in these cohorts have learned lists up to 500 words long.

The typical method for learning the vocabulary is through online flashcards such as Education Perfect or Quizlet, where students write or select single English meanings for each Latin item.

Multiple R = 0.4971
Multiple R = 0.4399

Single-item vocabulary tests like these have only a slightly stronger correlation to unseen translation performance than GA scores. The scatter is very wide, especially at the lower end of performance.

I was expecting vocabulary testing to correlate strongly with “diligence” or “motivation” and thus be at least somewhat closely correlated to unseen translation skills, but it seems that being able to do well on a vocab test is not a great predictor of unseen translation performance.

In my experience, when students encounter their vocab words in a real sentence, they have trouble making the right decisions as to what meaning that word should have in its context. I tend to be pretty lenient on vocab tests and allow them to just state one meaning of the word, but in an unseen translation, some meanings are definitely more appropriate than others. An additional difficulty is that students see vocab words in “dictionary format” (with all their principal parts) in vocab tests, but need to be able to recognise them in inflected forms in an unseen. When I experimented with giving vocab tests with inflected forms of the words students were learning on the lists (eg. cēpit instead of capiō, capere, cēpī, captum), vocab scores dropped by about 50% compared to traditional tests.

Having a large vocabulary is extremely important to being able to fluently read any piece of writing. However, single-definition traditional vocab tests do not very accurately capture a student’s ability to recognise vocabulary in a translation.

I am currently looking into better ways of encouraging the development of vocabulary that responds well in context. Some of this includes experimenting with picture vocab lists where students must answer a question out loud in Latin which contains the target vocabulary feature.

Oral

My oral assessment involves getting the student to read aloud a passage of Latin, with attention to accuracy of pronunciation as well as dramatic expression. I mark students on a rubric containing five criteria: consonants, vowels, accentuation, fluency, and expression. During Covid years (2020-2021), I required students to create voice recordings, but during pre-Covid years, I listened to them perform individually.

The oral assessments were essentially just testing pronunciation, with a bit of performance flair thrown in. It was not a communicative task in the sense of simulating a live conversation in Latin.

Students were given oral assessments in years 10 and 11, but not in year 12.

Multiple R = 0.3788
Multiple R = 0.6114

Scores in oral assessment were not good predictors of student achievement in unseen translation within the same year, but intriguingly, they were somewhat correlated with unseen translation in the following year. (Multiple R = 0.6114)

Why might a student who did well on an oral one year not get the benefit of that oral in the same year but have a lag-time when they would do better on translations the next year?

I’m speaking speculatively, but it would make sense that good pronunciation might have a positive effect on contextual vocabulary acquisition.

If a student worked hard to improve their pronunciation before an oral assessment, in theory they would spend the rest of the year reading Latin with a clearer sound assigned to the words. In addition, the effort of making their performance more dramatic would involve thinking carefully about how words should be naturally grouped together into phrases, which would help in later recognising phrase boundaries in readings.

The cumulative effect of having better pronunciation could be improvements in reading that eventually improve their unseen translation abilities.

I say all this knowing that my oral assessments are quite limited in scope and my data is still pretty fuzzy, but it makes me stop and think about the significance of oral assessments if they have more of an impact on the next year’s translation ability than last year’s vocab tests and GA scores.

English to Latin translation (E2L)

Students are presented sentences in English which they must then accurately translate into Latin. The most common errors involve adjective agreement, but often there are little rules like “this verb takes the dative not the accusative” that continually pop up. Ambiguity in English is often a source of angst – should you use “suus” or “eius” for “his”? It is also brutal if students haven’t completely internalised all the types of clauses, such as which ones use the subjunctive and which ones use accusative and infinitive. This is a high-stress assessment to pretty much all students.

Students were tested on E2L in years 10 and 11, but not in year 12.

Multiple R = 0.6522
Multiple R = 0.4901

English to Latin translation (E2L) is legitimately a very challenging task. Unlike a vocab test or a grammar analysis, simple rote-learning will not cut it here. This test cannot be brute-forced with memorisation: similar to unseen translation, it requires students to really, really know their stuff.

Interestingly, E2L scores are relatively strongly correlated to unseen translation within the same year (Multiple R = 0.6522) but not as strongly correlated to the unseen skills for the following year (Multiple R = 0.4901).

I can speculate as to why the E2L didn’t have as strong a correlation to the following year’s unseen translation progress. The last E2L test my students ever face is in the first semester of Year 11, and it typically excludes some final pieces of challenging grammar such as conditionals and gerundives. While studying for their final E2L, they work hard on consolidating previously learned grammar structures such as the uses of the subjunctive, and the accusative and infinitive phrases. This is very useful to the unseen translations in year 11, which are typically a bit easier than the year 12 unseens. As unseens get filled with harder grammar not covered by their final E2L test, the correlation between how they did back then and how they do with the last pieces of grammar becomes weaker.

That may be one possibility among many. I don’t quite know what is going on, because it could also be possible that E2L is a really good measure of the same qualities (diligence, motivation) that produce students with good unseen skills. It could also be a measure of their internalisation of the Latin language, or conversely their ablity to consciously apply learned grammar rules.

Ranking the correlations

These are the test scores in order of most correlated to least correlated to performance on unseen translations:

Multiple R
Unseen Previous vs. Next year0.724301
E2L same year0.652156
Oral previous year0.611353
Vocab same year0.497062
E2L previous year0.4901
Vocab previous year0.439875
GA same year0.403941
Oral same year0.378783
GA previous year0.364568

Grammar Analysis ranks very low in this list, suggesting that it is not drawing from the same skills or characteristics that make a student good at translating unseens.

The most highly correlated factor was how well students did on unseen translation in the previous year.

The second most highly correlated factor was how well students performed in English to Latin translation in the same year.

The third most highly correlated factor was how well students performed in an oral performance of Latin in the previous year – their pronunciation and ability to add dramatic expression to a passage of Latin.

Conclusions

Greater attention needs to be given to whether we are testing the right things, and whether the tasks we set are really helpful to improving our students’ ability to read Latin in context.

I don’t like how much pressure and stress English-To-Latin translation puts on students, and it is no longer a requirement in the new study design for year 11; however, it seems to have been measuring something that really correlated closely to student performance in unseens, at least within the same year. This something could have been the students’ internalisation of the language, or it could have been their ability to monitor grammar-rules, or it could even have been their diligence and motivation.

Oral performance and pronunciation could play a much larger role in improving unseen translation skills than Victorian Latin teachers have given credit. The oral assessment is no longer a requirement in the new study design for year 11, but most teachers I have spoken to had already treated it as a very minor task or not even a formal assessment. The benefits of good pronunciation, if my limited data is showing something, are most observed in the longer term as students become better at acquiring language features. If this is the case, then we should especially emphasise oral skills and pronunciation in the early years so that the process of internalising Latin and gaining reading skills is enhanced throughout the course of study.

Grammar Analysis has very little correlation with students’ ability to translate unseens, both in the same year and in future years. It is one of the least (if not the least) legitimate language test in terms of predicting ability in unseens.

Similarly, but not quite as bad, are traditional vocabulary tests (“provide one definition for the Latin word”). Even if these are very useful words taken from large, frequency-ordered lists, doing well in a vocab test is not a good predictor for how well students can recognise the words in real contexts.

Why I’ve changed my mind on Comprehensible Input (but still can’t stand LLPSI)

Posted on

I used to seriously question the Comprehensible Input method for learning ancient languages, but now I strongly recommend it as fully authentic and the best method for gaining reading fluency and becoming a lifelong learner of ancient languages.

However, I still cannot bring myself to recommend the Lingua Latina Per Se Illustrata textbook (LLPSI) as a standalone Latin course. I understand that it works. And I am glad that many motivated people have acquired Latin through it. However, I cannot bring myself to like it, even when I try as hard as I can.

Let me start with how I had formerly come to reject Comprehensible Input together with LLPSI. I learned Latin through a reading method with the Cambridge Latin Course. In practice, my teachers taught me Latin through a mixture of reading pleasurable stories, getting us to orally perform the stories, and drilling our grammar with grammar-translation exercises. I was initially intrigued by LLPSI because I had heard it was a more intuitive and reading-based course than the one I had learnt from, and I believed reading – any kind of reading – was inherently more enjoyable than grammar-translation.

However, when I read LLPSI myself, or used it as a resource for Latin tutoring, both I and my students found it very hard to endure. I couldn’t put my finger on it, but something seemed very unpleasant about the stories. The narrator seemed to dwell on bodily harm in chapter 3 when he wrote the onomatopoeia “tuxtax, tuxtax, tuxtax…” to the sound of a child being beaten (the illustration showed a man with a face wrinkled with anger, hitting the boy on his lap with a stick). The incident was neither funny nor tragic. The boy’s torment simply happened. The tone seemed to lack human empathy, but presented all things, whether good or evil, in clinical detail. Frequently the stories did not follow the conventions of narrative, of motivated characters driving a plotline (such as the stories in the CLC), but instead were written like information dumps, with the story being merely a framing device for saying more words, more horrible words. My dislike of the book’s construction hardened; it grew into a burning hatred the more I wanted to like it, and the more I read of it (except for the chapter on pirates – that was actually very interesting. There actually were a couple chapters in the middle that weren’t too bad).

Meanwhile, I was tutoring a lot of students from a nearby school who had significant gaps in their Latin learning because for a whole year in middle school, a Latin teacher there had taught them from LLPSI. Maybe this wasn’t LLPSI’s fault – perhaps the teacher or the students were sloppy, or the class was noisy, or it was because of the sudden change of Latin courses. Who knows? But I certainly did not hear a ringing endorsement of the LLPSI course from the boys I tutored. They said they were told to “work out the general meaning of the sentence from context” when reading LLPSI, and be content with not having every word exactly right (as I later learned, this is not how CI should be implemented). I asked them to interpret sentences from LLPSI aloud for me and they were getting every second word wrong. I drilled them on grammar to fill their gaps, and they learned the rest of their Latin through a mixture of reading and grammar-translation methods – the Oxford Latin Course.

As this was happening, I was reading only ringing endorsements of LLPSI on the internet from people who supported Comprehensible Input as a learning method. I couldn’t reconcile this with my experience. I came to the conclusion that these internet reviewers were extremely motivated adults who didn’t mind reading stories about boys being whipped, or info-dumps spelling out things like “it is necessary to breathe to be alive”. These self-learners could not access the Cambridge Latin Course or Oxford Latin Course due to the steep price (at roughly $60 per book, the OLC would set you back at $180, while $50 a book for the CLC makes it $250). Their only affordable alternative to LLPSI was the driest of all textbooks – the extremely dense grammar-exercise courses like Wheelock’s Latin, Jenny’s Latin, So You Really Want to Learn Latin, Latin for Dummies, and so on. Given that choice, I thought, it’s no wonder they supported LLPSI. I concluded that all the twaddle about Comprehensible Input was people justifying to themselves why they should keep persisting with LLPSI and making themselves feel like they are accomplishing more than other Latin learners, as a motivation tool to get through LLPSI.

Fast forward to this year. I recently came across a very well executed Comprehensible Input (CI) ancient language course – Aleph with Beth. This is a biblical Hebrew course taught entirely in spoken Hebrew, with props, gestures, and stock footage of animals, people, and natural scenery. The videos are very simple in some ways and very complex in others. They are not structured as stories, but as an instructor telling you joyfully about the things that exist. As I watch these videos, I feel like I am discovering the Creation afresh. She holds up a realistic miniature horse figurine and says “SUS!”, and now I know, that is indeed a “sus”. She points at two horses together and says “SUSIM!” and I know, yes, they are “susim”. And then I am hooked to the screen. I am mimicking what she is saying. I am asking and answering her questions out loud. I am listening and echoing and acquiring the Hebrew language, and just by paying attention, I am 99% confident of the meaning of the words she says.

This is going to sound strange, but I never considered myself a “gifted with languages” person or a “language lover”. I liked Latin specifically, and then Greek specifically, and then got into self-teaching Hebrew through a grammar method. Each time, there was a lot of hard work and joy mixed together, but it was worth the hard work because I then got to read ancient texts in their original languages. Most of the time the real payoff for my efforts would come months or years down the line.

Aleph with Beth is inherently engaging from the start. It does not feel like work, and it is not boring either. It respects you and doesn’t condescend, but it also takes as much time and joyful repetition as you need to learn each word and structure one at a time. After watching the videos my brain feels tired, but while watching them, I feel energised.

I have been inspired by Aleph with Beth to work CI strategies into my Latin teaching practice, with very pleasing results. I might share these strategies in more detail in the next few blog posts, but in summary, I have been showing CI Latin videos from ScorpioMartianus that were adaptations of LLPSI. I have also implemented the 7-step-Ranieri-rereading method with a variety of year levels, from total beginners in year 7 to our final year classes in year 12 studying Vergil’s Aeneid. I have made a vocabulary flash card set for my year 9 class with only pictures and prompting questions (eg. a picture of a book with the question, “quid est?”) so that students see the picture and have to orally answer the question by saying the target vocabulary word. I have been rewording grammar explanations to be more about using Latin phrases to explain other Latin phrases. All of these things have been immediately showing good results – students internalising and recalling the vocabulary, reading prepared stories at the speed of speech, understanding even complex grammatical constructs without first having to force them into an English pidgin.

I was so happy with the results of CI strategies that I felt that I should pick up LLPSI again and see if it actually makes sense as a course now. I liked the LLPSI adaptation videos by ScorpioMartianus. I liked Aleph with Beth. I respect Luke Ranieri as a very capable instructor, and he recommends LLPSI as the best Latin course ever made. Surely now I can actually put aside my prejudices and finally like LLPSI.

I read a few chapters again and think “Dear God, I really hate LLPSI. What is wrong with me?”

The position I am left in is this: I love Comprehensible Input, and I can’t stand LLPSI.

And now I think I know why.

When you watch Aleph with Beth, Beth sits in a clean and bright room. She holds one prop in front of her. There are no distractions. She shows you the concept – a horse. She asks what it is. Then she says it is a “sus”. She repeats “sus” a couple more times, gesturing to this new discovery. Later in the lesson, a video (or two, or three) of a horse galloping across a field flashes brightly on the screen. After our eyes register this, she says “sus” again, but I am already starting to say “sus”. She asks her assistant – her husband Abram – to show her “sus” from among several other animal models. He points at it and is praised, or else he points at the wrong animal and is gently corrected.

In Aleph with Beth, you are shown the concept before learning the word for it. You see a horse and then are told it is “sus”. You are shown 3 or more images of the same thing. The props are lifted up, and you see her tangibly interact with them in three dimensions. You have a very vivid image of the thing that is being talked about.

When I read LLPSI, it feels like I’m seeing about 20 new words per paragraph. Some of these have pictures somewhere else on the page. But what predominates is the text itself. You are supposed to read the text first and then use it and your intuition to work out the meanings of the unknown words and structures. Some of the glossed definitions are written in other Latin words. Sentences with new vocabulary are just as long as sentences with old vocabulary.

The problem here is that LLPSI tells you the word first, which at first is nothing but a set of letters, and then you have to exert effort to find the meaning. Aleph with Beth shows you the actual thing first, and only then do you learn its name. In Aleph with Beth, the meaning comes before the language.

I find that this mimics the process of a child acquiring terms in their first language, especially in what I have seen of the Montessori style of education. First the child is shown a set of wooden cubes arranged according to size. They pick up a bigger one and feel its greater weight, then pick up a smaller one and feel its lightness. Only after that experience are they eventually taught the words “big”, and “small”, gradually moving to “bigger”, “biggest”, “smaller”, “smallest.” The meaning precedes the terminology.

A second practical problem I find with LLPSI is the printed format of its question-answer dialogues inside of the main chapter stories. For example, a paragraph could consist of the following sentences printed together:

Num Crēta oppidum est? Crēta oppidum nōn est! Quid est Crēta? Crēta īnsula est. Num Sparta īnsula est? Sparta nōn est īnsula! Quid est Sparta? Sparta oppidum est. Rhēnus quid est? Rhēnus est magnus fluvius. Num ōceanus Atlanticus parvus est? nōn parvus, sed magnus est oceanus.

In a video, these questions and answers work well, because you don’t see the answer right next to the question. You have a moment, however brief, to interpret the question and start to say the answer, before the narrator actually gives the answer. But when you read this, the answer is already there. I find I skim through these written questions and answers very quickly and without much motivation, but when the same dialogue is presented in a video format, I try to get the answer before the narrator can. I find a similar thing with my students when I show them CI videos: a number of them will try to say the answer out loud ahead of the narration. It’s a thrill to beat the narrator. It is no joy whatsoever to have to read an answer printed inline next to a question.

I feel like the video adaptations of LLPSI have achieved what LLPSI in itself has struggled to do. The course cannot be blamed so heavily, as it is a remarkable thing to have held up over 50 years. I think the greatest problem is the physical limitations of a book format: there is a limit to how many repetitions of words and pictures can fit in a printed book without it starting to seem wasteful to the environment, but a video series can take its time on repetitions of minute details one by one, and show a thousand times as many images in the same period of time.

For those who are currently using the LLPSI, I would personally recommend listening to an audiobook of the passages (anything people have read aloud of it on youtube or elsewhere) and would encourage you not to be ashamed to look up a translation of LLPSI in your native language to ease the process of initially comprehending the text, if necessary. Re-reading the Latin while re-listening to LLPSI passages would certainly be a good reinforcement for the target language features, even if the meaning was initially found through your first language.

I greatly look forward to seeing more comprehensible Latin videos being made, and I would also like contribute to this effort by making some CI videos of my own. I have seen the results of well-implement CI methods and I cannot endorse them enough.

When ‘comprehensible input’ is not enough

Posted on

I’ve been on Latin reddit discussions for years and one single educational theory comes up again and again as if it were the only way to learn a language: Krashen and his comprehensible input hypothesis.

Put simply, a learner should be introduced to the each feature of the language incrementally, by receiving input that contains their previous level of competence plus the next feature (i + 1).

This is extremely similar to Vygotsky’s Zone of Proximal Development which can be summarised as “for optimal learning, give a student enough challenge but not too much”. And “if a challenge is too great, give them scaffolding (i.e. help) until they can do it with the scaffold, then gradually take the scaffold away.”

There is sound wisdom in advice as basic as “give them just enough challenge” – that it is what we aim to do in teaching every subject, not just languages.

The error comes in believing that if a student successfully translates a sentence which happens to contain the target grammar feature, they necessarily understood the target grammar feature.

I see this all the time when teaching the accusative case to a beginner level class of mixed ability.

A lot of textbooks, when introducing the accusative, just keep on piling on sentence after sentence in the format of Subject-Object-Verb (SOV):

puer puellam videt.

The boy sees the girl.

As teachers we try to verbally explain that the ending of “puella” changed here, and this is the accusative case and you should pay atttention to it. That’s all well and good, but the experiential learning is more powerful for the student, and it often overpowers the verbal, abstract, theoretical explanation. What the student learns experientially from seeing SOV sentence after SOV sentence is that “the first noun I see is the subject, the second noun I see is the object”. By contrast, a single-letter difference on the end of the second word seems of minor importance. “Oh, so the second word gets a spelling change. Cool. Well I already guessed it was the object – correctly! – without noticing the one letter difference. I don’t need that letter.” And the more SOV sentences they see, the more their reliance on assuming the SOV word order is reinforced.

Step back and think how little a change an -m is on the end of puellam. When you are reading, you don’t read every single letter in every word. You glance at each word and it triggers recognition. Ifm there’sm one differentm letter sometimesm you can still recognise what word it’s supposed to be, and ignore the extra letter, just like you ignored the extra -m’s in this sentence. Seeing puella and puellam as variant spellings of the same word does not seem like that big of a deal to a lot of beginner Latin students, if they even registered there was an -m at all.

So now, I’m going to use Lingua Latina Per Se Illustrata (LLPSI) as an example of how ‘comprehensible input’ can reinforce an unintended habit. I chose this because the Krashen crowd likes this book. However, it is far from the only textbook that does this. I could have chosen almost any.

Chapter 3 introduces the accusative case. Therefore, I tallied up the word order in all the 41 sentences containing an accusative case noun in Ch. 3 (but I excluded sentences near the end which contained the relative pronoun, which I would treat as a seperate and more difficult topic).

These are the stats: a student can correctly assume that the first noun is the subject in 36 out of the 41 sentences. That’s 88% for just assuming English-like word order.

In the remaining five sentences, a student can try translating the first noun in each phrase as the subject and not necessarily believe they made a mistake from the context. If the teacher does not immediately correct them, many students will work on the assumption that these answers could be correct:

Quīntus īrātus est et Mārcum pulsat.

*Quintus is angry and Marcus punches. (in the next sentence, Marcus indeed punches Quintus, and Julia then complains that Marcus punches Quintus.)

Iūlia plōrat et Aemiliam vocat: “Mamma!”

*Julia cries and Aemilia calls. “Mamma!” (It makes sense that the mum might call to her daughter to figure out why she’s crying. Then again, it would be odd for the next bit of speech to be Julia’s, but sometimes the word for “she says” is left out. And sometimes students just forget who is meant to be the mum or daughter.)

pater dormit neque tē audit

*Father is sleeping and you don’t hear [him]. (That’s why you didn’t know he was asleep, you just didn’t hear whether he was awake or asleep)

Iūlius Quīntum videt eumque interrogat

*Julius sees Quintus and he asks (the father really does do the asking in the next sentence.)

Mārcum verberat, quia puer improbus est.

*Marcus is hitting [people], because he is a rude boy. (Rude boys do beat up people.)

Students can use an incorrect method to translate sentences that happen to contain the accusative case, and apparently get it right almost all the time, reinforcing their erroneous assumptions. The input is comprehensible and it ‘contains’ that one extra grammar feature (i + 1), but it does so in such a way that it gives the same reward and reinforcement to ignoring the accusative case ending as it does to actually paying attention to the endings.

There are a variety of solutions to this particular problem – for example, a teacher can train students on three word sentences with totally random word order (SOV, OSV, SVO, OVS), showing that they must rely on word endings alone for meaning. I don’t prefer this approach personally, because it gives the impression that Latin word order is random, as if it were concocted to trip up students like a puzzle.

Another approach is to give a “fill in the blanks” exercise to students where they have to put the nominative and accusative endings on the Latin words to show they know they’re supposed to be there. Unless you’re training students to compose in Latin (in which case this is a good scaffolding step), I think from a reader’s perspective, this type of exercise just reinforces to students that they can still read and understand the meaning of the Latin, even with the endings removed, so endings are really unnecessary for comprehension. No wonder Latin lost its endings in the Romance languages, if you can read “Marc__ Iuli__ pulsat” and already guess what it’s supposed to say! Noun endings are clearly superfluous to getting the meaning.

My favoured approach to teaching students to really heed the accusative case is to give them many examples of a sentence type I call “SOV-OV” and work on them in an environment that gives instant feedback:

The word order of the first part is Subject-Object-Verb (SOV), while the second part is Object-Verb (OV) with an implied subject carried over from the start of the sentence.

In this sentence, the format of the question shows that the answer must either be “and the daughter greets” or “and greets the daughter”.

The words have to be carefully chosen. The subject and object must be people (who, unlike inanimate objects, are capable of both acting and receiving action), and the verb must make sense as an action either person could do to the other. The student makes a choice not based on what they could assume from context, but only on the grammatical information in the word they are focusing on, and instantly, the computer tells them if their choice was correct.

Also, the word order in this sentence is something they are likely to encounter in real stories, as the implied subject is carried over without it needing to be restated. This is more natural than getting sentences in totally jumbled word orders, or having to insert an unknown “he/she” at the start of a sentence without a stated subject.

Just so that students don’t learn “the word in this position must be an object”, I sprinkle its opposite version amongst the example questions too. This is the SOV-SV sentence that would accompany the above example:

The instant feedback is key, or else students will believe they were right when they were actually wrong. I’ve been making these as online exercises via the platform Education Perfect, so that as soon as students select their answer, they are given instant feedback (and an explanation for why they were wrong, if they were wrong). You could also convert these into multiple choice questions on a free platform like Quizlet. If I have to teach in a tech-free classroom (a rarity in this pandemic now), I give students an answer key to their practice sentences and instruct them to correct their sentences as soon as they finish each question.

Habits (like assuming the first noun is a subject) are not broken by verbally explaining this to young students and hoping they do it on their own when they are caught up in the flow of reading. Students learn more powerfully by “doing”, and a poorly designed experience can teach them things we don’t intend. When designing comprehensible input for a target grammar feature, the reading activity has to reward only the right reasoning, or it will unintentionally reinforce a lazy and erroneous assumption.

Getting better at teaching scansion

Posted on

scansion4

Previously, I had taught the scansion of dactylic hexameter by giving a lecture of the whole system then getting students to have a crack at it with a copy of Latin text – and with no macrons printed either.

(I quickly learned not to use the opening of Aeneid 1 as the starter material, because line 2 is so aberrant – you could go hundreds of lines without seeing an –ia ending turn into a –ja like that! It’s awful! If I do use the opening of the Aeneid for learning scansion, I tell students to skip line 2, and all the while it stares at us awkwardly on the board.)

Now I’ve just introduced scansion with my year 11s, who are translating a section of Ovid’s Metamorphoses, and I’ve worked out a way to make the first introduction to scansion less of a lecture, and more of an intuitive workshop.

icarusanddaedalus_painting

Icarus and Daedalus – our set passage from Ovid, and perhaps a metaphor for how students feel when they first see hexameter scansion. [source]

My goal in this session is not to give students a step-by-step formula for answering exam questions – not yet. My goal is to bring them to understand scansion from first principles.

Continue reading →

‘Palatina Medea’ or ‘Medea Palatina’? A preference for adjective-noun word order in Latin

Posted on

We’ve been told that adjectives in Latin ‘tend to’ or ‘prefer to’ follow the nouns they describe.

But on the contrary, the statistical evidence shows that Caesar and Cicero actually preferred putting adjectives before nouns.

We didn’t learn that ‘noun then adjective’ rule from reading unadapted Latin. We didn’t discover it from real usage. We learned it from our textbooks and from word of mouth. ‘They’ said that it was true and ‘they’ must right!

So when Cicero dramatically calls Clodia a ‘Medea of the Palatine’, we think he should naturally put his adjective after the noun, but he doesn’t:

507px-George_Romney_-_Lady_Hamilton_as_Medea

George Romney’s painting of ‘Lady Hamilton as Medea’. Source: [link]

Sic enim, iudices, reperietis … hanc Palatinam Medeam migrationemque hanc adulescenti causam sive malorum omnium sive potius sermonum fuisse.

For indeed, judges, you will find out… that this Palatine Medea and this migration of hers [to Caelius’s neighbourhood] has been the cause of either all the ills or, rather, all the rumours for the young man. 

However, examples like this don’t change our minds about adjective-noun order. Since they told us that adjectives ‘usually follow’ the noun, that must mean that Cicero was violating the ‘natural word order’ for dramatic effect.

It doesn’t matter how many anecdotal counter-examples we read when actually encountering the Latin; ‘this was the rule’ when we learned Latin, therefore whatever we see later confirms it. Whenever an author puts a noun first, it’s because of the rule; and whenever an author puts an adjective first, it’s because he does it for effect; ‘the exception proves the rule’ (!!).

So let’s get into the actual statistics.

Continue reading →

Speed-running putting on a toga

Posted on

What can you do stuck at home inside? See how fast you can get yourself dressed for the Senate of course.

I started this challenge when someone said, “You need slaves or helpers to get yourself dressed in a toga – it’s so long and difficult to drape.” I understand that in practice, a wealthy senator probably did get a slave to help him dress; but I wanted to point out that it isn’t that hard to put it on quickly by yourself, and you aren’t required to be helped.

I think it is a matter of familiarity. We don’t wear togas much these days, and so we don’t have the years of practice that would make it come as second nature. But we succeed today in say, tying ties, tying shoelaces, slipping into tight dresses, and all other awkward clothes-wearing feats – as long as we do it often enough.

More videos to come in this series – I’ll post my personal best run soon!

Tabula: a strategic 2-player Roman board and dice game

Picture yourself planning for a Year 8 class in the final period of the day. It’s the second last week of term and everyone has finished their exams and have mentally started their holidays already. The class includes several (loud, influential) students who are not continuing Latin next year. They’ve probably been watching videos all day and I don’t want to contribute more to that. What do you do with them?

35efffc8260c89c01aaacc4121ad70e9

Pieces from the Gloucester Tabula board. Source: [Link]

How about play a strategic ancient Roman board and dice game?

Tabula was enjoyed for hundreds of years during the Roman period. It similar to Backgammon and relatively easy to explain.

For your convenience I have found here a version of Tabula which is playable, fun, and well explained. I’ve also made a printable board for it.

Continue reading →

The accent of words ending in -que

TL;DR: Latin words ending in -que should be accented on the syllable before -que only if that syllable is (or has become) heavy; otherwise, the word should retain its original accent. If this sounds new to you, that’s probably because you’ve been following Allen & Greenough and other nineteenth century scholarship.

SPQR Roman inscription augustus imperator

The rules of accent in Classical Latin are usually very simple. Almost all words follow the formula of the ‘Penultimate Law’, which states that the accent in multiple syllable words falls on the second-last (the penultimate) syllable if this is of heavy quantity, and otherwise on the third-last (the antepenultimate).[1]

But the enclitic –que (and the other enclitics, –ve, –ne, –ce) complicates these rules.

Continue reading →

Latin tutoring: Practice literary and context questions for Aeneid VI

The golden bough, ticket to the underworld in Aeneid VI. I painted this in my Year 12 - check it out at my deviantArt gallery

The golden bough, ticket to the underworld in Aeneid VI.
I painted this in my Year 12 – check it out at my deviantArt gallery

I’ve been captivated once again by the wonderful style and substance of Vergil’s Aeneid.

But this year I’ve been particularly nerding out because my three Year 12 Latin tutoring students are all studying book VI, the journey to the Underworld, which was the book I studied when I was in Year 12.

Continue reading →