OCR – Emma Williams: Latin tutor

Ghosts in the Machine

The last time I wrote about the professional use of generative AI was exactly a year ago. I had discovered that Chat GPT was surprisingly effective at constructing Latin sentences and was delighted that I could use it for such tasks as churning out examples of the ablative absolute. That’s about as far as I expected its capabilities to go and it already felt like a jolly useful gift.

Since then, I have made what I now realise is shamefully limited use of this quite remarkable tool. Only now am I just beginning to understand its full potential and the first epiphany has been that to experience those, one has to pay for it. Who knew? The fee is not huge, in fact I am currently working on the most affordable package which costs me £7 per month. With the work that it helped me to produce in just a couple of hours, it has already paid for itself one hundred times over.

Teachers, you must listen: if you have not already done so, make exploring the capacities of AI for resource-building a priority this summer. It will pay dividends. As this year’s GCSE Latin exams draw to a close and my attention is turned to the new set texts for the coming years, I promised myself that I would explore the possibility of using AI to generate materials. I have been blown away by what it can do.

Recently, I agreed with one of my lovely contacts that, during the summer holidays, he would give me some guidance on how to use AI to do this. Although I have started to have a go on my own, I still plan to meet with him, as the benefit of talking to someone who is much further down the road with something than you are is enormous. Simon Beasley, a fellow tutor and classroom teacher, has been using AI to generate resources, reduce workload and raise standards within his department for some time. I now understand how it is that the new generation of Classics teachers are pumping out such huge quantities of high-quality resources in their limited spare time. In my day (she says, waving her stick), I was doing so from scratch and I was pretty prolific in churning out stuff; these days, the sheer amount that my successors are able to produce is simply jaw-dropping. I had thought that they just had even less of a life than I did during my early years of teaching: I now appreciate that it is all thanks to AI.

So, what can it do and what are the pitfalls? Both questions are hugely important, because the use of AI is frankly dangerous unless you know what you’re doing. I don’t mean “know what you’re doing” when it comes to the AI itself — frankly, I haven’t got a clue what’s going on, how it does it and whether we are soon facing James Cameron’s rise of the machines. I’ll cross that bridge when we come to it. What I mean is, you need to know what you’re doing in the subject you are asking it to help you with. Because, fellow humans, AI definitely can and does hallucinate. Ignore this at your peril.

Firstly, what it can do. I have been using Chat GPT to help me to prepare practice papers and revision questions for the new set texts. Rather wonderfully, it told me how to train it to do so. I asked it what to do and it told me that I need to give it as many examples as possible of past papers, which is no problem as I have them all: this in itself was the first revelation of being a paid user — you can upload a seemingly limitless number of files and it will read and analyse them in milliseconds. What then happens is you enter into an extended dialogue with it as you invite it to start generating what you want it to create for you. You critique its output and it learns as you do so. Even more wonderfully, it evidentially starts to develop its own instincts for what it is that you’re looking for and reflects this back to you. Furthermore, due to its capacity to “look” at multiple examples of past papers alongside each other in a way that the human brain struggles to do, it is able to point out patterns, quirks and minor changes in how the board have presented and structured the papers over the years, some of which I had never even noticed. The fact that it can do so in seconds is merely an added bonus. To be clear, it is actually generating the resources for you, right down to the fact that it can produce them as a Word document for you to download and edit. It’s unbelievable. I am still reeling from the excitement at what it can do.

Now, on to the pitfalls. There are three things that I think new users of AI need to be aware of in order to make safe and productive use of it. Firstly, Chat GPT is only as good as the instructions and the material that you provide it with. Computer scientists back in the 20th century had already developed the acronym GIGO, standing for “garbage in, garbage out”, summing up the principle that if flawed, incomplete or incorrect data is entered into a system, the resulting output will inevitably be just as flawed and useless, sometimes more so. If you’re not clear and explicit, and/or if you provide it with limited exemplars, you won’t get anything good out of your project. So far, so obvious.

Secondly, Chat GPT can and does hallucinate. I don’t know enough about how AI works to understand why, but it is the spookiest thing about it. In some ways, I guess it reflects how it is not merely a computer but a generative entity with the capacity to learn, one which mimics human thinking; humans go off at wild tangents and get things spectacularly wrong, and so does Chat GPT. Being of the capacity that it is, way beyond what a single human brain is capable of, it does so on a spectacular scale and you have to be aware of this.

For example, for my first project, I asked it to help me to develop some practice papers and questions for one of the new prose texts, which is an extract from Apuleius’ Metamorphoses, the tale of Cupid and Psyche. I provided it with all the past papers from previous years on different texts, it analysed the board’s questioning style and created a style guide for how OCR formulates its exam papers. After a bit of back and forth with me pointing out what I felt it had got wrong, we were ready for it to start producing something for the new texts, so I provided it with the exact lines of the text that the board have prescribed. I anticipated no issue with this, having pasted the lines directly into the chat. Yet, when we were part way through the second paper, it told me that it thought “Juno’s speech” would be a good choice for the 8-mark style question. Now, I have prepared the new texts, but I do not yet know them well; despite this, I felt sure that to my memory there was no speech by Juno in the prescribed lines. I hunted through the published version of the prescribed text and then asked Chat GPT where it was getting what it called “Sections 14 and 15” from, since the board have divided the text into 13 sections only. It then cheerfully informed me that I was right: it had made up Sections 14 and 15.

This brings me to my third and final warning about my new best friend. It is remarkably charming and winsome. It is built to be agreeable and it rarely pushes back against you. It also presents its hallucinations as facts and, when you point out that it is talking out of its proverbial derrière, it cheerfully agrees with you. There is a danger here, perhaps in the same way that there is a danger when working with excessively amenable colleagues: you can end up in serious trouble, because there is nobody willing and able to tell you when you’re getting it wrong. To mitigate against this, I have learned is to poke it and challenge it robustly, to play devil’s advocate. Otherwise its capacity for people-pleasing could be quite catastrophic.

In my darker moments, I suspect that this is how it will take over the world. With a smile on its face.

Cambridge hangovers

The Cambridge Latin Course: love it or hate it, you can’t ignore it. Longterm readers of my blog and listeners of my podcast will be aware that I have been quite critical of the CLC in the past, despite the fact that it did form the backdrop to my classroom teaching for most of my career. While I continued to use the stories (albeit adjusted) and the characters from the course, I moved further and further away from its approach to grammar during my time at the chalkface and rejected its underlying principles (show, don’t tell) pretty early on. Towards the end I had completely re-written the curriculum and had stopped using the text books altogether.

Now, as a full-time tutor, I am increasingly aware of the legacy that the CLC has left Latin teaching and I am genuinely curious to know how long this legacy will last. Whilst many schools have ostensibly stopped using the CLC, its influence on teachers’ approach remains apparent in ways that many of them are perhaps not even aware of. In this blog post I hope to reveal some of the habitual oversights that classroom teachers of Latin are making as a result of what I believe is a hangover from the CLC curriculum.

One key blind spot for classroom teachers aiming to prepare their students for the OCR examination is a failure to teach the verb malo at the same time as they teach volo and nolo. I cannot explain this, other than a legacy of the fact that malo is not taught in the CLC when volo and nolo are taught. Taylor & Cullen introduce malo at the same time (in chapter 7 of their text book), but the overwhelming majority of students that I teach are reasonably well-drilled on volo and nolo but have never been taught the verb malo. Students following the WJEC/Eduqas syllabus do not need to know malo, but those aiming at the OCR examination need to know it, so to miss this tricky verb out of one’s teaching is a major oversight. I believe that this is purely and simply because schools are following curricula that were originally built around the CLC, which makes a big deal out of volo and nolo in Book 2, but never mentions malo.

Another legacy from the CLC which I have written about before is the decision to teach the purpose clause before the indirect command. It was many years ago now when it suddenly hit me what a massive mistake this was. I asked myself why students were so wedded to the habit of translating ut as “in order to” whenever they see it and realised that it is because this is how they first see it and after that they can’t let it go. I have yet to meet a single student who has been taught the indirect command prior to the purpose clause unless they have been taught by me, and this is genuinely fascinating. Every single Latin teacher seems to assume that it is a good idea to teach the purpose clause first, and I believe that the all-pervasive influence of the Cambridge Latin Course is partly to blame. Even Taylor & Cullen do in Latin to GCSE: despite mixing up the approach taken by the CLC (they teach ut clauses first, leaving cum clauses and the indirect question until later), they still take the decision to teach purpose clauses first. In my experience, this is a massive error, and leaves students convinced that ut always means “in order to” when in fact it only means this when it’s used in a purpose clause.

My final grammar-based concern when it comes to school curricula being based around the legacy of the CLC is that teachers are still teaching the perfect active participle as if it is a broad grammar feature. This is done in the CLC, which for some extraordinary reason introduces PAPs towards the beginning of Book 3, long before deponent verbs are even mentioned in Book 4. Students really struggle as a result, since they form the understandable belief that the perfect active participle is a grammar feature that is common to all verbs. They thus struggle with the concept that most verbs have a perfect passive participle because they have not been taught that perfect active participles only exist because of deponent verbs. I have to spend a great deal of time unpicking students’ misapprehensions and misconceptions about this, teaching them in detail about deponent verbs and their features and then mapping this onto their participle. It takes so much time to dispel these misunderstandings, which would never be there in the first place were schools to adjust the curriculum to introduce the perfect active participle solely as a feature of deponent verbs.

It is genuinely fascinating to observe the fallout from text book use and to be able to identify where students’ misconceptions are coming from as a direct result of the curriculum that many schools are adhering to. I do find it worrying that so few schools are asking themselves why they are using text books that are not built around the examination that their students are aiming at, not least because the vocabulary in those text books is quite often a monumental waste of time. While the 5th edition of the CLC goes some way towards addressing this, it doesn’t solve the problem entirely and too much of its old stucture and principles remains for the problem to be solved in its entirety.

GCSE Latin set texts – why students struggle

Few things risk being so damning as the insight of a one-to-one tutor. As an ex-classroom teacher myself, I am painfully aware just what a difficult job teaching is, and how it is entirely possible to leave some students behind, despite your best efforts. It is from this perspective that I come to this topic.

It is obvious and undeniable that many of the students I work with have been well-taught: they have simply lost their way or misunderstood for a variety of complex reasons. Others, I must confess, I do wonder what’s been happening in their classroom. Whatever the truth of the situation, once a student has indeed lost their way with their studies, it can be a Sisyphean endeavour for them to rejoin the road to success without support. As I write these reflections on what the students I am paid to help have missed and misunderstood about set text work, it is in the full consciousness that there will have been some members of my own classes over the years that became lost by the wayside. A classroom teacher who can claim otherwise is a rare creature indeed.

Set text work remains one of the biggest challenges that students face when they reach GCSE level in their Latin studies. Suddenly, there’s a whole new world of real, unedited Latin in front of you, some of it in verse. The expectation we place upon students to cope with this is frankly mind-boggling. Imagine asking a student of French to study Molière, Maupassant or Descartes at GCSE level: this is what we are asking students to do in Latin. The whole thing is frankly ridiculous, and I have written before about what a pointless exercise the whole business is, but given that the exam boards resolutely refuse to change their approach, we’re stuck with it. What follows are some observations about students who struggle with this element of the exam.

Perhaps the most striking thing I notice about some students’ understanding of the literature is the fact that those who are struggling with the set texts cannot articulate the very basics of what they are about. Teachers are often under enormous time pressure when it comes to the huge swathes of literature they must plough through, and – as a result – they often dive straight in to working through the text line by line, and do not find the time to ensure that their students understand the basic meaning and purpose of the text.

Currently, this is manifesting itself most strikingly with the Virgil text prescribed last year and this year for OCR (selections from the opening of Aeneid 1) and the Love & Marriage texts for Eduqas. For one student studying the latter, it took me more than one session with her to establish which texts she was studying, so non-existent was her grasp of what had been covered. With the Virgil, teachers have a particularly difficult task: how much to tell students who may have little to no knowledge of epic and/or mythological stories in general? Aside from this, however, is notable that not one single student that I have worked with during the last 18 months has had even the slightest inkling of an idea that Carthage had significance for a Roman audience. I find this genuinely sad. I cannot think of anything more important than explaining to them that the Carthaginian empire was a rival superpower that the Romans had overturned some 150 years before Virgil was writing. In a series of three conflicts between Rome and Carthage, Rome was ultimately victorious and utterly destroyed Carthage in 146 BCE. While the wars themselves were history to someone writing in Virgil’s time (the 1^st century AD), the experience and trauma of these conflicts, especially the long and harrowing campaigns of Hannibal, were a central and formative part of Roman collective memory and crucial to their self-definition. The Carthaginian Wars quite literally defined them as indefatigable warriors and the global superpower of their age.

Beyond this surely fundamental understanding of why Virgil is banging on about Carthage at the start of his epic work, no student that I have worked with understands or can define what an epic work is. I cling to the notion that they must have been taught this, but I can only assume that they are given this information in lesson one and that their teachers then assume that it has stuck. Such things are crying out to be used as a regular Do Now or similar quick retrieval task: what is an epic? Who was Homer? How is Virgil imitating him? A student should be able to tell us that an epic is a lengthy poem, written to be publicly performed, and focusing traditionally on tales of battle and self-definition; they should also understand that the gods and destiny play an important role in epic and that epic is a genre that evolved through the Greek oral tradition and that Virgil is doing something rather special by canonising this into a definitive Roman origin story in Latin. These basic notions really need to be revisited regularly to ensure that students remember them.

Beyond the fundamentals, the biggest mistake made by classroom teachers in my experience is their excessive focus on style, over and above teaching students how to learn the text. At this point, we come to the crushing reality and the reason why I believe that set text work is such a monumentally pointless waste of students’ time: the Latin is too hard for GCSE-level students to grasp in full, meaning that their only option is to rote-learn the text in English. Few classroom teachers labour under the illusion that this is not the case, but few also realise just how much guidance students need in order to do this necessary and time-consuming task successfully. When I was teaching, I learned to drill students on the best methodology for rote-learning, modelled it for them and then gave them short bursts of classroom time to start doing so, while I monitored them. It was essential, in my view, for me to see it demonstrated that students had understood the methods I had shown them and were trying them out. Students can be remarkably stubborn when it comes to study skills, and unless it is literally demonstrated to them that a method works, they will ignore your advice and go it alone. As a result, they will fail. Students who have been shown how to learn the text successfully come to realise that the demonstrated methods work and will stick with them.

The final issue with classroom set-text teaching arises out of a combination of two issues I have already raised: teachers being under time pressure to push ahead with the text line by line, combined with an excessive focus on stylistic features. What this means is that teachers generally introduce a new bit of text and talk about its stylistic features at the same time. The reality for novices is that this will be impossible to follow. My advice to students is always to attempt to get ahead of the class with the rote-learning, so that they are looking at a section of the text that they understand when their teacher starts talking about style. This gives them a better chance of following what the teacher is saying. When I was in the classroom, I would take the students through the meaning of the text and set them to learn it before I said anything about its stylistic features. It worked infinitely better than expecting them to follow what I was saying when working through a new bit of text.

Fundamentally, classroom teachers must remind themselves that students can achieve around 80% in the exam with only the haziest of grasps when it comes to the stylistic features of the text. The vast majority of their marks come from knowing the text, and yet this aspect of their studies is given the least amount of focus in the classroom. In their anxiety to help students with the most difficult aspects of the examination, many classroom teachers overlook the low-hanging fruit: how to help them to achieve the bulk of their marks.

Fake news: ancient style

The notion of “fake news” is generally considered to be a feature of the modern world. Yet, while the proliferation of false narratives and the digital means to both invent and spread them at high speed is indeed a modern problem, the issue of fake news is not in itself entirely new.

I have recently been ploughing my way through the four new Latin set texts listed by OCR for GCSE examination in 2026 and 2027. One of the prescriptions included two prose texts, one a letter by Pliny the Younger and another an extract from Attic Nights by Aullus Gellius. Both stories involve wild animals and I was vaguely familiar with both of them, indeed the one by Aullus Gellius is quite remarkably famous. Pliny claims that his story about dolphins is “true, but very similar to fiction” — the modern saying “stranger than fiction” would perhaps express the sentiment he is going for. He claims that the source of his story is unquestionably reliable and there is good reason to believe that the source is Pliny the Elder, who was the younger Pliny’s uncle and in many ways a father to the younger Pliny. Pliny the Elder tells a very similar story in his own work, so it seems plausible that the passionate academic and naturalist was indeed the source of this story. A shame, because the story is clearly hugely exaggerated.

The tale reports that a city in the province of North Africa had a large estuary, which the locals used for fishing, sailing and swimming. One day, some boys in the lake were joined by a dolphin, who engaged with them and played. So far, so plausible: dolphins are indeed famously intelligent and sociable and there are many documented modern accounts of dolphins playing with humans in the water. The truth of the story is stretched somewhat when the dolphin takes a boy on his back and carries him out to sea, but even this I could just about accept. What I cannot accept is that eventually the dolphin becomes so enamoured with the boy that he regularly hauls himself out of the water to spend time with him on land, returning to the water only “when he gets too hot”. No, Pliny, that didn’t happen. Fake news.

The tale by Aullus Gellius is the one about Androcles and the lion, a famous story about a man who escapes the dreadful fate of being killed by wild beasts for the entertainment of the Roman crowd. When Androcles is approached by the lion, it turns out to be the very same lion that he had helped with an injury and befriended whilst fleeing cruel treatment by his master, again in the province of North Africa. The story was later adapted and turned into a Christian triumphalist tale, with various versions popping up and placing a different protagonist at the centre of the lion-taming. The whole story is beyond ludicrous, but Aullus Gellius claims that his source (who is named as Apion Plistonices, a Hellenised Egyptian and the author of a work on all things Egyptian) “saw the event as an eyewitness” in the city of Rome. The event he claims to have seen was the lion refusing to attack Androcles, which in fact was not uncommon in wild beast hunts. The animals were starved but terrified and their handlers had little to no idea how to look after them. Most of them died en route from Africa and those that did survive usually had to be goaded into attacking their victims. So, the notion of a lion not leaping enthusiastically on its prey is perhaps entirely plausible; however, the idea that it “gradually and calmly approached the man as if recognising him, then moved his tail in the manner and way of dogs showing affection” does not sound in any way plausible to me.

There is no doubt that authors in the ancient world struggled with a significant amount of ignorance, thus they were vulnerable to fake news just as much as we are. Ironically, their vulnerability to fake news and propaganda was perhaps caused by the exact opposite problem to the one that we face: in the ancient world, news spread incredibly slowly and came largely through word of mouth. Was it easier for an emperor to maintain an aura of mystique when nobody saw him, except perhaps as an etching on a coin? Certainly, we have evidence that emperor worship was more common in the distant provinces, places where the reality of the man would never be seen.

Obviously, it was not just the way that news was proliferated in the ancient world that was the problem, it was their relative ignorance when it came to matters of science. While philosophy, studied and practised rigorously by many academics in the ancient world, was the mother of science and Aristotle is rightly hailed as the father of the scientific method, he and other philosophers had little past knowledge to build upon in the field of biology. While the ancient thinkers made quite remarkable observations in the fields of what we would now call mathematics and astronomy, they really were a right bunch of dullards when it came to biology, I’m afraid. I would strongly advise you not to take any of their health advice, particularly when it comes to female biology!

One of the most depressing things about the ancient philosophers is how tightly constrained they were by their own cultural blindness. For me, it is a sobering lesson that even the most brilliant minds in the history of time have been deadened by their cultural milieu. Take someone like Aristotle: I would argue that his intellect is one of the greatest that man has ever seen. His fascination with everything, his breadth of knowledge and interests, his capacity for learning, his ability to understand that morality is complex and nuanced (something really not grasped by any of his philosophical predecessors) and his tentative forays into what we would call the scientific method, all of these things and more make him a genuine phenomenon. Yet this man argued doggedly that some people are “natural slaves” and wrote a whole treatise on what a jolly good idea slavery was. Slavery in the ancient world wasn’t really questioned by any of the great thinkers. Some, particularly the Stoics, argued against the cruel and unusual treatment of slaves, but none really made the case that slavery in and of itself was an aberration of morality. This, more than anything, should prove to us that people are always at the mercy of the time in which they are born: however great their intellect, it remains very difficult for them to leap outside of the assumptions that they have been presented with as the cultural norm.

One of the many reasons for studying the ancient world is to be able to view a whole society through a dispassionate lens. It is much easier, for example, to talk about the concept of slavery when you are talking about something that happened in a society that existed 2000 years ago, whose ills do not feel like your responsibility on any level. It is much safer than discussing the much more recent transatlantic slave trade, for example, indeed I know plenty of modern historians who do not consider it to be an appropriate topic for younger students: it is, quite simply, too raw. Further than this, what few people seem capable of grasping is that the ancient world should give us cause to reflect on our own ignorance. If some of the greatest minds that have existed since the dawn of time could get some things so spectacularly wrong, then what concepts are we failing to grasp? What will our successors be horrified by in the future? What will they laugh at? How will we seem uninformed? What are our inevitable blind spots? What, indeed, are we lying about?

Photo of a recent anti-Trump protest at LA International Airport by Kayla Velasquez on Unsplash; nice to see the correct use of the subjunctive on a placard!

Who needs decent resources?

It is an absolute miracle. For the first time in the history of the subject, a publisher has produced a complete Anthology, containing all of the OCR GCSE Latin set texts for examination in 2027 and 2028. In an unprecedented move, someone has had the ground-breaking idea of actually publishing the resources that OCR wish teachers to teach and children to study. Such radical thinking can only be attributed to a stroke of genius.

Previously, it may surprise non-specialists to know, only some of the GCSE Latin literature texts were published in a modern format and only some of those publications were formally ratified by OCR. What an incredible leap of imagination it must have taken for the intelligentsia behind the wheel at OCR to think of the idea of a published Anthology of all the texts that they have selected, in the fancy modern format of a book! To be fair, they have been very busy coming up with their dramatic new rebrand, an imminent name switch from “OCR” to “Cambridge OCR”, billed in an email they sent me this week as “an exciting change”. Fundamentally, it means that a group currently called OCR, which stands for “Oxford, Cambridge and Royal Society of Arts” will now become “Cambridge Oxford, Cambridge and Royal Society of Arts”. I hope that’s clear.

Anyway, back to the majestic leap of imagination that is the new Latin Anthology. Not only has someone printed the texts out, they have even glued the pages together! It really is quite the thing. And get this. You can buy it through the publisher, you can buy it through bookstores, you can even buy it on Amazon! Did you know that you can purchase books on Amazon? Imagine my excitement. What relief and joy this publication will bring! Obviously, it will be aimed at students, will it not? Or perhaps aimed rather at teachers, as a complete preparation tool? I was breathless with anticipation. However, within five minutes of glancing through my much-anticipated purchase, it became apparent that this Anthology was an attempt at both of these things and a success at neither.

The first thing to note about the publication is the distinctly bizarre “endorsement statement” from OCR (soon to be Cambridge OCR) at the beginning. It states that while “the teaching content of this resource is endorsed by OCR” (for which I read that they’ve managed to select the correct bits of the text) we are told that “all references to assessment, including assessment preparation and practice questions of any format/style, are the publisher’s interpretation of the specification and are not endorsed by OCR.” Erm, okay. There follows some further language of accountability avoidance that goes on for quite some time, but the general gist is a clear and rather anxiety-inducing attempt by the board to distance themselves from the statement printed on the front cover, which is that the book is “endorsed by OCR”. Does this even happen in other subjects?! Maybe it does, but it seems distinctly odd. Either the book is endorsed or it isn’t, surely?

Things then get worse. The preface and “how to use this book” both seem to slide and shift constantly between the implication that the resource is aimed at teachers for preparation purposes and that it is aimed at students as a workbook. The result of this apparent attempt at dual purpose (or perhaps confusion/indecision as to the purpose at all), is unsurprising: the Anthology fails in its attempt to achieve either of these things. Whether this is the fault of the publisher or the authors is impossible to tell, but it really is a tangible fail.

So far, I have only worked through the Virgil text (extracts from Book II of the Aeneid), and I am already half way to despair. Firstly, despite its promise in the preface to students and non-specialist teachers that the book “aims primarily to help readers understand what the Latin means” there is one rather glaring omission. The authors do not provide an English translation of the texts. For the love of God, why not?! As a friend and fellow tutor put it to me in a message last week, “If [OCR are] going to be so picky as to what they allow … they might as well provide [a translation] and put everyone out of their misery.” Exactly this. In mark schemes over the years I have frequently seen phrases such as “do not accept [perfectly legitimate translation of the word in my reasonably well-informed opinion]”. So, teachers are still expected to somehow divine what it is that examiners will and will not consider to be an acceptable translation of every single word and phrase in every single text. It is genuinely exhausting and I simply do not understand why we have to play this game every single year. Just give us the translation that you approve of, for crying out loud.

The authors’ (or perhaps OCR’s) decision not to provide a translation causes further, compounding inadequacies in their notes, since they frequently fail to give sufficient thought to their suggestions for the translation of individual words. For example, they suggest the translation “waves” for both undas and fluctus, when those two words occur very close together and surely need differing translations to avoid confusion and to mimic the original Latin; in the same lines, the authors provide “raised” for the participle arrecta, then “rise above” for superant, which comes very soon after it. Following their instructions, this would render the lines:

pectora quorum inter fluctus arrecta iubaeque
Their chests raised above the waves and their blood-red crests

sanguinae superant undas.
rising above the waves.

Not only does this fail to do Virgil any kind of justice, it lacks clarity for the novice reader. The authors’ failure to sit down and decide how they would render a full and competent translation of the lines in their entirety (a task which will be asked of the 16-year-old novices who will be examined on this text) leads inevitably to some thoroughly confusing suggestions on their part for the translation of individual words. This is merely one example, but I found multiple cases throughout the Anthology which evidenced this lack of coordinated thinking.

In addition to the conspicuous omission of an approved translation and the knock-on effect that this has on the notes, the notes are disappointing in other ways. While some of them provide useful textual support, there have been times when I have wanted to wail in frustration. My exasperation stems from the authors’ palpable lack of clarity about the purpose of this Anthology, their inability to decide their target audience. Here is just one example of what I mean: at the end of the first section of the Virgil text, Aeneas claims reluctance to recount the painful story of how the Greeks sacked Troy. He says, quamquam animus meminisse horret luctuque refugit, incipiam: “although my mind shudders to remember and recoils in grief, I shall begin.” I would love someone to explain to me the purpose of the facing note in the Anthology, which relates to the final word of this section: it says, “what tense is incipiam?” Ugh. Obviously, I can tell you what tense incipiam is, because I am a Latin teacher: I do not need help with recognising the future tense. But if I were needing help with this (for example, if I were a student, or if I were a non-specialist who was wrestling with the material), then what is the point of asking me a question to which I may not know the correct answer? This is exactly the kind of infuriatingly pointless annotation that is useful to precisely nobody. For a subject expert, it is superfluous; for a novice, it is maddeningly unhelpful.

I am honestly quite a cheerful person, with a positive outlook. Yet, with so many people in institutions that have power and influence over my own working life so unrelentingly mediocre at what they do, it is becoming increasingly difficult to remain sanguine.

What GCSE students don’t know about the Aeneid

Since last week, when I wrote again about the power of one-to-one tutoring, I have had even further cause to reflect on its essential benefits.

It would come as shock, I suspect, to most classroom teachers, the extent to which students forget, misinterpret or loftily ignore what they have no doubt been taught in school. I say “no doubt” because I refuse to believe that students have never been taught the basic background to the texts that they are studying, despite their protestations.

What does happen, I believe, is that teachers over-estimate students’ ability to absorb and remember complex material. It certainly came as a shock to me when I started to read more about how memory works (a criminally overlooked field of study in my training) and came to realise just how much repetition is required for students to grasp the basics. In this blog post, I plan to outline the opening few lines of one of the current OCR set texts and explore the things that have puzzled, baffled and troubled the students I have worked with this year. I hope that this will enlighten readers as to the extent that some students struggle with complex material.

One of this year’s texts is taken from Virgil’s Aeneid Book 1. It starts at line 13, so as close to the beginning of the text as one could wish for. This potentially makes for a much easier life than the times when a set text has been taken from Book 10 or Book 12. One would have thought that it would be an easy task to get students to comprehend the basic facts of what the text is about and its core purpose. Well, one would have thought wrongly. With only one exception, the students requesting my help with the Virgil text this year have not been able to define what an epic is, nor were they able to say what Virgil’s purpose was in writing the Aeneid. Most of them swore blind that they’d never been taught the definition of an epic. Beyond this, they have all been baffled to the point of total and utter confusion as to who the Trojans were and what on earth they had to do with the Romans and their self-definition. So, let’s look at some extracts from the opening lines of the text and see in more detail what’s been troubling my charges.

urbs antiqua fuit, Tyrii tenuere coloni,
There was an ancient city, [which] Tyrian settlers inhabited,

Karthago, Italiam contra Tiberinaque longe
Carthage, opposite Italy and the far-distant mouth of the Tiber,

ostia, dives opum studiisque asperrima belli;
rich in resources and most formidable in the practices of war;

Out of those who have requested help with the Virgil, most of them were unable to tell me where Carthage was and why it’s described as a formidable stronghold. None of them – genuinely no exceptions – understood the historical fact that the Romans had destroyed Carthage over 100 years before Virgil was writing. While I would not for one moment expect any of them to have detailed knowledge of the three Punic Wars, I was a little surprised that none of them seemed to be conscious of the fact that Virgil was writing in a world in which this rival superpower had been razed to the ground decades earlier, and that this was a crucially important part of how the Romans defined themselves. Does it seem likely that this was never mentioned by any of their teachers? I think probably not. Is it likely, however, that this was perhaps mentioned once in the first lesson and then rarely – if ever – reiterated? That, I’m afraid, seems plausible. I think teachers need to think very hard about what’s happening in the first couple of lessons of set text work. When you present the students with the text, their minds are completely preoccupied with the length of it and how on earth they are going to cope with learning it; they are thus even less likely to absorb any background information you’re giving them.

Very few students were able to tell me what the Tiber is (a river in Rome, as iconic to the Romans as the Thames is to Londoners) and none of them seemed to understand how Carthage is “opposite” Italy. Carthage lay on the other side of the Mediterranean sea, located on the coast of north Africa, in what we now call Tunisia – indeed, it kind of bulges out into the sea and looks to be the bit of land mass in Africa that is closest to Italy. Perhaps it is because my own sense of direction and general geography is so embarrassingly poor that I always look all of these places and features up on a map and contextualise them for myself in detail. Do teachers assume that their students’ knowledge of geography is as sound as their own? Maybe so, and if so, I guess my advantage is that my own geography is so awful that I assume absolutely nothing! Anyway, the text and the description of Carthage continues:

quam Iuno fertur terris magis omnibus unam
[one] which Juno is said to have cherished more [than] all [other] lands,

posthabita coluisse Samo; hic illius arma,
valuing [even] Samos the less;

Now we’re getting on to the meat of the text and what Virgil is building up to in this opening section. He sets out to explain why Juno, the queen of the gods (most students didn’t know that, by the way), has a massive beef with the Trojans. Here, he highlights the fact that Juno values Carthage even more than Samos. What’s he on about? My students didn’t know. Samos, an island off the coast of modern-day Turkey, was the birthplace of Juno and a centre of her worship. The fact that she values it less than Carthage highlights the importance of Carthage to her and hence her overwhelming desire to protect it. This is why Virgil mentions Samos.

progeniem sed enim Troiano a sanguine duci
But indeed she had heard [that] a breed [would] arise from Trojan blood,

audierat, Tyrias olim quae verteret arces;
which would one day overturn the Tyrian stronghold;

hinc populum late regem belloque superbum
from this would come a nation, wide-ruling and superior in war,

venturum excidio Libyae: sic volvere Parcas.
for the destruction of Libya: thus were the Fates unrolling.

I have asked all of my students to tell me who “the breed that would arise from Trojan blood” are, which could absolutely come up as a one-mark question in the exam. Until I explained, very few of them understood that it was the Romans. They seemed genuinely unsure about the point of the Aeneid‘s opening, which is to highlight how difficult the goddess Juno made it for the Trojans to make it to Italy, which was their destiny. Why were they headed to Italy? Again, when asked, students had not grasped the fact that Aeneas and the rest of the Trojans were refugees, survivors of the Trojan War and in search of a new city now that theirs had been destroyed. It seems remarkable given current events in both Europe and beyond that students seem to find this resonant fact so easy to forget. Has the analogy with modern refugees setting sail across dangerous waters ever been drawn for them? I do hope that is has, but again, maybe that’s happened only once. Students had failed to grasp that the Trojans are trying to get to Italy and that Juno is trying to prevent this because she is trying to prevent the Roman empire from existing and thus to prevent the destruction of Carthage by the Romans. Now, here’s what’s really interesting: I have explained this multiple times and in multiple ways to several different students individually, and most of them have really struggled to grasp it. I suspect it’s partly because they are having to think about multiple timelines and this is difficult for younger people; I also think it might have something to do with the fact that some of what they are being told is historical fact and some of it is legend – they genuinely find it difficult to get a handle on what it all means and how it fits together. I am still thinking about how it could be better explained in the future, since it’s clearly a lot more difficult to understand than those of us who are subject experts realise.

necdum etiam causae irarum saevique dolores
not even now had the causes of [her] resentment and bitter griefs

exciderant animo: manet alta mente repostum
left [her] heart: deep in her mind remained the far-off

iudicium Paridis spretaeque iniuria formae,
judgement of Paris and the insult of her beauty scorned,

et genus invisum, et rapti Ganymedis honores.
and her enmity towards the tribe and the honours paid to the stolen Ganymede.

Here, Virgil lists the reasons that Juno has for hating the Trojans. It seems that students find this really difficult, too. This is perhaps because they must grasp two separate things: firstly, they must understand that Juno’s over-arching reason for hating the Trojans is that they are destined to give rise to the Romans, who will eventually destroy her beloved Carthage. They find this really difficult to grasp, as I explained above. In addition, they must also understand that Juno has some other more petty reasons for hating the Trojans, mentioned here by Virgil. She has a general enmity towards the tribe because it is descended from someone called Dardanus, who was the son of her husband Jupiter as a result of one of his numerous extra-marital affairs. Thus, the existence of the entire Trojan race was an insult to Juno. In addition (and this is the only story that most of the students seemed familiar with) there was the beauty contest between three goddesses that Paris, a Trojan prince, was given the dubious task of judging. His choice was ultimately the cause of the Trojan War, since the bribe he was offered by the winner (Venus) was the most beautiful woman in the world, which was Helen, who happened to be married to a Greek. Hence, when Paris claimed his prize, the Greek tribes waged war upon the Trojans. More importantly for our purposes, the fact that Juno was not selected as the winner of the contest was yet another slight against her by a Trojan. The third petty reason mentioned, the “honours paid to the stolen Ganymede” is all about Jupiter’s promiscuity again. Ganymede was a handsome Trojan that Jupiter took a fancy to and abducted, yet another insult to his wife. (Note: Ganymede was not, as one of my students was absolutely convinced of, a horse. Not that taking a fancy to a horse was beyond Jupiter, miind you, but that isn’t what happened in the story).

his accensa super, iactatos aequore toto
Inflamed further by these [things], she kept the Trojans [who were] left by the Danaans

Troas, reliquias Danaum atque immitis Achilli,
and by ruthless Achilles far-distant from Latium, storm-tossed in every corner of the sea;

arcebat longe Latio, multosque per annos
and for many years

errabant, acti fatis, maria omnia circum.
they wandered around all the oceans by an act of fate.

tantae molis erat Romanam condere gentem!
Such a great undertaking it was to found the Roman race!

Here, Virgil sums up his overall point: that it is Juno’s hatred of the Trojans and her fear of their impending destiny, which causes her to work against their journey and to thus postpone their fulfilment of fate. One of the final things that I have noticed students really struggle to grasp is that fact that Juno knows full well that she won’t succeed: as a goddess, she can see the past, the present and the future, and she knows that the destruction of Carthage by the Romans is fated and inevitable. Still, she’s going to do everything in her power to prevent, or at least delay, the inevitable. I find it interesting that young people should struggle to understand this very human kind of motivation – that we might still strive for something that we already know is doomed to failure in the longterm. I guess they haven’t had experience of it yet.

Before teachers feel too dismal, I should point out that I do tend to specialise in working with students who really struggle with the subject. That said, what has been interesting this year is that almost all of my students have struggled with this text, even the high-fliers. I hope that this post has given some food for thought. It is so easy to assume that students have understood what we have told them, so easy to imagine they are following what we say. Until we delve a little deeper – one of the immense joys of working one-to-one as I do now – we can delude ourselves that they have understood the point of a text and are following its meaning.

Latin language GCSE

Tomorrow the several thousand students studying Latin across the UK will sit their language examination. The Boards clearly collaborate when it comes to exam timetabling, so both OCR and Eduqas/WJEC have their Latin language examinations on the timetable tomorrow, both of them setting a paper that lasts and hour and a half.

Having worked with both Boards for three years now and having worked proactively through the existing past papers for both Boards in somewhat obsessive detail, I consider myself something of an expert on the quirks of each. Broadly, the Boards take a markedly different approach to examination, although they have a couple of interesting quirks in common. For example, both Boards seem somewhat obsessed with candidates noticing whether or not an adjective is in the superlative, including when that superlative is irregular. Personally, I don’t really understand why it is so crucially important that candidates translate plurimi as “very many” rather than just “many”, but for whatever reason, both Boards are very keen on it. Neither Board lists the irregular comparatives and superlatives as separate vocabulary, which given their obsession with their accurate translation seems a drastic oversight to me.

The language exam for OCR has a much longer history than Eduqas, which is the relatively new kid on the block for Latin. Those who taught the subject prior to the examination reforms in 2018 will understand why the OCR paper is divided into two sections, which bear no relation to each other: Section A represents what used to be Paper 1 and Section B what used to be Paper 2: the Board have simply merged what used to be two language papers into one, which I remember thinking at the time was quite extraordinarily lazy and has made for the current exam seeming bizarrely disjointed. Section A, worth 30 marks, consists of a 16-mark comprehension, a 4-mark derivatives question and then a choice between some grammar questions based on the comprehension passage or three English to Latin sentences. I always advise candidates to attempt the grammar questions as these are relatively straightforward (although considerably harder than the ones on the Eduqas paper). The grammar questions are quite ridiculously predictable and it is easy to drill even the weakest candidates to get full marks or close to full marks on this section. Section B of the OCR paper starts with a completely new story and contains a longer comprehension followed by a translation, which is worth 50% of the candidates’ overall marks. Section B is considerably harder than Section A and candidates do need to be aware that 50% of their overall mark is represented by that final translation passage.

Eduqas takes a completely different approach, one which followed the spirit of traditional “momentum” tests of old: the same storyline is maintained throughout most of the paper (which seems much more sensible), and what is labelled “Section A” is 90% of the paper: it consists of a short passage for comprehension, two short passages for translation and then a longer comprehension at the end; because the story is continuous, candidates benefit from completing the paper in order. Section B is worth only 10% and consists of a choice between some English into Latin sentences or some quite remarkably simple grammar questions, based on a very short and very simple passage of Latin, which is not even close to the complexity of the rest of the paper. As for OCR, the grammar questions are repetitive and predictable, thus it is easy once again to drill candidates to gain full marks on this section.

One notable quirk of Eduqas, and it is one I dislike, is that they seem particularly keen on candidates being able to follow the story. The reason I dislike this is I feel it advantages students who come from a background of traditional schooling, who may know the story involved. Candidates are often asked to infer things that are not actually contained in the passage and I find this unfair as those who have spent time in the prep school system or know the ancient stories from general interest may well find themselves better off. Another thing I dislike about the way that Eduqas examines candidates is that it uses a huge number of multiple choice questions, many of which seem specifically designed to trick candidates. They will, for example, encourage candidates to select the wrong meaning of words that are easy to mix up. That said, their approach to derivatives is much more benevolent: OCR seem ludicrously wedded to the idea of forcing candidates to define the meaning of the derivative they select, which I simply do not understand. I generally dislike questions about derivatives as again I feel they disadvantage candidates from certain backgrounds; they certainly disadvantage those for whom English is their second language, especially if that language is not European.

I am reaching the point where I know the vocabulary lists pretty well for both Boards, and there is roughly a 90% crossover. If anything, Eduqas has more words that are easy to mix up due to its inclusion of adiuvo (often confused with audio) and pareo (often confused with paro). That said, OCR included the word liber (book) as well as liberi (children), whereas Eduqas only has the latter. Both Boards have both iacio and iaceo, a nightmare to distinguish, and they also both have puto as well as peto, neco as well as nescio. All of these are regular traps that candidates fall into. When it comes to irregular verbs, OCR has more of these and includes the particularly awkward verb malo, which in my experience is massively undertaught in schools, which all focus on volo and nolo (as per the Cambridge Latin Course) and do not appear to teach malo discretely at all. Eduqas do not included it on their list.

As candidates make their final preparations for the exam one can only hope, as ever, that we have prepared them for the relevant pitfalls to the best of our ability. Michael Gove once said that he wanted to eliminate teachers’ ability to teach to the test but I’m afraid he has failed dismally in that department. While results continue to matter, teachers will continue to prepare candidates for the specific exam that they are facing. Not to do so would be sheer negligence.

Vocabulary acquisition

An essential challenge faced by students and teachers alike is the acquisition of vocabulary. I have written before on the best methods that students can employ when tackling vocabulary learning, so I do not plan to reiterate those here. What follows are rather some observations and musings about what we’re getting wrong in the Latin classroom when it comes to vocabulary acquisition, especially when compared to our counterparts in modern languages.

In my experience to date, supporting students in the accretion of vocabulary is a responsibility undertaken more effectively and proactively by modern language teachers than by those of us who specialise in Latin. It is possible that Latinists are under more time pressure in the curriculum and thus have no choice but to place the responsibility for vocabulary learning onto our students, but I think it more likely that we are simply less well trained in how to go about it than our colleagues in MFL. Classicists suffer from the fact that our training is somewhat broad – a qualified Classics teacher will necessarily have spread their training time across Ancient History and Classical Civilisation subjects, dramatically reducing the time that they spend focused purely on the teaching of the Latin language. I have little to no recollection of being given any significant guidance on how to help my students to develop their knowledge of vocabulary, so all my knowledge in this area has come later – through experience and through reading.

One of the many differences between the manner in which ancient languages are taught compared to modern ones is in the presentation of vocabulary to students. While modern linguists favour grouping words into themes or topics (e.g. “going to the shops” or “hobbies”), Latin teachers tend to present vocabulary in the following ways:

By chapters in a text book (e.g. Cambridge Latin Course, Suburani, De Romanis or Taylor & Cullen). Sometimes these may have a loose theme, but it’s generally pretty tenuous.
As one long alphabetical list (e.g. OCR GCSE or Eduqas GCSE).
In parts of speech. Some teachers invite students to learn the GCSE list in types of words, e.g. 1^st declension nouns, 2^nd declension nouns etc.

Each of these approaches has its drawbacks, so let’s consider those one by one. First of all, let us consider the approach of learning vocabulary by text book chapter. If one were to use Taylor & Cullen for this purpose, one would at least be learning the set vocabulary for OCR and thus there is some longterm justification for the approach. The vocabulary also reflects what is being introduced in each chapter and therefore there is some pedagogical justification for students learning it as they go. All of that said, you wouldn’t believe how few schools are actually doing this and to date I’m not sure I have met a single student that is working systematically through the chapters of Taylor & Cullen and learning the vocabulary as they go: some students are being tested on the chapters retrospectively, but I have not worked with any who are using the text book as it was designed. This is most likely because Taylor & Cullen is an ab initio course and thus the early chapters are not suitable for use with Year 10s who have studied Latin in Years 7-9. Why don’t schools use it during those years? Well, I’m assuming that its somewhat sombre presentation and lack of colour pictures puts teachers off the idea of using it a basis for KS3, when (to be frank) they are under pressure to recruit bums onto seats for KS4 or else find themselves out of a job. This means that there is no text book explicitly aimed at preparing students for a specific GCSE exam board being made wide use of in schools.

None of the text books commonly used in schools at KS3 build vocabulary that is explicitly and exclusively aimed at a particular GCSE course. While Suburani is supposedly linked to the Eduqas course, it diverts from using the vocabulary that is relevant to this in favour of what suits its own narrative. For example, students of Suburani will be deeply familiar with the word popina as meaning “bar” (not on the GCSE list for either OCR or Eduqas but used widely throughout the first few chapters), yet they are not introduced to the word taberna meaning “tavern” or “shop” (on the GCSE list for both boards) until chapter 12. Similar problems occur in terms of the thematic focus of Suburani: because it focuses on the life of the poor in Rome, students are taught that insula means “block of flats”. While it does mean this, I have never seen it used in this way on a GCSE paper – the word is used exclusively by both boards in a context in which the only sensible translation is “island”. I shall say more about the problem of words with multiple meanings later on.

Presenting words in an alphabetical list seems to be the practice used by most schools when students reach Years 10 and 11 and are embarking on their GCSE studies. Most students that I have worked with are told to learn a certain number of words from the alphabetical list and are thus tested on multiple words that have nothing in common, either in terms of their meaning or their grammatical form. One advantage of this is that students are forced to look at words with similar appearance but different meaning. However, multiple and in my opinion worse problems arise from this method. Students learning the vocabulary in alphabetical order give little thought to what type of word they are looking at (e.g. whether it is a noun or a verb) or to its morphology. This means that students do not learn the principal parts of their verbs, nor do they learn the stem changes of nouns and adjectives. This can cause considerable frustration and demotivation when students struggle to recognise the words that they have supposedly learnt when those words appear in different forms. Teachers could mitigate against this by testing students on those forms, but most seem reluctant to do so. Do they think it’s too hard?

The method I used was to present the GCSE list in parts of speech and invite students to learn different types of words in groups: all the 1^st declension nouns, all the 2^nd declension nouns etc. The advantage of this method is that it allows for the opportunity to link the vocabulary to the grammar. For example, the first vocabulary learning task I used to set my Year 10s in September was to learn/revise all the 1^st declension nouns (in theory they knew most of them already from KS3) and to revise the endings of the 1^st declension. In the test, they were expected to be able to give the meaning of the nouns I selected for testing and they were expected to be able to write out their endings also. I felt (and still feel, on the whole) that this was the best approach, but that does not mean that it does not have its own disadvantages. Firstly, it made some learning tasks excessively onerous and others too easy: for example, that task of learning the 1^st declension nouns was very easy (because most of the words were already familiar and the forms of the nouns are very simple) but the task of learning 3^rd conjugation verbs was much harder (fewer of them were previously known and their principal parts are a nightmare). This meant that students were often hit with homework that turned out to be extremely difficult at what might not have been the ideal time for them. A second disadvantage was that it was impossible to give students a translation test, because one could not create sentences out of a set of words which all belong to one category. Thirdly, and related to that point, testing according to parts of speech made it very difficult to link vocabulary learning to classroom teaching in any meaningful way: in class, we might be studying the uses of the subjunctive, and that could not necessarily be linked to the homework task that was next on the list. This is something that I have been thinking about more and more in recent years as a massive problem in Latin teaching – a disconnect between what students are learning in the classroom and the vocabulary they are invited to learn for homework. The more I think about it, the more I believe this is a fundamental problem which requires a complete curriculum re-think.

The difficulty of linking vocabulary learning to explicit classroom teaching is something that modern language teachers would probably be very puzzled by. Modern linguists are way ahead when it comes to tying vocabulary learning to what’s happening in their classroom and to the relevant grammar. Given this, imagine my excitement when one of my tutees shared with me that she has been presented with the OCR vocabulary list in themes! I was full of anticipation as to how her school was planning to test their students on those themes. For example, one theme might be “fighting and military language”, within which students learn nouns such as “battle” and “war” alongside verbs such as “fight” and attack”. Call me daft, but I hoped and expected that she would be tested using some simple sentences, which would afford teachers the opportunity to observe students’ (hopefully) increasing understanding of grammar and morphology alongside the acquisition of the relevant vocabulary. Surely no teacher would have gone to the trouble of dividing up 450 words into a set of themes unless they were going to make use of some innovative testing methodologies? No? Well … actually, no. The school are testing the students on a list of words, with no link made between the meanings of those words and the learning that is going on in classroom. I have absolutely no idea what the point of this is. Maybe somebody in the department has read somewhere that “themes” is a good way to classify vocabulary and I am sure it is – but I’d place a hefty bet that there is no tangible pedagogical gain unless that learning is linked to the use of those words in sentence-structures, the kind of approach favoured by Gianfranco Conti.

I said that I would come back to the issue of words with multiple meanings, and that is something I have noted with interest from my tutee’s themed list. Words with multiple meanings appear more than once on the different lists, with their meanings edited to suit the theme of that list. This is an interesting idea and I am still pondering whether or not I think it is a good one. Multiple meanings are a real menace, particularly when the most obvious meaning (i.e. the one which is a derivative) is the least essential. For example, on the GCSE list for both boards is the word imperium, which can mean “empire” and all students immediately plump for that meaning as it is an obvious derivative. However, the word is more commonly used on language papers to mean “command” or “power” – it is therefore those meanings that must be prioritised when a student is learning the word. Similarly, all students need to be drilled on the fact that while imperator does come to mean “emperor” in time, it originally meant “general” and is usually used in that way on exam papers. Even worse is a nightmare word such as peto, which is listed on both boards as meaning anything from “make for”, “head for”, “seek” and “attack”. Students really struggle with learning all of its multiple possible meanings and it is important to show them multiple sentences with the verb being used in lots of different contexts so that they can grasp all of the possibilities.

As so often, I reach the end of my musings having criticised much and resolved little. I am thankful to be working in a one-to-one setting, in which I can support students with vocabulary learning in a proactive and detailed way, one which goes way beyond what is possible in the mainstream classroom and supports their learning in a way that simply cannot be expected of a classroom teacher. I shall continue to ponder what I would do were I in a position to re-shape the curriculum all over again, but I fear that this would entail writing an entire text book from scratch. Many have tried to do this, and even those who have made it to publication remain flawed: I have no conviction that I could do any better.