Challenging the System Part 6: Next Steps...

 In a previous post, I offered a vision of what a functioning languages ecosystem in the English state sector could look like. With a positive experience of languages in Primary and Secondary School, mainstream pathways post-16 and beyond. A contrast with the stark reality of the current situation, where you could hardly design a worse system for teaching and learning languages.

The question is, what practical steps can we take to move away from where we are now. And this is really serious. We are facing a crisis where the impending lack of language teachers will only be managed by reductions to incentives for schools to teach languages, leading to total collapse of the system. This excellent article by Juliette Claro (click the link) shows the unsustainable situation.

What steps can we take?

Fortunately there is one cost-free and immediate step that will have a huge impact. GCSE grading. At the stroke of a pen, we can eliminate the unfair disincentive for pupils to take languages. There is no reason for it to be on average a whole grade harder to get a good grade in languages. It is simply a historical anomaly that we have got stuck with. Ofqual's brief is to maintain the "value" of grades in languages year on year, not to compare it to grades in other subjects. In most subjects, if an exam is too hard one year, they use grade boundaries to bring grades into line. Or if an exam is too easy, then they use the grade boundaries to bring it into line. Except for languages, where grade boundaries are used every year to keep grades out of line with other subjects. 

The only people who supposedly would suffer from ending this would be pupils in previous years whose grade in languages would look bad compared to pupils going forward. This argument is bogus. In fact, it would be an overdue recognition for all those pupils over the years, that their languages grade that looked lower than their other subjects was not in fact any reflection on them. Just a reflection of how grades were allocated in different subjects.

And of course schools are sensitive to these figures. The government has taken to using school performance measures such as the ebacc and changes to Progress 8 to encourage schools to promote certain subjects. Ending unfair grading would feed into this system of controlling what subjects are valued and incentivised in school.

It would help address the narrative of "failure" that we see in the national press when it comes to languages. And the culture wars which distort language teaching initiatives, driven by the perceived "failure" to recruit pupils to an unfairly graded subject. The reputation for "difficulty" of languages GCSE isn't based on the content of the course or of the exams. It's an artificial outcome of the decision to grade languages unfairly compared to other subjects.

And if anyone tries to tell you that it's because of the quality of teaching, you can tell them that that calibration has not been made. At no point did they make that assessment and apply it to how grades should be set. And if they tell you it's because it's hard to recruit language teachers so there must be an impact on teaching quality, you can ask them why that doesn't apply to maths. No-one is going around suggesting maths grading should be made harsher. And if languages grades were in line with other subjects, no-one would be clamouring for them to be made harsher either. Going back through the history of grading, it's been like that since the 1970s and we've just got stuck with it.

It would end the blight on the careers of language teachers, tired out by the moral bankrupcy of having to hide the truth of the grading from pupils, and scorned as a blot on the school's results in performance tables.

It's free, it's transformative, it's easy. To maintain the unfair grading would be perverse.

Another area currently under review is Primary language teaching. The CAR review seems to have gone with the "substantial progress in one language" model and to want to tackle transition from KS2 to KS3 head-on. Their ambition is for KS2-KS3 to see uninterrupted progression.

This is a hugely ambitious demand. I think that whatever demands the government makes of schools, the teachers will perform miracles towards making them work. Despite the system (in the case of languages) working against them.

What is happening here, is the government doubling down on the demands. While doing nothing to solve the systemic issues. We need to have as our starting point that transition from KS2 to KS3 will be disruptive. 

Pupils will go into classes in Year 7 where there will be pupils who are continuing with the same language, and other pupils who previously studied a different language. All at different levels. Primary school recruitment is not built around who can teach the language taught at one or other of the local Secondary schools. Secondary school timetables are not built around what languages different pupils studied in Primary school.

Recognising that there is a hiatus here is not necessarily a bad thing. What it means is that Primary languages needs to be valued on its own merits. Not just seen as a preparation for Secondary school. We can stress the lived experience of meeting new languages, rather than some imposed concept of "progression". Pupils' immediate initial focus is on learning to say and understand things, rather than on grammar. By exploring other languages, they inevitably make discoveries about the nature of language including learning that English isn't the natural innate and default language of thought and communication, and that other languages work differently. They can learn through rhyme, song, stories, routines, games and cultural aspects, seeing how languages and cultures are part of the world they are growing up in.

This approach isn't inimical to the idea of progression in a language. But it puts the starting point perhaps more comfortably in a style of learning more suitable to the age group and the confidence of teachers in their own language ability. It's absolutely vital that the voices of Primary teachers are heard here. They are the ones making this work, and finding what works is fundamental to what we should be developing. And of course, when we remove unfair grading at GCSE, we will see more recruits into Primary teaching who have a GCSE in a language.

Recruiting teachers is the key to the impending crisis in languages teaching. Currently we depend on non specialists in Primary and increasingly in Secondary, alongside immigration to boost numbers. Without changes to the system, these are only temporary emergency sticking plasters. So what else can we do to make things work?

The answer is NOT to try to get more people to study A Level Languages or do Philology degrees at University. 

I am not unsympathetic to the demise of Language degrees for the tiny minority who want specialist academic study of the language through linguistics and through literary and other texts and its socio-historical context with essays (in English) and translation. Indeed, if we can solve the systemic issues, then there may be a wider pool of takers for such courses. 

But these are not mainstream language learning pathways. And the idea that language learning is only valid if it has the added intellectual heft of essays, literature, grammar, linguistics is dangerously inimical to language learning in this country. It means as well as being harshly graded at GCSE, languages is a dead-end subject. With no suitable pathways post 16 for the vast majority of pupils. Even if we somehow doubled the number of people taking languages A Level, it would still be all but irrelevant. And who are these people for whom it would be suitable? A Level languages feels like a fraud perpetrated on the few students who do take it, with the workload and harsh grading driving most into disillusion despite their love of the subject and the huge progress they make. In my school, we are into the last cohort to take A Level languages, and it will be a huge relief not to feel guilty for letting keen bright students commit themselves to something they will regret.

So what is the alternative? How do we have a post-16 offer which means pupils can enjoy continuing with a language that they studied at GCSE (in greater numbers once unfair grading is removed), or start a new language or multiple languages? One answer is staring us in the face. Because it's already vastly more popular than A Level: online apps and AI. British young people who supposedly don't study languages, are avid users of Duolingo.

Is it OK for sixth forms and colleges to abdicate responsibility for language learning to an app? I am uneasy about this. But could we or should we be trying to replicate this within the education system? The aim would be for students to study the range of languages they want at the level and pace that suits them. Just as at Primary school, rather than focusing on "progress", we should be looking at experience and attitudes. And with a large number studying online, more formal learning may perhaps follow?

Could colleges do more to incentivise this? To create the expectation that learning a language independently is something to be shared, documented and accredited. Even if they can't themselves offer lessons on demand in Korean or Japanese. And Universities. It must be possible for Universities to send a stronger message to applicants: When you come to university, you are stepping into an international community with international horizons, and we will take an interest in evidence of language learning you have done. And when you come to university, you will have the possibility, whatever your subject, of studying abroad. And alongside your degree subject, you can study a language either as part of your degree or as an add on.

And this should be huge: Erasmus.

When we took pupils on the Spanish exchange, our kids were always stunned at how the Spanish kids saw this as part of a strategic plan. The exchange in Year 10. Spending a year across the border as a boarder in a French lycée. Doing an Erasmus year in England. And a masters in the States. All planned out from age 14.

How do we inspire this ambition in our pupils?

Erasmus+ has to be a huge part of this. In Primary and Secondary it funds teachers and pupils to collaborate and travel internationally. We need to make all the things we say in the blurb on our Programmes of Study actually come true. Languages needs to be a normal thing in a world where we work with people around the globe. In person and virtually. We need to convince our Year 9 pupils so that they arrive at university demanding that they can study abroad for a year, whether they do History or Engineering or Architecture or Maths. We have to create this expectation, because otherwise University departments won't bother. Last time we were in Erasmus, although students from other countries and all disciplines came to study here, the majority of British young people who went abroad were studying languages. So limited to a tiny minority.

And then we will have cracked it. We will have enough people who have continued with a language post 16, many of whom will have lived and studied abroad for a year. They may not have written essays on Fray Luis de León or André Gide. But they will have the confidence they can pass on to young people the real-world benefits of learning a language.

Nothing pie in the sky. Easy changes to the system and its demands and incentives. Then you can let teachers and learners get on with the job. I promise you they won't let you down.

Challenging the System. Part 5. The Vocabulary (and Grammar) Curriculum: A Failed Experiment

 Actually, that title is unfair. In science an experiment is never a failure. If you discover something doesn't work, then you have learned something. Unless you fail to learn, in which case, yes, it's a failed experiment.

There are three main reasons why it failed. One because of teaching. Two, because of learning. And thirdly because of politics.

In terms of teaching, it seems like a great idea to have a curriculum where pupils learn and are tested on a defined body of language. A curriculum where they meet the words repeatedly at carefully planned and tracked intervals in different contexts so just as words were about to be forgotten, they are reinforced. And knowledge accumulates.

But in practice, it's impossible for the classroom teacher to construct that meticulous fine tuned instrument. I do not have the wherewithal (is that really one word?!) to create such a thing. What I can create is a core snowball of pupils' language where their repertoire rolls from one unit to another and picks up more and more language once it has something to stick to. For me, the vocabulary list approach means trying to create an even covering of language. And an even covering will melt much quicker than a snowball. The incredibly perfect layering required to make an even covering that will survive, is beyond me to resource and teach. We need something much more robust, without fear of some of the things we teach melting, and something which won't fall apart in the real world of the classroom.

In fact, I can't even write single texts based on this approach, let alone a whole curriculum. Because when I sit down to create a text for a resource, I don't know what words are or are not on the list. Or what words are due for a fresh encounter. I write the text using the words that the piece of writing needs. But how to write a text about Marie Curie if we don't have the words Polish or chemistry? Of course I can gloss them. But that's not the idea of what we are supposed to be doing. It's what we've ended up doing, especially when pupils are asked questions such as "What do you like to do in your free time?" or "What do you like to eat when you go to a restaurant?" Because they can't all say "rabbit". So we have made the vocabulary approach work. By ignoring it. And teaching the words pupils need.

That's from the teaching point of view. And already that last point is moving into the learner's point of view. But it's not just about having to teach vocabulary that's not on the list. It's about the whole perversion of language learning. Language is not about making sentences for the sake of practising prescribed words. Language learning is about saying things you want to say. Fortunately our learners always, even when you try to make them create sentences from a predetermined list, always try to say things you weren't expecting. But most importantly, learning that way doesn't actually work.

I have an example in this post where pupils who had written some practice sentences in Year 10, when it came to Year 11 were basically starting from scratch. The pupil who had written a true story remembered it clearly. Not word by word. But he could retell the story fluently a year later. There are two sides to this. Firstly it made the language memorable when he wrote it, because he was deploying his full knowledge of how the language worked, challenging himself to articulate bits of knowledge into a something that made sense. Secondly, it was memorable when he was rewriting it a year later. Because it had personal meaning. You can see from the original post that the two versions a year apart are quite different. I am not talking about memorisation here. I am talking about consolidated language learning that has long lasting impact on his ability to deploy his knowledge to express himself. In a way that the pupils who had just done some sentences to practise the Spanish could not keep up with. Meaning matters. We have to start with meaning, not with a list of words to practise until they are memorised.

This language learning focused argument is of course the most powerful argument. But there is also a political side. Because the "learn the prescribed content and be graded on it" approach is part of the right wing perversion of the knowledge curriculum idea. It sees learning as accurate recall of important knowledge, in order to meet the demands of the gatekeepers. It denies pupils agency, creativity, expression. You can see this is a philosophical rephrasing of the very real issues outlined above in the sections on Teaching and on Learning. And you can see the political assumptions it is based on.

So...

Imagine if we are given a list of words to be learned at KS3. And this is then policed for conformity as to whether pupils know those words. What would happen to our carefully constructed and robust schemes of work and resources then? What would happen to the careful balance of communication and grammatical progression? Can you imagine having to throw out everything that works? And "deliver" something that has already failed? It would be heartbreaking and I think I would have to leave teaching.

This could only come to happen because of the narrative of "failure". Severe grading and other systemic failures are perfectly designed to make continuing with language learning impossible for all but a tiny minority. We need to challenge these systemic failures, and challenge the narrative that something is "wrong" with teaching. With the utmost urgency and forthrightness. Before they make things even worse.

Challenging the System Part 4: Are we ripe for AI to take over in language teaching?

 Here's a scary thought. Is MFL teaching in exactly the sort of place where AI could take over and do the job instead of us?

Firstly there's the nature of language learning. One view has it that language learning consists of meeting words and vocabulary in different contexts in a meticulously constructed and tracked series of encounters with words being met and recalled at intervals. AI can generate sentences to practise this and could supposedly track success and failure in recalling words.

Then there's the systemic failures of languages in the English system:

A Primary sector under resourced to meet the vision of pupils in this age group learning languages through listening and joining in with routines, games, songs, stories and cross curricular opportunities. A secondary sector being pushed towards grammar and vocabulary. Unfair grading and the subsequent narrative that there must be something wrong with language teaching. A sixth form offering that does not meet the demands of learners, with more young people using Duolingo than taking the niche academic A Level. Universties where the tragic demise of philology degrees risks dragging down with it the provision of non specialist language learning centres for students of all disciplines.

I don't think we would set out to replace Primary teaching of languages with AI. But it may happen by the back door. Teachers who may feel ill-equipped to teach a language, may readily turn to AI for planning, resources, and even as an interactive teaching tool. They may welcome a source of content which is readily accessible and solves problems of resources and confidence in the language. They may feel that given the circumstances, AI enables them to do better. In teacher tapp surveys, when teachers are asked what aspect of the curriculum they would rather leave to someone else, Languages often tops the vote. If a safe AI interface came along, that pupils could engage with directly, it might feel like a step forward.

Likewise at secondary, where there are staffing issues, AI could present a solution. And elsewhere it could be pushed as a tool for pupils to be more independent in their learning and for learning to be more personalised. Two buzz words that are easy for policy makers to eagerly pick up on.

Post-16 where schools and colleges have no mainstream language learning pathways, AI can solve the problem of the need for individualised language learning at different levels and in a range of different languages for different learners. Some may have a GCSE in a language, some may not. Some may want to continue with the same language, some may want to start a new one. It's unlikely that sixth forms could provide individualised teaching of a range of languages. But Duolingo already does. And AI could...

Is this desirable? The deficiencies of our system could be mitigated by use of AI. Were it to one day be reliable and safe.

There are two dangers here. One is that in the face of a defective system, this does nothing to do what needs doing. What needs to be done is to deal with the systemic failure. Underfunding at Primary, unfair grading at GCSE, lack of mainstream language learning post-16. AI might help with some of these disasters. But it would be much better to tackle the real problems in the system.

Secondly, it may offer the hope of improving on what can be done in a broken system. But it doesn't come close to what could be done in a working system.

Language learning has to be more than gameified recall of vocabulary and grammar. Or interacting with a machine. Points, levels, streaks... these are all aspects of addiction to screens that we see bringing a deterioration of young people's lives. They are no replacement for what should be the positives of language learning. At Primary, AI would take us away from the magic of pupils using language in songs, games, stories, creativity, routines, daily life. It would take us away from ownership of language and identity as a language learner. It would take us away from communication.

As language teachers, we can respond to AI with all the things we do best. International and intercultural communication, self expression and imagination. The things stopping us are built into the system. And AI would do nothing to remove those obstacles. We need to challenge the system and change the narrative before someone proposes AI as the wrong answer to the wrong problem.

AI will kill your children and abandon their Granny. Don't use it.

 The worst thing AI has done for me is this:

I was testing its problem solving skills by asking it how to cut a cheesecake into 5 equal pieces. It has given me many absurd answers even including cutting the cheesecake horizontally and then reassembling portions of crust and cream. As everyone knows, the correct answer is to cut it into 6 and make the extra slice disappear before serving the 5 perfectly equal slices. On this occasion it was giving me a spurious method based on the Golden Ratio. I was laughing at this when I noticed the diagram it was giving me to illustrate its method. A guide to the different forms of Female Genital Mutilation. How it had bundled FGM in with How to Cut Cheesecake as if it were instructions isn't a joke. It's abhorrent.

Does this mean that we shouldn't use AI? Well everyone knows not to use it for "factual" information. Because it just makes things up:

You can have fun with variations of this question, inserting your own choice of minor celebrity. What it reveals is how it's never really answering YOUR question. It's trotting out a concocted text that your question may or may not be relevant to. In this case it can answer the chicken and the egg paradox. But not with the addition of a world class athlete. Common questions with an easy answer you already know are more likely to be answered correctly. But why would you be asking for answers you already know?

The same applies to problem solving. AI is NOT thinking about your problem. It is not analysing your situation. It is writing something plausible that sounds as if it might make a good answer.

I tried it with a few scenarios.

I hit refresh on this many times or tried clearer prompts. Each time it gave me a different answer. Its times made no sense. It didn't get me to the beach on time. And it failed to spot that I had frozen items that I needed to get in the freezer.

It can't deal with the simple logistics of getting to the beach on time. Let alone pick up on issues that ought to be dealt with. Are we really suggesting it would be a good idea to delegate planning and decisions to AI? It is NOT making decisions or analysing the situation. It is just cobbling together some words.

It can't do the maths. Can it respond to a simple decision that doesn't involve calculations?

It variously suggested knocking on the door to wake my neighbours up to tell them I was going to be cutting the grass, or practising my drums first...

All very amusing. But how far will its cheerfully delivered bad advice go?

Here it is failing to respond to very young kids being left in a hot car.

I tried this one with mistral chat, Gemini, chatgpt, copilot and perplexity. All showed different degrees of mock horror or glib answers about the ice-cream or the upholstery of the car. Several commented about the kids making a mess. They all gave varying estimates of the temperature from 40° to 60°. None suggested I should check on the kids.

Here it is showing similar disregard for my very elderly mother in law.

It can't process the simple maths of logistics problems and it can't grasp the human dimension of a situation. Because it's not actually doing any analysis or solving. It's just doing words.

So maybe if we don't use it for decision making, we could perhaps use it for drafting?

Like this you mean?

It has no clue.

Why are my experiences with AI so consistently bad? Because I am trapping it into mistakes by asking it things that are a little bit complicated or unexpected. OK then. Use it for highly predictable easy things that you already know the answer to. But what is the point of that? And also think what these experiments are really showing: It does not have the intelligence to think about and solve your problem. All it has is the power to put words together to fob you off. If that's what you want and need in your life and your business, then go ahead and use it!

Challenging the System Part 3: Transition from KS2 to KS3

 Sorting out unfair grading at GCSE, with its immediate impact on pupil motivation to take a language, incentivisation for schools, and on the narrative of national "failure" is the easiest part of the failed system to put right. The tricky ones are What to do about Transition from KS2 to KS3 and How to Offer Mainstream Language Learning Post-16. And writing about KS2 is tricky for me. Because Primary teachers don't need Secondary teachers stepping in to opine on things they have no experience of.

So I need to make it clear that I am not commenting on teaching. The problems are not with teaching. The problems are systemic. And teachers are the ones making things work, even when the system is designed to stop people learning a language. 

What are some of the issues at Transition from KS2 to KS3?

• Pupils start again in KS3, ignoring what was done in KS2, repeating basic content.
• Pupils start a new language in KS3, abandoning what was done in KS2.
• Pupils arrive at Secondary School from many different Primaries, all with different experiences of languages in KS2, including the study of different languages.

These systemic failures are worsened by the fragmentation of education provision and competition between schools which is a deliberate feature of the English education system, in the belief that this pushes up standards. And which teachers constantly have to try to overcome in order to provide a coherent experience for pupils.

This is the current situation: teachers in Primary and Secondary trying to reach out to each other to make things work. In a system that constantly erodes their efforts. Attempts to create a joined up 9 year language learning experience from KS2 to KS4 in a system designed to disrupt collaboration and planning. Decisions made in one school on things like curriculum options, timings, staffing, choice of language, resources, assessment... all have knock on effects across other schools' attempts to create coherence. Managing a joined up learning experience across multiple competing schools can see things fall apart as quickly as they are put in place. A small Primary School appoints someone who happens to speak Spanish. Or the German teacher leaves. Or the school aligns itself with one favoured destination Secondary, based maybe on being part of the same Trust or another school belonging to a rival Trust. A Secondary School decides to stop offering Spanish in Year 7 in order to focus curriculum time on French. Or decides to offer Mandarin as part of a national initiative. Or has different halves of the year group take different languages in an attempt to keep two languages going. A Sixth Form stops offering German, so other schools have to wonder if that means they should stop offering it at GCSE. These decisions are taken based on shifting priorities within an individual school and have challenging implications for language teaching within the school, let alone how to manage the implications for other schools and transition. Top down decisions are not made with languages as the top priority, nor with implications for other schools as the top priority.

In aiming to make a coherent experience of language learning, what chance does something put in place for pupils in Year 3 have of still being relevant wherever they end up 9 years later? There will have been so many changes, transitions and disruptions. A vision of coherence has to be more robust, more purposeful and more fundamental than individual teachers occasionally meeting to have a conversation about topics.

Language teachers are trying to make this work. But the system is against them. As quickly as you put something in place to try to make it work more smoothly, something somewhere changes or doesn't quite work out as planned. And nothing can overcome a situation where fragmentation means there's no coherence in what language pupils study, let alone a coherent experience of progression in the language.

This is not to say that we shouldn't keep talking to colleagues across sectors. But if teachers aren't managing to make something work, it's because it's unworkable.

What options are there to make the system more workable? What would be more robust than asking teachers to constantly shore up transition as it is washed away by the tides?

There is hope that politicians might begin to see the harm done to children's lives by a regime of high stakes competition, commercialisation and fragmentation. But that isn't going to happen quickly. We need to think about how different models can work within the landscape of disruption between schools and between sectors.

One model is to have "substantial progress" in one language at Primary School. The proposals in the Curriculum and Assessment Review would see all pupils in all Primaries having a common grammar and vocabulary curriculum. When they arrive at Secondary School, schools could assume "coverage" of this content and plan accordingly. That plan would still have to cope with individual pupils having different levels of attainment and having studied different languages. But it could reduce the level of variation that currently makes planning for transition so hard. This could reduce the amount of starting from scratch and repetition of content already seen in Primary school. This model could make the work done in Primary harder to "ignore" on arrival in Secondary. But starting a new language on arrival in Secondary would seem a total disruption to this model. These are possible outcomes, and they are a question of degree - aiming to reduce some of the issues rather than removing them. And potentially reducing one issue while compounding another. It seems linked to the idea of learning as "progress" rather than experience. And it sets schools the challenge of having staffing, resources, curriculum and transition in place to deliver the ambition of substantial progress. Is specifying content enough of a contribution to overcoming the system that currently defeats the best efforts of teachers?

Another model is to have more than one language taught in Primary. In itself this seems positive - the more languages pupils have some knowledge of the better! It also allows Primaries to deploy staff with different languages, and potentially lessens the level of language demanded of staff. By design it harnesses the synergies of how learning one language helps learn another, through learning about how languages work and how language learning works. It combines the idea of progress in the languages studied, with a deliberate awareness of the importance of the experience of language learning and learning about languages. This could be a more robust model for transition, where Secondary teachers know that pupils arrive as experienced language learners, with some knowledge of more than just one language.

There is also the Language Awareness model which proposes explicit teaching about language, languages and linguistics. This could remove the issue of pupils arriving at Secondary with progress in a language which is then ignored or abandoned. It would emphasise the nature of other languages being all around us in our community, and the wider world, and also within the English language itself. It would open pupils up to a world where cultures and language are explored together. Would it also include the idea that learning at this stage can be through stories, songs, rhymes, games, listening and joining in, routines and links to other subjects? Or in fact, would a better way to learn about language and linguistics be to focus on learning a language after all?

Reading through these 3 options, I think it's clear that as much as possible needs to be common to all of them. An experience of language learning that is an experience of learning through listening and joining in; that furthers understanding of language in our world and our world through language; that develops understanding of language learning by learning a language. Whether the headline is substantial progress or language awareness, the underlying experiences have much in common. Is this a ridiculous idea?

We are all familiar with these 3 options for Primary languages. The neglected one is the What Happens in Year 7 question. How can we make sure we are building on pupils' learning in KS2, consolidating the basics without repeating content or ignoring their knowledge? Can we have a much more original and creative start to KS3? What would this look like? I would want to see consolidation of phonics, ground rules for active participation in speaking, a focus on developing pupils' abilty to use their language, through cultural content and contact. Our Year 7 French Art Exhibition is an example of how we try to deliver these aspects in a way that will be different to what pupils did at KS2.

The government seems to have gone for the Substantial Progress in One Language model. If anything, I think this is the one that does least to remove the systemic barriers. In fact it may exaggerate them if pupils have made substantial progress in a language that they then do not continue with. It is ambitious in its aims and therefore in the challenges it sets Primary schools, even if these challenges are now to be more clearly defined. Is the danger that in our fragmented system, Primary schools will focus even more on meeting this defined content, rather than on the experience of learning a language and continuity into KS3?

I am glad the government has identified KS2 to KS3 Transition as one of the greatest issues. Their response seems to be to challenge teachers to do the same thing better. This feels unwise. Teachers are the ones who make the unworkable almost work. Teachers are the ones seeing their work eroded in the Sisyphean task of building progression only to see it crumble away in the face of the system. I can see that they have gone for the option they see as the most ambitious for pupil progress. Is there wisdom in this choice? Ambition is good. Workable is also good. The calculation has to include both. My instinct is that if teachers are barely managing to make something work, it's because it's unworkable.

This post is in a series called "Challenging the System". I don't seriously think we can change the system in that pupils will inevitably continue to move from Primary to Secondary. There is disruption at this stage, with cohorts splitting up and new intakes of pupils coming together with different previous experiences. We need to accept that this disruption is real. And choose models in KS2 and in KS3 that recognise this disruption and make the most of what really matters in learning languages. Academic assessed and measured "progress" in grammar and vocabulary may not be the only factor alongside experience, attitudes, understanding, communication, ambition and enjoyment.

Challenging the System. Part 2: A Vision of a System that Works

 In an earlier post, I looked at how the system could hardly be worse for encouraging people to continue learning languages: Disjointed Transition at age 11, Unfair Grading at GCSE, and lack of mainstream language learning post-16. Within each of these stages, teachers and learners do a good job. But the system is designed to stymie them at every turn. It is not a failure of teaching. Good teaching is making the best of an unworkable system.

So what would a working system look like? Wouldn't this be nice:

Primary Languages. Not my area of expertise. But we are told that this is the age where pupils learn languages through hearing and joining in, with songs, chants, routines and stories. Where language can be used throughout the day, integrated with everyday activities and across curriculum subjects. We are going to see a greater definition of the detail of formal learning and progression. Pupils and teachers should be confident that languages are a natural part of life in and out of school. It should be a positive experience of openness and celebration of linguistic diversity.

Secondary Languages. An experience of languages which combines all the facets of language learning: communication, grammar, culture, linguistics, self-expression, transactional language. Accumulating more language and knowledge of language will go hand in hand with a growing ability to use the language with increasing independence, sophistication and accuracy. This will be boosted by international ambition where schools use Erasmus + strategically to make cooperation with schools abroad part of school life. An end to unfair grading at GCSE will be transformative in terms of numbers and in the narrative of our success in language learning. And teachers will be able to tell pupils that if they go to University or take an Apprenticeship, they can look forward to studying and working abroad, whatever their chosen path.

Post-16 Languages. The norm and the expectation will be that people learn languages for pleasure, work and study. The promise and the premise should be that the future (University, work, apprenticeship, life) will mean interaction with the international world. That an outward looking perspective will take you places. A Level languages may continue, or academic study of philology may be delayed until university level. Instead, sixth forms and colleges will offer language learning opportunities at different levels and in different languages to all students.

University. The greater number of students continuing at Sixth Form with language study will see a boost for specialist philology degrees. But Universities will stress to all applicants for every subject, that going to university means entering a global environment, working alongside people from around the world in research that transcends borders. It will be normal to study a language alongside other courses. It will be the expectation that students of all disciplines take advantage of Erasmus to study abroad, not just students studying philology degrees.

Hopefully you agree with some of this or have your own ideas that I have missed. And the point is that this vision is actually very close to what teachers are aiming to achieve within our current broken system. We don't need to change teaching. We need to change the system. The question of course, is how to make it work. I don't have answers but I do have thoughts. Will post them soon...

Challenging the System: Part 1 Do you need a degree in philology to teach languages in school?

 In a previous post, I looked at how we appear to have a system designed to stop people learning languages in school. So we need to challenge the system. Here's the first challenge: Does the system exist simply to be a pipeline to produce language teachers to maintain the system? Do we have to base our decisions around the school curriculum and the languages curriculum so that 2 pupils per secondary school per year can end up doing French A Level, then a fraction of them can end up doing a French degree, and then a fraction of them can end up becoming teachers? Is this the raison d'etre for the system we have?

Let's ask the challenging questions... I warn you that the writing here is not well-formed answers. It is nibbling away at the presumptions we bring to language learning in our current broken system.

Do you really need a degree in philology to be a language teacher in English schools?

Philology is the academic study of a language through literary criticism and analysis of socio-cultural texts, history of the language and linguistics.

Quite clearly the answer is no. Primary teachers teach languages, sometimes without even a GCSE. And they make a good job of it. Alongside teaching English, maths, geography, history, music, science and a whole list of other subjects they don't have a degree in. At an appropriate level and ideally with resources and a well-planned curriculum.

Does this apply also to secondary school? Yes it does. The expansion of Spanish is seen as one of the success stories of the last 30 years. Done in the first instance by French teachers who learned enough Spanish to introduce it into the curriculum. Without a degree in Spanish themselves.

Who else? Well, teachers. I have worked with excellent teachers who didn't have a languages degree and either moved into language teaching to fill a gap or qualified as language teachers based on having another degree plus some knowledge of the language.

(We also need to ask the question of just how good the language level of some philology graduates might be. I used to interview and mark subject knowledge tests for a PGCE course. All of the applicants had a degree in languages. Not all of them made it onto the course. Language learning doesn't have to be through A Levels and degrees in literature, linguistics and essays.)

Native speakers. The All Party Parliamentary Group for Modern Languages recently discussed this as a solution to the "pipeline" issues. They pointed out the absurdities of paying bursaries for international trainees who were then unable to take up a post because of immigration hurdles. Although I see the government has now resolved this contradiction by ending the payment of bursaries to overseas trainees.

Who else could teach a language? Well, if we can do away with the ridiculous systemic barriers to language learning, then we could have a wide pool of applicants. Much wider than the tiny numbers who are ever going to study a philology degree. (Although of course, having more people continuing with languages would also mean a wider pool of potential candidates wanting to study philology.)

How would this look?

If it became the norm to study languages (by ending unfair grading at GCSE and offering mainstream language opportunities post-16) we would have a wide pool of people who will have studied a language or languages throughout their school life, going on to University where they could take a language course either as part of their degree or as a stand alone. They could study abroad for a year now we are returning to Erasmus. And then maybe work abroad before thinking of returning to England and becoming teachers.

Another thing. If languages are being taught by normal people, rather than a tiny elite who did A Level and a philology degree, would that have implications for the teaching and learning? Quite possibly. Seeing languages as something normal rather than vanishingly rarified is what we are aiming for.

Could you design a worse system for making sure people don't study languages?

 If you wanted to design a system for making sure as few people as possible learn a language, our school system is perfect:

• "Transition" from KS2 to KS3 where pupils either start all over again in the language they have been studying, or abandon it and start another language.

• Cowardly tinkering with performance measures supposedly to support take up at KS4, while the biggest headline performance measures insidiously disincentivise schools from promoting languages.

• Unfair grading at GCSE, where lower grades are given out in languages. This leads to a bogus narrative that either teaching isn't good enough in languages, or as a nation our pupils are not good enough.

• No mainstream options for continuing to study a language in college 16-18. Just a bizarre niche A Level that is barely suitable even for the tiny minority that do take it. Meanwhile, our young people are avid users of apps such as Duolingo. But in the education system, they are unable to continue with language learning.

• Universities offering degree courses to a minority of the minority who took A Level. Cold spots emerging where courses are no longer offered. Universities offering language courses to students of other disciplines. But who have had a 2 year hiatus post GCSE, or potentially since Year 9.

I have heard Professor Charles Forsdick of the British Academy and Cambridge University refer to the system as a series of "cliff edges". After Year 6, at Year 9 options, and after Year 11 there are built in car crashes where numbers fall off a cliff. But the carnage is not driven by the numbers. It's no accident. The road heads straight for the cliff. Transition. Unfair grading. No mainstream post 16 options. Language learning isn't in jeopardy because not enough people are continuing. It's the other way round. There are reasons why people don't continue. In fact it's a miracle anyone survives through to a degree in languages at all.

What can we do about this? I will examine how we can challenge this system in future posts, starting here by asking if we really need degrees in languages in order to provide language teachers. And this one looking at what teachers might be trying to achieve were the system not stopping them.

But meanwhile one thing is for certain: this is not about the quality or nature of teaching and learning. One look at the broken system tells you that we are not going to solve this by claiming that this or that aspect of language teaching will motivate pupils more. Teachers make each section of the system work as well as it can work. What we need to do is challenge the systemic issues that make the learning a dead-end for the vast majority.

GCSE Options, Take-up, Ebacc? Pupil survey.

 

Year 9 Survey April 2026

We have surveyed our Year 9 pupils after they picked their GCSE options, to see if we can find useful information on what shaped their choices and their attitudes to French. We are a school which did not go down the road of compulsory languages at KS4 for any pupils. Or even any guided pathways for certain groups of pupils. So with the removal of the ebacc, our experiences may turn out to be relevant to not just our school, but to other schools suddently finding the rules of the game have changed.

Our school is very supportive of languages. But not over and above the other subjects. And we historically have a very strong ethos of creative subjects, dating back to our former Specialist Status in literacy, performance and art. These are all aspects that schools will need to take into account, as they juggle the impact of the end of the ebacc alongside new "buckets" for creative subjects. So again, our school may be an interesting one to look at when it comes to the fate of languages.

We normally have about 25% of pupils picking a language at KS4. I don't know if this is what awaits other schools with the changes to performance measures. You would have hoped that schools who told pupils taking a language was important did so because they believed it. Not because they were chasing a bogus performance measure. But stories are certainly starting to circulate of some schools changing their options system. And it turns out that this is the single most significant thing in determining take up of languages. Not the quality of the curriculum or the teaching, or the intake, or parental attitudes. Where schools had high take-up of languages, it was generally because the school made it compulsory for some or all pupils.

What might some of the lessons be from a school that never went down this route?

Firstly, there is no longer a pupil who "should" be doing a language. That high flyer, successful in languages, ambitious and keen: they don't feel they "should" be doing a language. And they quite probably won't. All the other subjects on the list are also perfectly valid choices. And we find we generally have more lower prior attainment pupils picking French.

Furthermore, if this is reproduced across the country, with schools no longer pushing "higher set" pupils into languages, it could completely change the profile of the candidates sitting MFL. And further muddy the waters over the shape of the "curve". As the graph below from FFT Datalab shows, the profile of pupils with lower prior attainment currently taking GCSE is unlike that of other subjects. The last time we had a sudden change to the nature of the cohort, the consequences for MFL grading were chaotic, as shown in this post about the grading of pupils in 2005.

And what about the survey of attitudes of our Year 9 pupils? The graph below would seem to show that lower take-up of languages doesn't mean an outright rejection. It is not the "failure" that has been characterised in some reports. Most pupils were considering taking a language and may have even put it as a second choice. Schools where take up was "low" weren't failing in languages. Take-up was just at a natural level, given the number of subjects on offer. Pupils were seriously considering taking French, but didn't see themselves as "I am the sort of pupil who ought to be taking a language". That idea just does not apply. It exists only in the heads of people who went to school in the olden days, and we need to let go of it.

Year 9 Survey April 2026

This graph is to some extent misleading. Only 50 pupils out of 180 responded to the survey. And it is over represented by the blue section. We do NOT have 38% of pupils taking French next year. The true figure is closer to 25%. This is in line with previous years, except that this year we are no longer offering Spanish, and only a few of the pupils who were hoping to take Spanish have opted for French. So although it means numbers for languages has stayed around 25%, it does mean an increase in the number of pupils opting for French.

So what are the kind of things that pupils are taking into consideration when it comes to options? Is it the usefulness of languages for careers, university applications or travel? Is it their experience of language learning in KS2 and KS3 and the progress they feel they have made? Is it their knowledge of the GCSE exam, perhaps relayed to them by older pupils?

Here's what they said:

Factors influencing my decision NOT to pick French

Factors influencing my decision to pick French

The first graph shows the pupils who did not pick French, including the ones who were seriously considering it or put it as a second choice. The second graph shows the pupils who did pick French.

First of all, "Plans I have for the future" stands out as a reason for NOT picking French. I have written before about how early our pupils seem to have very specific ideas of what they plan to do in the future. And in most cases, a language isn't relevant to their plans. It seems from comparing the two graphs that we are more likely to attract pupils who are more open minded about their future.

Secondly, experience and sense of progress in French so far doesn't come out strongly as a negative influence. It's more likely to be a positive factor.

And overall, "interests and passions" seems an important factor.

It is interesting that the GCSE and grading don't seem to come out as an important factor. But as I explained here, we are careful to hide the truth of GCSE grading from pupils. If we told them it's harder to get a GCSE in languages, it would of course be off putting. In fact it's the assumption that grading is fair, that leads people to the conclusion that we must be bad at learning languages or bad at teaching languages.

The survey also fits in with conversations we had with pupils around their options. So I know those pupils who "should" be taking a language but eventually decided not to. Because they came to talk to me about it. For example our blurb suggests that if you want to be a Primary teacher you should seriously consider taking a language and one pupil wanted to explore whether not taking a language would be an obstacle to them entering the profession. Of course we had to say that it would not. And given that their French is already of this standard (open link for examples) in Year 9, plus they study Spanish on Duolingo, I think they would be well placed to teach a language in KS2. And I have other stories of pupils whose experience of language learning has already been important in their lives. It has taken them places and given them experiences they otherwise wouldn't have had. They have learned how language learning works, and they continue to study different languages online. They have achieved a level in KS3 that would have been GCSE for their parents' generation. (Yes, I taught many of their parents.) And they have made carefully considered decisions based on personal interest and future plans.

What should we conclude from this? We shall see the fallout from changes to the performance measures. And maybe start to think more deeply about the purpose and place of language learning in schools. And how the current structures are not really fit for purpose. But that is for another post...

Tiering Decisions. Pupils suffer. Teachers get the blame.

 The situation we find ourselves in with regard to Tier decisions for the new GCSE is unconscionable.

Deciding whether to enter a pupil for Higher or Foundation Tier will have an effect on their grade. A pupil capable of getting a grade 6 will be limited to a grade 5 if entered for Foundation. A pupil who falls short on Higher is in real danger of getting a U.

In a normal year, this is a tricky situation. And the particularly low number of pupils getting a grade 6 in languages is a sign that schools know they have to play it safe and enter pupils for Foundation Tier. Especially schools who have had their fingers burned with pupils getting a U.

But this year is different. Very different. And much, much worse. It is an impossible situation for schools, and deeply unfair for pupils.

We have a new exam. With new style papers, different content, but the same standards. The Sample Assessment Materials seem much more accessible than the old past papers, but we have no grade boundaries to determine what the pupils' marks mean or what Tier they should be entered for.

What can schools do?

1. On Speaking and Writing, you can gauge pupils' performance against the standards stated in the specification (although this has been the subject of much prevarication) and make judgments of parity with expectations in the old exam. Can pupils speak and write coherently and accurately at a level that we recognise as matching a certain grade? We can make some judgements here.

2. On Listening and Reading, you can invent grade boundaries for the Sample Assessment Materials exam papers. You can adjust these to be more cautious given that the papers seem more accessible, especially the listening. And you can back this up on hearing anecdotal evidence from teachers at schools where large numbers of pupils are scoring full marks or close to full marks on the papers. You can use your impressionistic knowledge of your pupils and how they compare to previous cohorts. You can moderate with other schools and trusts to see if your invented grade boundaries deliver reasonable outcomes when you plot larger numbers of pupils onto a curve. Ofqual have optimistically offered guidance on the shape of the curve, even though they know that this information is useless for individual schools.

3. You can identify the "cross over" questions which appear on both Higher and Foundation and check how pupils are doing on these key questions, to see if they would perform well at the lower end of Higher Tier, which presumably is aimed at grades 4 to 5.

4. You can give pupils old style past papers, for which we do have reliable grade boundaries. Because although the content is different and the exams are very different, the standards should not have changed.

5. You can look at pupils' "targets" and decide if you think they are working well and would be on track to meet expectations.

What happens when you do this?

1. Pupils perform well on Speaking and Writing. But we know that our pupils have always had a "spiky profile", doing better on some papers and worse on others. While encouraging that our pupils can speak and write well, if other schools' pupils also perform at this level, then this is not what is going to determine their grade. I currently have pupils with very strong Speaking and Writing who are entered for Foundation Tier because their grade will depend on the Listening and Reading.

2. Moderation with other schools confirms that the papers are more accessible. And that we need to be conservative with our grade boundaries, avoiding allocating high grades on Listening to pupils who come out with marks much higher than in previous years. While useful in demonstrating that we have done our best to deal with the impossible situation, it doesn't give enough confidence to trust the grade boundaries when it comes to making Tier decisions that could limit or wipe out a pupil's grade.

3. Looking at the cross over questions: do we really make decisions on individual pupils' tier of entry based on a couple of questions rather than their overall mark? What if pupils seem to do about as well on these questions as on the others on the paper? Does this mean we should calibrate downwards their grade because they are underperforming on these key questions? Or calibrate it upwards because they are doing as well on some "fully Higher" questions as they are on the cross-over questions? If these questions are aimed at pupils in the grade range of 4-5, these pupils are not the ones who should be entered for Higher Tier anyway. This exercise just seems to confirm that pupils are finding the SAMs accessible and that the grade boundaries this year could be ridiculously high if the real exams are anything like the SAMs.

4. There is no point giving pupils the old style Higher Listening paper. As the marks for all pupils from grade 4 to grade 7 all tend to fall into a narrow range of marks. A few marks either way would totally change the grade. So it's not a reliable indicator of anything useful. Therefore we are dependent on the old style Higher Reading paper to try to calibrate pupil performance. Firstly, it becomes painfully obvious how bad the old papers were. Pupils who know the vocabulary and grammar are bamboozled (quite deliberately - that's how the papers are set up) and come out with very low marks. Secondly, how do you use the information? Do you use the old Reading grade to entirely set the expectation of the individual pupil's overall expected grade? Do you use it to specifically check against the individual pupil's SAMs Reading grade and calibrate that? Do you use your cohort's performance on the old paper to get a general calibration of how far all of your grades are (or are not) in line with your invented grades based on the SAMs? How many goes do you have at old past papers and which of those do you take as significant if the grades come out different every time? When does it start to become depressing for the pupils because of the negative experience? And is it a waste of time because you should be practising the new style papers? If we were to go by the old style Reading, we would be entering nearly all of our pupils for Foundation Tier.

5. Pupils and parents look at the "targets" and the teacher's report saying that the pupil is working well. And wonder why they are not being entered for Higher Tier if their target is a 6. So we have to explain that the targets this year were homemade because the pupils didn't do SATs in KS2. And that pupils this year have been given the same target in all subjects, regardless of the fact that the grades given out in languages are a grade lower in languages than in other subjects.

In other words, we are doing an incredible amount of hard work but we are still completely in the dark in making a decision which will directly affect pupils' grades, and which could result in them getting a U.

There is nothing more that teachers can do. We have been badly let down by the exam boards, ofqual and the DfE. There are whisperings that something is going to be done about grading. Which would be welcome. But just throws another unknown into the mix. People are saying that it's going to be really embarrasing if pupils score high % marks but get low grades or no grade at all. We need absolute clarity on this now, so we can enter pupils for the correct tier. If I enter them for Higher on the basis of these whispers, I am gambling with their grade. If I enter them for Foundation, I am limiting them to a grade 5.

Either grades are going to be more generous. Because everyone knows they have been unfair for years. And not having the baseline of SATs is an opportunity for the DfE and ofqual to tweak this. As well as the declared aims of the new GCSE to have a better curriculum and better teaching, which ought to be rewarded with better outcomes. In which case if we put pupils in for Foundation, we will be penalised.

Or grading standards will be maintained. Because that is ofqual's brief. In which case there will be a bloodbath of pupils getting a grade U because their teacher made the decision to enter them for Higher Tier based on their high % score on the SAMs.

So we have to guess and gamble one way or the other.

When the old new GCSE came in, for the first year there was a wider grade 3 safety net on the Higher Tier. The fact that we have not been given a safety net this time round, even though the situation is even trickier, is a dereliction of duty between the exam boards, ofqual and the DfE, with no-one taking responsiblity. They are all waiting to see how things turn out. And teachers will get the blame and pupils suffer the consequences. This is disgraceful.

Language World 2026

 Really enjoyed Language World 2026. Meeting old friends and online friends and new friends. It's always about recharging batteries and making connections, chatting over breakfast or standing in a queue.

My talk was about how languages has the benefit of many complementary facets: grammar, communication, usefulness, linguistics, culture, memorisation, self expression, risk taking, accuracy... All in half an hour.

Here it is. My manifesto for how to teach in the messy middle between the extremes where we are told everything we do is wrong:

I showed examples from the classroom and of pupils' work. But it may leave you with questions about exactly how we get Y9 working on spontaneous extended answers in speaking and writing.

Things like:

I can tell you. But it won't fit into half an hour!

If you really want to, you could catch up with my 2020 Language World talk that takes you through the sorts of things we do:

Can I find examples to explain why step one of the knowledge approach is flawed?

 In a post on the knowledge theory of learning called "Spherical Children in a Vacuum", I made this quick characterisation of Direct Instruction/Knowledge based teaching:

You break down the knowledge into carefully chosen chunks. You make sure that each chunk is carefully defined. You use regular low stakes recall to make sure that memorisation is happening over time. You make sure explanations take account of all possible misconceptions, with words and images carefully chosen to be clear rather than decorative or distracting. You deliberately contrast and discriminate between important concepts so they are not confused. You meticulously order the content of the curriculum so nothing is extraneous or redundant, and everything is consolidated and built upon. You make sure pupils can recite the best definition of a concept and can quote verbatim the best evidence.

And I suggested that it is fundamentally flawed. How can it be flawed? Are all the things on this list not A Good Thing? They certainly seem sensible or even important. But what if there is a mistaken assumption in the very first step. "You break down the knowledge." How can this be wrong? What other approach is there?

I am going to draw first on my experience of language learners. These could be hypothetical situations to make a point, but they are also true personal stories.

Imagine you are tasked with teaching the bosses of a small high tech business in a converted old school building in a village in Switzerland who speak no English how to read documents from the British Ministry of Defence on the specification of some diving equipment that doesn't make bubbles because it recycles and remixes the exhaled gases. (They initially wanted me to translate the classified documents but didn't realise that I couldn't do this without being allowed to read them.) How would you set about teaching them English? Maybe you would start with grammar. You could chop up the knowledge in a logical way. They could start with the present tense of regular verbs. Then some irregular verbs. Then some other tenses introduced one by one until they built up the overall picture of the language.  Ensuring mastery at every step through a "novice" stage in which knowledge is introduced, consolidated and built on. And eventually they would reach the "expert" level where they could read the secret documents.

Now imagine you are living with a large family in Mexico who speak no English. An English friend comes out to stay. They have a C in O Level French and they teach themselves colours, numbers, days of the week in Spanish on the aeroplane. What would their language learning experience be?

The family could have fun explicitly teaching them. Names of foods, how to give opinions, aspects of daily routine and interaction, some rhymes, songs and sayings. But with no thought of how the conceptualisation of the language is building up. No careful breaking down of the overall grammar system. Instead, a direct response to the individual learner and the situation they find themselves in.

And in interactions outside the house. Immersive exposure to Spanish in social, transactional and professional situations. How much of this input is Comprehensible Input that the human brain can learn from? Almost all of it. Because Comprehension isn't a function of the words and the grammar. It's a function of the interaction. You can tell by what the situation is, what the interaction is, what people are showing you, tone of voice, other people's response, the reaction you get when you act on what you think is being said. We need to bear this in mind when we swap to talking about the classroom. One thing we know the human brain can do is acquire languages. Especially when the language is being used to communicate, not as a chopped up thing to be learned. But the debate about whether that is relevant to the classroom situation continues...

Just to satisfy your curiosity, my Swiss employers never did learn enough English to read those documents for themselves. My friend stayed on in Mexico after I returned to England, and within 6 months had Spanish fluent enough to deliver post-graduate science presentations.

What could I have done with those Swiss engineers? Get hold of some similar (non classified) documents and read through with them, showing them how their technical understanding of the content equipped them to extract meaning. Diagrams, technical terms, cognates, figures, and what made sense given their knowledge of the subject. Despite their lack of English, their engagement with the meaning of the texts would have allowed them to recognise important information. And to start to engage with learning the English they needed.

My first point in all this is to illustrate the difference between learning that starts with chopping up the knowledge, and learning that starts with what the learner will immediately grasp hold of. Are you teaching grammar step by step so one day the learner will have the full picture and be able to communicate? Or are you teaching the learner what they need to communicate, and leaving the overall grammatical picture to emerge once they have enough language for patterns to appear?

My second point is that we can't answer that by chosing one of the approaches. It would seem sensible to attempt something that combines the two. And my biggest point is that this overlapping combination doesn't emerge from thinking about what should work. We have to start from the classroom and what we find actually does work.

And here we find things get messy. Learning is not neat and tidy or uniform. And it is not purely cognitive. Whole dimensions of relevance, meaning, importance, enjoyment make a real difference. We saw in this post that using language to say things that have meaning for them (rather than just to practise the language) has an impact on learning. We saw in this post how pupils remember chien and chat even though they never need them in lessons again, but mess up j'ai and je even though they see them every lesson. And pupils who can tell you the difference between travailler and travaillé get it wrong as soon as they are focused on genuinely telling you what job they would like to do.

What does this tell us? If you take a chopping up the knowledge approach, you could say this means we should double down on drilling the difference between travailler and travaillé and make sure you avoid letting pupils express themselves until they are ready. This is an approach based on determining what you think should stick and not moving on until it has.

If you take an approach based on what the learner knows and can do, you observe what sticks and what doesn't, what is salient, relevant, useful. You observe how the learner's body of language grows, and you build a syllabus that deliberately focuses on how the learner's snowball of French is growing, not on making sure your knowledge is delivered in an even covering. As we know when we look out of the window the day after the snowfall, the snowball outlasts the even covering.

You don't need to agree with me, but I hope I have gone some way to making the distinction clear. And I am sure we all agree that it's what happens in the classroom that matters, not on paper.

Does this apply only to language learning?

What about say science? The version of the knowledge approach we see in some schools would start by making sure pupils know definitions and rules. They learn them and then see them in action and then apply them to exam questions. For example making sure that pupils know that an object remains at rest or in uniform motion at constant speed and in a straight line unless acted upon by a force. You can teach pupils this. Make sure they understand it, can repeat it. And then see if they can apply it to what happens when a person A takes their foot off a skateboard B while standing on a slope of gradient y.

What would the opposite approach be? And here the "we never ask pupils questions they haven't been told the answer to" or the "discovery learning doesn't work" people would be horrified. But how about asking pupils what is going on when you push a smallish lump of rock and when you push a marble. Do things keep moving when you stop pushing them? You need to find out if pupils think that things stop moving when you stop pushing them and there must be something strange going on with the marble, maybe to do with its rotating motion. And that maybe different rules govern different things of different weights. The rules seem complicated and to apply differently to different objects. Then, you can show them that there's actually a different way of seeing things that suddenly simplifies everything and covers everything. Things stay still until something moves them. Things stay moving until something stops them. This simple rule explains the rock and the marble.

Isn't this an example of how chopping up the knowledge, giving clear definitions and making sure pupils "know" it and can repeat it, is a partial and defective approach? The question needs to be how we can make something relevant, important and revelatory to pupils, by engaging with their actual thinking, not what they ought to know.

All the steps after the first statement (in bold) are still desirable. 

You break down the knowledge into carefully chosen chunks. You make sure that each chunk is carefully defined. You use regular low stakes recall to make sure that memorisation is happening over time. You make sure explanations take account of all possible misconceptions, with words and images carefully chosen to be clear rather than decorative or distracting. You deliberately contrast and discriminate between important concepts so they are not confused. You meticulously order the content of the curriculum so nothing is extraneous or redundant, and everything is consolidated and built upon. You make sure pupils can recite the best definition of a concept and can quote verbatim the best evidence.

But the first step, with its assumption that the aim of teaching is to break knowledge into chunks and then reconstruct it in the learner's brain, is bogus. Learning is much messier and much more active. Learning isn't flatpack assembly. Learners engage much more creatively with learning. They assign importance, use or ignore features, transform, link, make associations and extrapolations, imagine implications, create, question, contort and pervert the knowledge they are given.

Assuming it's all about the knowledge leads to attempts to quash pupils' agency and individuality. When we know learning works best when pupils think, find and give meaning, and take ownership.

Spherical Children in a Vacuum.

 I do like the adage that, "There's nothing so practical as a good theory." In CPD, an insightful understanding that equips you to plan and understand what is going on, is much better than a bunch of lesson activities to try. And we don't learn items of knowledge in isolation. We learn by attaching new knowledge to concepts we already have. If that sounds simplistic, don't worry; it's meant to.

We have lived through a decade where a theory of learning and knowledge has been in vogue. And if you were to come up with a theory of how to teach spherical pupils in a vacuum, then it's a great theory. You break down the knowledge into carefully chosen chunks. You make sure that each chunk is carefully defined. You use regular low stakes recall to make sure that memorisation is happening over time. You make sure explanations take account of all possible misconceptions, with words and images carefully chosen to be clear rather than decorative or distracting. You deliberately contrast and discriminate between important concepts so they are not confused. You meticulously order the content of the curriculum so nothing is extraneous or redundant, and everything is consolidated and built upon. You make sure pupils can recite the best definition of a concept and can quote verbatim the best evidence.

Absolutely fantastic as a theory.

Where it falls down is the amount of effort it requires to make our pupils conform to being spherical and in a vacuum. Some of the measures are more reasonable than others. Spherical pupils have regulation haircuts, carry their bag in the correct hand, follow the text with a ruler, answer the teacher in well projected full sentences in standard English, keep eye contact with the teacher at all times, smile and say good morning when bidden, copy knowledge organisers respecting the exact format. Pupils in a vacuum walk along a painted line in silence between lessons, have their phones locked away or preferably do not have phones at all, cannot visit the shops on their way home, resist the "degenerate" culture of their milieu, are marked for their conversation on set topics at the table, are exposed only to the authoritative voices of the "best". And are considered "novices" not yet capable of creativity, expression or original thought.

The sheer amount of effort put in to making pupils conform to being spherical and in a vacuum ought to tell us that the theory is great... as a theory.

In fact, on contact with real pupils, it turns out it may be a terrible theory. We know that the pupils who learn best are the pupils who think about what they are learning. The pupils who question what they are being told. The pupils who think of links, implications, associations, imaginative tangents. Pupils who know they have the agency to examine new knowledge in the light of what they already know. Pupils who reconsider their existing ideas in the light of new knowledge. Pupils who take risks and learn from their mistakes. Learning is an on-going process that they are in charge of. We know this is what our best learners do.

When I sit in a CPD session taking notes on what the important man is saying, I am not writing down what he says! I am writing down what the things he says make me think. Or what I think about the things he says. If he's lucky. I may be planning my lessons for tomorrow and trying to block out his predictable nonsense. And I am an excellent learner.

And we know that the pupils who struggle with learning are the ones who have been made to feel that they don't have that agency. Their thoughts were wrong. Their ideas were unwanted. Their associations and conceptualisations were invalid. Learning now consists of having to ingest more stuff and be tested on it. Real learning, in the areas where they have been made to feel like this, has almost stopped. The emotional link between getting a question right/wrong and being morally right/wrong is painful. They have been labelled as bad and their agency has been taken away.

Yes. All this is another theory or model. But it's also observed in the reality in the classroom. Which the knowledge theory tacitly acknowledges by its insistence on making all pupils spherical rather than individuals.

Having a theory that works fine for pupils who conform to how they "should" be rather than how they are, is harmful. It's the thoughts of "should, should, should" that have to be teased out and confronted once pupils' mental health starts to suffer.

Let's see if we can start to get to somewhere more positive and see how we can successfully bring the knowledge theory into contact with reality.

Writer and CPD expert Sue Cowley alerted me on bluesky to this example where Ofsted start from what they think is a logical chunking of the knowledge, rather than from an expert understanding of children and learning:

It reminded me of this Ofsted video:

The image illustrates their idea that knowledge preceeds skills. I can't think of a worse example. It immediately debunks their whole idea. Finding out what shapes fit where is a perfect example of where learning can happen through play... even maybe by... discovery. A child who arrives in a school setting without any knowledge of shapes, may lack the cultural capital of having played with a shape sorter. Not the cultural capital of having had direct instruction of what shape goes where.

In language learning, we have Scott Thornbury's idea of the omelette. (Make your own link to spherical chickens.) You do not teach someone how to make an omelette by chopping up a cold dead omelette and asking them to reassemble it. You make an omelette from raw ingredients. Stirring, adjusting the heat, paying close attention to what is happening in the pan.

It's the same for language-learning. To change metaphors again, we are helping pupils acquire their own snowball of French. With slow accretion, where more French sticks to what they already have, rolling from one topic to another getting bigger and bigger. We are not aiming for an even coverage, with everything planned out and programmed. A snowball will outlast the even coverage once the snow has all melted away.

The problem with the knowledge theory was the very first line in the description near the start of this post. Now you know why I put it in bold type. The very first step is a mistake: You break down the knowledge into carefully chosen chunks.

Chopping up the knowledge into logical chunks, is not the same as building up knowledge, concepts, skills. The process of building up the pupil's accretion of knowledge and skills is different to chopping up the knowledge.

A starfish when it regenerates does not necessarily regrow an arm. It could be that the arm regrows the starfish. When learners have a chunk of knowledge, but don't yet have the whole starfish, their brains still see how it can grow, what shapes it can take, what it can graft onto, how it can feed, move, reproduce. It's no good getting cross because they were meant to wait for when you thought they were ready for the next chunk to be attached to the first. If they did that, the first bit of starfish would be as dead as a cold omelette before they get your next bit.

Stop pulling apart dead starfish and posting them into children's brains like shape sorters. Start looking at what's growing. [Regretting this metaphor.] Instead of starting from pulling apart the knowledge, we should be looking at how learning happens, how learning grows organically.

This is what the Frameworks of Objectives did if you used it intelligently rather than as a tickbox. This is what APP did if you were allowed to use it intelligently rather than just for assessment. This may even be what David Didau did in English language, with a focus on skills broken down and built up.

In languages, breaking down the grammar does not reflect how pupils learn. Firstly, beginners make strong links between the French they are learning and the real world. Links within the language come much later. So when we teach pets, I know that j'ai, je n'ai pas, un, une and the rules for plurals, the phonics in oiseau or poisson are all going to be much more important than chien or chat. But the learners are focused on the meaning of the words. They will remember chien even though we never use it. But they will mix up je/j'ai lesson after lesson. We're not teaching from how the content "should" be broken down. We're working with what the pupils are learning.

I have fantastic ways of introducing the perfect tense in French. I can explain it formally, I can introduce it through cheats like j'ai décidé de, I can talk about fiancé meaning engaged, I can contrast mange/manger/mangé, I can physically get pupils to change the form... But as soon as pupils are let loose on their own writing, they are not thinking in the same way as someone who has a complete coherent picture of the grammar and wants to show it. They are concentrating on saying things they want to say. Not on the forms of the endings. As human beings, they are approaching language from a different angle. Their partial interlanguage is forming from raw ingredients, not from chopped up things I gave them. 

Please do click on some of the links to see practical examples of me engaging with this in the classroom. It will probably make more sense than spherical chickens and snowballs and eggs.

Talking to pupils about their accretion of learning: https://whoteacheslanguages.blogspot.com/2026/02/reflecting-with-pupils-on-beliefs-about.html

Why meaning matters even in terms of "change in long term memory": https://whoteacheslanguages.blogspot.com/2025/12/what-do-we-mean-by-meaning-and-does-it.html

Here's a final positive message about how cognitive science does make total sense in the classroom:

If you are a teacher who wonders about the fact you seem to be trying things out, watching what happens, trying again... and you wonder if there's a theory that tells you what you SHOULD be doing. Well there is. It's called cognitive science or cognitive load theory. It tells you that pupils learn by thinking about what they are learning. It tells you that pupils learn by making links and networks of knowledge. It tells you that there's a sweetspot between making things engaging and making things distracting. There's a sweetspot between being challenging and being overwhelming. It doesn't tell you where that sweetspot is. It just tells you it's important. So what all of this is telling you is that you were doing the right thing all along. Teaching happens in the classroom with real children and a teacher who constantly monitors, adjusts and reflects.

I have written a follow up post to try to give examples of what I mean by teaching the knowledge versus teaching the pupil. Click here to see if it makes sense: https://whoteacheslanguages.blogspot.com/2026/02/can-i-find-examples-to-explain-why-step.html