A reform that deserves a much harder look

This article was prompted by a recent post by Steve Smith on FrenchTeacher.net questioning the direction in which modern language teaching in England has travelled under the influence of the traditionalist, knowledge-rich movement. His post made me reflect again on something I have been thinking about for some time.

What follows is my interpretation, not necessarily Steve’s. And I should make one caveat from the outset: I cannot prove a simple causal relationship between curriculum reform and the decline of French and German. Educational systems are far too complicated for that.

But the facts themselves are becoming very difficult to ignore.

In 2025, Spanish finally overtook French as the most popular GCSE language. Spanish entries increased by 2.6%, French fell by 1.4%, and German fell by a further 7%. Spanish entries had risen from around 109,600 in 2020 to almost 137,000 in 2025, while French stood at around 133,000. German’s decline has been much more severe. The British Council’s 2026 Language Trends findings confirm the direction of travel: Spanish remains the most popular GCSE language; French fell again, albeit slightly; and German entries were 21.7% below their 2019 level. (British Council)

So here is my deliberately provocative hypothesis:

The knowledge-rich reform that was supposed to strengthen MFL may have helped put the final nail in the coffin of French and German.

Not because grammar is bad. Not because phonics is bad. Not because vocabulary knowledge is unimportant.

But because we may have foregrounded precisely those aspects of French and German which make children feel that these languages are difficult, at exactly the moment when what they desperately need is to experience success.

We have attacked the wrong problem

For years, research and teacher surveys have told us essentially the same thing about pupils who abandon languages.

They perceive them as difficult. They believe they require disproportionate amounts of work. And many fail to see their relevance to their lives and futures.

The broader national picture certainly gives little cause for complacency. The government’s EBacc ambition was for 90% of pupils to study the relevant combination of subjects by 2025. Yet Language Trends 2025 found that teachers in responding schools reported an average of only 53% of Year 11 pupils learning a language for GCSE. The socioeconomic divide is enormous: GCSE language uptake is more than 20 percentage points higher in the most affluent state schools. (British Council)

If difficulty and relevance are central problems, what should language education try to create first?

Surely a sense of:

I understand this. I can say things. I can communicate. I am becoming somebody who can speak another language.

Instead, the knowledge-rich movement has increasingly conceptualised beginner language learning through the explicit acquisition of three bodies of knowledge: vocabulary, grammar and phonics.

The revised GCSE itself reflects that conception. The new subject content for French, German and Spanish, first taught from 2024, formalises a curriculum built strongly around vocabulary, grammar and sound-symbol correspondences. (GOV.UK)

All three matter.

The question is not whether pupils should learn them.

The question is whether making them so visible, explicit and separable is the best way of creating motivated young linguists.

I increasingly suspect that it isn’t.

Why French and German may pay the highest price

This matters because languages are not equally affected by such a curriculum.

French presents beginners with a particularly awkward relationship between written and spoken language. Letters disappear. Word boundaries blur in connected speech. Liaison changes what pupils hear. A word confidently recognised on a page may seem to vanish completely when spoken naturally.

That does not make French intrinsically “too difficult”. But it does mean that foregrounding phonics and orthographic decoding can make one of French’s most intimidating features unusually salient to a beginner. Add to this that the new French curriculum is very grammar heavy and accuracy-orientated, with GCSE course books covering myriads of often challenging structures at break-neck speed. French grammar’s notorious difficulty gets amplified by the current curriculum in a way which is likely to alienate the typical non-resilient, easily distracted and short-term gratification driven teen. Not to mention the issue of relevance: surely, if relevance is key in intrinsic motivation, a grammar heavy curriculum is unlikely to appeal to today’s language learners…

German presents a different problem. Its sound-writing relationship is comparatively transparent. The difficulty lies much more in morphology, grammar and syntax: gender and case, adjective endings, word order, verb placement and the interaction between them.

Again, these things are learnable.

But if we present language primarily as an accumulation of explicit knowledge, French pupils encounter conspicuous phonological complexity and German pupils conspicuous grammatical complexity.

Spanish has an accidental advantage.

Its sound-symbol relationships are considerably more transparent for beginners, and much of its early morphology allows learners to perceive patterns relatively quickly. A knowledge-heavy curriculum therefore risks creating a profoundly ironic outcome: the reform is notionally language-neutral, but its effects may not be.

It may disproportionately expose precisely the features that make French and German feel harder.

And perceptions matter enormously.

A thirteen-year-old deciding whether to continue a language is not conducting a psycholinguistic analysis.

They are asking:

Am I any good at this?

If the answer feels like no, we have probably already lost them.

Where is the “can-do” factor?

This, for me, is the central weakness in the current settlement.

Language learning is unusual because knowledge and performance are psychologically inseparable.

Knowing ten grammatical rules does not necessarily make a child feel like a French speaker. Understanding a phonics rule does not necessarily make them feel capable of understanding French.

But understanding a teacher speaking French does.

Following a story does.

Understanding a joke does.

Reading something they could not have read six months earlier does.

Having somebody say something to them in another language and discovering that they understand it does.

Those experiences generate what might loosely be called the can-do factor.

And that emotional dimension isn’t fluffy decoration around learning. It influences whether pupils persevere with something difficult.

This is why I worry that MFL reform has confused what expert analysis tells us language consists of with what novice learners need language learning to feel like.

Of course language contains grammar, vocabulary and phonology.

But a cathedral contains bricks.

That does not mean the best way to make somebody fall in love with architecture is to begin by sorting the bricks.

And then there is EPI

There is an important complication here, and it should not be brushed aside.

Across England, large numbers of classroom practitioners report considerable success with Extensive Processing Instruction (EPI) and related approaches. Whatever one’s theoretical position on EPI, its influence on current MFL practice is undeniable.

This is important because EPI represents, in several respects, a very different response to the motivational problem. It places considerable emphasis on modelling, highly scaffolded input, repeated processing, extensive exposure and building confidence before expecting pupils to perform language independently.

Language Trends has documented the influence of EPI and changing pedagogical practice among responding schools.

That deserves attention.

But we must distinguish two entirely different claims.

One is:

“Teachers using this approach report better classroom experiences.”

The other is:

“National MFL policy has reversed the decline of language learning.”

Those are not remotely the same proposition.

The national indicators remain deeply troubling. Spanish’s growth is encouraging because it demonstrates that pupils have not simply rejected language learning altogether. But French continues to lose relative ground and German’s position is increasingly precarious. Language Trends 2025 described access to German as close to disappearing in some less affluent settings. (All Languages)

Indeed, the success reported by EPI practitioners may actually strengthen the argument for reconsidering the reform.

If approaches emphasising rich input, extensive processing, scaffolding and early communicative confidence are producing enthusiastic teachers and successful pupils, perhaps policymakers should be asking why.

The primary problem: we may be making a bad situation worse

Then we come to primary languages.

French remains the most widely taught language in English primary schools. Yet transition remains extraordinarily weak. Language Trends 2026 found that almost seven in ten state secondary schools could not organise Year 7 classes according to the language children had learned at primary school. (British Council)

There is already a structural problem here.

Now imagine adding an increasingly explicit knowledge-rich interpretation of language teaching.

We are asking primary teachers — many of whom are not language specialists and may themselves possess relatively limited confidence in French — to teach pronunciation, phonics, grammar and vocabulary explicitly and accurately.

That is an enormous ask.

French pronunciation is precisely the area in which a teacher with limited proficiency is likely to feel least secure.

And explicit grammar instruction requires sufficient linguistic understanding to answer the child’s devastatingly simple follow-up question:

“But why?”

A specialist can move flexibly between explanation, examples and abundant natural language.

A nervous non-specialist cannot always do that.

The danger is that the curriculum therefore becomes increasingly dependent on slides, isolated activities, scripts and packaged materials rather than the teacher’s ability to flood the classroom with comprehensible, engaging French.

And that takes us to a deeper problem which predates the present reform.

Primary French itself may be helping to undermine French

Even if we completely remove the knowledge-rich debate from the equation, I think we need to confront an uncomfortable possibility.

The way French is delivered in some primary schools may itself be contributing to its later decline.

That is not an attack on primary teachers.

Quite the opposite.

We have given teachers a specialist job without consistently giving them specialist linguistic preparation.

Many understandably depend heavily on ready-made schemes of work and online resources — for example, Language Angels. Such resources can offer useful structure and reduce teacher workload. But in my view, reliance on lower-quality or heavily scripted resources cannot substitute for a linguistically confident teacher providing large amounts of accurate, compelling spoken input and genuine interaction.

Language is acquired through enormous exposure to language.

Children need to hear it.

Understand it.

Respond to it.

Play with it.

Read it.

Encounter it again and again in meaningful contexts.

And listening and spontaneous interaction are precisely the areas in which a non-specialist teacher is most likely to lack confidence.

So imagine the child’s trajectory.

French is their first foreign language. They encounter it in primary school. Their experience is fragmented or overly resource-led. They arrive in Year 7 without a strong sense of achievement — and, because transition is so poor, may effectively start again.

Then somebody offers them Spanish.

New language. New beginning. More transparent pronunciation.

Which would you choose?

I suspect that what we sometimes interpret as attraction towards Spanish may partly be flight from an unsuccessful previous experience of French.

That hypothesis urgently deserves proper research.

Spanish has something else: cultural heat

None of this means pedagogy explains everything.

There is another elephant dancing rather cheerfully around the MFL classroom.

Spanish is culturally hot.

French, at least for many English teenagers at present, isn’t.

Spanish has enormous global popular-cultural visibility: Latin music, globally successful artists such as Bad Bunny, football, La Liga, streaming television and social media.

And then there is Spain itself.

For enormous numbers of British families — including working-class families for whom an abstract argument about “global citizenship” may mean very little — Spain is not abstract at all.

It is somewhere they actually go.

The airport announcements are in Spanish. The menus are in Spanish. The waiter speaks Spanish. Their parents attempt Spanish. Their friends go to Spain. They may return year after year.

The British Council itself pointed to Spanish’s growing popularity and cultural appeal when it overtook French at GCSE in 2025, while commentary around the figures highlighted tourism, football, music and the perceived practical reach of Spanish. (British Council)

That is relevance made tangible.

French remains an enormously important world language. France is our neighbour. Francophone Africa will matter increasingly economically and geopolitically. French has extraordinary cultural capital.

But that is not necessarily the France or the French-speaking world being experienced by a fourteen-year-old choosing GCSE options.

If pupils already regard languages as difficult, cultural desirability becomes educationally important.

Spanish begins with a motivational tailwind.

French increasingly begins with a headwind.

And what about senior leadership?

There is another factor that appears repeatedly in conversations among language teachers: the attitude of senior leadership.

This is harder to quantify nationally, and we should be careful not to pretend otherwise.

But it should not be dismissed.

Teachers frequently describe languages being squeezed by option blocks, timetable decisions, staffing shortages, minimum class-size requirements and assumptions about which subjects pupils actually “need”.

The 2026 Language Trends findings show how structural those pressures can become: nearly half of state secondary schools reported being unable to run A-level classes when minimum pupil numbers were not reached, while recruitment problems remained substantial, particularly around German and dual linguists. (British Council)

Once a language becomes marginal, a vicious circle begins.

Numbers fall.

Leadership reduces provision.

Reduced provision makes the subject less visible.

Lower visibility reduces numbers further.

The department shrinks.

Then somebody says:

“There simply isn’t demand.”

But demand is not independent of provision.

Sometimes institutions manufacture the decline they later cite as justification for further contraction.

German should make us particularly nervous here.

This is bigger than one teaching method

None of these explanations works alone.

That is precisely the point.

French and German may be caught in a perfect storm:

• a curriculum philosophy that foregrounds explicit knowledge and may inadvertently magnify pupils’ perceptions of difficulty;
• French phonology and German grammar making those effects potentially language-specific;
• insufficient emphasis on early communicative confidence and the feeling of “I can do this”;
• uneven primary provision, frequently delivered by non-specialists;
• poor primary-secondary continuity;
• excessive reliance in some settings on ready-made schemes and resources rather than abundant high-quality target-language input;
• cultural currents strongly favouring Spanish;
• staffing problems, especially in German;
• timetabling and financial pressures;
• and, in some schools, leadership cultures in which MFL is not sufficiently protected or valued.

Meanwhile, Spanish grows.

That last fact matters enormously.

If all languages were collapsing equally, we might reasonably conclude that English children simply do not want to learn foreign languages.

But they aren’t.

In 2010, government figures show that 25% of pupils at the end of Key Stage 4 entered French, compared with only 9% entering Spanish. By 2020, French had fallen to 20% while Spanish had risen to 17%. Then, in 2025, Spanish finally overtook French outright. (UK Parliament)

Something has changed.

We should be intellectually curious enough to ask what.

Ten years on, where is the transformation?

This is ultimately the question policymakers and advocates of the knowledge-rich turn need to answer.

Not:

Can pupils explain phonics?

Not:

Have departments rewritten their schemes of work?

Not even:

Is the new GCSE theoretically better sequenced?

The question is:

Do more children want to learn languages?

Because if our curriculum becomes more theoretically elegant while fewer pupils want the product, we have solved the wrong problem.

The evidence does not show the motivational transformation that England needs. The government’s EBacc aspirations remain far from reality. Socioeconomic inequalities remain enormous. German continues to contract. French has lost its position as the country’s leading GCSE language. Primary-secondary transition remains badly fractured. (British Council)

And yet Spanish demonstrates that decline is not inevitable.

That should be the most interesting finding of all.

Perhaps we should start with success

None of this requires abandoning grammar.

I teach grammar.

Nor does it require abandoning phonics.

Pronunciation matters enormously.

And nobody seriously disputes that vocabulary knowledge is fundamental.

The argument is about what occupies the foreground of the learner’s experience.

The child should experience language before constantly being invited to anatomise it.

They should hear enormous quantities of comprehensible language.

They should understand messages.

They should read.

They should interact.

They should encounter grammar repeatedly in meaningful language and then have their attention drawn to patterns when doing so helps them.

They should gradually discover that what once sounded like noise now carries meaning.

Because that moment is intoxicating.

It is one of the reasons people fall in love with languages.

And perhaps we have underestimated it.

The uncomfortable conclusion

The decline of French and German cannot honestly be attributed to one policy, one pedagogy, one government or one curriculum reform.

The evidence does not permit that conclusion.

But neither should that complexity become an excuse for refusing to interrogate the dominant policy assumptions of the last decade.

My interpretation is that the evidence now gives us good reason to take a much more uncomfortable hypothesis seriously.

We have attempted to make languages more rigorous by defining more precisely the knowledge pupils should acquire.

But in doing so, we may have made the experience of learning French and German feel more like learning about a language and less like acquiring one.

We may have made their hardest features more visible.

We may have underestimated motivation.

We may have undervalued comprehensible input.

We may have confused explicit knowledge with the psychological experience of becoming a linguist.

And we may have forgotten that before children choose French or German at fourteen, they first have to believe something much simpler:

I can do this.

Spanish currently seems better able to generate that belief — through linguistic accessibility, cultural relevance, travel, popular culture and, perhaps, the simple psychological attraction of a fresh start.

French and German have no inherent reason to die in English schools.

But if we continue treating their decline as an unfortunate mystery rather than asking whether some of our own educational choices have contributed to it, we may eventually make that decline irreversible.

The reform that was supposed to save languages has certainly not saved French and German.

The more disturbing possibility is that it has helped hasten their decline.

And after nearly a decade of travelling down this road, we should at least be brave enough to ask whether it is time to change direction.