The Language I Could Speak but Couldn't Think In

For years, some of the words in my head meant the wrong thing.

Not dramatically wrong. That would have been easier. If I had thought water meant fire, somebody would have corrected me before I could build anything on top of it. The dangerous mistakes were smaller. A word meant something close to what I thought it meant, close enough to survive a classroom, an exam, sometimes several years. I could place it in a sentence. I could recognize it on a page. I could repeat the definition the teacher had given us and receive the mark that said I understood.

Then, much later, I would meet the word again in a context where the old meaning could not hold, and I would discover that it was narrower than I had thought, or wider, or that it carried a distinction my first understanding had erased. The correction did not repair one word. It opened a hole beneath every idea I had built using it. I could suddenly see the crooked first brick and the years of wall balanced above it.

Nobody had noticed the hole, including me, because I spoke Arabic fluently.

That sentence is the whole problem. I could speak the language, read it, write it, answer questions in it, and pass through an education conducted in it. By every measurement available to the school, the language had reached me. What the measurements could not see was that I had not reached the world through the language.

In the last writeup, I wrote about the final victory of conquest: teaching a people to forget their ancestors. This is the same argument seen from inside one child’s head. A history can be removed from a schoolbook in a year. Removing the language through which the child might have questioned that book takes longer. It happens one lesson at a time.

I could speak Arabic. I could not think in it without first crossing a distance.

Before a child has vocabulary, he has a world#

I was born in the south of Morocco, near Agadir, and the first world I knew spoke both Tachelhit and Darija. I did not meet either of them as a subject. I learned them from the street: from family and neighbors, from games, arguments, errands, weddings, jokes, warnings, and all the ordinary exchanges through which a child slowly discovers what kind of world he has entered.

I do not mean only that people around me used Amazigh words for objects. A first language is not a box of labels a child sticks onto a finished reality. The language participates in making the reality. It tells him which differences matter enough to receive different names. It teaches him what can be left unsaid, what tone can carry, what a pause means, which joke becomes funnier when the sentence ends early because everyone already knows the missing word.

Tachelhit and Darija arrived attached to faces, places, and consequences. A word for care came with the person who cared for me. A word for danger came with the change in an adult’s voice that meant I should stop before I understood why. Respect was not a definition. It was a map of how to stand near an elder, how to answer, when not to answer, and how the same silence could mean obedience in one moment and hurt in another. These languages entered through an entire body before either of them ever entered through text.

The neighborhood spoke them. The weddings spoke them. The women preparing food, the men moving chairs, the children disappearing into a crowd of relatives and neighbors who all had temporary authority over them, all of it spoke through them. Music did not translate itself before reaching me. The rhythm, the line of poetry, the reply from the other side of the ahwach, the laughter that arrived before I was old enough to understand every layer of the joke—these things taught me that meaning could be shared before it was explained.

This is the first education every child receives, and it is the most successful one. Nobody gives him a grammar lesson, yet he discovers the grammar. Nobody tests his vocabulary, yet he adds to it every day. He learns intention, causality, social rank, danger, love, deception, humor, time, and the weather through a language nobody has yet told him is unsuitable for knowledge.

Then he arrives at school, and the institution behaves as if his mind has arrived empty.

The door of the school was a border#

The classroom spoke Arabic.

We could speak Arabic. This needs to be repeated because the obvious objection to this story is that perhaps we simply did not know the language. We knew it. We were not sitting in total incomprehension while a teacher produced foreign sounds. Arabic was on television, in religion, in public life, and around us in different forms. We could form sentences. We could follow instructions. We could become very good students in it.

But the Arabic of school was not the Tachelhit through which the world first became coherent, and it was not quite the Darija through which ordinary Moroccan life moved either. The language of the textbook arrived already standing above us. It carried correctness. It carried the teacher, the answer key, the exam, the government, the newsreader, the sermon, and the red pen. We did not meet it as another way to say what we knew. We met it as the place where valid knowledge lived.

The first lesson was therefore not mathematics or reading. The first lesson was a hierarchy of minds.

At home, Darija could tell me that somebody was hungry, angry, generous, ashamed, ill, or lying. It could carry a wedding and settle an argument. But the school quietly assigned it a boundary. It was good enough to live in and not good enough to learn through. Formal Arabic was where an idea became serious. A fact said in Darija belonged to the family. The same fact printed in Arabic belonged to knowledge.

No teacher had to say this aloud. Institutions teach most powerfully through arrangement. Put one language in the mouth of the child and another in every book, every examination, every certificate, every path out of poverty, and the child will understand the ranking. He may never develop the vocabulary to describe what happened. He will still learn which part of himself must wait outside.

So I crossed the border every morning.

Fluency is a terrible witness#

The strange thing about language difficulty is that the better you become at hiding it, the less help you receive.

A beginner is visible. He stops, asks, mispronounces, produces the kind of error that interrupts the room. Everyone can see that language stands between him and the lesson. A fluent student is more complicated. He has learned how to keep the room moving. When he does not fully know a word, he infers from context. When a sentence is dense, he holds the beginning in memory until the end makes it probable. When he cannot express the exact question in his head, he asks the smaller question he can build safely. When none of this works, he stays quiet.

The silence looks like understanding.

Fluency measures movement. It tells you that a person can travel through a language at a useful speed. It says almost nothing about what the journey costs him. Two students can give the same correct answer while one of them traveled ten meters and the other crossed a border, translated the signposts, checked whether the translation was socially acceptable, and arrived with no breath left to wonder why the answer was true.

That was the part nobody could see: a fraction of the mind was always occupied by the crossing.

When the teacher asked a question, the work began before the subject began. What does this particular word mean here? Is it the word I know from religion, or does science use it differently? What is the Darija equivalent? Is there a Tachelhit idea nearby that makes this concrete? Can I answer with the words I have, or will my language make a correct thought sound childish? By the time the question reached the part of me capable of solving it, another student was already raising a hand.

Speed is easily mistaken for intelligence in a classroom. The student whose language sits closest to the language of instruction appears to understand faster because he is doing less work before understanding can begin. The student translating invisibly appears slower because only the final stage of his labor is graded. The school times both runners from the same starting gun and refuses to notice that one began a kilometer behind it.

This does not mean every difficulty I had was linguistic. That would be another flattering lie. Sometimes I did not study enough. Sometimes the idea was hard. Sometimes I was wrong in every language available to me. But the existence of ordinary failure does not erase the additional tax. It only makes the tax harder to isolate, because every bad result can be blamed on the child before anyone asks how much of the examination was secretly an examination in translation.

A definition can pass an exam without entering a mind#

Schools are very good at rewarding the appearance of transfer.

The teacher gives a definition. The student copies it. The student memorizes the sequence, reproduces it under pressure, and receives a mark. The words have moved from the board to the notebook to the answer sheet. Because the words moved, we say knowledge moved with them.

But knowledge is not the sentence. Knowledge is what the sentence connects to.

If I learn the definition of erosion but cannot attach it to the land I have seen change after rain, I own a password, not a concept. If I can state a rule of biology but cannot use it to reorganize the living things already present in my experience, then the rule is stored as text waiting for the exam that will retrieve it. It has not joined the rest of my mind. It cannot easily combine with anything. It cannot surprise me.

A first language is powerful partly because it contains so many existing hooks. New knowledge has somewhere to attach. A scientific idea can find the childhood observation it explains, the tool a parent used, the behavior of an animal, the season when the field changed, the household rule that suddenly reveals a hidden logic. The child does not need to create the world and learn the idea at the same time. The world is already there, dense with associations, waiting for a better explanation.

In a less intimate language, the same definition can float. Every individual word is known, but the sentence has no roots. You can carry it for years because the school keeps asking for the same sentence back. Then someone asks you to use the idea somewhere unfamiliar and it collapses. Memorization was not a moral failure. It was the most rational response to a system in which words were easier to transport than meaning.

This is how a student can be fluent, hardworking, successful, and full of holes. The transcript records the floors he reached. It does not record the missing stairs inside them.

The language we borrow when we are ashamed#

The distance did not exist only in school. It entered the most private parts of life too.

There is a habit I began to notice around subjects considered 7chouma: sex, love, desire, the body, and sometimes even ordinary emotion. People who would normally speak Darija or Tachelhit would suddenly reach for Arabic to explain them. The language became more formal at the exact moment the subject became more intimate. It was as if a borrowed register could place a clean sheet between the speaker and what was being said.

This does not mean Arabic is incapable of intimacy. It means we often used it as distance. A word learned from a textbook, a sermon, television, or an expert could be pronounced without feeling as exposed as the equivalent that belonged to the home. Formal language made the subject sound educational, medical, or moral. It allowed us to discuss the human body while pretending that no human body was present.

But the protection has a cost. When the language used to explain an experience is separated from the language in which the experience is actually felt, the explanation can remain outside the person. A young person may learn the approved term for sex without learning how to speak about consent, fear, tenderness, or confusion. He may know a grand word for love and still be unable to tell someone close to him that he is lonely. He may recognize the name of an emotion on a page while lacking words he can use when that emotion enters the room.

Then silence grows in the space between vocabulary and life. What cannot be said naturally cannot be questioned naturally. What cannot be questioned is left to rumor, shame, jokes, warnings, and whatever the street is willing to teach without responsibility. We think we have protected people from the taboo, but often we have protected the taboo from understanding.

This is another way a language hierarchy enters the body. One tongue carries the life itself; another is summoned to make that life respectable enough to mention. Between them stands the person, able to repeat an explanation yet unable to place himself inside it.

The teacher translated too#

There was another clue that the official arrangement was not working: the teacher often had to leave it in order to teach.

A difficult idea would be presented in formal Arabic. Faces would remain still. The teacher would try again, perhaps more slowly, using nearly the same words. Then, at some point, the language would loosen. Darija would enter. In an Amazigh area, Tachelhit might enter if the teacher shared it. A familiar example would appear. The room would change. Shoulders relaxed. Someone laughed. Hands rose. The concept had not become simpler. It had become connected.

Then, before we wrote anything down, the idea had to change clothes again.

The explanation that made us understand was often not the language permitted in the answer. We could discover the concept in Darija or Tachelhit, but we had to return it to Arabic for certification. The living bridge was treated as a temporary classroom trick. The official language remained the only one allowed to appear in the record, so the record showed a system succeeding in Arabic after the teacher had quietly rescued it through another tongue.

This produces a peculiar dishonesty. Everyone in the room knows that the lesson needed translation. The teacher knows. The students know. The examiner may know because the examiner once sat in a similar room. But the institution records only the final Arabic, and therefore never has to admit how much understanding was built outside it.

Darija performs this rescue across Morocco and is then dismissed as informal, broken, merely spoken, not a real language. Tachelhit performs it where teachers and students share the language and then disappears completely from the answer sheet. The languages doing the cognitive labor receive none of the institutional credit.

Then science changed languages again#

Just as Arabic became familiar enough to carry more of the load, another border appeared.

French did not arrive merely as a subject. It arrived attached to science, medicine, engineering, administration, higher education, employment, and the class of people who seemed to move easily through all of them. The exact point of transition has shifted across Moroccan reforms and generations, but the underlying experience is familiar: concepts learned under one vocabulary later reappear under another, and the student is expected to treat the switch as if only the labels changed.

Only the labels did not simply change. The whole map changed with them.

Mathematics is often described as a universal language, which is true only after you have reached the mathematics. The explanation surrounding the equation is not universal. The word problem, the instruction, the distinction between prove, calculate, infer, approximate, and justify, the sentence that tells you what the symbols are doing—these belong to a human language, and misunderstanding one can make you look incapable of the universal thing beneath it.

Science became a test of how rapidly we could rebuild the same conceptual house under French names. Some students had been living near that language for years, through private schooling, family, media, or class position. Others encountered it mainly as schoolwork. Again the runners lined up at a painted line that hid how far each had already traveled. Again linguistic proximity appeared on paper as intelligence.

Morocco has spent decades moving instructional authority between Arabic and French, particularly in science, while Amazigh and Darija remain largely outside the formal contest. The debate usually asks which prestigious language should own the advanced subject. It asks much less often what happens to the child who must repeatedly rebuild his understanding as the ownership changes.

Later, English opened another world for me: research, computing, papers, people, and ideas I might never have reached otherwise. I love what additional languages gave me. I am writing this in English. That fact should prevent an easy conclusion in which the answer is to close every door except the first one.

The problem was never that I learned too many languages. The problem was that I had to abandon the ground each time I entered a new one.

Multilingualism and repeated displacement are not the same thing#

People defend this system by saying languages are an advantage, and they are right. A new language is not only a new vocabulary. It is access to another archive, another humor, another set of intuitions, another population of minds. Arabic connected me to religion, literature, history, and hundreds of millions of people. French opened institutions and texts that Morocco had placed behind it. English opened much of the scientific and technical present. I would not give any of them back.

But multilingualism should mean accumulation. What we often practiced was displacement.

The distinction is simple. In an additive system, the child begins from the language in which he understands most deeply, learns through it, develops literacy and conceptual confidence, and adds other languages as bridges to more worlds. Each language increases his reach. In a subtractive system, every new language becomes evidence that the previous one should retreat. The child does not add a tool. He changes the part of himself allowed to be intelligent.

We congratulate the final multilingual adult and ignore how many children were lost during the crossings.

The survivor makes the system look generous. Look, he speaks four languages. Look how adaptable Moroccans are. What the survivor cannot show us is the classmate who understood the subject but could not make the linguistic transition quickly enough, the student who decided science was not for her when French was the real barrier, the child who concluded he was stupid because the only intelligence the school could hear arrived in a language he did not yet command.

Adaptability is admirable. Requiring it before a child is permitted to learn is not.

The language ladder is also a class ladder#

Languages in Morocco are not arranged only by function. They are arranged by power.

An accent can reveal where someone grew up, what kind of school their parents could afford, which rooms they have already been trained to enter, and whether they expect the person across the desk to accommodate them. A French sentence pronounced with ease can be treated as evidence of competence before its content has been examined. Tachelhit in the wrong room can make the same mind appear provincial before it has said anything at all.

This is why telling children simply to work harder at languages misses the structure. Some children receive the language of later opportunity early, in small classrooms, through books, conversation, travel, and parents who can correct them without humiliation. Others first encounter it as the condition for understanding a subject they are already being graded on. One child learns French and then learns science. Another is expected to do both simultaneously, and the resulting gap is recorded under science.

The language can then launder inequality. The final selection appears neutral: everyone sat the same examination. But sameness at the end does not correct unequal preparation at the beginning. It hides it beautifully.

And because linguistic prestige is audible, the hierarchy reproduces itself in ordinary life. Parents who want to protect their children begin speaking the language of mobility at home, even imperfectly. Tachelhit becomes something the grandparents use with each other. The child understands but answers in Darija. Then the next child understands less. Nobody decided to kill the language. Each family made a practical decision under a system that had already assigned the prices.

Language loss rarely arrives looking like hatred. It arrives looking like good parenting.

What cannot be said cannot join the institution#

The deepest cost is not what the student fails to receive. It is what the country fails to hear from him.

Knowledge does not flow in only one direction, from textbook to child. Every child arrives carrying observations the institution does not yet have. Every community contains expertise formed outside formal education: about land, water, plants, animals, birth, construction, conflict, trade, care, weather, memory, and the local forms of a problem that a national policy can see only from far away.

But for an observation to become recognized knowledge, it must cross into the approved language. The farmer may understand the soil and not possess the French in which an agronomist will publish. The grandmother may understand human behavior with a precision no questionnaire can reproduce and not possess the Arabic in which her insight would be considered quotable. A child may notice a pattern and lack the classroom vocabulary to make the pattern sound intelligent.

We call the failure to cross a lack of knowledge.

This reverses reality. The knowledge existed. The institution lacked the ears for it.

Imagine what it does to a civilization when millions of people must translate an insight before they are allowed to contribute it. Translation itself is not the enemy; human knowledge depends on it. The danger is making translation the price of admission and then confusing failure to pay with having nothing to offer. Every border loses travelers. Every linguistic border around knowledge loses ideas.

Over generations, the loss changes what people attempt. If Tachelhit is never where mathematics happens, nobody develops the mathematical register it needs. Then the absence of vocabulary is cited as proof that mathematics cannot happen there. If the language is not used for scientific writing, scientific terms do not become ordinary. Then its lack of ordinary scientific terms is treated as a natural limitation rather than the predictable result of institutional neglect.

A language is accused of being unable to carry the weight it has never been allowed to lift.

We were allowed to feel in it#

This is the arrangement that troubles me most.

Tachelhit was not completely forbidden. A completely forbidden language creates resistance because the violence is visible. Ours was allowed to remain in the places power considered harmless. It could carry affection. It could carry a joke, a wedding, an old song, a conversation with a grandmother, a memory of the village. It could decorate a festival and appear beneath Arabic on a public sign. It could be heritage.

It was simply denied the future tense.

When a language is allowed to hold only the past, even love can become part of its burial. We speak warmly about preserving it while making every practical decision in favor of something else. We record the old singer but do not create the conditions for the next poet to speak about artificial intelligence, climate change, loneliness in a modern city, or a scientific discovery in the same language. We teach the child that Tachelhit is who his ancestors were and leave him no image of what he might become through it.

Affection without authority turns a language into an heirloom. You handle it carefully, show it to visitors, and never use it to build anything.

I do not want Tachelhit saved only because it is old. Age is not its strongest claim. I want it alive because there are thoughts people can form more honestly inside it now. Because a child should not have to distance himself from his first mind before he is allowed to develop a second one. Because the language that can carry grief and humor can also carry a theorem if we invest in the words, teachers, books, and patience required. Every technical language was once considered incapable of technical thought before people chose to do the work in it.

Beginning from home does not mean staying there#

The answer is not to place an Amazigh child inside one language and lock the door.

He should learn Arabic well. He should learn French if it remains useful to the world he must navigate. He should learn English and any other language that expands what he can read, make, and become. A child near Agadir deserves the Quran, Arabic poetry, Francophone literature, global science, and the internet. He should inherit more worlds than I did, not fewer.

But he should enter them from somewhere.

Education can begin with the language in which the child’s concepts are densest and build deliberate bridges outward. Teach literacy in the home language while introducing the national and international languages systematically. Let the teacher move between languages openly rather than pretending the translation is a failure of discipline. Build bilingual terminology. Allow a student to show the concept before linguistic polish becomes the next objective. Train teachers who can teach across the actual languages in the room, not the imaginary monolingual classroom described by policy.

This is not a sentimental theory. UNESCO’s guidance on multilingual education recommends beginning learning in the language children understand best and adding further languages through trained teachers and context-specific bridges. The principle is almost embarrassingly obvious: understanding is the foundation on which additional languages should be built, not the prize awarded after a child survives them.

Morocco’s difficulty is real. Tachelhit is not the only Amazigh variety. Darija itself varies. Teachers, materials, terminology, scripts, mobility between regions, and the need for shared national languages make implementation hard. There is no switch that produces a perfectly multilingual education system on Monday morning.

But difficulty is not an argument for preserving the current damage. It is an argument for beginning the work honestly. We have spent generations asking children to solve the complexity privately inside their heads. The system saves itself the inconvenience and sends the entire cognitive bill to a six-year-old.

The words I learned too late#

I sometimes wonder which version of me would exist if knowledge had first been allowed to meet me where I was.

Not a version who spoke only Tachelhit. Not a version protected from every difficult word. Difficulty is not the wound, and struggle is part of learning. I mean a version who could spend the struggle on the idea itself. A boy allowed to wrestle with mathematics instead of mathematics plus Arabic, then science plus French, then research plus English. A boy whose first language was treated as cognitive capital rather than baggage he would eventually learn to hide.

Perhaps the result would not be dramatically different. I made it through. I learned the languages. I crossed the borders and built a life in fields my childhood could not name. It is tempting to use that survival as proof that no harm was done.

But survival is a poor measurement of a system. It tells you who arrived, not what they had to leave behind or who disappeared from the road.

I can write in English now. I can think in it, though not in exactly the same way I think in the languages beneath it. Arabic no longer feels like the same wall it once did. French became a tool. Darija remains the fast, improvised country between formal worlds. Tachelhit is still the place where some meanings arrive before their explanation.

The languages do not sit in separate rooms anymore. They interrupt one another, lend one another words, reveal gaps, and sometimes rescue a thought the others could not complete. This is what multilingualism can become when it stops being a sequence of expulsions: not four incomplete selves, but one mind with more than one way home.

What I mourn is that the system made this integration an individual accident. It did not help us build it. It tested whether we could build it alone.

And I keep returning to those words I understood incorrectly for years. I do not remember the first one. That feels appropriate. The injury was never one word large enough to preserve as a story. It was a thousand almost-correct meanings, each too small to trigger an alarm, together shaping the borders of what I could understand and what I believed I was capable of understanding.

They let us speak Arabic and called us fluent. They let us reproduce knowledge and called us educated. They never asked whether the language had become a place where knowledge could live inside us.

I could speak it. I could pass through it. I could make myself legible to every institution that required it.

But before I learned to think in it, I first had to learn that the language I already thought in was not allowed to count.

← all writeups