Some people think that children should be taught to write computer programs from primary school. Others believe that time would be better spent on reading and writing.

Some people believe that children should start to learn programming when they are still in primary school, while others think that reading and writing are more important at this age. In my opinion, basic skills must come first.

On the one hand, technology is everywhere in modern life. Almost every job today uses computers, and in the future the number of jobs connected with programming will be even bigger. If a child starts early, the subject seems normal and not frightening for him, and he can build a strong base for university. Programming also teaches logical thinking, because a child must divide a big problem into small steps and find his mistakes.

On the other hand, reading and writing are the base of all education. If a ten-year-old child cannot read well, he cannot study history, science or even mathematics, because all these subjects need reading. A child who has problems with reading in primary school usually has problems in every subject later, and it is very difficult to correct this after some years.

In addition, the school day has a limited number of hours. If we add programming lessons, we must remove something else, and probably it will be reading or writing time.

In conclusion, although programming is a useful skill for the future, I believe that primary schools should give priority to reading and writing, and teach programming later, in secondary school.

Opinion is divided over whether primary schools should teach programming or concentrate on reading and writing. Since the school day is fixed, this is a genuine choice rather than a rhetorical one, and in my view literacy must take priority — though the two are less separable than the question implies.

The argument for early programming is not really about producing programmers. Its stronger form is that coding teaches a transferable habit of mind: breaking a problem into steps, testing an idea, and locating your own error rather than waiting to be told where it is. Children rarely encounter that combination elsewhere in a primary curriculum. There is also a fairness argument, since children who meet technical subjects only at fifteen have often already decided such things are not for them, and that decision divides sharply by background and gender.

The case for literacy, however, rests on something the coding argument cannot match. Reading is not one subject among several; it is the medium through which every other subject is taught, so a child who reads poorly at ten is disadvantaged in history, science and mathematics simultaneously. These gaps are cumulative — each year of weak comprehension makes the next year's material harder — and they become extremely expensive to remedy later. A child who begins programming at fourteen loses very little; a child who reaches fourteen reading badly has lost almost everything.

In conclusion, where the two genuinely compete for the same hours, reading and writing should win, because their absence damages all later learning rather than one part of it. Well-taught programming, however, involves a great deal of precise reading and clear expression, so the choice is not as stark as it is usually presented.

Because a school day cannot be extended, arguments about adding programming to primary education are really arguments about what to remove. That makes the opportunity cost the heart of the question, and it is what most discussions of it avoid. My position is that literacy has the stronger claim, for a reason that has nothing to do with the relative usefulness of the two subjects.

The case for early coding is better than its caricature. Nobody serious expects nine-year-olds to become software engineers; the claim is that programming teaches decomposition, hypothesis-testing and self-diagnosis of error — a child debugging a program is doing something almost no other primary activity demands, which is locating their own mistake without an adult identifying it first. The equity argument is more powerful still. Children who first encounter technical subjects in their teens have frequently already concluded that such work belongs to other people, a conclusion that tracks gender and family background closely, and early exposure disrupts that gatekeeping before it hardens.

What decides the question, though, is not which subject is more valuable but how the two behave when delayed. Literacy is the substrate of every other subject: a child who reads poorly at ten is not weak at reading, they are weak at history, science and word problems in mathematics, because all of it arrives through text. Worse, the deficit compounds — each year of shaky comprehension makes the following year's material less accessible, so the gap widens without anyone doing anything wrong. Programming has no comparable property. A fourteen-year-old with strong reading and reasoning can acquire it quickly, and adults do so routinely; nobody acquires fluent reading at thirty with anything like the same ease. Where a delay is recoverable in one case and cumulative in the other, the choice is not close.

Having said that, the dichotomy is partly an artefact of how both subjects are imagined. Coding taught properly is an unusually demanding literacy activity — writing precise instructions, reading unfamiliar notation, expressing a process unambiguously — and reading taught properly involves inference and structure rather than decoding alone. Much of the apparent conflict comes from picturing programming as clicking and reading as reciting.

My conclusion is therefore firm but narrow. Where the hours genuinely compete, literacy must win, because its failures are the only ones that make everything else unteachable. But schools should stop treating the two as rivals, since the version of programming worth teaching young children is largely a way of practising the same precision that reading and writing require.