Your mistakes are a short list

The list is shorter than it feels

Being corrected feels like being told you're wrong about the whole language. It almost never is. Sit down with a few weeks of someone's writing and the errors collapse into a set you can count on your fingers twice over: a gender they never get right, one preposition standing in for three, a tense that's always slightly off, word order that reverts under pressure, an article that shows up where the language doesn't want one.

That's not a small number because the learner is good. It's small because that's how interference works. Your first language has a fixed set of habits, the new one breaks them in a fixed set of places, and you keep arriving at the same handful of junctions and taking the same wrong turn. New vocabulary doesn't add new categories of error, it just gives the old ones more places to happen.

Which is good news, and nobody uses it

A list of fifteen things is a to-do list. It's finite, it can be worked through, and progress on it is visible. Compare that with the feeling most learners carry around instead, which is a vague sense of being generally bad, unattached to anything specific enough to fix.

So why does the list survive for years? Because nothing writes it down. Corrections arrive one at a time, from different places: a teacher on Tuesday, a friend who winces on Friday, a red underline in a message. Each one is about a single sentence, and each one disappears the moment it's been read. Nobody is keeping the tally, so nobody notices that this is the same error for the ninth time.

The result is a strange kind of blindness. Ask a learner what they get wrong most often and you'll usually get one honest example, sometimes two. The rest of the list is invisible to them, even though it's been in plain sight of every person who has ever corrected them.

Writing it down changes the size of the problem

Once the list exists, three things get easier. You can practise deliberately, because you know what to aim at. You can tell progress from noise, because a mistake that stops appearing is a measurable thing. And you can stop treating every correction as evidence about your general ability, because it's now evidence about item four.

You can keep the list by hand. A notebook where every correction gets a line, and every repeat gets a tally mark next to the line it belongs to, will do the whole job. Almost nobody sustains this, for the same reason almost nobody sustains a food diary: the work lands at exactly the moment you least want more work, right after being told you were wrong.

What a list has to do to be useful

It has to group, not accumulate. Twenty separate notes saying "wrong preposition" are still twenty notes; one entry that says "wrong preposition, seen twenty times, here are the last four" is a diagnosis. Grouping is the whole value, and it's the part that's tedious by hand.

It has to survive your enthusiasm. A list you keep for three good weeks tells you about three good weeks. The interesting information is what's still there in month four, because that's the difference between an error you've beaten and an error that was quiet for a while.

And it has to know when to let something go. An entry that hasn't reappeared in a month of regular writing has probably been fixed, and a list that never forgets anything eventually reads like an accusation instead of a plan.

None of this requires cleverness. It requires that something is keeping count while you write, so the list is a by-product of the practice rather than another chore stacked on top of it. Get that far and the shortness of the list stops being a consolation and starts being the reason you can finish it.