What makes a language learner

Ask a hundred people why they never learned a second language and most will say they have no talent for it. The research is unkind to that explanation. Aptitude differences exist and are real, but across large studies they account for a small share of the variation in outcomes. Three other factors matter more, and all three can be changed.

The first is quantity of contact with the language. Hours spent reading, listening and speaking predict progress better than the method used to deliver them. Estimates of the time needed to reach a working professional level from a related language cluster around six hundred to seven hundred and fifty classroom hours, with roughly the same again in private study. For a distant language the figure can be three times higher. These numbers are rough, but they establish the scale: the task is a marathon, and plans that ignore total hours fail regardless of their cleverness.

The second is the difficulty of the input. Material slightly beyond current ability — where most words are known and a few are not — produces faster gains than either comfortable or impossible text. The practical signal is simple: if you must stop to look up more than about one word in twenty, the text is teaching you less than it appears to, because the sentence structure dissolves while you search.

The third is production under feedback. Reading and listening build recognition; speaking and writing build the ability to retrieve. A learner who only reads will understand far more than they can say, a pattern so common that teachers treat it as normal. Retrieval improves only when it is practised and corrected, which is why keeping a record of your own errors is worth more than reading another page.

The tempting conclusion is that method does not matter. That goes too far. Spaced review demonstrably beats massed review for vocabulary; teaching the most frequent two thousand words first gives faster access to real text than alphabetical lists; explicit grammar helps adults more than it helps young children. Method matters at the margin, and margins accumulate over hundreds of hours.

What the evidence does not support is the belief that some people simply cannot. Most people who fail to learn a language did not fail; they stopped, usually in the third month, when the early sense of rapid progress disappears and the long middle begins.