Is It Too Late to Learn English at 40?

You heard it from a teacher back in school, and now you repeat it yourself: you learn a language as a kid or you never learn it. So you opened an app at 38, quit three weeks later, and filed English under trains you missed. The line is wrong.

You didn't miss anything. The research on age and second languages points at something far more specific: what gets hard with age is sounding exactly like a native speaker, not learning the language. The largest study ever run on the question, with 669,498 participants, puts the onset of decline in grammar-learning ability at around 17.4 years old, and that decline is gradual, not a door slamming shut. In the early stages, a 40-year-old learns faster than a child. Hours spent speaking decide your outcome; the years you've lived don't.

The "critical period" was never a door that closes#

The idea of a biological deadline doesn't survive contact with large datasets. Kenji Hakuta, Ellen Bialystok, and Edward Wiley analyzed 2.3 million Spanish- and Chinese-speaking immigrants in the US census, hunting for the abrupt break a biological window would have to produce. It wasn't there. Their conclusion: "the pattern of decline in second-language acquisition failed to produce the necessary discontinuity that is an essential hallmark of a critical period". They found a smooth, continuous slope instead, plus an effect of education level as large as the effect of age of arrival.

Reanalyze the classic datasets with better statistics and the pattern still collapses. Jan Vanhove, of the University of Fribourg, reprocessed two datasets using piecewise regression in PLoS ONE and closed his paper this way: "it is concluded that age patterns in second language acquisition are not governed by a critical period". Age matters. It just doesn't work like a switch.

An adult starts faster than a child#

Adults win the early rounds, and it shows up in the lab and in the classroom alike. The authors of the 669,498-person study put it bluntly: "Studies that compare children and adults exposed to comparable material in the lab or during the initial months of an immersion program show that adults perform better, not worse, than children". Your 40-year-old head carries tools an 8-year-old's doesn't: you know what a subject is, you spot a pattern on sight, and you can study on purpose.

The most useful data for a Spanish speaker comes from Spain. The Barcelona Age Factor project followed roughly 2,000 Catalan-Spanish learners of English who started at ages 4, 8, 11, 14, and 18+, comparing them by accumulated hours of instruction rather than calendar years. At 200 hours and again at 416, the later starters were ahead. Carmen Muñoz, who led the project at the Universitat de Barcelona, sums it up: "The older starters advance faster in the first stages of the process of L2 acquisition, which makes them more efficient learners in the short term".

Your accent is the part that's actually age-sensitive#

Accent and grammar don't age the same way, and conflating them is what keeps the myth alive. James Flege, Grace Yeni-Komshian, and Serena Liu tested 240 native Korean speakers who arrived in the United States between age 1 and 23 and had lived there fifteen years on average, rating their accents and giving them a 144-item grammaticality judgment test. The two measures came apart: "unlike the case for the foreign accent ratings, the effect of AOA on the grammaticality judgment test scores became nonsignificant when variables confounded with AOA were controlled".

A Spanish accent will stay with you, and it costs you nothing that counts. It doesn't stop you from negotiating a contract, following a meeting, or making friends. Vocabulary, grammar, comprehension: the parts that decide whether you can actually work in the language are the parts age barely touches. Adults reach professional working fluency. What they rarely reach is a native-sounding accent, and those are not the same goal.

The clock that counts is hours, not your age#

The useful question isn't "how old am I?" but "how many hours am I in?". Cambridge English puts a number on it: "It takes approximately 200 guided learning hours for a language learner to progress from one level of the Common European Framework of Reference (CEFR) to the next". From zero to a solid B2 runs 500 to 600 guided hours. The US Foreign Service Institute works with comparable figures for adults learning a language close to their own: 24 to 30 weeks, 552 to 690 class hours, to reach professional working level.

Outside the lab, the highest scores belong to working adults. The EF English Proficiency Index 2025, which tested 2.2 million people, reports that in Latin America the youngest test-takers are not the strongest: "the youngest cohort trails working adults by almost 100 points in several countries—the widest gap between age cohorts in the world". Twenty-somethings aren't winning on youth.

Understanding English and speaking it are two different workouts#

The habit that actually stalls you has nothing to do with your age: you train the wrong skill. Listening to podcasts trains you to listen to podcasts. Robert DeKeyser and Yuichi Suzuki, in their chapter on skill acquisition theory, compress decades of experiments into one uncomfortable sentence: "performance was far worse when participants were tested on the reverse skill than when they were tested on the practiced skill". Comprehension practice does not make you fluent at speaking. Production is trained by producing, out loud and under time pressure.

Count the English hours you logged last year, then count how many of them you spent talking. That subtraction explains your plateau far better than your birth date does, and closing it is the whole point of Tabla: every exercise asks you to say the phrase out loud against a speech recognizer instead of picking it out of a list.

The adult brain still remodels itself for this work. A team at Lund University used MRI to measure young adults in intensive language training and found growth in the hippocampus and several cortical areas after just three months; the control group, studying equally demanding non-language subjects, showed no change. Johan Mårtensson, one of the researchers, put it this way: "there is a lot to suggest that learning languages is a good way to keep the brain in shape".

Forty isn't the problem. The problem is starting in January, quitting in February, and starting over the next January, so the hours never add up. Pick a daily slot small enough to survive a bad week, then defend it.

Try this today: set a timer for five minutes and tell the wall what you did yesterday, in English, without stopping to look up words. Write down the three that escaped you.