How Long Does It Take to Learn a Language?
Many people begin a language learning journey at some point in their lives, whether it’s taking Spanish 1 in high school, having parents who force you to attend Chinese School, or partaking in the journey on their own volition. On the contrary, very few people actually finish the journey; almost everyone quits at some point in the middle, whether they give up after three days or reach a point of fluency deemed acceptable, or good enough. But how long does it actually take to reach complete fluency, whatever that means?
Big O
Big O notation describes the growth rate of functions. Some common examples include O(1), O(log(n)), O(n), O(nᵏ), and O(2ⁿ). It describes the asymptotic behavior of functions as the number of inputs approaches infinity, and is extremely helpful to classify algorithms in computer science. For example, finding the maximum value in a sorted array of numbers has a time complexity, or Big O, of O(1). Binary search has a time complexity of O(log(n)), merge sort has a time complexity of O(nlog(n)), and solving the traveling salesman problem by brute force has a time complexity of O(n!).
But what does this have to do with languages?
We can observe the language learning process with a similar approach. For languages with an alphabet foreign to English speakers, such as Arabic, Japanese, Korean, or Thai, it may seem daunting at first, but in reality, learning an alphabet has a time complexity of O(n) to learn n characters, Chinese being an exception due to the number of characters. However, from a practical standpoint, learning an alphabet has a time complexity of O(1); it takes a fixed amount of time to learn, and recognizing any character in the future takes a constant amount of time.
Learning vocabulary is a different story, however. If there are V vocabulary items, then theoretically, learning vocabulary would have a time complexity of O(V). However, factoring in forgetfulness makes it take longer, so in reality, the time complexity would be around O(V⋅f(T)), where f(T) depends on how you review. Grammar and hanzi (Chinese characters) have a similar time complexity due to the requirement to review in order not to forget.
FSI
But time complexity doesn’t tell you the absolute time required to learn any specific language. And plus, who gets to decide whether the time complexities stated above are correct, anyway?
For absolute times required, we can look at another resource: the U.S. Department of State Foreign Service Institute.
The FSI classifies languages into categories based on how much time it takes to learn a language for English speakers. Languages “closer” to English are easier to learn and languages “farther” from English are harder to learn. The tiers are as follows:
- Category I Languages (552-690 class hours): This category includes languages that are similar to English such as Spanish, French, Portuguese, and Danish.
- Category II Languages (828 class hours): This category includes German, Haitian Creole, Indonesian, Malay, and Swahili.
- Category III Languages (1,012 class hours): This includes most languages not in the other categories. Languages present in this category include Burmese, Estonian, Farsi, Hebrew, Russian, and Thai.
- Category IV Languages (2200 class hours): This category consists of the hardest languages for English speakers to learn. It consists of Arabic, Cantonese, Mandarin, Japanese, and Korean.
These numbers hopefully give you a clearer insight into what to expect when learning a language. You won’t get far if the only time you spend is one hour each school day in Spanish class. Even if you have a foreign language class, if you barely put in any effort, you’ll stay stuck at the “¿Dónde está el baño?” level for years.
Alternatively...
If you use the legendary “I’ll do it tomorrow” strategy for language learning, it’ll take you infinite time to learn a language!
Life Update!
I don’t think I’ve mentioned my classes, so I’ll talk about them here.
AP Lit is fun. I think I got lucky with a great teacher; she’s really flexible, the assignments are challenging yet fun, and her feedback is actually helpful! I struggle an unreasonable amount with analyzing poetry and prose, so the class is probably my hardest class as of now.
AP Gov seems like a standard class; we learn about the government, the functions of the branches and a few critical documents, and we have a few assignments. Nothing difficult, but nothing trivial either.
Classical humanities is a class I took merely for the credits, but it’s actually somewhat fun! I usually have a bit time to work on other assignments, and my teacher studied math in college, so we covered a proof of the Pythagorean Theorem as well as a few basics of logical operations. On the logic assignment, the examples of statements or propositions given were the irrationality of the square root of two, the polynomial n² − n + 41, and the Goldbach Conjecture.
AP Stats is probably my most disappointing course; I went into it being told it was more of an English class than a math class, and I was still disappointed. There are so many small details that can lose you points, and for the free-response questions, there are certain types of questions where you paste a sentence and insert specifics pertaining to the problem. “[r² value expressed as a percentage] of the variation in [y-variable] can be explained by the linear relationship with [x-variable]” or “[Yes/No], a linear model [is/is not] appropriate because the data shows a [linear/nonlinear] pattern on the scatterplot, the residual plot has a [random/clear] pattern, and the r² value is [high/low].” Absolutely zero thought has to be put into the FRQs or even the MCQs, but at the same time, you must stay alert the entire time because even the tiniest of mistakes can lose you points. The lack of rigor is also painful. There’s no way to definitively show whether a scatterplot shows a linear trend between two variables other than “It looks somewhat linear”. There’s no specific range for a high r², a moderate r², and a low r². Instead of learning the formula for a normal distribution, we have a table full of areas under a curve from negative infinity to a specific z-score. There’s no actual math involved, just pain.
wow!
ReplyDeletefrom the multilinguist himself
ReplyDelete6 or 7 months
ReplyDelete