Effective language-learning for effective altruists

By Tao @ 2024-03-19T00:42 (+43)

Epistemic status: Tentative — I have thought about this for some time (~2 years) and have firsthand experience, but have done minimal research into the literature.

TL;DR: Language learning is probably not the best use of your time. Some exceptions might be (1) learning English as a non-native speaker, (2) if you are particularly apt at learning languages, (3) if you see it as leisure and so minimize opportunity costs, (4) if you are aiming at regional specialist roles (e.g., China specialist) and are playing the long game, and more. If you still want to do it, I propose some ways of greatly speeding up the process: practicing artificial immersion by maximizing exposure and language input, learning a few principles of linguistics (e.g., IPA, arbitrariness), learning vocabulary through spaced repetition and active recall (e.g., with Anki), and more.

Motivation: I'd bet that EAs are unusually interested in learning languages (definitely compared to the general population, probably compared to demographically similar populations). This raises two big questions: (1) Does learning a foreign language make sense, from an impact perspective? (2) If it does, how does one do it most effectively?

My goals are: 

• To dissuade most EAs from learning a random language without a clear understanding of the (opportunity) costs.
• To encourage the comparatively few for which language-learning makes sense, and to give them some tips to do so faster and better.

Is this a draft? The reason I am publishing this (late!) on Draft Amnesty Week is that I believe a quality post on effective language learning should draw from the second language acquisition (SLA) literature and make evidence-based claims. I don't have time to do this, so this post is based almost entirely on my own experience and learning from successful polyglots (see "learn from others" below). Still, I think most people approach language learning in such an inefficient way that this post will be valuable to many.

Who am I to say? Spanish is my native language. I have learned two foreign languages: English to level C2 and German to level B2.^[1] I learned both of these faster than my peers,^[2] which I mostly attribute to using the principles detailed below. Many readers will have much more experience learning languages, so  I encourage you to add useful tips or challenge mine in the comments!

What are some costs and benefits?

This is not an exhaustive list!

Benefits: 

•  Access to new jobs, jobs in new regions, or higher likelihood of being hired for certain jobs. This is only the case if you reach an advanced level (probably C1 or C2, at least B2), and is most relevant if you are learning English.
• Access to more resources and news. If you plan to be, say, a regional foreign policy expert, learning the region's language(s) can be necessary.
• Good signaling of conscientiousness and intelligence.
• Cognitive benefits? Language learning purportedly benefits memory, IQ, creativity, and slows down cognitive aging — but I have not gone into this literature and so am not confident either way.
• Greater ability to form social connections. Speaking someone's language and knowing about their culture is a great introduction.

Costs: 

• A LOT of time (depends on the language, the learner, and the method), attention and effort; large opportunity costs. There are ways of speeding up the process, but it is still a particularly costly endeavor.
  • Seriously, think about all you could learn instead. In the same amount of time, you could learn to program in Python, or you could read several textbooks on biotechnology, or get up to speed on AI safety research, or start a blog to increase your visibility, or work part-time, or really anything else.
• (!) Don't be fooled by the many bullet points in the "benefits" section versus only one in the "costs" section. I think the costs likely far outweigh the benefits for most EAs in most circumstances (more on this below).

So when should an EA learn a language?

Here are some factors that contribute to language-learning making sense for someone:

• When the language is itself worth learning for instrumental reasons. Learning English as a second language seems very important. 
  • Other languages that seem particularly useful, depending on your goals and circumstances: Mandarin Chinese, Russian, Spanish, French, Standard Arabic, Hindi. Still, these seem much less valuable for most people than English (if it's not their first language).
  • Consider that some languages will be much harder to learn than others, depending on the languages you already speak. For native English speakers, it is estimated that Spanish and French are relatively easy, Russian and Hindi are about twice as difficult, and Standard Arabic and Mandarin Chinese are about three times as difficult.^[3]
• For better (and more impactful) work opportunities (e.g., learning English).
• For very specific career goals, like becoming a regional specialist. For example, if you are playing the long game and trying to become a China specialist, it may make sense for you to study Mandarin Chinese. Try to be honest with yourself and create a clear theory of change about how learning Language X will be crucial for you.
• When they are talented at learning languages. Please interpret this with caution. I think most people who think they are bad at languages have just been approaching the task the wrong way. Still, as with any cognitively demanding task, some people are just better at it than others.
• When they enjoy it greatly and are energized by it. Some people (myself included) just don't budget language-learning time as "work." If you genuinely enjoy it and are energized by it, then your opportunity costs decrease sharply. If language-learning does not take from your limited attentional resources, it makes more sense to do it.
  • Useful questions to ask yourself are: If you started learning Language X, would you have to stop doing Valuable Activity Y? If you did NOT start learning Language X, would you have the energy and motivation to start doing Valuable Activity Y? Have I wanted to learn Language X for a very long time?
• When they have a unique opportunity. You might have close family members, friends, or a partner that speak your target language (TL), or you might already live in a country that speaks your TL, or you might have a citizenship that allows you to work in a country where your TL is spoken, etc. These factors will make language-learning easier and more engaging.

Principles of Effective Language-Learning

If learning a language makes sense for you after all, here are some ideas to speed up the process (and make it more enjoyable).

• The intuitions
  • I have often found that people have "bad intuitions" about language-learning, particularly monolingual English speakers. Having "better" intuitions is a great way to develop a sense for language-learning.

• Get the basics and create structure.
  • If you're a complete beginner, you'll want to get to a level where you can sort-of understand very basic media in your target language (TL) as fast as possible. You can't jump right into reading a children's book in French without knowing the basics. Try enrolling in a course in your university, local community college, or any other language school, even online ones. Program quality varies a lot, so try to find a good place.
    • Don't worry if you don't like this approach, you'll only have to do it for a short period of time. This is the hardest part of the process — getting from zero comprehension to minimal comprehension.
  • Don't rely exclusively on Duolingo or other language-learning apps to get you out of beginner stage, especially not with only 15 minutes of practice a day. You may use Duolingo if you are seriously going to spend a lot of time per day, but I don't think it's an efficient way to go about getting the basics.
• Do a linguistics crash-course.
  • This is one of the points that I consider most important, and that I see most neglected. You do not need to take a whole course in linguistics like I did. Most of the value comes from (1) some high-level ideas that will give you good intuitions, and (2) a basic understanding of phonetics.
  • Knowing a bit about syntax, specifically, will provide useful schemas for you to organize and retain new knowledge. Example: being able to quickly identify subject, direct/indirect object, verb, adverb, etc. will be advantageous when learning languages with grammatical cases (German, Slavic languages, and many others).
  • On phonetics: this is "a branch of linguistics that studies how humans produce and perceive sounds" (Wikipedia). I cannot emphasize enough how helpful studying phonetics is for language-learning. Here are the things that you should learn:
    • A basic introduction to phonetics, from any source.
    • Your target language's phonetic inventory (a full list of sounds that the language's speakers use to communicate). This will help you differentiate written characters from correct pronunciation. 
      • Notice, for example, the difference in "i" sounds between "leave" and "live." Native Spanish speakers will often pronounce both the same, since our phonetic inventory has only one "i" sound. Native German speakers will pronounce them correctly, because their phonetic inventory includes both sounds.
    • How sound production works. The reason "leave" sounds different from "live" is that the tongue occupies slightly different positions for each. Learn about place of articulation (where the sound is produced), manner of articulation (how the sound is produced), and voicing (whether you vibrate your vocal cords or not).
    • See below a phonetic inventory of American English (which will make sense if you spend even a little bit of time studying phonetics):

•  
  • My recommendation is at the very least watching the Crash Course Linguistics playlist on YouTube and paying special attention to videos #8 and #9 on phonetics. Then you can play around with the interactive IPA chart (International Phonetic Alphabet). If this seems valuable to you, I would  recommend digging deeper independently.
  • Final point on linguistics: although I think this is important, please do not get stuck on only studying linguistics. This is meant as a tool to make your learning more efficient; it can't supplant your TL learning.
• Move on to (artificial) immersion.
  • We all know that the best way to learn a language is to spend time in the country where the language is spoken... or is it? What are the unique benefits of it? Breathing French air won't help you speak French. It is the reading signs on the street and listening to random conversations in French, turning on the TV and it being in French, having to order in French at a coffee shop, having to text a local friend in French, etc. that will increase your language competency.
  • What I described above is called "immersion." But most people can't just move to another country. The good news is, many of these things can be replicated while in your country of origin. This is called "artificial immersion."
    • Ideas: Make it a rule to only watch TL movies and TV shows (or dubbed versions). If it's too hard, start with TL audio and NL subtitles, then switch to TL audio and TL subtitles (but do make the switch as quickly as possible!). Read only books in your TL (there are beginner's books with very simple language and built-in dictionaries). Only listen to music in your TL. Only listen to podcasts in your TL. Switch your phone and laptop's language to your TL. Of course, you'll only be able to do some of these, depending on your current ability to comprehend your TL.
  • The core insight is that people acquire languages mainly through consuming lots and lots of input (listening and reading). Speaking practice and studying grammar will certainly help, but I think people's bottleneck in 90% of the cases is a lack of TL input.
  • Immersion isn't meant to be comfortable. You'll inevitably experience the discomfort of understanding only X% of what is being said in you TL TV show or book. Don't be discouraged. Even if you don't really get the narrative or nuance, you are both picking up on new words, new syntax patterns, intonation patterns, mannerisms, etc. even if you don't realize it. The more input you consume, the more you will understand.
  • Go for L+1 input. This refers to "level +1," or consuming input that is always one level above your comprehension level. This makes the input challenging but somewhat comprehensible. In my experience, you'd want to shoot for 60-90% comprehension, mostly 80%. When you're closer to 60%, you're listening for intonation patterns, the words that you do know and how they're used, trying to identify the individual words when someone is speaking, etc. When you're closer to 90%, you're listening for advanced words, nuances in meaning from conjugations and modal particles, unusual syntax, unknown verbs, etc. People's level of comfort in this regard varies, but the general idea is that discomfort is good — it means you're pushing yourself to learn.
• Ensure recall of the core vocabulary with spaced repetition (Anki!).
  • This, along with practicing artificial immersion, is one of them most important parts of the method. People are often bottlenecked by having poor retrieval of core vocabulary. That is, it's not that they can't understand X conversation or can't speak fluently because of some fuzzy "low language ability" — it's often poor recall of the core vocabulary.
  • By "core vocabulary" I mean the most commonly used words in the language. The more extensively a word is used in natural speech, the more important it is that you know it so well you won't even hesitate when retrieving it.
  • And when I say "recall" I really mean active recall — being able to retrieve this word in conversation or when testing yourself, unaided— not just being able to recognize the word  and think "yeah I knew that word" while staring at the translation.
  • Your approach to learning vocabulary should be to focus single-mindedly on learning the core vocabulary. Try learning the most frequently used 200 words, then the next 300, then the next 500, and so on. When it comes to vocabulary, the Pareto principle applies.
  • So how should you do it? The answer is Anki. Anki is a flashcard learning platform that allows you to create flashcard decks or download pre-made decks. Anki's strength comes from (1) its customizability, but most importantly (2) its built-in spaced repetition algorithm. Spaced repetition, in this context, means re-learning flashcards at increasing time intervals. For example, you might be shown Flashcard X now, then in 10 minutes, then tomorrow, then in three days, then in a week, then in a month, and then in six months, and so on. If you guess the flashcard right every time, you'll keep making the intervals longer. If you fail a flashcard, you'll regress and have to repeat the flashcard again soon. This method quite literally forces you to have flawless recall of vocab. Try looking for pre-made Anki decks of core vocabulary in your TL online, or try making one yourself from a list of the most frequently used words.
• Study grammar.
  • Another reason why people might not understand X Conversation or be able to speak fluently is that they haven't internalized basic grammar knowledge.
  • I think people usually focus too much on explicit grammar instruction, and that most grammar will be acquired intuitively by just consuming lots and lots of input. But input itself can be made more comprehensible by familiarizing yourself with basic grammar.
  • For this, chapters of textbooks might be the best option. Do some explicit grammar learning, and try to notice these patterns when reading natural language or listening to input.
• Learn from others.
  • I learned many of these principles by tinkering on my own, but I owe most of it to learning from others like Stephen Krashen, Steve Kaufman, and Matt vs Japan (especially his website Refold.la, which I highly encourage you to visit).

1. ^{^}

   The standard scale for European languages is the CEFR (Common European Framework of Reference). Language levels range from A1 (complete beginner) to C2 (better than the average native speaker). B2 is high-intermediate. For more information: https://www.languagetesting.com/cefr-scale. 

2. ^{^}

   Just to establish some ethos: I set my high school's highest score on the standardized Cambridge English exam. I later received a diploma for having one of the highest regional scores. After taking two intro German courses in college I skipped a full year of German and took an advanced (C1)  course. At least for German, this seems almost entirely like a product of using the right learning method (for which I advocate here), NOT inherent ability.

3. ^{^}

   See the Defense Language Institute's widely used categorization for how much time it takes to learn a language to the same operational standard: https://www.ausa.org/articles/dlis-language-guidelines. The exact amount of time will vary depending on your time commitment per week and desired proficiency, so focus on the relative differences in expected time commitment.

4. ^{^}