2/20: Investigating how we learn language

If you've studied a language, whether at Brown or in high school, think back to your first day of language class. If you were studying here in the US, your instructor probably spoke in  English. They may have introduced you to a few words in the target language. But now, maybe a semester and a half or years later, your course probably no longer has much English spoken, if at all. Though it might seem obvious what changed—you learned the language—it still makes me wonder. How do we make progress learning a language, starting from a baseline of nothing at all? 

When we learn our first language as infants, we are exposed to hundreds of words, in mostly opaque contexts, and without clear 'translations' of their meanings to objects or concepts in the physical and emotional worlds. Eventually, somehow, we learn to understand and speak in our first language and pick up new words from conversations with other people, or at a certain point, from learning to read. 

But language learning, at least the form institutionalized in the education system, obviously doesn't mimic this process. Imagine if you attended a Portuguese 100 course, not knowing any Portuguese, and the professor and a few other native speakers spent each class talking with each other and to you in fluent Portuguese. What would happen after a semester? What would happen if you participated in this situation for hours every day? It's unclear. But unlike infants, as a student or adult, we already have a default 'reference' for the world: our native language, which might get in the way of this kind of learning by exposure.

Regardless of the possibilities of this 'immersive' language learning, our current model instead focuses on gradually building tiers of vocabulary and sentence structures, after some basic introduction to sound, pronunciation, and whatever else may be needed for that particular language, like tonal sounds or a new writing system.  

But if starting from no language knowledge at all, how do instructors or textbooks choose what words we should acquire, and in what order? One pedagogical method common for teaching vocabulary to students in primary schools involves dividing the language into approximately three tiers of vocabulary words. Tier One words typically have fixed meanings, are commonly used, and are more basic, like book, car, dog, red. Tier Three words are complex, context specific and less common unless in that specific context, like isotope, economics, bookbinding. And the sweet spot, Tier Two, contains words that are both more complex and used with higher frequency by mature speakers, like permission, coincidence, masterpiece, and unfortunate. 

From taking note of the structures of my language courses in particular, the earliest stages seem to involve presenting useful Tier One words (especially words that are self-referentially useful for a language learning classroom, like book, study, school, grammar, homework, etc.). But they also involve teaching a more specific type of word: in linguistic terms, functors. These functors, also called functional words or closed-class words, have little concrete meaning, yet are essential to building meaning in combination with other vocabulary—in English, words like of, the, and, to, and for. As students advance past the first level of stringing together words into a meaningful sentence and understanding basic phrases, Tier 1 and 2 words get mixed together on vocab lists, occasionally sprinkled with a more specific Tier 3 words. Of course, this process certainly has differences across different languages. 

But if you're uninterested in vocabulary classifications, this structure does leave a few, more interesting questions. Are phrases and idioms that can't be expressed as just one word considered part of this vocabulary system? How can students learn not just the word as a unit of meaning translated from their native language, but as a unique unit of meaning within the framework of another language and culture? And what about slang? 

The debate over how slang—the most colloquial of language—should be taught to second language learners is ongoing. Slang words are unlikely to be included in most textbook vocabulary lists; they are also unlikely to be used by an instructor. But students who travel abroad and actually speak the language will certainly encounter them. Having no exposure might hinder communication. At the same time, explicitly teaching slang terms might tend toward teaching outdated phrases or overstating how a particular phrase is actually used. Slang is very context-dependent, not only based on its surrounding words, but also on the people speaking it and their circumstances. Internalizing this context and how it relates to language usage seems to require a different type of learning than simple vocabulary matching. 

While this was just a sample of the complicated decisions and questions behind the language learning we experience as students, I hope it encourages you tothink more curiously about how the language learning process is constructed. For me, this type of reflection also helps me put in perspective how far I've come from that first day of class, knowing nothing about the new language—I hope it can do the same for you. 

For a more in-depth exploration, check out this article about using film and tv to teach slang or this paper about the role of WeChat in language learning (from an eco-sociological perspective).