Bootstrapping and Finding the Fun in Language Learning

For my language learners and linguistics friends…

Bootstrapping

I’ve been thinking a lot lately about how I study and learn language versus how it was and is taught. I’ve been looking at a ton of language study apps (for myself and also my wife), and a lot of them have some of the main keys — learning words through multiple vectors and from different direction, interval training, etc — but they all feel pretty terrible at introducing NEW things.

So as I’ve been thinking about how I study when I make flashcards versus how these apps EXPECT me to study, I’ve come down to what I’m calling “bootstrapping” (term stolen from programming probably). That is, when you have no basis for building understanding, how do you get from there to HAVING understanding and having a foundation for building on more knowledge. It’s this idea of bootstrapping that I think most neglected — at least in the discourse of Japanese pedagogy. (There, I’ve gotten the big academic words out of the way, so I can write like a normal person again.)

Please allow me to start with an example of the problem:

I am an intermediate-level Japanese reader, and possibly an advanced-level speaker. But I have holes in my knowledge. Sometimes large, gaping, embarrassing holes (like I only within the last month learned the verb “to guide” and it’s one they use all the time, and now suddenly I hear it EVERYWHERE). So I’ve been trying to fill those holes by starting at the JLPT N5 word list and studying up so I know that I know the whole thing. (N5 is the beginning Japanese proficiency level according to the nationally recognized JLPT system — Japanese Language Proficiency Test. You have to get at least a C on each level to be awarded that level. It goes 5, 4, 3, 2, 1 in increasing proficiency. If you are an N1 speaker, you are probably better than most Japanese at speaking the language and could likely teach their kids how to speak it better.)

After completing out my N5 list one afternoon, I excitedly switched over to the N4 list, ready to finally feel like my knowledge of the language was moving forward. I switched to the new list in the app I was using (Sticky Study, one of the better flashcard study apps I’ve found) and started the process of learning some new words. Unlike in the previous list, N4 had a lot of words I just plain didn’t recognize, neither from kanji nor from pronunciation. And I very quickly became frustrated (I’ll get back to the frustration part). It would show me a handful of words, maybe 5 or 6, then it would review some words I had marked as already knowing, and then when it brought me back around to the new words from before, they looked exactly as foreign to me as they did when I first saw them. There were no lingering pronunciations or meanings trying to become attached to these new words. It became disheartening and I gave up for the day.

And this is a problem. When you are handed 5 new things that don’t have any attachment or reference points in your mind, there’s nothing for them to stick to — no memories or ideas for those new words to affix themselves to. I envision knowledge like a web or lattice system: the more similar things a new piece of information has to attach to, the more stable that information becomes. Also, the longer you spend reinforcing the connections, the longer it sticks together (and over time if you never come back, the connections will wither away. I guess they need time to set at each layer before you can apply the next coating or something. It’s not a perfect analogy.)

So these words, all I had was having seen them on the screen and read the meaning once. I had no memories of seeing the characters before, I had no feeling like I’d ever heard people SAYING the words before. There was essentially nothing to attach them to, and so no memory could be formed.

This is where I feel like we need to improve the bootstrapping of new words and grammar. “Bootstrapping” here (which, I mean, I could be wrong), I am using to mean the process of starting something up from nothing. No foundation, no similarities to other things. Functionally nothing. You give people a vocab list, and the words they are most likely to remember are the cognates first, followed by maybe either the really common words they hear all the time or the ones that they’ve come up with fun mnemonics for. But there always seem to be some set of words that people have a hard time with — words that sound nothing like their own language, or whose meanings are more subtly different and difficult to translate. (Japanese has a lot of both, btw.)

I feel like I have, on my own, been able to overcome this problem pretty regularly, but the technique I use doesn’t seem to be represented anywhere in the literature I’ve read (and certainly isn’t present in the tools I’ve tried).

Basically, I take two arbitrary items and compare them to each other. I build up the framework by hand. I look at one crazy character or word, acknowledge its meaning, pronunciation, etc; and then I look at a second one. I see how it looks different, sounds different, and how its meaning is different. Then, before I’ve had a chance to really forget the first one, I go back to it. First, second, first, second. I do that a few times, and then I add in a third. The third isn’t very difficult because I’ve already learned the first two, so at first it’s just “the new one”. And I go through those three for a while before adding a fourth, and then a fifth, and so on. Sometimes I’ll add more than one, but I never add a new word until I’ve got a pretty solid sense of the difference between the last new one and the other ones I’ve already studied. (Sometimes I call them targets, since “word” is maybe not always the best, er, word to use — maybe morpheme would be more accurate here? But “target” in that it’s the thing you’re aiming at to learn, yeah?)

And after you’ve done this for maybe 5 or 10 or 30 words, you start feeling less like you have nothing going for you and more like you have quite a lot going for you. I used this technique every other day in college to memorize (and let’s not kid ourselves, what I’m talking about here is memorization, not quite learning yet) sometimes as many as 60 new words or more to ace the morning’s vocab quiz.

My tactic was also assisted by a kind of defacto interval training. The better I knew the word, the further back I would put it in the list of vocab cards, so words I had trouble with would come back sooner, ones I already knew would recur less frequently.

The thing that I want to emphasize is that the bootstrapping process in learning new words — that route by which you take something completely alien and make it familiar by comparing it to something else completely alien — seem both vitally necessary and severely underrepresented in the work I’ve seen on language learning.

Finding the Fun

In my career as a game designer, there was a mantra that everyone has used: “Find the fun”. If a game mechanic isn’t fun, then you don’t have a good game. Period. End of line. The single most vitally important thing in making fun games is for the designers to “find the fun”. This is both harder and easier than it seems (depending on how you look at it), but for now, I’ll just leave it at this is the most fundamental job that a game designer has. Everything else is predicated upon knowing that fun exists, and where it is and understanding it.

I think that this should apply to how we live our lives, as well. From a personal and professional standpoint, whether dealing with hardships or learning something new.

Remember how I said I would get back to the frustration I felt? If something isn’t fun, we tend to stop doing it. Or we do it begrudgingly. I don’t want to hear any old-person jibber-jabber about “sometimes you have to do stuff that isn’t fun, it just needs to be done.” I think their definition of fun is either too narrow, or they’ve gone about doing things in a way that makes them more miserable. (Which is totally a thing that people do. Maybe another blog post sometime about that one.)

When I was studying and I came back around to the new words, not only could I not remember them, the only thing I DID remember was that there was a set of words, and visually, yeah, maybe this was one of them. I guess. It was frustrating, and also it wasn’t very fun. So sue me if I want to spend my life enjoying the things I do (wow, I must have some repressed something about that fun thing).

Basically, nobody needs to learn a language that isn’t their native language — and native languages are kind of acquired without having to try too hard. That is, little kids try REALLY hard, but they’re at a stage in their life where just learning and making connections at whatever pace works for them is not only fun, it’s also TOTALLY OKAY for them to be having fun. They haven’t had the sense of “it’s okay to have fun” beaten out of them yet, and so they slowly build up their working knowledge. But for the “grown-ups” of us out there, yeah, we probably don’t NEED to learn another language. You can see this all over the world by looking at immigrants and refugees who never learn the language very well for where they are now living: Spanish-speakers in the US, English speakers in Japan, German speakers in Brazil… You can get by with surprisingly little native language understanding.

So when we say we want to learn a language, we should acknowledge that what we’re setting out to do is, functionally, completely optional (even for schools that require it, you can still get a simple job if you fail out of school, it happens all the time). And if something is optional AND unenjoyable (or regularly frustrating), we’re almost certainly going to give up.

So could you go through the list of five new words over and over again and eventually pick them up? Yeah, probably. And you would almost certainly learn one of them first, know which one that was in order, be looking for it, and then learn the others in the context of knowing that one. But in the meantime, you’ll likely be frustrated and (like I did) put it down.

Another thing I learned as a game designer is that once someone puts something down, if there isn’t a strong draw to come back, they won’t. Once the game is off, it’s gone. Once you stop studying, you’re done. Unless the task was REALLY fun, or you want to learn the language SO MUCH that it overcomes how boring and tedious and frustrating the task it, you’ll give up. I can’t tell you how often I’ve given up and come back and given up again.

Basically, what I’m trying to say is that language learning should always be fun — ALL learning should be fun — or else you’re going to give up and you may never come back.

Level Design

When designing content for a game — levels and puzzles and so forth — your main goal is to push the player without frustrating them. You take what they already know and you add some new thing to it to make it just a little bit trickier, or to make it just a little bit newer or different. In this way, you’re always showing them something new and challenging them just a little bit. There’s a sweet spot for video games between making players do the same thing over and over again, they get bored; and forcing them to grow in skill too quickly, and the game becomes hard and many players will give up. But if the game does it just right, then you can start from not knowing how to do anything at all in the game to now pulling off crazy skilled jumps and attacks that, if you’d been asked to do them from the start, would have frustrated you and you would have likely given up.

You can probably tell where I’m going with this one.

I think we should be applying the same tactics to language instruction that we do to level design. Introduce things slowly. Give them some time to play around with that things and become comfortable and good at it. Then, right when they’re starting to get a little bored, you give them something new.

This, in a simple way, is an advantage that I think the system I use on myself has — at least in the sphere of learning new vocabulary targets. You play around in the space that you know until you’re comfortable with how to recognize it against the backdrop of all the other things you already know about. Then you add something new, and play with that. Eventually, you don’t have to do a direct comparison against EVERY other thing because you’ve built up that framework of knowledge. Though sometimes you do get thrown for a loop by targets that are surprisingly similar, or whose meanings get crossed somehow.

(I had that exact problem with the characters for big sister, little sister, and little brother in Japanese — I could easily recognize that each of them was one of the words in that group, and when placed next to each other I could have likely told you which was which; but when seen individually, I was just awful at remembering quickly which one had what meaning.)

A topic for another time (since I’ve written over 2,000 words already and I’m not QUITE done yet) is how we should take what I’m saying here and then train against things that are cognitively similar for whatever reason, to learn to differentiate them from one another.

Anyway, what I’m trying to say in this section is that, when you introduce topics or new words slowly and give people time to play and understand it, they will feel like they’re GOOD at it all, and they will want to do more. Whereas if you just hand them a huge list and say “go”, they’d probably do it ineffectively and they’ll feel like they’re bad at it and give up and (possibly) think themselves lesser than people who just naturally figure out how to study better for themselves earlier.

Memorization vs. Learning

Lastly, I just want to touch briefly on “learning” versus “memorization”. Obviously, the goal is learning. Learning is what we do with our native languages. It’s that instinctual kind of knowledge where you have a non-verbal thought and the words and grammar just materialize and you say them. Memorizing is more like having a high-speed look-up table. Like a dictionary in your head. And when you have a thought, you have to formalize it into something you understand (words, meta-thoughts, something weird like that) and then “look up” or “figure out” what those things are in the target language.

Memorization can happen very quickly, but it also goes away very quickly. It’s part of why I still got a D in that class where I was rockstar-memorizing over 100 words each week and acing morning quizzes. I did the quizzes, but then by the time class was out, the words were gone. The step I lacked was coming back to what I’d studied later, after a break, and RE-memorizing the ones I’d lost. It’s the classic long-tailed second half of recurrence learning. It’s the part that everyone already has figured out in all of their flashcard apps.

The thing they’re missing is that “getting started” bit right at the beginning.

And that little part is why I’ve started working on a flashcard app of my own. Because I’m tired of having to deal with poor bootstrapping of new vocabulary. Because it’s not fun, and I think it wastes my time.

Advertisements