Saturday, April 24, 2010

Getting to Grammar: If you want accurate grammar quickly, Steve Kaufmann's method is not for you

This is the best our graphics department could do.
he great language-learning blogosphere battle of the day has been Steve "The Inputter" Kaufmann v. Benny "The Haxor". The latest salvo in this battle comes from Steve:
In my view, there are three divergent approaches to language learning, divergent in terms of their emphasis or principal focus. This is true whether we learn in the classroom, online or on the street. One approach focuses on input, another on output, and a third on what I would call shortcuts and some people call language-hacking techniques. These techniques include grammar study, studying vocab lists and phrase books, heavy use of flash cards, "deconstructing the language", memory techniques, and so forth.
I don't think Steve's division has it right at all. As I noted in my last post, output is input. In other words, it's all just exposure. From there, the only thing you need to think about is what kind of exposure you need to get in order to burn the language into your brain as efficiently as possible.

And efficiency leads me to one of my main points of disagreement with Steve: grammar.

In a nutshell, here's my understanding of Steve's approach to grammar:
  1. Spend lots of time getting input.
  2. If you figure out the grammar rules, great. If not, don't worry about it.
  3. When you feel like it, try to fill in those gaps in your knowledge by looking up the rules.
Steve's German is good enough to listen to and read fairly tough materials, and yet he still get criticized for screwing up the cases. That tells me that his method is not working as far as his ability to produce correct grammar. (Of course, I'm assuming those criticisms are correct, as I've never heard or read his German myself. If that's not right, I'm sure I'll be hearing about it in the comments soon enough.)

This is a close adult approximation of the inductive "learn like a child" method. Children get years of exposure to a language and still make lots of grammar mistakes, until years of schooling finally iron out the wrinkles. Without extra efforts beyond mere exposure, they end up being able to understand just about everything, even while they may still be speaking incorrectly. Steve's results seem to match that pattern.

And, after all, is it really surprising that a method that focuses on input results in you having a good understanding of input without being able to produce accurate output?

I think a different approach can get you much better results. Here's the rough outline of how I approach grammar:
  1. Get the rule in your head.
  2. Get exposure to the rule in use. Because the rule's in your head, you're seeing the rule in action rather than trying to puzzle out what the rule is.
  3. If you forget the rule, or if you're exposed to something that doesn't fit into your understanding of the rule, go over the rule again.
During the bulk of my German studies, I actually followed a method that was much closer to Steve's, and it resulted in German being one of the weaker languages that I can actually communicate in with some degree of proficiency. (So, yes, I screw up the cases, and probably much worse than Steve.) I'd love to find the time to attack it again, using the approach that I laid out above.

As a final note, I'd also say that I think Steve's and my divergent approaches to grammar may stem in part from our divergent goals in language learning. Steve's goal seems to be enjoying literature and whatever else he feels like enjoying in the language. For me, that's a means to an end while my actual goal is being able to prepare business documents, contracts, etc. I need to obtain a higher degree of accuracy in a shorter amount of time that Steve's method will allow, while Steve can quickly reach his own goal of jumping into content he enjoys without worrying about whether he's producing correct grammar.


  1. I fully agree with your approach to Grammar German is my native tongue and I am at the moment learning Serbian. I have spent and wasted too much time and money with fellow students who are not willing to learn Grammar and teachers who cannot explain it either! I do not know how you could learn a new language which is very different from your own without knowing the Grammar very well, also the Grammar of your mother tongue. We have lots of problems here with children of the second generation of migrant workers who can neither speak the language of their parents very well neither the language of this country. If it was so easy to "pick up the language" why do we have these problems?
    As far as reading literature goes, also here you need a profound knowledge of the grammar to understand what you are reading. As a matter of fact the deeper I am "diving" into the Serbian language the more I am also confronted with linguistic issues. I really enjoyed reading Berlitz´s book "the wunderful world of languages" it helped me a lot getting more insight into other languages.
    BTW: Thank you for putting all this work into your blog. I enjoy reading it and it gave me many ideas!

  2. Thanks for the detailed post - the main image is funny :P

    I agree absolutely that output IS input. I don't like this whole debate at all of me representing "output" as if all I do is speak all day and never listen or study.

    And your analogy with children learning is important because one crucial difference that does not allow adults to learn like children (although I think we should as much as possible) is our native tongue screwing up how we think something *should* be said. This is not an issue when learning your native language, but grammar helps you "unlearn" your native language when you get to the second one.

    I also think goals are important. Steve's method works great for him and I'd never argue against that (despite the many times he seems to imply that my method might not even work for ME), but your purposes and my purposes for learning languages are different and more time sensitive so we need to do it quickly and efficiently.

  3. I fully agree with you.

    If you read "En face d'elle, il y a du café et beaucoup de thé. Voici aussi du sucre. Je prends aussi un peu de fromage.", do you really understand the differences between "de" and "du"?

    If you go with Steve's "If you figure out the grammar rules, great. If not, don't worry about it.", you will never speak or write correctly.

    Learn grammar, use a lot of input (Steve), use it and start speaking (Benny) - the mixture is the key.

  4. What about this method?

    1) Get exposure to the grammar rule in a sentence/phrase you already understand the meaning of. This presumes that you already know at least a few basic phrases.
    2) Notice that there is some grammar at work there…a case ending, a different form of a pronoun, whatever.
    3) Read a clear explanation of the rule so you understand exactly how it works.
    4) Get more exposure to (examples of) the rule in action.
    5) Practice using the grammar yourself.

    You can fit that all in one 45-minute lesson.

    For example:
    1) You know that in Hungarian jó means good, nap means day, and “Jó napot kivánok” means I wish you good day.
    2) You notice that the phrase uses “napot” instead of “nap.” What’s that –ot doing there?
    3) You read the rule about the –t ending for the accusative case. Sometimes it’s just –t, sometimes you need to add a certain vowel, sometimes there's a vowel change, and sometimes the word’s form changes. There are rules for all this and you study them carefully. But oy, does it seem complicated!
    4) Through input you see and hear things like kéz/kezet, alma/almát, ló/lovat, eper/epret and so on. Lots of variety, but actually, it’s pretty regular and predictable. Now you really “get it.” And you'd never say "kézet" because it just sounds wrong.
    5) You speak and write and people politely correct you when you error. Well, ideally.

    One plus here is that at step 3 you get that rewarding “Ah-ha! moment” of insight, which reinforces your memory of the rule. It also makes your students happy. :-) Once you get the idea of the rule, it’s a matter of repeating steps 4 and 5 in various ways to get better and better at it.

  5. I used to be a big grammar fan and struggled for years trying to learn languages but still spoke very broken, slow and inaccurate. Some how you develop a filter that prevents smooth output because you keep checking back through it automatically and so cannot speak smoothly.

    From my investigations of all the great polyglots its a common thing that they all got to being fluent FIRST, and then ironed out their grammar after reaching a more fluent and commanding level of the language. Perhaps Steve couldn't be bothered to correct his grammar, but remember he can speak fluently and smoothly. I'm sure if he wanted to he could improve grammar with some concentrated studies, and the good thing is that he has already passed the fluency barrier that stops many people from continuing.

    I'm pretty sure if you have lived in China you would know that tons of Hong Kong people spend 20+ years just learning grammar but can't speak smoothly to save themselves.

    Anyway, I'd rather speak fluently on all topics with 80% accurate grammar over limited and broken but 100% accurate grammar.

  6. Whoa, so glad you commented on Steve's blog, or I wouldn't have found this. Greast post. I really like this technique that you describe.

  7. Maybe I am missing the point and I have only focused on the Chinese aspects but I have heard a lot of content on-line youtube and MP3 where Steve SPEAKS Chinese on Radio shows etc, with Chinese people, having real conversations over a long period. Granted his Chinese may not be perfect and sometimes his written Chinese may not be completly natural but having read a bunch of Benny's stuff so far I haven't seen or heard any conversations, just Benny talking at me.

    Coff. Benny is the talker but where are the conversations, maybe I have not dug deep enough, links to them appreciated. Talking at people is not communication.

  8. The only stuff I heard from Benny are a few simple sentences in Thai in a video at the end of his Bangkok stay - after 3 months in Thailand this is something really everybody can achieve! Disappointing.

    Steve instead speaks very often in Japanese, Chinese, French, etc., e.g. in his YouTube videos!

  9. There's a further post digging more into the difference between my and Steve's method in respect of grammar here, and a more detailed look at my method here.

    @Benny: I'm happy to argue against Steve's method when I think it's wrong, as I think it is on grammar. While his method may work, I'm completely unsold on the idea that it works quickly or reaches the last bits of grammar you need to approach a native level.

    @Amelia: The method you propose overlaps largely with my method, the major difference being that you seem to start with vocab and then move to picking the grammar out of that. I think getting a broad understanding up front before implementing your method will make your method even more efficient.

    @Milan: While I'm not sure exactly how you studied, my impression from your comment above is that you focused too much on grammar. It's something that you should get through, but get through quickly. Anyone—including those people in Hong Kong that you mention—who spend 20+ years just learning grammar are going in the wrong direction. Heck, I'd say if you're spending even a couple months just learning grammar you're headed in the wrong direction.

    I would agree with "all the great polyglots" that you'll get pretty darn good at the language before you iron out your grammar issues, but that doesn't mean you shouldn't try to get an early understanding of those issues.

    @Chris / @Anonymous: I've actually seen more success with people using methods that are closer to Benny's than to Steve's, so even if you say Steve's method works for Steve and Benny's method doesn't work for Benny, that doesn't say much about how effective their methods are overall, although it is always nice to have a proof of concept.

  10. My experience with language learning is similar to yours. I used Michel Thomas with French and Spanish, then did my own grammar outlines (which is kind of creepy because you do the EXACT same thing. This was probably the most important thing I did because it gave me the structural framework for the language), then I started speaking and listening/watching podcasts, movies, and TV. I was immursed both times and spent as little time as possible in English. It was very effective. I learned fast--developing proficiency in the languages in about 4 months each.

    BUT ANYWAYS, I would like to hear your thoughts or those of other readers on this video. It is Stephen Krashen, who is pretty much the Michael Jordan of Second Language Acquisition Theory. It is longish (15 min), but I think its persuasive and worth it.

    If you dont want to watch the whole thing, check out start at 14:20 (but in doing this you will lose the large emphasis he puts on comprehensible input)

    If you dont want to watch it at all, here are the major ideas:
    1) You only learn when you have Comprehensible Input. There are no exceptions
    2) Everyone learns the same way
    3) Talking is not practicing
    -Talking only helps expose you to Comprehensible Input because your conversation partners respond
    4) Anxiety can prevent the language acquisition process

    Clearly to master a language you have to speak it and languages with sounds that dont exist in English (like the R in Chinese or the deep throat H’s and G’s in Arabic) need to be practiced. These well-supported theses, however, argue that what comes in seems much more important than what comes out.

    This has not been my personal experience, but I am not that experienced a language learner--I've only been really doing this for about a year--but it makes sense. I think that gettnig the grammatical framework is essential. Trying to piece things together is not efficient. Imagine trying to figure out when to use the subjunctive in Spanish and when not just by listening. That would take an eternity. But one the framework is there and strong--which should come from I believe more output than input because the framework is something explicit/learned so it should be practised as such--you should start challenging it with tons and tons of input. I picked up and imitated what I heard. Very rarely did I invent my own phrases.

    What do you think about Krashen's ideas. He has ALOT of research on his site to back it up. Look forward to hearing back from the forum

  11. Im not at all associated with Benny, but I like his site. I am 95% sure he was only in Thailand for two or three days, so your criticism doesnt hold water.

  12. As you know the issue of how to learn and to teach
    English grammar is contentious. Supporters of unconventional
    learning/teaching methods of English as a second/foreign language claim
    that conscious learning of English grammar is unnecessary and advocate
    intuitive learning of English grammar. I suggest that
    you thoroughly explore the following especially useful prominent websites dealing
    with practical issues of learning and teaching English grammar and include new pertinent research articles:

 me know your thoughts on this issue.