Want to study this episode as a lesson on LingQ? Give it a try!
Part 2 of Steve’s conversation with Paul Nation, a leading world expert on vocabulary acquisition, and language learning in general.
Steve: When I’m learning Russian all the words are new and strange to me and I have a tendency to confuse words that look or sound similar to me.
Some of these words are related and some are not.
Paul: That’s right.
Steve: Again, when I’m in our system — I’d be happy to have one of your students get on our system, it’s free by the way – and I’m reading through a text and I have my vocabulary section open on a different tab, so with your example I would say “s-t-o-n”.
See, I’m getting confused, “astonished”, “stone”, I don’t know if there are others.
“Spec” would be a good one, “specification”, “speculate”, “inspection”; they’re all kind of confusing to me.
I would then take “spec” and I would search for that component on my vocabulary list.
Especially in Russian, it will throw up 25 words and these are words that I’ve previously saved in the system, so I’ll see here’s eight of them that are related and here’s a few others that are not related and it’s just as good to know that these are not related so I can eventually train myself not to get confused over them.
Steve: So you’re looking at it as sort of the deliberate attempt to discover not cognates, but related words that are going to help you, it’s just as useful to get rid of those that are confusing you, if you follow what I’m saying.
Paul: I wouldn’t disagree with that.
The point of the research he’s doing is he’s saying once you know these first 1,000 how will they help.
It’s not as if you’re trying to learn them all at the same time.
The other thing in what you said that struck me, I agree with most of what you said.
Our system is very much based on sort of a deliberate learning of vocabulary and where I find crash and falls short is that it’s just not possible to go and just read and read and read, you need some help and you need to focus on vocabulary.
I find that learning the language through vocabulary, especially if you’re looking at all the types and you’re able to see how they work with different words, is very helpful.
One thing I heard you say about numbers…like I believe, certainly in my Russian study, I’ve gone for a long time without using the language.
I’ve been studying it on my own and I do an hour a day when I can.
I can listen to Tolstoy, I can read Tolstoy and so forth, I enjoy it tremendously, but numbers are difficult.
Numbers are very difficult for the simple reason that numbers are so strongly programmed in our brains in our own language.
If I’m listening to a text in say Russian history and they give me a date it’s only now that the date makes sense and this is after almost two years; it takes a long time.
If you force me to answer questions using numbers I find it stressful and annoying.
Granted I’m learning on my own and our learners are learning on their own, so you haven’t got me in a classroom where you can force me to do things, but on my own I don’t want to do things that I find annoying.
I would resist doing things where I can’t quickly find the answer; whereas if I’m just taking it in and taking it in I’m building up this familiarity with the language and a later stage when I’m more comfortable then I will be able to answer those questions.
Therefore, that’s where I question what you were saying about produce it right away.
Maybe in a classroom that’s a good thing to do, but if you’re learning on your own and you want to maintain the motivation and the fun that might just discourage some people.
For some of the fluency development, particularly with numbers, you’d probably need to have some other person helping you.
A very easy way to develop fluency with numbers is just to sit down say starting with the numbers 1 to 10 written in figures then your teacher or your partner just says the numbers quickly in random order and you point to the figure that’s being said.
Paul: Now you do that for 10 minutes and you’ll see your speed of recognition of that improves very much and that once you can do it with single digits do it with double digits, but you only do it for 5-10 minutes a day two or three days a week.
You’ll find that after two or three weeks these are no longer stressful, uncomfortable things for you, but things which you handle very easy and fluently.
There’s no point walking into a shop and saying how much does this cost and when the guy says $98.57 you make a stupid grin because even though you know all the numbers you can’t understand what he said.
Paul: That’s what fluency does, it just takes away that stress by getting good initially with a small group of things that are really important to be able to deal with quickly.
I’m sure it takes a little more discipline to do that, in a way.
It’s not also obviously, for example in my situation where I have no one to speak Russian to and I have no shops to go to, but I’m interested in being able to read the literature, so some of these abstract nouns, in fact, might be more important to me than numbers.
Everyone has a different purpose in learning the language.
Paul: The problems that you and others face is that when you learn languages, which are not English and perhaps not French or Spanish, is you don’t have a great deal of simplified and adapted material to give you good meaning-focused input and to give you good fluency development for reading.
People who learn English are really privileged in the sense that there’s probably more than 2,000 graded readers now at a whole range of levels, which they can read within the vocabulary that they know and that’s just tremendous for developing fluency, motivation and a whole pile of things.
It would be great if people would do this for Russian, Japanese and Chinese.
It would be tremendous if we had graded readers for those languages.
One thing we do at our system is we have a small amount of content that is deliberately sort of easy within the first 1,000 words or so.
Thereafter, because people are saving words to a database, the system gets to know each individual learner’s database and then the system will then grade the material to the vocabulary of the learner so that it will say this has 20% new words for you.
If I’m studying Tolstoy after a while my new word count in Tolstoy is going to be lower than if I went to another author, for example, or if I were into economics.
You can focus on one area and you can actually bring your new word count down and then if you move to another area then you’ll find your new…once you get past the first couple thousand words, we’re able to at least help the learner by telling him or her how many new words there are in a content item.
Paul: There’s good research evidence for that because one of our students did some research comparing an Economics textbook, which was 300,000 words long, with a 300,000 word collection of different texts and each of those different texts were about 2,000 words long.
The total vocabulary in word families of the Economics textbook was just over 5,000 words.
The total vocabulary of the mixed text of the same length as the textbook was over 12,000 different words.
That’s two and a half times the number of new words you can see, so the idea of following one author or following one topic is a really good way of reducing the vocabulary load and it’s really important in the early stages to reduce the vocabulary load.
Anyway, it’s interesting.
We’re not doing any research we’re sort of charging ahead with what we think makes sense, but I don’t think we’re that different from what you’re commenting.
It would be very interesting for one of your Ph.D.
students to get into our system — as I say it’s free to register — and just see what we’re doing and see if it’s of any relevance to the research that you’re doing.
Paul: Now if you employed me as an evaluator of your system, which I haven’t looked at so I can say this and you can not feel offended because I haven’t looked at it, what I would do would be to look at what are the different kinds of activities which are encouraged by your system.
I would classify each activity into each of the four strands and then I would see how much time is spent on each activity, if it was possible to do that, and I would see do we have a roughly equal balance across those four strands.
If we don’t then I’d be looking to see how do we redress that balance a bit.
If you had a course which is 90% language-focused learning and 10% of the other three strands you’ve really got a misbalance.
Similarly, if you have a course which is 100% message-focused and no deliberate learning then the learners are missing out on very efficient and effective ways of making quick progress.
As an evaluator that would be one of the things I’d be looking at to see is there a balance of learning opportunities across those four strands of meaning-focused input, meaning-focused output, language-focused learning and fluency development.
Steve: One of the problems we have is that a lot of the input and the sort of vocabulary-related meaning-focused learning is free, so most people do that.
Steve: The output, that is to say writing and having the writing corrected and speaking with our tutors, they have to pay for, so they do less of it.
We are introducing a bunch of improvements in our discussions with the tutors and in our writing correction whereby we integrate the words and phrases that they have trouble with so that they are more easily brought into their flashcards, the words and so forth, the meaning-based studying that they’re doing.
We’re still a few months away from having all of that integrated, but the bigger thing with us is that we’re dealing with, basically, independent learners, so they do whatever they want.
We’re not in a position to control what they do and, as I say, certain things are free so more people are going to do the things that are free then the things that they have to pay for.
Paul: Yeah, that’s right.
It would be good I guess if your learners realized that because then it’s partly their responsibility to try and address that balance between the learning.
Steve: You know it’s a good idea.
What I’m thinking as I hear you talk is that one of the things we have thought about doing as we go forward is to have courses.
One of the things we should strive for then is here’s a course and it’s balanced, so this is a program that includes a balance of the four strands.
We are not yet there in terms of the development of our program.
Paul: Yeah, that’s one direction to go, but I have no problem with the other direction if the learners are intelligent enough and motivated enough to take some responsibility for their own learning.
They can sort of see now how can I get these balanced and if you are offering a range of opportunities or directing people to opportunities then they can take some responsibility for that balance too.
I don’t advocate any particular method, I just like to see that there’s a good balance of learning opportunities.
The other thing I was going to suggest that I think would be quite useful is that I don’t know of frequency counts of languages like Russian and so on, but I’m sure they must be around somewhere.
Steve: Oh they all have them, yeah.
Paul: It would be quite useful if learners had some sort of access to that kind of information so that then they could make decisions in deliberate learning about which words are worth learning and which ones would be best left until later.
Steve: Well sure. What we do is every word has either four stars, three stars, two stars or one star.
Paul: Oh good.
Steve: In the case of English – I think it’s the most frequent – we have 2,000 words, in Russian we have to bump it up because it’s a very heavily-inflected language and we base it simply on the total content that we have in our library.
In the case of English we have over 2,000 items, we have like 30,000 different words, but we only deal with the first few thousand.
Beyond that then it really depends on the interest of the learner, but anytime they save a word it will show up as a four star, three star, two star or one star word.
One of the things we do is of the words that the learner has learned or has saved and is trying to learn, we will pick the 25 highest-frequency words and we call these their priority words, so these are the ones they’re supposed to be trying to learn.
As they learn those and the status of those words moves to known then other words come in, so they’re always focusing on the 25 highest-frequency words of the words that they’re trying to learn, so we definitely deal with the frequency issue.
I was going to say, one other point is that in terms of encouraging people to have a balanced approach, philosophically, the biggest thing is to keep people spending time with the language.
We encourage people to do whatever they like to do, so if a person likes to listen then listen, if you like to read, read.
If we look at the success of traditional language teaching in schools, here in Canada for example, the success of teaching French in the English language school system is just abysmal.
In the case of once province New Brunswick, which is actually the only bilingual province in Canada, 30% of the population is French-speaking.
They have French in the English learning school system 30 minutes a day every day for 12 years.
They did a survey and they found that after 12 years the number of graduates who had what they called an intermediate level of oral proficiency in French was 0.68%.
In other words, they might just as well not done anything because I think they would have had the same number of people who would have done it on their own.
Paul: The advantage you have is that the people who come to use your programs have a strong motivation.
Steve: That’s right.
Paul: That’s something you can then build on and work with.
Steve: Well that’s right, plus we try to encourage the motivation and try to make more people motivated.
Again, that’s why we say do what you like to do, but we can certainly advise them.
I certainly have gotten a lot out of this discussion and one of the things would be to try to encourage them to have a balance.
A lot of people will listen to guidance because people do want to improve in the language.
Paul: Yeah and there’s a lot of good things on the Web. Thanks Steve.
Steve: Thank you very, very much for this discussion.
I would appreciate it if someone there would have a look at what we’re doing.
We’re not going to use it as some kind of a recommendation of our system or anything like that, but I would be interested in any feedback.
Steve: And, as I say, registration is free.
Paul: Good, thanks Steve.
Steve: Thank you very much for this time I really appreciate it.
Paul: No problem, bye-bye.