319 messages over 40 pages: << Previous 1 2 3 4 5 6 7 ... 10 ... 39 40 Next >>
s_allard Triglot Senior Member Canada Joined 5427 days ago 2704 posts - 5425 votes Speaks: French*, English, Spanish Studies: Polish
| Message 73 of 319 10 April 2014 at 3:02pm | IP Logged |
I really don't disagree with Iversen. I would simplify it even more. Your vocabulary varies according to your needs. If
you want to read all Shakespeare, you need 29,000 words. If you don't want to read Shakespeare and instead want
to pass that C1 speaking exam maybe 1800 will probably do the trick. I even think most of these figures are
irrelevant. You just get the vocabulary you need. My only word of caution is against the belief that vocabulary alone
is the problem.
1 person has voted this message useful
| dampingwire Bilingual Triglot Senior Member United Kingdom Joined 4662 days ago 1185 posts - 1513 votes Speaks: English*, Italian*, French Studies: Japanese
| Message 74 of 319 10 April 2014 at 3:06pm | IP Logged |
s_allard wrote:
In my opinion the significance of all this is that it puts the lie to all these extravagant
claims about the number of
words one needs to pass these kinds of tests. Mutatis mutandi I would say that you could
probably pass a C1
oral proficiency language test using less than 1800 words. I said using not knowing.
|
|
|
s_allard wrote:
Regular readers of my posts know that I have been saying this all along. You don't have to
know a vast number of
words to speak a language. You have to master the core high-frequency elements that contain
most of the
grammatical function words and then add whatever low-frequency specialized and content-rich
words that are
needed.
|
|
|
That study shows that you can pass a C1 test using only unique 1800 words. That's
not the same (by any stretch of the imagination) as (actively) knowing only unique
1800 words. It just means that when tested today you only needed to use a specific 1800
words. As with most tests, you don't get to know beforehand which 1800 you'll need
on the day. That's the case with most tests. You're supposed to learn N things, we'll test
as much as we can in (say) 3 hours and we'll try to design the questions so that you can't
"study to the test".
I happened to remember that I had some mail files lying around that represent a reasonable
fraction of my home emails over the last few years or so plus a few years worth of emails
from when I was with a previous employer. I knocked up a very rough script to plough
through them and extract those that were written by me, to ignore binary attachments, to
discard quoted text. The script is rather too aggressive and discards some emails that I
did write and discards text that I wrote too (where, for example I no longer have my
original email but do have the reply where someone quoted it). The script turns what's left
into unique words and throws away any that are not in the dictionary file on this system.
The result is 6881 unique words.
If I had a way of screen scraping my HTLAL posts, I'd adapt the script and see what it
thinks of that. A better corpus might be the sum total of my usenet postings, but I can't
find any way to get google groups to give me that in any kind of electronic format. Close,
but no cigar google!
3 persons have voted this message useful
| s_allard Triglot Senior Member Canada Joined 5427 days ago 2704 posts - 5425 votes Speaks: French*, English, Spanish Studies: Polish
| Message 75 of 319 10 April 2014 at 6:47pm | IP Logged |
dampingwire wrote:
s_allard wrote:
In my opinion the significance of all this is that it puts the lie to all these extravagant
claims about the number of
words one needs to pass these kinds of tests. Mutatis mutandi I would say that you could
probably pass a C1
oral proficiency language test using less than 1800 words. I said using not knowing.
|
|
|
s_allard wrote:
Regular readers of my posts know that I have been saying this all along. You don't have to
know a vast number of
words to speak a language. You have to master the core high-frequency elements that contain
most of the
grammatical function words and then add whatever low-frequency specialized and content-rich
words that are
needed.
|
|
|
That study shows that you can pass a C1 test using only unique 1800 words. That's
not the same (by any stretch of the imagination) as (actively) knowing only unique
1800 words. It just means that when tested today you only needed to use a specific 1800
words. As with most tests, you don't get to know beforehand which 1800 you'll need
on the day. That's the case with most tests. You're supposed to learn N things, we'll test
as much as we can in (say) 3 hours and we'll try to design the questions so that you can't
"study to the test".
|
|
|
I totally agree. That is exactly why I wrote: "I said using not knowing."
Now the question is how many words does one have to know how to use. Since you don't know what words might
come up, then theoretically the number of possible words is very high. So how do we determine how many
words we need to prepare. Is it the 5,000 most common words or 10,000 or 20,000? And how do you go about
preparing? Do you make up 10,000 flashcards and spend your time going through them?
I think that people are making things too difficult for themselves. First, we have to keep in mind that neither the
IELTS or the CFER tests for vocabulary. There are no prescribed or even recommended numbers. It's the ability to
use the language that is being tested. You will not be asked technical questions such as how does an internal
combustion engine work. On the other hand you may be asked how do you think cars have changed society.
In fact, the IELTS guidelines state explicitly that you will be asked about topics such as hobbies and daily
activities. Look at the tasks mentioned in the Nation and Head study: Eating out, Reading a book, Language
learning, Describing a person. What the examiner is looking for here is the ability of the candidate to speak
about these topics at a certain level. Can you study for these topics and game the system? Can a Band 2
candidate not study a bunch of flashcards and outfox the examiner into believing that they are really a Band 8?
Anyone who has done a bit of language testing will tell you that you cannot easily bamboozle a live examiner.
It seems that the Band 8 candidates could perform satisfactorily with on the average less than 1500 words. Given
that we know what kind of topics will appear and in what form, could we not assume that a person with a solid
ability in only 1500 general words could handle any of these questions?
I'm not saying that 1500 is a guarantee of success. What I'm saying is that the study pointed out that the Band 8
users, in addition to knowing more words than lower-level uses demonstrated more sophisticated abilities.
Frankly, I don't think these candidates went into the examination deliberately limiting themselves to 1500 words
but I don't believe there is any proof that you absolutely need any higher number.
Edited by s_allard on 10 April 2014 at 6:51pm
1 person has voted this message useful
| s_allard Triglot Senior Member Canada Joined 5427 days ago 2704 posts - 5425 votes Speaks: French*, English, Spanish Studies: Polish
| Message 76 of 319 10 April 2014 at 7:06pm | IP Logged |
In my opinion, much of the debate we are having here stems from this mistaken belief that one needs a large
vocabulary to
do anything. It would seem that if you know less than 1500 words, all you can do is grunt, snort and point in
order
to communicate. The great thing about the Nation and Head study is that is has proved plain as day that you can
do
a hell of a lot with 1500 words. Could you do more with 15,000 words? Certainly. Will that automatically make
you
a better speaker of the language? I'm not so sure.
If it were all so simple, why not just have vocabulary tests on computers and do away with all the live examiners?
I once argued here that you could start speaking French with just 300 words. The reaction was such that I
thought I
was going to be tarred and feathered and run off the forum. Well, not exactly, but I reaped many a bushel of
incredulity and scorn. For some strange reason, many people thought that I had said that you shouldn't learn
more
than 300 words in French. The lesson I learned is that what you write is not always what people understand.
Edited by s_allard on 10 April 2014 at 7:08pm
1 person has voted this message useful
| Serpent Octoglot Senior Member Russian Federation serpent-849.livejour Joined 6594 days ago 9753 posts - 15779 votes 4 sounds Speaks: Russian*, English, FinnishC1, Latin, German, Italian, Spanish, Portuguese Studies: Danish, Romanian, Polish, Belarusian, Ukrainian, Croatian, Slovenian, Catalan, Czech, Galician, Dutch, Swedish
| Message 77 of 319 10 April 2014 at 7:20pm | IP Logged |
IELTS is only a CEFR-compatible test though, not designed with the scale in mind. Most of the topics you mentioned are too easy for C1.
More importantly, do you honestly think one can fit the CEFR guidelines with such a vocabulary? Especially in a language with few cognates/borrowings.
1 person has voted this message useful
| s_allard Triglot Senior Member Canada Joined 5427 days ago 2704 posts - 5425 votes Speaks: French*, English, Spanish Studies: Polish
| Message 78 of 319 10 April 2014 at 8:44pm | IP Logged |
I wouldn't even bother thinking about vocabulary size. I think that when one is preparing for a test, you should
above all look at the guidelines, requirements and recommendations from the test makers. And there are lots of
sample or past tests around.
The interesting thing is that most of the time we know what the tests consist of. For example, the Spanish DELE
C2 test require that you read and article, summarize before two examiners and engage in a debate over the
article for 20 minutes.
Instead of worrying about my vocabulary size, I would practice doing the above exercise with a tutor until I felt
really comfortable. Of course I would use materials similar to what I expect to get on the test.
I would certainly not try to learn vocabulary from 20 different technical areas just because I don't know what
questions will come up. Instead I would try to read widely so as to get a kind of superficial general knowledge.
1 person has voted this message useful
| Serpent Octoglot Senior Member Russian Federation serpent-849.livejour Joined 6594 days ago 9753 posts - 15779 votes 4 sounds Speaks: Russian*, English, FinnishC1, Latin, German, Italian, Spanish, Portuguese Studies: Danish, Romanian, Polish, Belarusian, Ukrainian, Croatian, Slovenian, Catalan, Czech, Galician, Dutch, Swedish
| Message 79 of 319 10 April 2014 at 9:30pm | IP Logged |
s_allard wrote:
I think that when one is preparing for a test, you should above all look at the guidelines, requirements and recommendations from the test makers. |
|
|
Yeah, unfortunately in terms of tests it's more efficient than actually reaching the level and learning the things you don't need in the test but only in real life.
2 persons have voted this message useful
|
DavidStyles Octoglot Pro Member United Kingdom Joined 3938 days ago 82 posts - 179 votes Speaks: English*, German, Italian, Spanish, Latin, French, Portuguese, Norwegian Studies: Mandarin, Russian, Swedish, Danish, Serbian, Arabic (Egyptian) Personal Language Map
| Message 80 of 319 10 April 2014 at 11:04pm | IP Logged |
s_allard wrote:
I once argued here that you could start speaking French with just 300 words. The reaction was such that I thought I was going to be tarred and feathered and run off the forum. |
|
|
Did you have a specific 300 words in mind? I'd be interested to see the list.
I compiled a list of around 300 words (I sometimes modify it slightly - last count was 313 that I consider essential to being able to speak English. With these words one can express most things, albeit in a clumsy fashion. I've long used the list and its directly translated equivalents for the purpose of giving myself or my students a swift leg-up into a new language, learning these words before anything else.
Nowadays, as part of a PhD study, I've taken the notion and applied it to a larger scale, and am creating a tiered corpus (so, tiered for what is needed for various "levels", with the CEFR levels being obvious examples of such) optimized for many languages (ie, so it doesn't have the shortcoming that my long-used English list does of being too anglocentric to the disfavour of other languages).
Edited by DavidStyles on 10 April 2014 at 11:19pm
1 person has voted this message useful
|
You cannot post new topics in this forum - You cannot reply to topics in this forum - You cannot delete your posts in this forum You cannot edit your posts in this forum - You cannot create polls in this forum - You cannot vote in polls in this forum
This page was generated in 0.4414 seconds.
DHTML Menu By Milonic JavaScript
|