Blog Archives

Question on lexical sufficiency


Joseph Conrad

Joseph Conrad Korzeniowski - the ultimate benchmark in mastery of an acquired language is surely that of having added to its artistic literature?

Reader (and poster) Bill_Sage667 from How-To-Learn-Any-Language.com’s forum wrote me the following question and agreed kindly to a public answer here:

Dunno whether u’ll be able to find the time to reply to this, 1 in a million chance lol……but I’ll write out my questions anyway lol

You said something about 15,000 words needed in order to achieve a good degree level in Russian. Are imperfective and perfective verbs considered separate words, as well as adjectives and verbs under the same lexemes (e.g. беремменость, беремменая, беремменеть, забеременнеть) when you were estimating the number of words?

And what if someday I want to attain the proficiency of an educated native speaker (might take me 20 yrs but oh well)? How many words am I supposed to know (for active and passive knowledge)? For Russian, that is. btw thanks for releasing the Gold List Method to the public for free!

Firstly, Bill, be careful about the number of ‘m’s and ‘n’s you have in those pregancy-related words. You have too many ‘m’s and not enough ‘n’s. I’ll leave you to review that one.

You’re very welcome about the Goldlist. As I say in the section I wrote in syzygycc’s The Polyglot Project, I’m just paying forward the favours I got from so many people when I was a young learner.

In my opinion 15,000 words, as long as they are properly selected, are perfectly adequate and in the headlist you would use all the forms initially as separate forms (but not the various conjugations and case endings, only the so-called ‘dictionary forms’) and you could soon condense them on distillation.

If you use the frequency distionary I am selling on www.oioioio.com you will be able to focus on commoner words first. Within the first 10,000 words you do get words that are already pretty specialised that you wouldn’t use maybe more than once a month or so even if you were a native, and so it continues over the next 5,000 as well. You’ll find 15,000 enough to read the great novels comfortably and to appreciate the poetry of Akhmatova, Tsvetaeva and even Strogonova (the last of which you will find uniquely published in this blog as a ‘page’. She is no poorer a poet than these well-known ones, only far less known.)

I would also like to draw people’s attention to something else I wrote about the 15,000 word ‘marathon’ in a thread over on the HTLAL forum:

What this Gladwell character [I’m referring to upstream discussion of someone who said you must have 10,000 hours of learning to become fully fluent, like a native – a claim almost unanimously rejected by every serious linguist and polyglot I know other than those who teach languages privately, as this idea is grist to their mill] needs to bear in mind is the Pareto rule. If it were true (which I dispute) that you need 10,000 hours to become as native (although how this deals with your accent is anyone’s guess) then you could get 80% of it in 20% of the time. That means you’d need to have 2,000 hours study to get to 80% of native fluency. Since that’s ludicrously overcautios, I’d suggest that the 10,000 hour target for full native fluency is overcautious.

The fact is, a person could be like Konrad Korzeniowski (Joseph Conrad) and already writing ground-breaking literature in the language he or she had learned and still have a strong enough accent to provoke politely meant but annoying compliments on the quality of his language by native speakers.

In the end you just have to accept what English speakers accepted for their own language in the main long ago – that as long as it doesn’t hinder comprehension, a foreigner’s accent in English is just as valid as a “native” accent. This is easily accepted by multi-national or mega-regional languages like Spanish, Russian, Chinese, etc, but in places like Poland as there is largely only one way of speaking, the bar is raised for their own language.

So in fact that means that the same n-thousand hours done by an Englishman in Russian could have the Russians noticing very little different about the foreigner, especially if he has a bad haircut. Whereas if he has a really bad haircut and the same n-thousand hours of Czech, the reaction will probably be “he looks like one of us, but our language is difficult and so we must forgive the way he sounds, although obviously we are frank and friendly people so we will tell him to his face at regular intervals that his Czech sucks bigtime.”

Given this subjectivity, I decided long ago never to walk in anyone’s linguistic shadow, but simply to set amounts of words as targets. 15,000 words is in language learning, to my mind, what the marathon is in athletics. If you’re fit, you can do it with patience and training. And if you can do it, nobody can say you’re not fit.

There are longer races, there are tougher events. But the marathon is the ‘classic’ and the marathon runner knows that it’s really a competition against yourself and not really against the runners alongside. Even people coming in at six hours are clapped and get a medal. So should language learning be.

If this article is of interest you can look up the article as plenty of people have some interesting stuff to say, both about the 10,000 hours nonsense and the number of words needed. I get into a discussion with “Lingua Frankly” blogger Niall Beag (known as Cainntear) on when the Pareto rule isn’t just a number like 10,000 with no real basis for being a law. There are also those who are ready to stand up for the honour of the number 10,000 and tell the detractors of 10,000 hours to mind their jolly manners. Excellent thread.

I’m going to add more thoughts there today.

%d bloggers like this: