Learn 12 languages in 12 months? Right.

[emk]

« I have to admit that I have given this much thought. I’ve heard a lot about it in the mainstream media but since I don’t have a technical background in the field I’m frankly sort of clueless about the whole thing. That said, I would think we have to start by defining thinking, intelligence and the various kinds of intelligence. Then how can we make artificial forms of these intelligences. Did you have particular concerns that you would want to share ? »

If I were a language examiner, I would say that an answer like that demonstrates a pretty good mastery of the language. There’s probably not much point in continuing.

This is an evasion strategy. And I understand the DELF/DALF examiners, for example, are specifically trained to notice these evasions, and to either deduct points for them, or to come at the topic from another direction. Otherwise students would just prepare and rehearse 20 different bits of boilerplate, plug in words, and ace the exams. They do actually expect you talk about the topic that got picked.

In the middle of my exam, one of the examiners asked a question about a genuine weakness in my proposal. I was frustrated, and I spontaneously replied, "Je ne peux pas résoudre tous les problèmes du monde aujourd'hui!" And the interviewing examiner laughed, while the silent note-taking examiner made a note. Oops. I don't think she was giving me points for that response, although it was fluent enough.

On a good exam, the examiners aren't just looking at 120 seconds of speech and seeing whether it sounds smooth. There's a specific set of things they want to see you do. From the CEFR levels:

B2. Can develop an argument systematically with appropriate highlighting of significant points, and relevant
supporting detail.
Can develop a clear argument, expanding and supporting their points of view at some length with
subsidiary points and relevant examples.
Can construct a chain of reasoned argument.
Can explain a viewpoint on a topical issue giving the advantages and disadvantages of various options.

[...]

Can take a series of follow-up questions with a degree of fluency and spontaneity which poses no strain for
either themselves or the audience.

So if we look at the response you proposed, it does not include an argument. There is no supporting detail. There are no examples. There is no discussion of the advantages and disadvantages. However, you do ask for followup questions, so that's good.

In most European countries, CEFR exams are used as a gateway to citizenship, to university, and to other real-world goals. Which is why they're typically so focused on using language to accomplish specific tasks.

Anyway, this argument isn't going to change anyone's mind this time, either.

[Severine]

I have assessed probably north of 2,500 adult English learners in the US, using both the Center for Applied Linguistics BEST evaluations and various institution-specific evaluations. Not CEFR, obviously, but I've certainly thought a lot about assessment. Any assessor with half a brain and some integrity will admit that, no matter how systematized one tries to make level evaluations, they're all to some extent subjective and fuzzy. Of course, everything is flawed, and just because something is hard (or even impossible) to measure doesn't mean we shouldn't try, but we need to know the limits of our systems.

When someone passes a B2 (speaking) test, it means they spoke at a B2 level that day according to their assessor. Someone who fails the same test couldn't meet that bar. That gives us plenty of useful information, sure, but competence and eloquence are fluid and situational and complex.

I have seen learners who I know have modest active vocabularies consistently outperform those with larger active vocabularies, in both testing and social situations, because they were able to better marshal and manipulate the words they knew, and because they were better able to perform competency and conceal their weaknesses. However, I have also seen these students stumble when they reached the limit of how far they could coast on talent.

I think of it in terms of probabilities. A vocabulary of 1000 words gives you a high chance of doing well on an A1 exam and a near-zero chance of doing well on a B2 or higher exam. If you want to pass a C1 exam, you might be able to pull it off with a vocabulary of 7,000 words and a shaky knowledge of conditionals, but you'll have a better chance with a vocabulary of 10,000 words and a solid mastery of grammar. The range of how many words are needed depending on the person is, I believe, larger than many people think.

Various aspects of learners' speech that we're expected to judge have been mentioned: appropriate vocabulary usage (a supposed indicator of vocabulary size, although it's seldom that simple, as mentioned above), ease of manipulating grammatical structures correctly, use of complex phrases, fluidity and ease of speech overall, use of expressions and idioms, organized and logical flow, etc. However, each one involves a subjective, human judgement. What does "ease" look like? What constitutes a complex phrase structure? Should errors common to native speakers count as errors for the learner? What does a coherent and organized flow of speech sound like? Should level of education be factored in, considering it's a semi-academic exercise? There are a thousand line-drawing exercises and opportunities for unconscious bias in every assessment. Assessors try to be fair, but are human.

People have said, in this thread, that in daily life we judge people's skill and fluency in a foreign language very quickly. This is true, but we're sometimes "wrong" in some sense. Some people are shy or anxious, for example, and I had students who presented, especially with strangers and in stressful situations, as having a small vocabulary or low levels of comprehension when, in fact, the problem was fear and freezing up. Conversely, I'm sure we've all had the experience of starting a conversation with someone, native speaker or not, who seemed confident and competent at the outset, only to realize after listening for a longer period of time that the person was just a fantastic bullshitter.

People who are confident and gregarious tend to do better on verbal assessments because, on some level, we're measuring how good that person is at talking. The question then is, is that okay? Learners who project self-confidence and personality are easier and more pleasant to talk to, for most people, so is an assessment system that privileges that demeanour actually a reasonable reflection of that person's "level" as our society judges it? Does having the capacity to do more count for much if you can't do it on command in a high-pressure situation?

[s_allard]

The fundamental problem is that many people believe that a language is a random bunch of words from a dictionary. Learning a language then means learning a ton of words, the more the better.

This is a bit of a non-sequitur. As it happens I don't think 'many believe language is a random bunch of words from a dictionary', but even if they did it wouldn't invalidate the truth that knowing more or 'a lot' of words common to different registers, and knowing how to use them, adds both considerable substance and flexibility to communication.

You seem to be making an argument for reducing communication to its lowest 'functional' threshold in the name of reducing both time and quantity of material and then insisting this qualifies as 'having learned a language. I don't even want to argue against the strong possibility that this base could be put into use and that over time, with lots of exposure, interaction and use it will develop into more. That is commonsensical, but the base is not 'having learned a language' unless you want to shift the definition goalposts. To my mind the process of expanding that elementary base over time is the actual 'learning a language'. The first bit was merely acquiring some basic tools.

There is this myth that you need a huge vocabulary for a C2 exam because you never know that subject might come up. If you intend to sound like an expert talking to another expert, then you will need expert vocabulary. But you are talking to a language examiner who is not interested in the subject but in how you can use the language to talk about it. This includes admitting your ignorance and talking about that.

What is 'huge'? You should know that you won't just be able to repurpose and apply a vocabulaire considered satisfactory for B1 or B2 to navigate a C2 exam. I'm finding it fairly difficult to know what you are saying exactly, because it looks like you're claiming it's possible to traverse all the CEFR levels with some kind of minimal core vocabulary and that this will serve you in all scenarios. Which I find strange after rather forcefully insisting that the CEFR measures some sort of level progress. So what is this progress scale? A measure of how someone manipulates the same words they had at A2 plus a small handful extra? And that this gets the name 'C2'?

Forgive my bluntness, but it sounds like nonsense.And the refrain about not bringing it up because it doesn't end well it actually quite annoying. How could it ever 'end well' when making such absurd and frivolous claims?

I forgive your bluntness but I will also point out that unlike most people around here I have actually sat three CEFR exams, two of which successfully : C2 level (Spanish) and B2 (German) both at first go. My B2 Italian was not a total sucess but I considered it more of a pleasant experiment. Now I’m thinking about having a go at C1 or C2 in June. So I think I know a thing or two about preparing for CEFR exams.

As an aside, I was very surprised by how few people were actually taking the exams. For the Spanish exam, out of a total of around 50 people for all the levels there was only one person at the C2 level, me. For Italian B2, I was with two teenage girls who were not interested in speaking to anyone not a teenager. I’m not sure if there was anybody for the C levels. For German, I paid extra for a private test session at my convenience at the local Goethe Institut. So let’s say I put my money where my mouth is.

As the old-timers around here like emk, iversen and iguanamon know, I’ve always argued that a relatively small vocabulary -- not limited to the 300 most common words – combined with impeccable grammar, idioms, formulaic speech patterns, discourse-making strategies, writing practice, good phonology, reading one fat book, and exam-focussed preparation makes for great results. For me.

In the end, it is all about what works for you. My strategy and taking the tests have worked wonders for my language skills. Now I spend more time enjoying my languages and less time arguing here on this forum. All I can say is go take some CEFR tests and let us know how it went.

[Le Baron]

I’ve always argued that a relatively small vocabulary -- not limited to the 300 most common words – combined with impeccable grammar, idioms, formulaic speech patterns, discourse-making strategies, writing practice, good phonology, reading one fat book, and exam-focussed preparation makes for great results. For me.

For anyone I'd think. It usually takes more than three weeks. And it did for you, as we both know.

In the end, it is all about what works for you. My strategy and taking the tests have worked wonders for my language skills. Now I spend more time enjoying my languages and less time arguing here on this forum. All I can say is go take some CEFR tests and let us know how it went.

Thanks for the information. I now know that I've sat twice as many CEFR-based exams as you have (and two early version ones). On most of them I didn't spend much time learning (as in doing 'reps') for vocabulary. In fact none at all. Just reading, random listening and doing the work. It didn't mean I relied on a small core vocabulary, it was just a different way of accruing it. The only one I sat most recently was the Spanish B1 at Instituto Cervantes, some time after arriving at this forum, because I wanted to test everyone's insistence here that this is a robust measure. I still think CEFR tests are like any exams. Since many people can pass from the same intake, but have wildly different competences, it's vague.

The Dutch I passed ages ago and if I compare my competence now with when I passed e.g. C1, The difference is palpable. It has nothing to do with the tests, it's about time. I learned this fact about second language competence before I was twenty.

[leosmith]

All I can say is go take some CEFR tests and let us know how it went.

:lol: Fair enough. All I can say is go study some difficult languages with the same method and let us know how it went.

[s_allard]

« I have to admit that I have given this much thought. I’ve heard a lot about it in the mainstream media but since I don’t have a technical background in the field I’m frankly sort of clueless about the whole thing. That said, I would think we have to start by defining thinking, intelligence and the various kinds of intelligence. Then how can we make artificial forms of these intelligences. Did you have particular concerns that you would want to share ? »

If I were a language examiner, I would say that an answer like that demonstrates a pretty good mastery of the language. There’s probably not much point in continuing.

This is an evasion strategy. And I understand the DELF/DALF examiners, for example, are specifically trained to notice these evasions, and to either deduct points for them, or to come at the topic from another direction. Otherwise students would just prepare and rehearse 20 different bits of boilerplate, plug in words, and ace the exams. They do actually expect you talk about the topic that got picked.

In the middle of my exam, one of the examiners asked a question about a genuine weakness in my proposal. I was frustrated, and I spontaneously replied, "Je ne peux pas résoudre tous les problèmes du monde aujourd'hui!" And the interviewing examiner laughed, while the silent note-taking examiner made a note. Oops. I don't think she was giving me points for that response, although it was fluent enough.

On a good exam, the examiners aren't just looking at 120 seconds of speech and seeing whether it sounds smooth. There's a specific set of things they want to see you do. From the CEFR levels:

B2. Can develop an argument systematically with appropriate highlighting of significant points, and relevant
supporting detail.
Can develop a clear argument, expanding and supporting their points of view at some length with
subsidiary points and relevant examples.
Can construct a chain of reasoned argument.
Can explain a viewpoint on a topical issue giving the advantages and disadvantages of various options.

[...]

Can take a series of follow-up questions with a degree of fluency and spontaneity which poses no strain for
either themselves or the audience.

So if we look at the response you proposed, it does not include an argument. There is no supporting detail. There are no examples. There is no discussion of the advantages and disadvantages. However, you do ask for followup questions, so that's good.

In most European countries, CEFR exams are used as a gateway to citizenship, to university, and to other real-world goals. Which is why they're typically so focused on using language to accomplish specific tasks.

Anyway, this argument isn't going to change anyone's mind this time, either.

The problem here is that we are confusing knowledge of the topic with ability to use the language. The latter is what we are trying to assess. The topic is just a pretext to see how well you can use the language. We are not expected to know a bit about everything in the world.

We are confronted with this all the time in our native language(s). When asked a question like : Should bitcoin mining be banned for environmental reasons ?, what are you supposed to answer if you’ve never heard of bitcoin mining in your life ? You don’t evade the answer, you tell the truth. You simply say you know nothing about the topic and ask for more information. The fact that you can’t answer the question directly doesn’t mean you don’t know the language at a B2 level.

What I find so astonishing here is this obsession with vocabulary size and so little emphasis on how vocabulary is actually used. Why not just test for vocabulary ? Here is a list of words, how many do you know ? 7500, then you must be B2.

There has barely been mention of grammar in the discussion here. For example, grammatical gender in French is a real killer for speakers of other languages including the other Romance languages. What about the pronouns en and y ? These are the things that demonstrate real mastery of a language.

[tastyonions]

And with a small vocabulary you almost certainly wouldn't know the word "mining" and might not even understand the question at all.

[tastyonions]

All I can say is go take some CEFR tests and let us know how it went.

:lol: Fair enough. All I can say is go study some difficult languages with the same method and let us know how it went.

The big elephant in the room with the vocabulary issue here is the fact of (native French + English) -> Spanish transparency.

Someone with that linguistic background can have the illusion that he did "little vocabulary work for Spanish" when in reality he already did years and years of it: when he was learning his native languages.

[emk]

The big elephant in the room with the vocabulary issue here is the fact of (native French + English) -> Spanish transparency.

Someone with that linguistic background can have the illusion that he did "little vocabulary work for Spanish" when in reality he already did years and years of it: when he was learning his native languages.

Even just native English → French transparency is huge. It's a standing joke in our household that you can just take any big English or French word, and change the pronunciation to get the other language.

(Native English + post-Super-Challenge French) → Spanish gives an even bigger discount. There's still plenty of vocabulary that I actually need to learn, as every morning's flash cards are happy to remind me. Or I can just pick up a novel, and struggle to get a few sentences here or there. But over time, the discount matters.

Middle Egyptian, on the other hand? You can count the cognates on the fingers of one hand, all borrowings related to Egyptian culture. Plus the weird coincidence of dšr.t "red desert". If you want to know a word, you've got to actually learn it.

One of the weird things about the idea of trying to tackle so many languages so quickly is that:

• You can't tackle a bunch of Category IV and V languages. (Or the equivalent, depending on your starting point. I'm using the FSI classification, which assumes a native English speaker as the base.) There's just too much to learn. If you say, "I'm going to learn A2 Japanese and Korean and Mandarin and Arabic and Persian and Gaelic and Finnish, etc., all in 1 month each," you're just being silly. People have apparently reached an A2 in Mandarian in 3 months, but they were mostly experienced polyglots spending every waking minute studying.
• But if you tackle Category I and II languages, you'll inevitably end up learning a tons of closely related languages.

So let's say you go into this challenge speaking English, French and German. You could choose all the major Category I languages, giving you Spanish, Portuguese, Italian, Romanian, Dutch, Danish, Norweigian, Sweedish, and Afrikaans. That gives you 9.

To get 3 more, you could either look at smaller Romance and Germanic languages, or you could dip into Category III. If you decide on the latter, that gives you Swahili, Indonesian and Malaysian. These are all languages where a majority of speakers use it as an L2. Sort of like international English or North African French, these languages are likely to be more forgiving.

So you're probably looking at (roughly) 4 Romance languages, 5 Germanic languages, and 3 regional linguae francae that are unusually tolerant of L2 speakers. You could replace Indonesian with Icelandic, which would make things easier. Or if you can keep a straight face, you could swap out Malaysian for something like Scotts. But at that point, even your YouTube viewers might start raising eyebrows.

Just keeping that many closely-related A2 languages straight seems like a nightmare by itself. My brain is trying to leak out my ears just imagining it.

[tractor]

As the old-timers around here like emk, iversen and iguanamon know, I’ve always argued that a relatively small vocabulary -- not limited to the 300 most common words – combined with impeccable grammar, idioms, formulaic speech patterns, discourse-making strategies, writing practice, good phonology, reading one fat book, and exam-focussed preparation makes for great results. For me.

After all those activities and preparations, you’d probably picked up a lot of words, and no longer had such a small vocabulary.