Hard to say. From what I've read and heard anecdotally from friends that speak 5+ languages, the get easier as you learn new ones. You begin to think about them structurally and pick them up in groups of closely related languages. At some point its just a matter of learning the idioms and building you word catalog, as you've learned the grammar in a similar/related language.
I think it is the same with programming. Once you understand the basic programming paradigms the rest becomes easy. I think in terms of language is Object, Verb, Subject order.
I'm concerned that there are often going to be weird little issues about particles and stuff. For example, a lot of people have a tidy notion about Romance language grammar -- subject-verb-object, conjugate the verb to show person, number, tense, and mood. But the different Romance languages have rather different tricks about what happens when the object of a verb is a pronoun. In some languages it then gets squashed onto the end of the verb (possibly changing the form of the verb), even into the middle of the verb before the endings (Portuguese mesoclisis), or placed before the verb. And the different languages have different preferences and possibilities about this.
Exactly, the more you see the common traits of human communications beneath local syntax and grammar rules. Also the phonetic drifts (rhotacism) that push languages away from each other, when revealed, help to recognize patterns that were once obscure.
It's not important to speak all of them at the same time or is it? If you need 2-4 weeks to get back to fluency in one you haven't used for a while should be quite okay.
And that softer requirement is also not cheating. If you speak many other languages you even need to reactivate your native language.
It's reasonable to ask what a person means when they "speak X languages", though. It's in itself more vague than one might think. If you ask a monolingual person he might consider a C2 (European Framework) level to not be "speak my language", while a language learner might consider a barely passed B2 and then never used again as "speaking that language".
"monolingual person he might consider a C2 (European Framework) level"
I speak one foreign language at a solid C2 level, but I also scored in the 99th percentile on the GRE verbal (USA's English Language grad school entrance exam for native English speakers). I never feel fully fluent in that foreign language, even when I can pass language exams most locals can't. There's simply to way for me to match my fluency and comfort with English.
So I agree; the standard for fluency is always relative. And -- contrary to superficial appearance -- even native speakers have a wide variety of levels of competence and you can always benefit from studying more.
If you have very stringent requirements for "learning a language", than barely anyone other than native speakers can quality. For example, I speak English for decades, but I know I don't speak exactly like a native.
My father speaks 7ish languages, and if you talk to him about it he says there are some problems. It definitely takes some time for him to get into the "mindset" of a language. He also says that vocabulary is the hardest part: he says that if he forgets the word for a thing in the language he's trying to speak, he'll find the word in one of his languages, so sometimes he embarrasses himself that way.
That said, it's certainly amazing what he can do. He taught himself Italian over the course of about a month, simply to read a single dissertation written in Italian. He spoke French already, and he asked me a few questions because I'm fluent in Spanish, because when he didn't find a word in his Italian/English dictionary he wanted to make a guess based on other romance languages. I think ultimately he wouldn't even consider himself at all competent in Italian, even though he was able to read doctorate-level writing for a few weeks.
> Those who knew three languages, however, were diagnosed 6.4 years later than monolinguals, while for those fluent in four or more languages, enjoyed an extra nine years of healthy cognition...
Glad to read this...
I speak ~4, english being the 3rd (far from native but good enough to do biz) and the forth being basic/intermediate level. I've met people fluent in 5 languages... but they start to struggle with the 6th onwards.
That being said, I think 30+ is a lot. I've never known anyone who speaks fluently more than 6, but hey, if they can... great!
This people must have lived in a lot of countries since their early years, or their parents were polyglots themselves. I think the key is early exposure to different languages. You can easily teach kids 5 languages before they're 12...
I know a hyper-polyglot and he says it's not possible to be fluent in all of them. But it ought to be quite easy to reactivate one you studied well enough.
I have a friend who seems to pick up languages by osmosis. We went for a meal and were served by someone from Croatia. My friend started chatting to him (in English) and asked him for a few words in Croatian (no more than about a dozen) then was able to start speaking to him in Croatian. Afterwards he explained the few words let him know some basic structure he could relate to other languages he knew and he could extrapolate from them and the new words enough to be able to have a conversation. It was both spooky and amazing at the same time. He has a library in dozens of languages and gets books 'just to see what the language is like' - it's like some sort of super-power.
In The Four Hour Chef, Tim Ferriss explains how he picks up languages and it's a very similar method. The structure of many languages are the same and so once you know the basic structure and a few helpful words, you can start to communicate. Fluency is a whole other thing but it changed how I view learning a language.
So, basically, he's an organic version of Star Trek's universal translator?
More interestingly, if your friend can do that, perhaps the universal translator is not such a magical concept after all. For terrestrial languages, at least.
I'm curious how much effort goes into maintenance. Even if you could learn 10+ languages, how much time per day/per language is necessary to retain fluency?
That must be why these people seem to constantly seek each other out.
I remember once reading an interview with a supposed hyperpolyglot, in which he revealed that he spent nearly all of his time (8+ hours/day) reviewing flashcards. I guess it's possible, if you really like doing flash cards.
These popular narratives about "hyperpolyglots" tend to have really naive views about what "fluency" in a language means. It's probably feasible to have rudimentary knowledge of many languages, but there are large practical limits as to how sophisticated your competence can be unless you're using that language regularly to communicate in a variety of domains.
When I met one of those I felt like I just received a license to say I speak 5 languages.
The 1st: my mother tongue.
The 2nd: I'm native-level, studied all my life, use it everyday.
The 3rd: I'm somewhere between advanced-survival and fluent-colloquial depending on my BAC; spoken in the country I live in so I don't put that much effort.
The 4th: just basic knowledge but really closely related to the 1st, so communication is fluent even if I'm making up words and abusing the rules of grammar.
The 5th: I study it every day since over a year, I speak retarded and understand correctly pronounced simple phrases 60% of the time. Can survive in the countries where it's spoken and it's a nice perk with people; they become surprisingly friendly because very few people speak the language other than natives.
Maintaining a language is a lot of work. That's why I've created iOS language apps with several games. My thinking is that to learn and remain effective in a language, it has to be done as some form of entertainment.
If people could give them a try and provide ideas and feedback, I'll do my best to incorporate it into a future release. I'm adding Spanish Verb Conjugation drills to the next release.
Speaking foreign languages opens truly opens many opportunity that most monolinguals (especially English-speaking ones) do not even suspect. Yes, my German colleagues and friends all speak English and we could get along fine using it. However we wouldn't be able to share our personalities and cultural differences as well as we do on a daily basis if we didn't share a native language (for them) spoken at a high level (for me).
People in general seem to underestimate how important it can be to speak the same language as a group[1], even when you happen to share another language. If I were to move to another country for many years, I would definitely start learning the most relevant language there. Even if ~98% spoke English.
Being able to speak something like English in some part of the non-English word might make you able to communicate with people there. But that doesn't necessarily mean that you'll be able to have meaningful connections compared to speaking the native tongue.
[1] Like the native language in some country or region.
It's not even just an opportunity thing. Becoming fluent in Spanish has really changed the way I think, particularly on political, social, and emotional issues. Spanish is really preferable for expressing many ideas--it has a richer vocabulary for certain topics and forces word patterns that encourage different ways of thinking. Thinking in Spanish changes not only how I think, but what I think. Being able to think both ways gives me the ability to approach problems from more directions.
I speak 8 languages, 4 super fluently, 2 sufficient for daily communication and 2 so-so for special occasions only, and learning continuously. They are all from different groups without many commonalities (Germanic, Romance, Slavic, Sino-Japanese). For Sino-Japanese there is the odd disparity between the ability to read and ability to write that doesn't exist in the other groups, as well as a single character having different sound depending on context. Slavic languages are crazy because they can form sentences in any order and the order changes meaning substantially, something that doesn't exist in orderly Germanic languages. Romance languages often add terms related to feelings that do not have any corresponding term in another languages.
I would disagree that learning one language helps you with another unless they are from some kind of language continuum - languages are so vastly different (especially depending on geographic distance) when you try to really master them, often full of idiomatic expressions and expressions you can't understand unless you grow in the historical context the language developed, and accurate translations are basically impossible. Some even use completely different causal logic (like Aymara), or miss elements considered basic in other languages (directions absent in Hawaiian) etc.
Learning languages is driven by curiosity, was handy a bit while traveling around the world and it helps to be near some language school, attend parties with people from all around the world and just have fun trying to chat together natively ;-)
A lot of this centers around what proficiency means. With a few weeks of training you can be a tourist in a foreign city just fine. Getting to the proficiency of a native 10 year old is vastly harder. Another huge jump is having the language skills to get a job more complex than dishwasher in another language. Not to mention being a poet.
IMO, basic proficiency is simply the ability to communicate unaided with effort on both parties. Which really only takes the language skills of ~5-7 year old native child.
> A lot of this centers around what proficiency means.
This is very true and this is what bothers me when reading about these supposed polyglots/hyperglots. As a Portuguese speaker, I get Spanish "for free" as in I'm able to read Spanish, follow a Mexican soap opera on TV, and stumble through an order at a Mexican restaurant. Does that mean I can add Spanish to the list of languages that I speak?
> IMO, basic proficiency is simply the ability to communicate unaided with effort on both parties. Which really only takes the language skills of ~5-7 year old native child.
A 5-7 year old child with normal language development will be fluent, as in mastery of the grammar. There's no comparison between an adult who can get by in a language and a native speaker. For example, a child who is a native speaker of a language with grammatical gender will never mess up gender agreement whereas an adult learner might struggle mightily with that.
Why group Sinitic and Japonic languages together? It's true that Japanese (and the other minor Japonic languages) borrows a lot of Chinese vocabulary but grammatically and phonetically they have practically nothing in common.
This way I could express that the majority of the speakers struggle with reading/writing asymmetry, whether they are Japanese, Mandarin, Cantonese speakers, in Taiwan or speak another related language.
Because a large part of Japanese language (and Korean and Vietnamese for that matter) is made of loanwords of Chinese origin (漢語) it made some sense to bring them together (albeit I suspect it is often done because Japanese use Chinese characters too). So while the grammar vastly differ, learning one is really useful to learn another because a lot of words are roughly the same.
As a side note I was reading in a book[1] the rather extreme view that the "corrupt" Japanese or Korean readings of characters maybe seen as a dialect of Chinese.
Big surprise, a book from the early 19th Century has a silly linguistic view. By this logic English is a romance language because it has a lot of French borrowings, despite not descending from Latin. As for Chinese characters, well, do we consider Vietnamese and English part of the same language family because both are written with the Latin alphabet?
I acknowledged, in the post you're replying to, the large body of borrowed vocabulary, but that's not how languages are classified/made into families. Chinese and Japanese do not have a common ancestor and have very, very different grammar and phonetics.
There is also the possibility that it does help, but not strictly with the grammar or the mechanics of the actual language. I can imagine that learning a foreign language with any proficiency is as much about learning how to teach your brain to break free of the constraints of knowing only your first language. For most people, thoughts are directly connected to language and sentence structure. Perhaps learning another language teaches you to think more in the concepts and be more flexible with your thought structure than thinking simply within the natural constraints of your first language.
For instance, if you can't wrap your head around why a language may be missing 'basic' elements, you would never really master that language.
.. and here I'm having difficulty plodding through Scala... :-(
I wonder what's the similarity between learning human languages, and computer languages. I think one can pick up a computer language fairly quickly if it's in the same paradigm (imperative, functional, logic, etc.)... but actually mastering it is much harder, because there are so many different ways to do ("say") the same thing.
You've got a valid point. It took me about a year to become really proficient in python, but after that a month to pick up C relatively well, and another 3 months to pick up Scala (probably because functional programming was something new to me). But after Scala, Haskell took about a week, so you might be on to something with similar paradigms being easier to pick up.
One significant difference is that programming languages have a very small vocabulary, while spoken languages usually have an enormous vocabulary. A programming language might have 10-20 reserved words (plus specific meanings of tokens like "++", "\", ":=" that you need to learn). On the other hand, yesterday I realized that I didn't know the word for "brick" in any foreign language, including languages I've studied and regularly used for a long time.
That's one factor that makes it plausible that a programmer would learn a new programming language in a weekend, but somewhat implausible that someone would learn a spoken language in a weekend (although you can get surprisingly far). On the other hand, Daniel Tammet apparently learned Icelandic in one week:
I was at the meeting here in Berlin and can only say that it was awesome. Such interesting talks and people! For those interested, the talks should be posted online soon at https://www.youtube.com/channel/UC_8BR59zKgBzeWPxFtmHSwA
I think that is better to speak one or a few well and take profit of the extra brain space.
People very specialised into translation problems (just a narrow class of problems) probably will score poorly with complex technical problems of a different nature. If your problem needs to be solved just one time and as soon as possible (as in the 99% of the cases) you probably should not really care to much about if is solved in chinese or Tamil. Is not really relevant if you think about it.
You just can hire a translator one or two weeks each year to act as megaphone at the end of the process and that's all. Translation uses just a small percentage of the project time, generally at the beginning and at the very end.
To hire the translator full time, 365 days/year is just easier for human resources, but not a guarantee that you are really targeting the solver that you need.
If you claim to master 30 languages you can of course hammer a nail, but not to be a professional piano player. Is very unlikely. Days are just too short.
In any case reading the link provided seems that the journalist translates directly 'how to do something' as 'its great' and 'you can do it!'... Much hype, few substance, as usual.
I am somewhat skeptical of these people's claims. Some of these people are pretty prominent on the Internet and watching videos of them speaking languages I'm more familiar with found their command of them to be more beginner to intermediate. That's still a kind of accomplishment but really becoming fluent in a language is a huge investment of time because you have to learn lots of vocabulary, idioms, and collocations, and I just don't see how you can do that without spending thousands of hours using it.
I wonder if there are very different notions of fluency at work in different contexts. For example, I can speak extemporaneously about many topics in Portuguese and only somewhat infrequently pause to think about or ask for vocabulary. But I need others speaking with me to be cooperative, speaking clearly and without slang (in which case I generally understand easily). I usually don't understand informal conversations between native speakers. This could be viewed as obviously fluency by some people, yet obviously not fluency by others.
My degree was in linguistics (although it is not my current profession), and I wholeheartedly agree with your comment.
My personal opinion is that claims of polyglotism are quite often (if not usually) exaggerated, whether intentionally or not. There are many different metrics for language proficiency, but they are all just samples of a person's linguistic competence rather than a complete picture. For that reason, they're prone to the same issues as general intelligence tests. A high score on a language test is no guarantee of fluency just as a high IQ is no guarantee of intelligence-- and this is all assuming the person claiming to be fluent has even taken such a test, which is rare. I think more often people have individual standards for "fluency", which vary wildly. Again, I'm not saying that people intentionally mislead others about their own language ability, or that there aren't any true hyperpolyglots-- just that such claims should be taken with a grain of salt.
Both the description of the critical hypothesis in the article and the person who described it as "a bunch of crap" are wrong. The hypothesis actually states that there is an ideal window for language learning that occurs at a young age, whereas the article presents it as "a narrow window during childhood in which we can pick up the nuances of a new language". No serious academic linguist that I have ever met subscribes to a definition that rigid. Adults absolutely can pick up on the nuances of a new language, but for the overwhelming majority it takes considerably more time and effort compared to a child in the critical period.
I do believe that language learning for adults can be made more efficient than it usually is (e.g. the classic cookie-cutter layout of every foreign language textbook I've ever read), and perhaps some of the strategies laid out in the article are effective, but that does not discredit the critical period hypothesis and it's still going to take a lot of effort for a typical adult to become fluent in a new language.
Well, also, empirically, adults who learn a foreign language can become quite good at it but even after decades of using the language often have accents or fossilized incorrect grammatical uses or uses of vocabulary. That doesn't really happen to children either.
Yes-- I thought it was reasonably well established that children have superior "plastic" or "fluid" intelligence in general, but apparently a few people feel that this doesn't apply to language (for some reason).
I agree. Totally. I am not a very intellectual person, and you could say I'm a slow learner. But I really cannot fathom how these people can learn that much in so little a time. I'm currently enrolled in http://preply.com/en/skype/english-native-speakers where I get to chat with native English speakers online. And even if I am in the advanced levels of English, I am still learning a lot from my instructors. This means that, even after more than 20 years of speaking the language, I still haven't mastered it yet. I don't know how these people are doing it. It's amazing if it's true. But right now, I'm not buying it.
I agree. Totally. I am not a very intellectual person, and you could say I'm a slow learner. But I really cannot fathom how these people can learn that much in so little a time. I'm currently enrolled in http://preply.com/en/skype/english-native-speakers where I get to chat with native English speakers online. And even if I am in the advanced levels of English, I am still learning a lot from my instructors. This means that, even after more than 20 years of speaking the language, I still haven't mastered it yet. I don't know how these people are doing it. It's amazing if it's true. But right now, I'm not buying it
55 comments
[ 2.9 ms ] story [ 126 ms ] threadAnd that softer requirement is also not cheating. If you speak many other languages you even need to reactivate your native language.
It's reasonable to ask what a person means when they "speak X languages", though. It's in itself more vague than one might think. If you ask a monolingual person he might consider a C2 (European Framework) level to not be "speak my language", while a language learner might consider a barely passed B2 and then never used again as "speaking that language".
I speak one foreign language at a solid C2 level, but I also scored in the 99th percentile on the GRE verbal (USA's English Language grad school entrance exam for native English speakers). I never feel fully fluent in that foreign language, even when I can pass language exams most locals can't. There's simply to way for me to match my fluency and comfort with English.
So I agree; the standard for fluency is always relative. And -- contrary to superficial appearance -- even native speakers have a wide variety of levels of competence and you can always benefit from studying more.
"If..., then..." and "I've spoken English for decades."
That said, it's certainly amazing what he can do. He taught himself Italian over the course of about a month, simply to read a single dissertation written in Italian. He spoke French already, and he asked me a few questions because I'm fluent in Spanish, because when he didn't find a word in his Italian/English dictionary he wanted to make a guess based on other romance languages. I think ultimately he wouldn't even consider himself at all competent in Italian, even though he was able to read doctorate-level writing for a few weeks.
Glad to read this...
I speak ~4, english being the 3rd (far from native but good enough to do biz) and the forth being basic/intermediate level. I've met people fluent in 5 languages... but they start to struggle with the 6th onwards.
That being said, I think 30+ is a lot. I've never known anyone who speaks fluently more than 6, but hey, if they can... great!
This people must have lived in a lot of countries since their early years, or their parents were polyglots themselves. I think the key is early exposure to different languages. You can easily teach kids 5 languages before they're 12...
More interestingly, if your friend can do that, perhaps the universal translator is not such a magical concept after all. For terrestrial languages, at least.
These popular narratives about "hyperpolyglots" tend to have really naive views about what "fluency" in a language means. It's probably feasible to have rudimentary knowledge of many languages, but there are large practical limits as to how sophisticated your competence can be unless you're using that language regularly to communicate in a variety of domains.
The 1st: my mother tongue.
The 2nd: I'm native-level, studied all my life, use it everyday.
The 3rd: I'm somewhere between advanced-survival and fluent-colloquial depending on my BAC; spoken in the country I live in so I don't put that much effort.
The 4th: just basic knowledge but really closely related to the 1st, so communication is fluent even if I'm making up words and abusing the rules of grammar.
The 5th: I study it every day since over a year, I speak retarded and understand correctly pronounced simple phrases 60% of the time. Can survive in the countries where it's spoken and it's a nice perk with people; they become surprisingly friendly because very few people speak the language other than natives.
http://appstore.com/h4labs
If people could give them a try and provide ideas and feedback, I'll do my best to incorporate it into a future release. I'm adding Spanish Verb Conjugation drills to the next release.
Being able to speak something like English in some part of the non-English word might make you able to communicate with people there. But that doesn't necessarily mean that you'll be able to have meaningful connections compared to speaking the native tongue.
[1] Like the native language in some country or region.
I would disagree that learning one language helps you with another unless they are from some kind of language continuum - languages are so vastly different (especially depending on geographic distance) when you try to really master them, often full of idiomatic expressions and expressions you can't understand unless you grow in the historical context the language developed, and accurate translations are basically impossible. Some even use completely different causal logic (like Aymara), or miss elements considered basic in other languages (directions absent in Hawaiian) etc.
Learning languages is driven by curiosity, was handy a bit while traveling around the world and it helps to be near some language school, attend parties with people from all around the world and just have fun trying to chat together natively ;-)
IMO, basic proficiency is simply the ability to communicate unaided with effort on both parties. Which really only takes the language skills of ~5-7 year old native child.
This is very true and this is what bothers me when reading about these supposed polyglots/hyperglots. As a Portuguese speaker, I get Spanish "for free" as in I'm able to read Spanish, follow a Mexican soap opera on TV, and stumble through an order at a Mexican restaurant. Does that mean I can add Spanish to the list of languages that I speak?
> IMO, basic proficiency is simply the ability to communicate unaided with effort on both parties. Which really only takes the language skills of ~5-7 year old native child.
A 5-7 year old child with normal language development will be fluent, as in mastery of the grammar. There's no comparison between an adult who can get by in a language and a native speaker. For example, a child who is a native speaker of a language with grammatical gender will never mess up gender agreement whereas an adult learner might struggle mightily with that.
http://en.wikipedia.org/wiki/Sino-Japanese
As a side note I was reading in a book[1] the rather extreme view that the "corrupt" Japanese or Korean readings of characters maybe seen as a dialect of Chinese.
[1] Élémens de la grammaire chinoise. Abel Rémusat (1822) https://play.google.com/books/reader?printsec=frontcover&out...
I acknowledged, in the post you're replying to, the large body of borrowed vocabulary, but that's not how languages are classified/made into families. Chinese and Japanese do not have a common ancestor and have very, very different grammar and phonetics.
For instance, if you can't wrap your head around why a language may be missing 'basic' elements, you would never really master that language.
It's not the language that helps, but having the experience of learning it.
I wonder what's the similarity between learning human languages, and computer languages. I think one can pick up a computer language fairly quickly if it's in the same paradigm (imperative, functional, logic, etc.)... but actually mastering it is much harder, because there are so many different ways to do ("say") the same thing.
That's one factor that makes it plausible that a programmer would learn a new programming language in a weekend, but somewhat implausible that someone would learn a spoken language in a weekend (although you can get surprisingly far). On the other hand, Daniel Tammet apparently learned Icelandic in one week:
https://en.wikipedia.org/wiki/Daniel_Tammet#Savantism
People very specialised into translation problems (just a narrow class of problems) probably will score poorly with complex technical problems of a different nature. If your problem needs to be solved just one time and as soon as possible (as in the 99% of the cases) you probably should not really care to much about if is solved in chinese or Tamil. Is not really relevant if you think about it.
You just can hire a translator one or two weeks each year to act as megaphone at the end of the process and that's all. Translation uses just a small percentage of the project time, generally at the beginning and at the very end.
To hire the translator full time, 365 days/year is just easier for human resources, but not a guarantee that you are really targeting the solver that you need.
In any case reading the link provided seems that the journalist translates directly 'how to do something' as 'its great' and 'you can do it!'... Much hype, few substance, as usual.
My personal opinion is that claims of polyglotism are quite often (if not usually) exaggerated, whether intentionally or not. There are many different metrics for language proficiency, but they are all just samples of a person's linguistic competence rather than a complete picture. For that reason, they're prone to the same issues as general intelligence tests. A high score on a language test is no guarantee of fluency just as a high IQ is no guarantee of intelligence-- and this is all assuming the person claiming to be fluent has even taken such a test, which is rare. I think more often people have individual standards for "fluency", which vary wildly. Again, I'm not saying that people intentionally mislead others about their own language ability, or that there aren't any true hyperpolyglots-- just that such claims should be taken with a grain of salt.
Both the description of the critical hypothesis in the article and the person who described it as "a bunch of crap" are wrong. The hypothesis actually states that there is an ideal window for language learning that occurs at a young age, whereas the article presents it as "a narrow window during childhood in which we can pick up the nuances of a new language". No serious academic linguist that I have ever met subscribes to a definition that rigid. Adults absolutely can pick up on the nuances of a new language, but for the overwhelming majority it takes considerably more time and effort compared to a child in the critical period.
I do believe that language learning for adults can be made more efficient than it usually is (e.g. the classic cookie-cutter layout of every foreign language textbook I've ever read), and perhaps some of the strategies laid out in the article are effective, but that does not discredit the critical period hypothesis and it's still going to take a lot of effort for a typical adult to become fluent in a new language.