[miss_represent]

The article attempts to relate several hypothetical anecdotes, informally, but is otherwise scant on actual evidence.

Here and there, are sprinkled vague nods towards unnamed "researchers" and "the new way" but not many specifics on who, when or how. Mostly it's just: Trust us, here's how it is now.

The article uses only one name, and only aims to chip away at it. It has no new celebrities, no names stepping forward to take credit for the incredible breakthroughs that destroy the old and oh-so-inferior.

The authors, and the editor are after something. But they offer nothing (or very little) in exchange, pre-supposing that it's just common sense which prevails over the tyrrany of something innately unlikable.

Not sure why linguistics ruffles some people's feathers. It's such a dry, imprecise sociological subject after all.

EDIT: they make a single mention, in passing, of research by Harvard psychologist Steven Pinker, but very briefly, and don't return to it afterward.

[mikekchar]

Even a single citation to a paper would have gone a long way.... sigh... From my perspective Universal Grammar is not even the biggest contribution to language acquisition that Chomsky made: it's the notion that the mechanism for language acquisition is the same in children as it is in adults. What's weird about this is that I have always been under the impression that Chomsky's opinion was that language acquisition was not a lock step uncovering of universal grammar, but that universal grammar was simply the best way to represent language. Indeed, the work that has been done showing that Krashen's natural order hypothesis is correct (that people acquire the grammar for a language in a certain particular order, independent of the order in which is is presented) give some credence to that notion. It doesn't necessarily follow that the brain actually holds a representation of universal grammar.

I came away from this article being confused as to whether I need to re-read Chomsky's work, or whether the author needs to do so.

[4ad]

> people acquire the grammar for a language in a certain particular order, independent of the order in which is is presented

I didn't know about this. It sure makes sense. Do you know of any institute, course, or individual that makes use of this fact in order to optimize for learning a new language?

I need to learn German, and let's just say that I find the traditional learning methods almost impossible. And I have been living in a German-speaking country for over 5 years now.

[ktRolster]

How much time every day do you spend on learning German?

[4ad]

Like read books or have someone tutor me? Right now I spend zero time doing this. I just gave up. I tried but nothing stuck, and it was so difficult for me on an intellectual level that quite literally after one hour of study I was so mentally worn out that I could not do my job. I had to sleep through the rest of the day and I literally felt like I was going crazy.

I don't get it, I can learn (and remember) a new programming language and have some practical level of competency in hours (even with Haskell, actually, Haskell was one of the easier languages), but with real languages I am totally inapt.

In school I studied 8 years of French, two or three classes per week, and it was a real course, nothing leisurely, lots of homework. But I can't speak a single word of French. Believe me, I worked really hard those 8 years just to get a passing grade, and if I weren't exceptional at Physics (won 1st prize at the National Physics Olympiad, and many other national contests), I'm pretty sure my teacher would have failed me.

I spoke my first word while being 3 months old, and at 6 months I was using coherent sentences and engaging in conversation with people. I guess I used up all my language acquisition skills in that stage of development.

What is going on?

[panglott]

Acquiring languages isn't about going through books and studying—these are methods that adults can use to make use of the metalinguistic knowledge acquired from their native language. There's a pretty good popular-science book on the topic, Becoming Fluent: How Cognitive Science Can Help Adults Learn a Foreign Language, by Richard Roberts and Roger Kreuk (MIT Press, 2015). https://www.amazon.com/Becoming-Fluent-Cognitive-Science-Lan...

People acquire languages, they don't study them. Study is just a way of speeding the process—with immersion you can gain a huge amount of passive knowledge of a language without any study or realizing you are doing so. Questions like "what books are you studying" or "how many hours did you do grammar drills" aren't very useful. I suspect that the typical high school foreign-language study is a sneaky way of enforcing the US as a society of Anglophone monolinguals. If the US wanted a multilingual population, they would permit dual-language immersion elementary schools, but those have only been popping up in the last decade or so.

How much time do you spend fumbling around in German without falling back on English? How often do you try to communicate with German monolinguals? Is your main problem comprehension or production? If you've been living in a German-speaking country for 5 years, your passive knowledge of German is probably better than you think. The heart of a language is in its words more than its grammar, but there are just a lot of them.

[4ad]

> Acquiring languages isn't about going through books and studying [...] People acquire languages, they don't study them.

I know. Absolutely agree.

> There's a pretty good popular-science book on the topic, Becoming Fluent: How Cognitive Science Can Help Adults Learn a Foreign Language, by Richard Roberts and Roger Kreuk (MIT Press, 2015).

Thanks, I will check it out.

> Is your main problem comprehension or production?

I think my main problem is production.

> If you've been living in a German-speaking country for 5 years, your passive knowledge of German is probably better than you think.

You might be right, I know a lot of words. When it comes to cooking stuff, sometimes I know words my german-speaking friends don't know (I like to cook and buy lots of less-than-usual ingredients).

But I can't create meaningful sentences.

[wolfgke]

> The heart of a language is in its words more than its grammar, but there are just a lot of them.

As a native German speaker I would say that grammar plays a much larger role in German than in English and I personally cannot even imagine how one can learn German without grasping the grammar (I wrote more about that in https://news.ycombinator.com/item?id=12328906 and https://news.ycombinator.com/item?id=12243298). So my recommendation is (as I wrote in https://news.ycombinator.com/item?id=12196101) to do brute training on the grammar until you are very confident that you know it inside out, since otherwise you will always have to stop in the middle of a sentence how this verb is conjugated or this adjective + noun is conjugated - not good for fluent conversation.

EDIT: I have read multiple times that many people have problems with learning German on Duolingo since there is too less focus on grammar. Doing this might work for English and perhaps for Spanish but surely not for German.

[soperj]

I sincerely doubt you spoke your first word at 3 months old, and there's no way in hell you were using coherent sentences at 6 months old. That being said. Have you tried Duolingo? It's free, and it's less than 10 minutes a day. So even if you don't learn a single german word (i've learnt enough in a year doing 10 minutes of german a day to teach my 3 year old a fair bit, and read basic kids books to him) at least you won't be so frustrated that you have to sleep for the rest of the day.

[4ad]

> Have you tried Duolingo? It's free, and it's less than 10 minutes a day.

Thanks, no, I haven't heard of this. I will try it for sure!

> I sincerely doubt you spoke your first word at 3 months old, and there's no way in hell you were using coherent sentences at 6 months old.

When my grandmother told me this, I called total bullshit. But then I asked around and all my relatives and neighbours corroborated this. Since this was so unusual, they were so impressed at the time that they could remember specific conversations we supposedly had. Yes, parents and auts exagerate these things, but random neighbours?

I still found this hard to believe, but there are documented cases about people doing this[1].

I was a very gifted child, I learned to read somewhere around two years of age by studying the book while my mother or grandmother read to me. I also learned arithmetic by watching people use cash at the store. Numbers were so fascinating to me that somehow my parents got me a cash register to play with.

These things I remember clearly. Nobody thought me these things, I figured them out myself.

I was fascinated with money. I had a bank (actually my cash register) where my mother would put all the change. I charged a fee for both deposit and withdrawl, but I paid interest too. I cut all the coupons from newspapers and magazines, and demanded that my parents use them. All the savings from the coupons went to me, and from that revenue stream I payed my parents compound interest.

I also had a store where every product cost $1. We would go to the supermarket, and buy all kinds of household stuff that then I resold to my parents for real cash. I kept very good records for my bank and for my store. It felt very important to me that every transaction be recorded. I always gave to the customers carbon copies, so they had a record too.

Maybe I should start a business instead of doing this computer nonsense.

Unfortunately my grandmother thought I had autism rather than exceptional intelligence, and dragged me from psychologist to psychologist, who gave me all kinds of tests. They started with IQ tests (always scored over 160), but once they found I was gifted, they put me do all other kind of tests relevant to their current research. My grandmother was not happy they didn't really treat me, so she sent me to various psychiatrists instead.

My family pretended we were really, really poor, even though we were not, and didn't buy me the books and computers I needed. They sent me to a school full of stupid, violent, and quite literally criminal children.

By age 8 or 9 I have mastered feynman lectures of physics and I was studying particle physics. But I was slowed down by the lack of books and learning material.

I hated school. What a waste of my life. I stumble upon all these child prodigies who finished college at 12 while I had to endure that stupid school and lacked the most basic education appropiate for myself.

Remembering all this made me feel depressed, so I will stop now. There was a point I was trying to make (how I learn everything by myself, inductive and top-down, and how it's hard to learn bottom-up), but it's too difficult to write all this.

[1] https://en.wikipedia.org/wiki/Michael_Kearney

[xapata]

Why do you want to converse with people in German, rather than English? If you don't have a satisfactory answer to that question, you probably won't be able to speak German. "Because I think I should" is not motivating.

People learn languages best by simply listening to others talk. Hang out with folks that are speaking German. Tell them you want to hear no English at all, all night. Do this for a few weeks, maybe months, and you'll be speaking German soon.

Edit: Maybe I should add that you should focus on comprehension before speaking. Try to simply understand what other people are saying, first. While at an intermediate level, you'll find that you can understand a friend well, but a stranger almost not at all. It'll take some time listening to a variety of people before you can understand many speaking patterns.

[tragomaskhalos]

There is the theory that areas of our brain that were used for language acquisition as young children are "repurposed" as we age. From an evolutionary perspective, it is advantageous to commit a lot of brainpower to acquiring language initially, but thereafter this is not a particularly critical skill (esp. in traditional small monolingual communities). So I'd suggest that in your case this process has just been carried out to an extreme degree.

[mikekchar]

I also had 13 years of french instruction and even lived with a francophone girlfriend for several years. I never learned to speak french. On the other hand, I have learned Japanese as an adult without taking any courses.

There are 2 main things to understand. First is that it takes a very long time to learn a language. Many people completely underestimate the task. From the age of 3 years old, humans acquire about 1000 word "families" per year (police, policeman, police truck, policing would all be one word "family". Conjugation (even irregular) is counted in a word family too). This happens up until they are 20. Most university educated adults have more than 25,000 word families in their vocabulary. Even at 5 the average 5 year old has 5,000 word families in their vocabulary.

While you may have been an early developer with language, I think it is likely that your impression of your early abilities is hampering your current development -- mainly because it sets unreasonable expectations. At 6 months old, you may have had a vocabulary of 500-1000 words (it is not unheard of). Most children usually have a vocabulary of 1500 word families by the age of 3 and 5000 by the age of 5.

I recommend talking to 3 year olds and 5 year olds. Their command of their native language is truly awful. And yet, instructors of various languages imply (and sometimes outright say) that their course with 500 words of vocabulary (and a smattering of grammar) will allow you to "learn" the language. Adult level fluency is somewhere on the order of 15,000 word families. Even advanced learners with 7-10K words of vocabulary often can not understand movies without subtitles!

The main problem that people tend to have is that there is a huge disconnect between what they expect to accomplish in a short time frame and what they actually can accomplish. People bandy around ideas like "You can have adult like fluency with 2000 words of vocabulary", but it is absolutely wrong.

Basically, if you learn 20 new words a day (which is not unreasonable with spaced repetition software), you can get to 15K words in 3-5 years but you have to study every single day without a break. More likely is that you can reach that kind of level in about the same time a child can -- 15 years.

Second issue is grammar. There are only about 1500 grammar rules in a typical language. You can rip through them in much less than a year. At this point you will have a fascinating understanding of the language and still be unable to order a drink at a fast food restaurant.

Language acquisition is different than learning (I think this is now the accepted theory, but it's actually a fairly recent development -- in the last 30 years or so). Originally people thought that you memorised language and then practiced it to get faster and faster. Current theory suggests that acquired language springs to mind without pre-planning (we have an associative memory after all). When you are in the correct context (i.e. you want to say something), the correct thing just pops in to your head. Similarly, when you listen to something, understanding occurs without logical analysis. If you have learned language, you are essentially using look up tables and logic to divine the meaning. If you have acquired language, then you go straight from language to meaning (or vice versa).

Studying grammar helps you learn grammar, but does not necessarily help with acquisition. The current accepted theory is that repeated exposure to language that you can understand (usually from context) leads to acquisition.

So what are the practical tips? First, having a course 2-3 times a week will lead to exactly the result you have described: you will not learn the language. You must study every single day (well, my experience has been that 5 days a week will still lead to very slow progress, 4 days will keep you at a standstill and anything less than that will be a treadmill of instantly forgetting what you learned).

Study does not need to be the kind of study you are used to. The absolute best way to learn a language (according to the literature) is free reading. Pick up something you want to read and read it. It should be level appropriate... which is a pain for any language other than English (for which there are about a billion graded readers).

I have found an interesting technique (which I did independently, but later found that it is a very common technique ;-) ). Pick up any book/magazine/whatever that you want to read. Skim it and write down any word you don't know. When you get to 20, look them all up in the dictionary. Use whatever technique you like (spaced repetition software is awesome) to memorise the definitions of the words. Once they are memorised, read the passage again. If you still can't understand it, look up any potential grammar issues in a grammar dictionary. Write down the sentence with the grammar you want to learn along with your own translation of the sentence (after you puzzle it out). Memorise that sentence. Rinse and repeat.

If you are the kind of person who doesn't mind rote work, you can simply memorise all the example sentences in a grammar dictionary (or text book -- although text books suck for the most part). But you still need to do free reading to encounter it in the wild.

For listening, watch TV (or video web sites). For speaking, I have 2 techniques. Read and memorise songs for any genre of music that you like using the technique above. Get recordings of those songs. Sing along. The other technique is to find a news web site that has video news stories as well as a printed version of the story. Read the story. Listen to the story. Then try to read the story at the same time as the news reporter. Try to match pacing, intonation, etc exactly. You can record your voice and compare with the original.

Finally, you need conversation partners. If you live in a large city, you will almost certainly be able to find language meetups in your area. Go and chat with people. Otherwise, there are language exchange websites where you use skype or whatever to chat to people. Being a fluent English speaker will guarantee that you can find as many partners are you like.

Hope that helps! I've had success with these techniques myself, and I have taught them to my students when I was teaching English as a foreign language. I find them very effective. Just temper your expectations and never take a day off. Also realise that courses can never teach you a language because the language is too huge. They best they can do is teach you techniques that will allow you to learn a language. If you are taking a course and it is all about the content, find another course.

[4ad]

Thank you very much for the extremely thoughtful and detailed response! It is really appreciated.

[ktRolster]

But I can't speak a single word of French.

Seriously? Not a single word? 0_O

[4ad]

> Seriously? Not a single word? 0_O

No, I can't speak a single word. I try to remember words, but it's just nothing there. I can't think of a single word.

If you show me some simple words, I can probably remember what some of them mean, but I can't retrieve any word from my mind. It's just like the sensation where you want to say something, but can't remember a word, and you try real hard to remember that word. Except that's all I feel, only that sensation. There are no words I can remember.

Of course now I opened a french website, and tried to read it. Yes, I recognise some words, and there is even a minimal level of understanding. But I could not think of those words before reading the text.

[brians]

Oui

[andreasvc]

The article does end with three references to books.

[panglott]

Yea, I was reading through the article and wondering why they didn't mention actual researchers. Where's the discussion of Tomasello's work, for example? Then down at the bottom, I saw:

More to Explore

Constructing a Language: A Usage-Based Theory of Language Acquisition. Michael Tomasello. Harvard University Press, 2003.

Constructions at Work: The Nature of Generalization in Language. Adele Goldberg. Oxford University Press, 2006.

Language, Usage and Cognition. Joan Bybee. Cambridge University Press, 2010.

Some of the usual suspects. Further discussion of these would have improved the article immensely. I generally agree with the article, but it has a lot of nits to pick—it mischaracterizes the debate over Pirahã and recursion, for example, and significant evidence against the pro-drop parameter actually came from Western European languages.

[MawNicker]

The ruffling seems to come from the term: universal grammar. Without a proper understanding the term can cause all sorts of misapprehensions. It sounds like a mandate; it makes people rebellious. Noam Chomsky specifically disavows any evolutionary theory. He just hand waves it for the establishment of the model. Many people who "refute" him simply attempt to provide this. All he intended to build was the simplest concievable system for language parsing. That's all he ever claims to have. When he responds to criticism it is always staunchly from this context. People just don't care to understand the propositions of the actual idea. I suspect they're intent on invalidating him to undermine his political views.

[pjlegato]

Many -- perhaps even most -- of his fiercest opponents within linguistics agree strongly with most of his political views. George Lakoff, for example, is well known as a leading anti-Chomskyite linguist, and also as a vocal far-leftist whose political views largely echo Chomsky's.

[dilemma]

I'm not so sure that it's correct to call Chomsky a leftist. He clearly is not a Marxist and finds the ideology ridiculous, at the same level as religion. If Lakoff is a Marxist or socialist, then the two clearly do not have political views that are compatible.

Chomsky is, I believe, best referred to as a social libertarian, which means that he opposes both state and enterprise power and see them as more or less the same, and instead favors the individual. It is clear here why Marxists don't like him, at all.

I think this is also why the authors of the OA challenge Chomsky without offering anything other than minor claims hidden away throughout the text that language is a product of social interaction rather than a capability that lies within the individual. It is about the collective versus the individual and as a fierce individualist, the authors disagree with Chomsky on political grounds, not scientific.

[griffinmahon]

He's probably better described as an anarcho-syndicalist (I don't think "social libertarian" does the radicalness of his politics justice): https://en.wikipedia.org/wiki/Anarcho-syndicalism#Noam_Choms....

[pjlegato]

Not all leftists are Marxists, though they have come to dominate the field. The broad anti-capitalism of anarcho-syndicalism, combined with his well known criticisms of American and western power for so called imperialism, definitely slot him into the "leftist" category in my political schema.

[praptak]

> Not sure why linguistics ruffles some people's feathers.

My hypothesis is that it's Chomsky personally who ruffles people's feathers. And this is not because his work on linguistics but rather his political views.

[mattkrause]

I really doubt it's his politics. George Layoff and Chomsky had (have?) a protracted, acrimonious battle over the relationship between syntax and semantics, but they're not miles apart, politically.

I find Chomsky frustrating because he comes up with these incredibly elegant theoretical machinery--and then is totally disinterested in empirically testing then.

For example, the "Language Acquisition Device" absolutely HAS to be the brain, but a lot of the discussion of it--and Universal Grammar--are totally untethered by psychology and neuroscience.

For a while, it seemed like he regarded recursion as a key aspect of human language, versus animal "communications." Fitch, Hauser, and Chomsky had a 2002 article in Science that seemed to argue as much, and they were excited to show that humans, but not monkeys, can learn center-embedded recursion patterns[1]. However, Chomsky seems to have subtly backed away from that and now says that recursion important but might not manifest itself in all human languages (or something along those lines).

[parenthephobia]

That somebody genuinely disagrees with Chomsky's linguistic theories on their own merit doesn't mean that other people don't want to prove his linguistic theories wrong because they disagree with his politics. Both sorts of people can exist.

[mattkrause]

Well sure. Unicorns could exist too.

George Lakoff is one of his fiercest critics and I have a really hard time believing that he is motivated by Chomsky's political views; Lakoff is involved with various progressive and socialist think tanks himself. I'm not sure about (e.g.) Postal, Ross, or McCawley but I've never heard anything about their political animosity to Chomsky either.

If one wanted to go after Chomsky's political views, falsifying Universal Grammar (whatever that means) seems like an oddly difficult and indirect way to do so. Do you really think that someone decided that a deep dive into recursive structures in Pirahã is the way to push for TPPA? I'm totally willing to believe that more conservative linguists might get a tiny bit more of a thrill out of needling him, but again, this seems really unlikely to be a driving force.

If there's any political force that drives people to go after Chomsky, it's the one driving the academic job market. He is A Big Name, and conclusively debunking his theories would set the debunker's career on the fast track. In fact, even a high profile debate with him, assuming it wasn't a complete rout, would probably benefit most careers.

[mannykannot]

I get the impression that it goes beyond a disinterest in empirical testing; he (and his acolytes) seem to be unreasonably hostile towards anyone who dares to question his elegant theories - or point out that his position has changed.

[ScottBurson]

It's "Lakoff", but I suspect you typed it correctly and the spelling corrector "fixed" it for you.

[mattkrause]

Yup--autocorrecto. Perhaps Chomsky was right about the probability of a sentence being meaningless after all. /s

[ktRolster]

Chomsky isn't afraid to pick a fight.......whether it's with BF Skinner, or George Lakoff, he hasn't backed down.

[metaphorm]

Chomsky's politics have gotten him nearly blacklisted by mainstream media sources. They have not impacted his academic career in any meaningful way that I can detect. In fact, they have almost certainly amplified his fame and notoriety amongst academic circles, including in Linguistics, which has similar politics to most of the rest of academia i.e. it is extremely left wing.

as for the man's personality and interpersonal relationships, he is quite calm, even-tempered, and respectful. he is not known for ad hominem attacks or public meltdowns. quite the opposite. he's a bit dry and boring to be honest.

[andreasvc]

He's hugely influential in the field; arguably dominant in a way that is unhealthy for a field. Acolytes will say he's influential because of merit, detractors will say it is because of rhetorical skill, sophistry. I hope there will come a time when press releases no longer breathlessly announce Chomsky's theories falsified for the umpteenth time, but rather actual progress.

[mark_l_watson]

Chomsky nails politics and 'how the world works.' It is easy to understand why he is censored out of the news media. Not to go off on a tangent, but a brief opinion: I have made it a hobby for two decades comparing the coverage of specific news stories in different countries. I have noticed the censorship in the USA has become a much worse problem in the last few years.

[searine]

>Chomsky nails politics and 'how the world works.'

Yeah, tell that to Cambodia.

[mark_l_watson]

I don't know about Cambodia and Chomsky. Link please (thanks).

[bzbarsky]

https://en.wikipedia.org/wiki/Cambodian_genocide_denial#Chom... is the first search engine hit on "chomsky cambodia".

[mark_l_watson]

Thanks, I read that. Perhaps I should have said Chomsky usually gets it right. Middle east, encroaching corporatism, etc.

[staticautomatic]

Also, some Balkan countries.

[jesuslop]

Some of the context of this can be found in the 'Poverty of stimulus' wikipedia entry. Chomsky argument was tried to be cast in theorem form in a paper of E. M. Gold to apparently settle the question (in favour of Chomsky), but despite that result the computational linguists have been busy making useful work in NLP with machine learning methods, and now the debate has shifted from armchair discussion to the question of the interest of the emerging new language technologies and the role of unsupervised learning. Less famous, Zellig Harris (Chomsky thesis advisor), is now cited in deep learning language modelling literature as a precursor of the distributional model pushed by Bengio that is now used in word2vec and what not nlp.

[chillacy]

I hope this isn't the first time you're hearing that UG is on its way out.. articles like this pop up quite frequently.