How multi-lingual children learn languages

I was born in to a multi-lingual environment, learning Dutch, Swedish and English since birth. Later in life I learned French and German due to my locality; I also took Latin and Greek in school and have been teaching myself Japanese just because I enjoy the challenge. Finally, I also have enough exposure to Spanish as well as Hebrew to be able to put together a decent conclusion as to what a conversation might be about. Each of these languages have been learnt at different parts of my life and I would say that the only one I am still fluent in speaking is English. My Dutch is rusty, but I can hold a conversation, though will stumble over words as I try to translate them (ironic since I used to think in Dutch and had the same issue translating what I wanted to say in to English). The interesting thing is, while I may not be fluent in speaking any of the other languages, I understand them. I can pick up on key words and can figure out what is being said. Due to the nature of languages and the etymology behind words - I can often pick up on conversations in other languages I have not learned.<p>In contrast of my multi-lingual background in the spoken form, I also try to diversify the programming languages I am familiar with. While the idea of a programming language and a spoken word language vastly differs, I still see it as just learning another language - that is the mindset I go in with. A spoken language has rules - wether they are related to phonetics, semantics or most importantly the idiosyncratic rules of gender and relationships. Programming languages have a much more defined ruleset at the core that they all share (to a certain degree) and the language itself on top of it follows many of the rules I would expect in a spoken language.<p>I don't know if there is a correlation to the exposure of languages at a young age - but in my experience it has helped me.

There isn't any citation to the (quite interesting) study finding mentioned as a premise in this article. My two older children are native speakers of BOTH English and Modern Standard Chinese, and they once spoke to each other mostly in Chinese, although in recent years they have much more occasion to use English. There should have been follow-up studies on what happens to native multilingual persons when they suffer strokes by now, and there could even be brain-imaging studies on first-language and second-language multilingual persons by now, perhaps without conclusive results.<p>This submission makes me curious about the author's premise than about his conclusion.

There are too many factors involved here.<p>For example, human languages are reinforced completely differently.<p>From birth, you will hear from inside the womb often one language. Later, you hear and speak either a couple or more languages but generally in different contexts, e.g. mostly one with parents, maybe one or more with friends or strangers and one with school and media.<p>The latter context is normally reinforced further with reading and writing, while the former rarely or not at all. I suspect, though without more studies it would be impossible to know, that it is the exclusively spoken languages that are most affected. The reading and writing languages, as well as being used in different contexts, are highly likely to be processed by the brain differently.<p>In addition, almost all of these human language uses will be tied to life events and (strong) emotional feelings.<p>So, the comparison to programming languages seems tenuous at best, especially since there are plenty of other contexts where one can misapply learned rule-sets in general. The word "language" is almost a misnomer across these two different fields, in my opinion.

Trying to extrapolate between learning natural languages and learning programming languages seems rather questionable to me.

The point about keeping languages separate is interesting. I know bits and pieces of many different languages, but I only know two reasonably well. I studied French for about six years in high school and college, but I haven't had much occasion to use it since and am somewhat rusty. Conversely, Japanese I hear quite a bit of these days so my experience with it is much more recent.<p>But sometimes when I try to remember the French word for something, I get a Japanese word instead, which is odd because the languages are nothing alike. Similar problems exist for programming languages, for example, trying to remember the Perl way to do something and getting the Ruby keyword instead.

<i>There are some languages that are so difficult that they have never been successfully learned as a foreign language and are only spoken by native speakers who started at birth.</i><p>I am enormously skeptical of this claim, and would like to see evidence for it - and it would have to show that people had made significant efforts to learn the language, had access to a level of resources similar to those available for other languages, etc. I suspect that the examples that would actually come up would be some dying tribal language where nobody has been able to learn it properly from the remaining three grandparents who speak it, which is in no way evidence for 'the language was simply too difficult'.

As a multilingual speaker (English, Mandarin, French, all fluent) who eventually let English take over as my main language here's what I think.<p>If you learned two languages since birth you develop two language sites. I read somewhere that the Wernicke's area is divided into two if you do this as well. Perhaps auditory memories are triggered based on language as well. You think and have internal dialogues in different languages which triggers completely different memories. This explains why one might lose one language, but keep the other one.<p>Whereas, for learning languages later in life, a person tends to translate the new language into the main language. The internal dialogue is always in the form of the main language. The memories are one contiguous experience based on one language. Often, when self examining during meditations, it is hard to make out what language is used. This explains why a brain problem will wipe out sections of speaking ability across all languages because they are all eventually translated from the main language.<p>Like people who started with C, we tend to think in C and see how new languages add on to C and just remembers the new rules of the new language relative to C instead of having a completely new life experience from ground up of learning the new programming language.<p>In short: Bane said: "I was born in darkness, you simply adapted"

Let me give you another tiny sample group for the study. Indians! I started learning Hindi, English and Bengali when I was four. Most of my fellow citizens(middle class) were in the same boat because there are 22 official languages here in India. After college I moved to South India, where there is another myriad family of completely different languages. I can barely keep up with them. Does it improve or harms my ability to learn new programming languages? Personally, no. Generally, well the answer to that question lies in the high quality programming services provided by Indians well recorded by thousands of forums across the web :). So, no I don't think the comparison has any merit. Edit: The no. of official languages is now 22.

I've always thought this subject was very interesting as when I was younger I lived in Germany and German was actually one of my first languages, I used it in school as well as with friends. However when we came back to the US there was simply no opportunity to use it so I have lost most of it. Same goes for Spanish that I used to use semi-regularly when I was in my teens.<p>Naturally when I joined college I started studying language classes, Japanese the first year and every year after, as well as concurrently learning Chinese starting the second, I also went to college in Japan for a year focused on upper level Japanese. The process was me religiously studying, watching tv shows, constantly writing essays on my thoughts about society, culture, news, Rubik's Cubes, artificial intelligence, etc. even blog daily. Also listen to music. I can't emphasize enough how much I'd write and from the beginning engage daily in only using Japanese and Chinese in conversation. It took a year and a half to be conversational in Japanese (largely due to the completely different grammar and kanji) and for Chinese only around a year. I have also recently started Korean, to find that the overlap of Chinese and Japanese make it almost trivial to learn. So language similarity is also important.<p>Also in my years of interacting with foreigners who are learning English. The best speakers are typically ones who came here younger, there is even a large difference to students who studied abroad as early as high school (versus college)<p>I can definitely contest that you can still learn a language to very technical competency as late as in college. It just takes a lot of effort and discipline, and perhaps a bit of accumulated skill. Finally, you certainly get better at learning languages with every new language as with learning any subject.

The difference between L1 and L2 language competency is deeper than whether you acquired multiple languages in "parallel" or "serially". The competency level, whether you are fluent or near-fluent, is based on the critical-learning period which ends roughly at the onset of puberty. Nothing rules out acquiring multiple L1s in serial, which I suspect to be the case in many instances of multilingual children.<p>Moreover, the author's anecdotal support is in contradiction with the fact that children do acquire multiple languages largely without problem. There are effects of one language on the other, but they are different in natural, and far less frequent, to crossover effects in L2 language learning, e.g. learning Spanish is high school.

Here's a thought from someone who's semi bi-lingual from a very young age. (Semi because I'm not completely fluent, but proficient.<p>Some people when speaking another language that's not native, will 'image' the english word when talking. Like, saying "ni hao" but in their mind it will be 'hello'.<p>However for people who've learned it from a young age, 'ni hao' is simply 'ni hao' and 'hello' is 'hello' but the meanings we recognize as the same. I don't remember the source to this, but take it as you will.

Stephen Pinker writes about this in <a href="http://en.wikipedia.org/wiki/The_Language_Instinct" rel="nofollow">http://en.wikipedia.org/wiki/The_Language_Instinct</a>

I suggest reading the review article "Bilingualism and the brain: myth and reality" published in Neurologia in 2010: <a href="http://www.elsevier.es/en/abstract/revistas/neurologia-295/bilingualism-and-the-brain-myth-and-reality-13156053-revisiones-2010" rel="nofollow">http://www.elsevier.es/en/abstract/revistas/neurologia-295/b...</a>

by attaching more than one label/tag to an inner representation?