For polyglots, there's something special about their native tongue | Hacker News Reader

For polyglots, there's something special about their native tongue

news.mit.edu

91 pointsnetfortius74 comments

In my observation of non-native speakers working & living in the US (a group to which I belong), there's also something special about number-words. If you are a non-native speaker, even if you boast near-perfect English pronunciation and fluency, and are accustomed to using English in all aspects of your life, not just for communication but for e.g. personal note-taking, and even if you haven't spoken/written your native language with anyone in months, the moment you need to count (a sizable enough collection), or you need to enter on your computer a numerical passcode received on your smartphone, you (most likely) find yourself reverting (in your (sub)vocaliztion) to the number-words of your mother tongue, because doing the same thing in English just feels so much more cumbersome.

Does it have to do with syllables?

I use English mostly (and am certainly a native English speaker), but use my mother tongue to memorize numbers, as mine (Cantonese) has single syllable numbers. I think I can keep 5-7 numbers in short-term memory in english, whereas I think it's closer to 10 or 12 in Cantonese.

I don't think so. Neither my native nor adopted tongues are English. My native tongue actually has more syllables than the adopted one, for some lowish numbers. Yet, I still count in my native tongue, e.g., when counting the reps while lifting weights.

I'd say I speak better my adopted tongue than my native one. People from my country say I speak with an accent, while locals are surprised to learn I'm a foreigner, since I speak with no accent. When I think about things, I also use my adopted language (I went to school here, so it's easier, since I've learned the concepts in this language). Hell, I sometimes use English for my internal thinking. Just not when counting or adjacent tasks.

While I don't belong to the same group, I've been working remotely for a bit over 2 decades now (not continuously at first, but yes for the last 15 years) so on weekdays I do end up speaking more in English than in Spanish (my native language) with other people, and I definitely have the same issues you mention with number-words.

I'm not sure if it's related but a similar thing happens with spelling. I guess because there's some ordinal relation to the letters in a word? But I feel whenever I need to spell my name in English it takes me about the same mental effort as giving my phone number, and I'm otherwise rather fluent in that language.

Sample size of 1, but I don't think in my mother-tongue anymore, and also don't count numbers in my mother-tongue and find them harder to recall at this point. But I don't speak/listen to it anymore, so that probably influences it.

(as a native English speaker who has spent some time in India) I have such a hard time thinking of numbers in terms of Lakh and Crore, I find even if I can remember the number of decimal places represented by each one, I almost have to visualize the number and move the commas in my mind to be able to make sense of them.

Similarly the only knowledge I've pretty much entirely forgotten about Hindi is the way the digits are written. Even going years without reading or writing in Hindi, the muscle memory comes back pretty much immediately for everything except for the digits.

Technically Hindi would be my mother tongue, but I wonder how all that plays out in terms of having initially been raised with it, then having entirely forgotten it and replaced with native level English for a few years, and then having relearned Hindi from scratch. I tend to think in a mix of both languages, and have a tendency to mix words from other languages into my thoughts too if I'm exposed to them often, even if I can barely speak those languages.

Chinese also splits numbers differently than English. In Chinese it's every 4 digits, instead of every 3 digits as in English. There is a word for 1e4 (萬), 1e5 is 10萬, 1e6 is 100萬, 1e7 is 1000萬, and there is a word again for 1e8 (億).

same for japan. ju hyaku sen man

I have been learning English for close to 18 years by now.

I still can't count in English.

Interesting. There's something about this.

Two days ago I was jump roping. Sometimes I was (mentally) counting in English, sometimes in my native language. I'm not sure which one I did more often, but when I paid attention, I noticed that I made fewer mistakes when counting in my native language. An explanation would be that my brain can do that more effortlessly. Despite the fact that I lived in the US for longer than I lived in my native country.

I am a non-native speaker living in Australia, I have been here for 12 years now; my experience is that this eventually goes away. Until about 5 years ago I was experiencing the same as what you are describing but now it no longer happens. I suspect that the timing depends on one's age, for example I moved here in my 30's and my progress was much slower than my daughter who moved here very young. Another factor is the relationship between your native language and your acquired one, for example I have observed that people whose native tongue is an Indo-European language don't have as much of a hard time switching to English compared to people from other linguistic backgrounds.

I've encountered that as well. Someone at work tells me a number of something to check in English, I have to write it down. I can't even see the number in my head. It's the complete opposite if someone tells me say a phone number in Polish. I lived less than a year outside of the US since October 1981.

What was emigrating from Poland in 1981 like?

Like the "Coffee, coffee, coffee" scene from "Moscow on the Hudson", I guess. And also very lonely with all their friends and family left behind.

I don't want to be off topic so in short you come-up with a plausible reason for travel to then Yugoslavia which takes you by rail through Vienna and spend time at Traiskirchen.

https://en.wikipedia.org/wiki/Traiskirchen#Refugee_Camp_(Bun...

Hijacking planes was another popular route.

https://en.m.wikipedia.org/wiki/LOT_Polish_Airlines_Flight_1...

Incredible - they guy got his freedom. These stories are as inspiring as they are tragic.

Unbelievable. Glad you made it.

My impression from having asked people about this exact problem—I meet plenty of recent immigrants here in Paris—is this goes away quickly if you have to, say, work a customer-facing job where handling money is part of your duties, but much slower otherwise.

(Of course, you can also simulate that experience by drilling this for ten minutes a day until it becomes second nature.)

Not really though in my personal experience. You only need to learn conventions about how to say currency values out loud because you enter the numbers at a register and it presents you with a total that you read out like, "That'll be twenty-one oh seven," or, "twenty one dollars and seven cents please." All the money handling and change making you do in numbers in your head.

This! Internally it's still not-English, while the result can fluently and easily be said in English.

I do find however that it also depends on "when the number was remembered". For example, relatives phone numbers I remember in my head in non-English, as I learned them in that language. But my credit card number I recite in my head in English in order to write it out on order forms.

And numbers for calculating and such for a very long time was still non-English but I think it's habit. In my head, nobody tells me "Sorry dude, what language is that? I can't understand a word" and so there's never any real incentive. It's also faster to do that part in your first language and only say the result out loud while thinking about words (especially entire sentences) in your first language and then translating and saying it in English makes for awkwardly long pauses or really bad sentence structure. I think that provides again more incentive to learn English sentence structure and to think in English to begin with, so that words come out "right away" as you think and formulate.

I live in a non-english speaking country and sometimes go days without english at all even though its my native tounge.

I agree, the one thing I tend to still do literal translations in my head for is for numbers and arithmetic. Its just so much more cumbersome mentally for some reason to do the math in my non-native language.

Linguist John McWhorter has a fantastic course on human language on The Great Courses / Wondrium / Audible.

He specifically talks about this! Apparently the spoken word, the written word, song lyrics, and numbers are all processed in different parts of the brain almost as if they were different languages. He even mentions the same phenomenon as you and I think he also said this is the first part that gets hardwired into your brain as a kid, leaving you stuck counting in your mother tongue forever.

As a native English speaker, I speak primarily Spanish in my day to day life (Spanish speaking wife and children) and have done so for about 15 years. I’ve had to purposefully build relationships in -both- languages in order to feel normal when speaking English with them.

I can’t even describe the absolute weirdness of saying “I love you” to my wife after building our relationship over years in only. Spanish. There is definitely some kind of personality disconnect when I learn language and culture by immersion as an adult.

And yes, numbers are problematic. Spanish me really, really, really sucks at math. Even trying to do basic subtraction or calculate a trivial tip ends up with internal translations to work out.

As a bizarre aside, Once in a while I can be nudged into a temporary unresponsive state by the right timing of a language change by my speaking partner.

When I “snap out of it” it’s like the words went into a que but nobody fetched them so I went into some kind of waiting condition. It’s really weird and makes me wonder if there might not be a lot of ways to trigger altered states through similar hacks.

I'm Dutch, my first language was Dutch, my second language was German and my third language was English.

I count and do all my numbers in English, UNLESS I am in The Netherlands and I have just been talking Dutch then I do it in Dutch. I do not switch to Dutch even internally for my internal monologue when I encounter numbers.

If you think English counting is bad you should see French!

Numbers in many languages tend to be harder to internalize, yes. Numbers in Danish are a whole other level, thou.

Consider e.g. the number 157. That'd be 100 + 7 + ((-0,5 + 3) x 20), only that they don't pronounce the "x 20" because then it is too long.

For me it's the ten hundreds or similar way to call the thousands. It's something that I have to picture the number in my mind and then translate it to the equivalent in thousands.

Eh. Once I start dreaming and thinking in a non-native language, the only things I revert to native on are things like philosophy where extra cognitive load actively hinders my thoughts. Numbers/money aren’t really expensive in the same way.

It might depend on whether you do a lot of counting in your non-native tongue. Since I studied maths in English (not my native language), I also do counting in English when I'm thinking in English because to me counting in English is just as easy as in my mother tongue.

This happens to me not in my native language but in the language in which I learned mathematics.

This makes sense. I have fairly good L2 German production, but I go from conversant to barely coherent when fatigued. My L1 English production diminishes, but not nearly as much. Likewise, my friends who speak English as an L2, even those who have spoken it with a strong command for over a decade, will struggle when very tired. Though we usually speak in English (her English is way better than my German), I've had a German friend ask me to switch a few times when she was very stressed or tired. She speaks at least five languages, with English being her most fluent after German.

That said, I'm most interested in the follow-up research mentioned at the end of the article.

> They also plan to study people who learned one language from infancy but moved to the United States at a very young age and began speaking English as their dominant language, while becoming less proficient in their native language, to help disentangle the effects of proficiency versus age of acquisition on brain responses.

Yeah, I have similar experiences. When fatigued, my English speech stops being as fluid and my stuttering intensifies (akin to higher stress levels). I've also observed that in those cases, mixing language becomes much more common, incorporating words from my native language in English, and vice-versa too.

Regarding the follow-up...

I grew up speaking only Russian. I moved to the US when I was 15. I had 7 years of deliberate English practice prior to the move and became fluid in it within a year of moving.

That was about 30 years ago. Because I moved alone, and didn't seek out Russian speakers for this 30 year duration, I only occasionally spoke Russian. The result is that now my Russian feels like a second language. When speaking it, I constantly struggle to translate from English. It doesn't feel like my Russian comprehension diminished, but speaking it is a struggle.

The shift to English being the primary language happened in stages. The last noticeable change for me was when I would need to count something. For a decade, maybe longer, I would default to Russian in my head when counting. The change was gradual, but now it's English all the way.

Fun story... During the first year after the move, I remember being freaked out when I had my first dream where I spoke English.

I learned English around 5-6, and what’s weird to me is I have vivid memories from before that where I’m speaking English but clearly wasn’t.

Memory is a funny thing - much of what we "remember" is created on the fly, based on aggregates. I also have a clear memory from around this age that is impossible.

Yeah, this is something that I'd love to see research on, though I imagine it might be difficult to find enough people to act as subjects.

That first dream in a target language is always so weird! It's one of those things people say happens when a language first "clicks" for you, but it always feels momentous, even if it happens earlier than expected.

> became fluid in it

became fluent in it

I think you want this ^^

When I was in 6th grade I had a classmate move from Canada (speaking French) to the US, and after a couple of years speaking English he said that he started to lose his French speaking pretty quickly.

Touche... As I was typing that up, I knew I would screw up somehow trying to demonstrate proficiency. :-) Leaving as is.

> It doesn't feel like my Russian comprehension diminished, but speaking it is a struggle.

I didn't know you could 'lose' a native language that you grew up with til near adulthood. Maybe speaking has more to do with muscle memory and hence easier to lose than a somewhat passive skill like listening and comprehension. Makes me wonder how language is stored and processed in the brain. Are there two complete separate regions for speaking and listening? That feels a bit inefficient.

Language attrition: what we don't practice, we forget. Even the mother tongue.

What is L1, L2 Proficiency?

L1 is first language, L2 is second, etc. Except for L1, which implies native proficiency, it doesn't say much about the level, just the order in which they are learned.

I think, the European A1, A2, B1.. is better way for descriping language proficency

If you learn any language as an adult, chances are you're never going to become as fluent as a native speaker, no matter how long you live in a second language environment. I say that as someone who has done it myself, and seen a lot of friends and family going through the same process.

Children are a different story. They can learn a language and start speaking just like native speakers, local accent and all, within months. It's absolutely amazing to see. My niece went to Australia when she was 5. She could speak Portuguese, our native language, as a native, obviously, and was very afraid to learn English. A few months later she was fluent in English and had the local accent from basically day 1 :D. I never absorbed the accent myself (moved to Australia as a 23 yo, over 20 years ago). She is 11 now and she now mostly speaks English, and it takes effort for her to answer to us in Portuguese when we try to get her to do so. Her dad, my brother, studied for years (even before moving) to try to speak somewhat fluent and correct English... but even now, more than 5 years later, he struggles. His wife barely speaks anything.

When I was new in the country, I could see that the younger the person, the faster and the easier it was to learn. No child I've seen ever had any difficulty learning a new language from scratch... they really are amazing at it!

I have been learning Swedish as I've moved to Sweden years ago, and I am in the same position as my brother is in Australia: I can only speak basic Swedish and I make mistakes and have to think a lot to say things coherently, specially without having to translate from English first - I've spoken English so long that it's my main language now... but I still understand Portuguese with less effort than English (and after just a few weeks speaking mostly Portuguese, as I do on some of my vacations, I become much better at speaking it again and words stop coming in English first, but that depends on the topic as well - if it's a topic I only ever speak about in English, like work, I find it horrible to speak about that in Portuguese!).

To think some people are able to learn 50 language (as mentioned in the post) blows my mind. I don't think I will ever speak my third language properly, let alone learn another dozen languages in the future :D

But this is another observation I've made over the years: a minority of people are very talented at languages and can indeed learn much more easily than others... though as far as I know, only children can learn within months and with perfect pronunciation.

Sorry to be a curmudgeon but I get tired of all these claims about how well young children speak second languages.

A young child saying "Vanilla ice cream please!" often sounds more like "Ganiwa ishe! cweam! pweese!".

A typical 4-year-old speaking English substitutes many sounds for another because they developmentally aren't able to make the right sounds yet. Their intonation is also all over the place.

If an adult had their accent it would be barely intelligible (and in fact many small children are hard to understand by people who don't spend a lot of time with them).

Here's one speech sound development guide for English: https://childdevelopment.com.au/resources/child-development-...

It is amazing to watch how fast small children grow and develop but we also (rightfully) grade them on a curve. It is when we compare them to adults that somehow we forget the curve being applied.

Just a thought from someone who speaks English and a smattering of Spanish.

If the child can correctly express the idea/intent they want and be understood then arguably they have learned that part of the language. Sure it might not sound how an adult would be likewise they probably don’t in their native language either right?

I doubt anyone is saying that kids learn to speak the language perfectly but clearly there is a pattern where by they are able to understand and express themselves faster than adults - even if they sound childish still.

The person you're responding to said that children are graded on a curve. Your response is that children should be graded on a curve. Fine, but only being understood by your mother when near an ice cream shop is not the goal for adults learning a second language.

I would actually argue that if a child and an adult spent the same amount of time engaged in language learning and were graded on the same metrics the adult would get farther. I don't think there's a magical cutoff where we suddenly can't learn languages anymore.

Rather, I think it is partly because we grade them on a curve that children are said to be amazing linguists

Children have many other advantages that also help them: * More free time * Fewer internal distractions (Did I lock the door? What am I making for dinner? etc.) * Not being afraid to make mistakes

My children went to kindergarten (barnehage) in Norway when they were about three and a half and speaking English well. Well with a year they were indistinguishable from the native children.

Right, but the parent comment is comparing children to adults.

Even a native speaking 3.5 year old actually speaks a language very poorly compared to a native speaking adult.

I've observed my kiddo when he was a toddler and I think there might be an inflection point when they are very young and don't speak so well, when they might be able to learn any language with ease, precisely because they don't recognize let alone speak even their own so well. I recall my kid was watching TV and all and at some point while eating he pointed to a drawing on a cocoa box and said "monkey!" in English, and indeed it was a monkey. Must have picked it up from cartoons or something.

But then there's a point of no return when they learn enough native language from which point on any attempt at a foreign one becomes deliberate practice that requires effort. And this point of no return follows very early from the point when they start to speak.

My example above was of a 5 year old. At that age, she already spoke her own language fluently so that anyone understood her well (of course, they still have a smaller vocabulary at this age as they haven't been able to talk about all sorts of topics yet). Do you think a normal 5yo normally still speaks blurry?? Or a 6yo? At this age, and even up to perhaps 12yo, as far as I can tell, they're able to absorb another language quite easily, and it only becomes harder around age 14 or so... though of course I haven't run an experiment or anything. I expect your learning curve becomes worse gradually between the ages of 14 and 30, maybe very slowly up to mid 40's.

If a non-native English speaking 5 year old said "I goed to school today" people would say they are native level. (Even though they used the non-word "goed" as the past tense of "to go" instead of "went")

If a non-native English speaking 15 year old said "I goed to school today" people wouldn't say, "wow, they are a native speaker" nor would they say "wow, they are natively speaking like a 5-year-old" - they would say they still needed practice.

That's my point - speaking like a 5 year old as an adult wouldn't be considered native by anyone but a 5 year old. As you get older the standard is higher: speaking like a 10 year old is much harder than speaking like a 5 year old.

So for a 10 year old just learning a new language it may take longer than the 5 year old to get to their age level. That doesn't necessarily mean the 10 year old is worse at learning languages though since they have so much more to learn to be considered "native".

> there is something unique about the first language one acquires, which allows the brain to process it with minimal effort

I am only able to overhear normal conversations between other people in public in my native tongue. With any of the other languages I have to actively concentrate on listening. I have been living abroad for almost 10 years and I'm very proficient in the language (not English) so proficiency shouldn't be the issue.

> each of whom had at least some degree of proficiency in five or more languages

Maybe the study clarifies this but the conclusion is exactly as I would expect if you use such a loose definition of polyglot.

Why? Would you expect something different for people who speak more languages, or speak them more proficiently?

The paper abstract ( https://academic.oup.com/cercor/article/34/3/bhae049/7625489 ) does into a little more detail on their findings already.

I mean, "some degree of proficiency" means the person self-reported "at least minimal proficiency." I could achieve minimal proficiency through a few hours of language tapes.

I see your point.

I’m good at learning languages, and 100 hours of self-study would make me sound “fluent” to most Americans. That said, I would self-rate myself as “minimally proficient, mostly at short simple conversations”.

> Maybe the study clarifies this

They do:

“The mean self-reported proficiency for the highest-proficiency language was 20 out of the maximum score of 20 (SD = 0; note that for three participants, the highest-proficiency language was not their native language; Table S1), 18.3 for the second most proficient language (SD = 2.45, range: 10–20), 15.8 for the third most proficient language (SD = 3.36, range: 8–20), 13.3 for the fourth most proficient language (SD = 4.034, range: 7–20), 11.2 for the fifth most proficient language (SD = 4.23, range: 4–20), and—for individuals with proficiency in more than five languages—10.33 for the sixth most proficient language (SD = 3.99, range: 4–18).”

These were folks who self-reported (perhaps inaccurately, but still…) at a relatively high level of proficiency.

I'm not sure I buy this. According to their definition I'm a polyglot, but I do not experience this difference in mental effort when speaking or listening. I often make jokes crossing two or three different languages, it feels completely natural. My proficiency, however, isn't the same in all of them but that's mainly tied to frequency of use and not lack of knowledge.

I think when languages are learnt plays a critical factor in the measurements. If one learns three languages as a young child at the same time, chances are the level of effort is roughly equivalent when speaking or conversing, and the results will be quite distinct from someone else learning different languages at different points in their life.

If I miss in a pick up basketball game an easy layup I can curse to my hearts content in English but so what, now If say "Por un puto carajo!" now there is catharsis.

I truly wonder what dedication one would have to have in order to learn 54 languages. That is just the most impressive even if you know the language at a basic level. Quote from the article: "including one who spoke 54 languages with at least some proficiency."

Once you know Russian, Polish and Serbian - you know all Slavic languages basically. Maybe Albanian will give you some problems. That's like 15 for the price of 3.

There's probably similar shortcuts in other language families.

Albanian is not a Slavic language at all.

Meant Bulgarian/Macedonian, my bad.

I've noticed something else particular to my native tongue. It lacks emotion. I just don't really know how to express myself in English. I can speak very technically and logically, but I'm stumped when it comes to matters of the heart.

My Japanese, on the other hand, isn't fluent, but I'm able to tap into way more emotion. It feels as if I'm singing when I nail a great sentence in nihongo. It really lands with the listener and I feel emotionally connected in that moment.

Could this be related to the increased brain activity required for foreign language?

I think it might be that English isn’t very expressive. It’s much more technical and to the point. I find Greek and Spanish significantly more expressive when I speak them and particularly when I hear natives speak the languages.

I’d be curious what would happen with my kids who grew up with pretty much equal levels of Spanish and English at home, although right now, they’re mere bilinguals (my ex-wife as a trilingual (Spanish-English-French), is a bit of a laggard among her family where many speak four or more languages).

Presumably they’ve adjusted for actual use?

I grew up with German as first but find English (2nd) easier thanks to more use. And same for third lang (Afrikaans) - ease feels directly correlated to usage not order

I think perhaps both use and order have an effect on the cognitive effort one has to put in to speak in a language.

You probably started learning English at a fairly young age (less than 15?), which certainly helped you cement it as a "native" language. Yet, it is interesting that you report Afrikaans as being easier than German. When did you start learning it?

You're right - was at a young age. School enforced 3x lang minimum.

>Afrikaans as being easier than German

It definitely is. Most places quote hours to learn as substantially higher (~30%) for German and in practice it feels even higher. Afrikaans average conversational level feels simpler too.

>When did you start learning it?

German 1st and then English and Afrikaans bit later maybe like 10? School medium of instruction alternated German/English

I noticed an interesting phenomenon when I was learning my 3rd language. Whenever I was struggling to find a word in the newest language, I was more likely to confuse it with a word in my 2nd language than with my 1st. For some reason, my brain had assigned language #1 as "primary" and then all subsequent languages as "secondary" leading to sometimes substituting words from one to the other under duress.

i only know 4.

Reply on news.ycombinator.com