Becoming an Amateur Polyglot
lesswrong.com
lesswrong.com
When I start learning new language, I first focus on speech. So just audio until my ears can pick things like intonation, tones, words, and I start to grasp the sentence structure. After that, get my reading to a basic level and learn the alphabet (if necessary). Only after that do I start with some formal study to build a solid foundation. Usually, some kind of textbook series that covers reading, writing speaking, and listening. After that, I try to immerse myself in a language and learn as necessary.
I have seen great counter examples of people that could partition their mental spaces and isolate the languages properly.
I had one Spanish teacher in Brazil that could do this. I’m sure she had proper training.
I also had a Spanish friend and her mother was complaining that she was forgetting her home language because she was living in Brazil for some time. This is the most common thing to happen.
Interestingly, as she was telling me this after visiting Spain for about a month, I was having more trouble understanding her than usual, it can go both ways.
I've said for a few years now my personality is slightly different when I speak English. I have a partner of several years that doesn't speak my language nor me hers and we communicate in English and even simple things like expressing affection feel different even after a decade.
At some point since I speak more English every day than my own language when I switched to "thinking in English" (not on purpose) I remember feeling very sad, having almost lost a part of me, and I spent some time trying to think what impact it was having on my own actions and behaviors.
Nothing major, but I am more assertive and outspoken in English than in my native language because things don't feel so "real" to say, so it's easier to not feel scared.
I feel very strange sharing this because when I tried I got strange looks but maybe someone recognizes themselves in this too.
Sometimes I translate in my head what I just said back to my native language and realise that chances are I wouldn't have said it if I had to actually say it in my native language.
It's weird. It's also not something I think about on a daily basis. It's simply how it works.
Could never put words on it but things don’t feel so real either.
The funniest part is now I speak my native language with my small kids and have fully relapsed to my original home town accent. I had abolished that accent years ago to avoid getting bullied at high school after relocating to a big city.
My kids are actively hearing and speaking 4 languages and one is with a thick accent!
Did you mean to say your partner doesn't speak your native language, or how does that work?
However: I am kind of a different person when I speak English. My personality changes. Not for the worse though (and that's where I don't agree with the "and not in a nicer way", that's not how it is for me).
The problem is that, as a non-native speaker, the words carry less emotional weight somehow than my native language. This is not specific to English, and it's not because of a lack of familiarity. It's most visible in swear words - I have no problem saying any swear word in English, while I still get slightly squeamish with certain swear words in my native language.
It's basically a part of how you can't really absorb certain cultural aspects of a language, even if you learn about them. They're never going to be a part of your own internal culture if you didn't grow up with them.
- study flashcards from l3 <-> l4 to cut down on time - practice switching between the languages and distinguishing between them (I use interleaved flashcards on phrasing) - have a really painless audio review (to get a lot of your reviews done quickly) - listen to input every now and then (tv, podcast, something). A little bit every now and then goes a long way
Like that Alan Perlis quote about programming languages.
In the ever-growing geo-restrictions world, this is no longer as easy as it was in the days of yore.
For example, if you're learning Norwegian and you want to watch anything on NRK, you need to get a BankId[1] - for which you need to be a resident _in_ Norway to have.
The alternative, of course, is buying DVDs/Blu-Rays on Amazon -- but you have to know what you want to watch and what is "good" before buying it.
It's the producers of the tv shows which need to license their product to a worldwide broadcaster. NRK is only georestricting because the licenses forbids them from not doing it.
I talk about this very often because I'm learning a language. To me the easiest and best options is to watch series produced by the streaming platforms. They own their worldwide rights so they can let anyone anywhere watch any show in any language.
I'm talking about Prime and Netflix originals. But the absolute best streaming platform to learn a language is Disney+. Every content on Disney+ is available in all languages it was translated too, from anywhere. It absolutely changed my life.
This is the same thing.
https://en.wikipedia.org/wiki/Distinction_without_a_differen...
But I agree with what you said, I just think it is an interesting topic and so the distinction matters to me more than it may to you.
Things like the above are infuriating to me as someone who appreciates the position content creators and licensees are in, I just don’t have much patience with the failure modes of copyright.
I would be more inclined to say that they're doing it to be "safe". For example, their immediate neighbour, Sweden, has a selection "Kan ses utomlands"[2] -- which, roughly, means can see outside of the country.
Presumably, like elsewhere in the world, not everything NRK is showing is licensed only for Norway.
To the original point: The issue isn't that BankId is being used, it's saying you can just watch the originals, in their mother-tongue (person-tongue?).
For example, NRK produces some originals, like Vikingane[3] that eventually became Norsemen[4] on Netflix.
Clearly, that's been licensed for semi-global (if not global) consumption via Netflix; however, as you can see from the NRK link, you still need to login via BankId to watch it in it's original form.
Again: None of this is a problem, per se, but saying that one can just watch original shows - in their original language - isn't congruent at all with reality, these days.
[2] - https://www.svt.se/kontakt/anvand-svt-play-utomlands
[3] - https://tv.nrk.no/serie/vikingane
[4] - https://www.netflix.com/ie/title/80180182
Edit: List formatting fix.
I think often fiction-type shows like Vikingane has complicated licensing as it is normally produced by some sort of third-party, but other programs like news and the mintioned gameshow is produced in-house and NRK has all the rights. There is a very nice page with an archive of old programs [2], I think most of this should be globally available.
I of course agree in general that geo-restrictions and localized distribution rights are stupid concepts.
[1]: https://tv.nrk.no/serie/nytt-paa-nytt-quiz [2]: https://tv.nrk.no/programmer/nrk-arkivet
For example, the most popular show I've heard of, Skram[1], requires a login -- but Storst av alt (another I've heard of) does not.
NRK doesn't buy world-wide licenses, even if the same show/program is available everywhere. They still can't distribute that. What they can distribute is what's 100% produced by NRK. If there are collaborators, they can't.
I totally agree with despising these restrictions, but isn't it likely that more total material is available worldwide than ever before?
My nephew was learning Norwegian for a while and we easily found Norwegian newspapers and magazines online to attempt to read together. It doesn't seem like finding such reading material would have been so quick and easy in the United States thirty or forty years ago.
Even for video, he readily found lots of material on YouTube that interested him, including lots of contemporary Norwegian music videos. That also would probably not have been straightforward to find a few decades ago!
https://tv.nrk.no/serie/dagsrevyen-21/202405/NNFA21051224/av...
I didn't bother looking further into it but I would expect lots of other programs to be available without BankId.
Most apps focus too much on impeccable grammar (even if they usually restrict you to a few tenses/moods) and rote, non-contextual vocabulary learning. Both are, well, useless. Having perfect grammar but not knowing when to use which verb in which context won't get you far. It's like trying to learn how to code in some programming language by studying the formal syntax and rote memorizing a bunch of core functions from the standard library. You miss out on everything that actually makes a language.
All in all, it feels like this post was written by someone who mainly communicates (or learns to communicate) in foreign in writing.
> Forget intonation if it’s not entirely necessary. Especially in French, there is no need to learn if a word takes a ` or ‘.
I can't let that one go, sorry. You don't need to learn perfect spelling, but it's going to be obvious most of the time if it's é or è when you hear the word. And if you pronounce it weird, you will get confused looks. In English, do you "need" to learn if a word takes "o" or "u"? People who were raised with certain native languages cannot even hear the difference anyway, so why bother, I guess? I hope this makes my point clear.
Having taken some French in high school, formally learning grammar and such seemed more useful than what the apps do. Of course, doesn’t make for high entertainment.
I feel the apps only try to sell the idea of learning a language, not the actual ability to do so.
Accents aigu et grave sound completely different, and stand in for a variety of disambiguating spellings you might use (including the grave even for non loanwords: 'bless-id' vs 'blesst', for example) you could never confuse the pronunciation, they serve the opposite purpose.
In fact that example makes it seem more like a point about spelling, don't worry about writing the wrong accent. (Yeah, people will figure it out I'm sure; in English too 'bless-ayed' is nonsense, so the presence of any accent would look like 'bless-id' was intended.)
What I've noticed with most French words is that initial-è seems to be phonotactically disfavored if not outright forbidden. A quick scan of my dictionary suggests the only word with initial-è is ès which is archaic. In general, I associate è with a sort of final position, using je préfère as the mnemonic (since it has both the accents in that conjugation).
This is like the c++ pointers guy complaining "how can this guy write so ugly PHP". I'm the PHP guy and I'm a fan of the way @levelsio is doing things in a simple and effective way that makes others annoyed. But in reality his projects work. And I chat with french friends :)
Besides, learning a language isn't about exams or certifications. It seemed to me that we have very different ideas of what learning to speak a language means, and your answers confirm it. That's fine, but I'm not sure what there is left to discuss, then.
English semi-famously remains comprehensible if you replace every vowel with a schwa. So sure you'll have an extremely weird accent, but this is acceptable according to the OP's definition of "speaking a language" from the first paragraph of the article.
Back before widespread internet, I spent two weeks in Korea, and by the end I could basically read & pronounce it (without really understanding anything) because it was that critical to getting around.
Here is how the method is generally used: https://m.youtube.com/watch?v=Ar04zvpoylU
With all the other language courses I’ve done it has always been tons of grammar lessons, some basic classroom discussion, and making vocab decks for something like Anki. With Assimil you’re immediately just inside the language, learning it almost passively. I was able to pick up easy native content books — like Camus in the case of French — after getting through the full Assimil course in a few months. Of course it’s still slow going and takes lots of effort, but the faster you can get to native content that is interesting to you the better I think.
During the pandemic when everyone was trying to learn languages they were popping off. Like thousands would be in servers and you could just chat with people whenever you wanted. I would spend every minute commuting, cleaning, or any time I would've usually listened to podcasts bullshitting with random people and learning how to actually speak.
I've tried italki I've tried other platforms where it's bumble for language learning. They didn't even come close to how quickly I learned there. And you can't beat free.
I joined one for Spanish and French. I'm sure they exist for others.
There is no future tense in Finnish. Or gendered personal pronouns. Oh and there’s about a gazillion words for snow. And god help you if you need to articulate that your spruce is returning.
As an English speaker myself it's astonishing the number and range of features that other languages have built in, so to speak. Years ago I took Latin and the idea of declension just blew my mind. And that's basic in the grand scheme.
It's also trickier with more unusual languages because there aren't resources out there to the same extent as for more widely-spoken languages. I don't think there is an 'official' Memrise course for Finnish for example.
i have used Anki, Supermemo, Flashcards, Watched Series, Read Books and the only thing that seems to work is:
- Spaced Repetition - Something you actually enjoy(learning should be fun) - Comprehensible Input
anything else has not worked for me.
{{c1::answer::hint}}
* https://docs.ankiweb.net/templates/fields.html#hint-fields
* https://docs.ankiweb.net/editing.html#cloze-deletion:~:text=...
Disclaimer: AnkiDroid maintainer
Besides, as a professional (computational) linguist, I have met many trained/professional linguists that know only one language - typically English - well - I think that is professionally totally unacceptable when your mission is to study language (in abstract, intentionally without article here => linguistics); the way a language structures the world for you in the way its vocab influences the way you think in concepts (Sapir-Whorf hypothesis, AKA linguistic relativity => https://en.wikipedia.org/wiki/Linguistic_relativity ) is something that you should at least occasionally step out of, so that you don't take it for granted, and for that, you need to master a second language.
You can even see it in this article :)
"Learn only 4 tenses. Past for one time events, past continuous, present, future. These are enough to go by and explain yourself."
Some lanuages don't even have 4 tenses. Polish has 3. And Past Continuous isn't a thing, we have aspects instead.
BTW I don't intend to criticize the author, they're far more of a polyglot than I am (I only speak 4 languages, and only 2 well). It's just funny how we take some things for granted.
Native speakers of slavic languages tend to believe that there are three tenses. Try asking that someone who learned it as a second language and you'll realize that native french/english speakers think that slavic languages have 6+ tenses to them. Aspekt dokonany i niedokonany basically doubles number of tenses.
I think a reasonable interpretation of the statement is "learn how to use verbs such that you can talk about doing something now, doing something in the future, having done something and continuing to do it, and having some something and since stopped doing it."
In every language humans are capable of expressing those distinct concepts. But boy, that was an awful lot wordier than the way the author put it.
I agree that knowing more languages can only help, but it's an awfully large time investment for that hypothetical benefit. Unless you're moving to a place or in love with a person that speaks it, learning a language seems pretty far down the list of time-efficient ways to improve your "professional and private" life. That's not to say people shouldn't pursue it, as hobbies go it seems like an edifying one, but unless you think it's fun, it doesn't seem very worth it.
As a trained (and formerly professional) linguist, I vehemently disagree.
(1) Linguistics, as you seem to allude to, is the study of language as a phenomenon, not the study of languages. It includes fields like the study of acoustics (the intersection of the physics of sound and the biology of the ear / vocal cords), and neurolinguistics (the neuroscience of language) that have nothing to do with sociolinguistics or second language acquisition. [0]
(2) The Sapir-Whorf hypothesis in its strong formulation is pseudoscience, and is rejected as such by the academic mainstream of linguistics. Journalists love it because it flatters them in their role as word-smiths, so there is sadly no end to the ink being spilled over it.
(3) Most crucially, any linguist worth their salt can work in multiple languages without necessarily speaking them. I can understand the sandhi phenomenon in Vedic Sanskrit without speaking Sanskrit, because that's sort of the job of a linguist? People can talk about 'sun and moon letters' in Arabic, and I can go - cool, assimilation of coronal consonants. I don't need to weep over Arabic poetry before realising this, this is just the basic tradecraft of a linguist. Same as I can understand lambda functions as a concept, without needing to know how they're implemented in <insert programming language here>.
In my experience, linguists enjoy other languages and often speak them, as I do, but it's rather ignorant to assert it's "professionally totally unacceptable" for, say, a neurolinguist to be monolingual. It suggests an exceedingly narrow view of the field.
[0] Here's a fun example of neurolinguistics. Consider the sentence 'I drinks a tall.' Approx 100-200 ms after reading this, an electrical signal (simplifying for clarity) fired off in your left frontal lobe, near the section of your brain called Broca's area. That signal, the Early Left Anterior Negativity (ELAN), is associated with syntactic violations (in this case, the use of 'tall' without an accompanying noun). Approx. 300-500 ms after reading this sentence, a second impulse, the Left Anterior Negativity (LAN) fired off in the same area of your brain. This impulse is associated with morphological violations (in this case, the use of 'I drinks' rather than the expected 'I drink'). If I said 'He drinks a tall', you'd just get the ELAN, but no LAN. If I said 'I drinks a tall glass of water.', you'd get the LAN but no ELAN. Fascinatingly, you haven't actually parsed the meaning of the sentence at the time either of these fired - that comes later, at around 600 ms (the P600 signal, in a different part of the brain). This kind of work is very much the domain of linguistics, and clearly not contingent on learning lots and lots of languages. (Enjoyable though that may be!)
Either the path I chose is too complex, or age is beginning to impact my cognitive speed.
Has anyone feeling this? or maybe I overestimated the time I should have been investing into this?
So many words in English are from Latin origin, so you got so many shortcuts when learning English (and I’m sure you are exposed to it for more time than you realize, as time spent passively learning a language dominates in the end).
It took me some time to realize the amount of work (in terms of hours put into it) that are needed to learn Japanese.
But then what I look for in language learning is not what polyglots look for. I want to understand the language deeply to appreciate its beauty fully while communicating with natives in a deep level, so I’m OK with putting the hours and leaning it within many years, even if it means it is the last language I learn. I’m in for the depth, not for the number of languages.
Would love any tips from someone who has mastered all those characters without full immersion or a structured University curriculum.
I went from just knowing how to read hira/katakana and say a few sentences (after studying it on my own for many years) to being able to hold a conversation for 1 hour after just a couple of years.
Now I'm moving onto playing easier games for kids which seem fun to me too in Japanese (e.g. some pokemon games). I pick up a lot from animes already but they can be tough, depending on the target audience.
頑張って！
-don't just lock in on one learning app or method but always be open to find better tools. Every language has different options. But don't waste too much time on that either - that's the language equivalent of trying every JS framework before starting your personal website
-identify what the hard and the easy bits of the language are. Ask yourself which of those you absolutely need in order to progress and which can wait. Then tackle the low hanging fruit while you start nibbling at the hard-but-necessary ones and leave the hard-but-less-necessary ones for later - by which time they will be easier because you'll be more familiar with them. Your perception of what the hard and the easy parts of the language are will evolve with time, stay open to that.
-adding to the last point, identify your strengths and weaknesses towards the language, this includes understanding what advantages your current stock of languages give you. Like, how Japanese pronunciation is easy for a Spanish speaker, but vocabulary acquisition isn't. Or how if you have very good memory but poor social skills you'll be good at building your vocabulary but speaking will require extra effort.
The problem with language learning in my opinion is, that learning the vocabulary just cannot be skipped. I've had German for 6 years (2-3 lessons per week) in highschool, just to find out that I don't understand words on 100th positions on the frequency lists. We've mostly focused on memorizing crazy complex grammar structures.
I've also tried passively listening to podcasts and watching movies/series I've already knew, but that would result in at most 1-2 words per episode, which just wouldn't work, if you need at least 5000 words to understand reasonably well.
(As for English, I learned all my vocabulary by reading. I was reading for the purpose of reading, learning English was not the goal at all)
Languages, their libraries, and specifically the variety of frameworks kind of require wandering to see what delivers on it's promises in the immediate and forseeable future/direction.
The point in asking? I only know a single language that is spoken outside of schools, I'm curious if those dead languages still qualify for "polyglot" status. Seems the definition is nebulous.
Dead languages, might not be dead for very long, machines do better in grammar and syntax nowadays.
"The speed and precision of vocabulary acquisition leaves no real alternative to the conclusion that the child somehow has the concepts available before experience with language and is basically learning labels for concepts that are already a part of his or her conceptual apparatus" -- Noam Chomsky
It's a question that is hotly contended, however.