|
 |
The Accent Conjecture | ||
Tags: Language Acquisition Theory | Accent | ||
Share with:
| ||
 Language Learning Forum : General discussion |  | |
| Medulin Tetraglot Senior Member Croatia Joined 4671 days ago 1199 posts - 2192 votes  Speaks: Croatian*, English, Spanish, Portuguese Studies: Norwegian, Hindi, Nepali | Message 25 of 37 17 January 2015 at 3:19am | IP Logged | ||
|
Why should a white or a black person desperately try sounding (like a ) native while speaking a South/East Asian language if they will never be mistaken for a native speaker because of their skin?
Spy agencies and translation agencies stick to the "native speaker principle", meaning: open your eyes and your ears and close your mouth, the only quality translation service done is that from a foreign language into your native language. French people and Indian people are proud of their accent while speaking English, and to them, putting on a native English/American/Canadian etc. accent is an epitome of faking. Getting all your phonemes and allophones right is the easiest thing to do, though, connected speech, focus and speech melody / intonation, this is where the problems begin, that's why even talented phonologists will hardly be mistaken for native speakers of Norwegian, Swedish, Croatian, Cantonese or Malayalam... Even the Swedes, who brag about their ''impeccable'' English, are easy to ''expose'' as non-natives because most of them have the [z]/[s] merger, so they pronounce losing like loosing, zit as sit, music as ['mjuːsɪk] etc... Other voiced consonants are also frequently unvoiced: The funniest thing is ''I'm joking'' getting pronounced as ''I'm choking'' :p Edited by Medulin on 17 January 2015 at 3:42am 4 persons have voted this message useful | |||
| robarb Nonaglot Senior Member United States languagenpluson Joined 5062 days ago 361 posts - 921 votes Speaks: Portuguese, English*, German, Italian, Spanish, Dutch, Swedish, Esperanto, French Studies: Mandarin, Danish, Russian, Norwegian, Cantonese, Japanese, Korean, Polish, Greek, Latin, Nepali, Modern Hebrew | Message 26 of 37 17 January 2015 at 5:00am | IP Logged | ||
|
For most people I don't see the value of being mistaken for a native speaker, in fact I'd rather have a really good
accent that's extremely close but identifiably nonnative. Today I mistakenly thought a Finn was a native speaker of English (I could tell something was off about the accent, but it was subtle and she didn't make any grammar or usage errors, so I passed it off as a regional native accent or idiosyncratic speech). Then when I found out she was from Finland I was much more impressed, because instead of being someone who talks a little funny, she suddenly became to me someone whose L2 English is way better than the typical immigrant! So if you want to be perceived as extremely competent you want to be close but not quite native-like. 5 persons have voted this message useful | |||
| Ari Heptaglot Senior Member Norway Joined 6585 days ago 2314 posts - 5695 votes Speaks: Swedish*, English, French, Spanish, Portuguese, Mandarin, Cantonese Studies: Czech, Latin, German | Message 27 of 37 17 January 2015 at 8:02am | IP Logged | ||
|
When I'm talking about cost I'm not talking about cost to the individual. I'm talking about cost to the species. For example, we don't have three eyes, because the benefits (slightly larger field of vision, maybe) don't match the costs (extra point sensible to infection and damage, extra energy required for processing the information).
So, seeing as children have the ability to hear and reproduce all sounds that exist in human languages, the question is: what is the reason for this ability going away, leaving us with just the phoneme inventory of the languages we learned as children? There might be a developmental reason behind this. Maybe it's difficult or costly to learn to speak a language whilst keeping a large phoneme inventory. But bilingual children have significantly larger phoneme inventories that monolingual ones, and they don't seem to suffer from any issues. The other reason why we lose the ability might be that it's been beneficial to our ancestors to lose it. The cost to the individual making decisions (proximate cause) is different from the cost to the genes reproducing (ultimate cause). 2 persons have voted this message useful | |||
| daegga Tetraglot Senior Member Austria lang-8.com/553301 Joined 4524 days ago 1076 posts - 1792 votes Speaks: German*, EnglishC2, Swedish, Norwegian Studies: Danish, French, Finnish, Icelandic | Message 28 of 37 17 January 2015 at 12:28pm | IP Logged | ||
I wouldn't even limit the cost of recognizing too many phonemes to a language penalty. It looks like a general prioritization strategy. If you want more classes for your classifier, you need more neurons to do so, maybe even exponentially so. These neurons might better be used for different skills, which are more important at a certain age (like walking for a child). There certainly is an upper limit on how many neurons we can have at a certain age/brain volume. Then there also is the issue of how these neurons are connected. Basically, you have to find a good compromise between different necessary functions. If you don't happen to need more than x different classes, why waste resources and add more? On a side node: the probability for neighboring languages having roughly the same phoneme inventory as our own should be pretty high ... at least for most of the time in human history. One reason less for going for a universal phoneme system, trading with neighbor tribes works well enough without. Edited by daegga on 17 January 2015 at 12:28pm 1 person has voted this message useful | |||
| Ari Heptaglot Senior Member Norway Joined 6585 days ago 2314 posts - 5695 votes Speaks: Swedish*, English, French, Spanish, Portuguese, Mandarin, Cantonese Studies: Czech, Latin, German | Message 29 of 37 17 January 2015 at 2:48pm | IP Logged | ||
But even with languages very closely related, few are those who can pass for native speakers. We can understand speakers with very different phoneme inventories, too, so this doesn't seem to relevant to me. As regards to your comments on neuron space, this sounds like "attic theory" to me. As far as I understand, the brain doesn't really have limited space (or, it does, but this limit is so large that no human is likely to come near it). Rather, knowing more things makes it easier to learn new things. But I can see that there might be a reason for us to forget the categories not used in our native language. Sure. But why do we lose the ability for effectively make new categories? When we learn a new language, who can't we use the same system we used as children to accomodate some new categories? Children can be bilingual wothout problems and accomodate larger phoneme inventories, so why can't we expand them later in life? This ability to create new phoneme categories doesn't seem very costly to me. 1 person has voted this message useful | |||
| Bao Diglot Senior Member Germany tinyurl.com/pe4kqe5 Joined 5769 days ago 2256 posts - 4046 votes Speaks: German*, English Studies: French, Spanish, Japanese, Mandarin | Message 30 of 37 17 January 2015 at 3:54pm | IP Logged | ||
The issue lies not in acquisition - for all I know with the right training any adult can learn new phonemes - the issue lies in prioritization, in the management of those phonemes. While theoretically it may be possible to use different neural networks for the phonology of different languages they all have to use the same interface to the same sensory system, meaning there is a bottleneck. This bottleneck is smaller when it comes to the production of sound when compared to listening comprehension, and it exists even in natively bilingual speakers. I think it was Leah Fabiano-Smith who showed in her work that Puertorican children with native level in English and Spanish used only about half of the variation in allophones in either language compared to the monolingual control groups. 3 persons have voted this message useful | |||
| daegga Tetraglot Senior Member Austria lang-8.com/553301 Joined 4524 days ago 1076 posts - 1792 votes Speaks: German*, EnglishC2, Swedish, Norwegian Studies: Danish, French, Finnish, Icelandic | Message 31 of 37 17 January 2015 at 4:09pm | IP Logged | ||
|
As far as I can see it, there are 3 ways to store new information:
1. make new neurons 2. rewire existing neurons 3. change firing thresholds in existing neurons (ie. adjusting parameters) and of course combinations thereof. As long as you grow, (1) and (2) make you learn quite fast. Lot's of free space to fill. As an adult, you mostly do (3) and some (2). (1) is then mostly replacement action for dead tissue. Changing stuff has a side effect: it alters you knowledge (things getting blurry after a while). There must be some method in place in order to protect your most important memories/knowledge. Useless stuff (like phoneme boundaries you don't need) will not be as protected. You gradually lose it. So why lose the ability to make new categories? Well, we don't, it's just harder. It would require some more severe changes than independent information I guess, even playing with your abilities in your native language. So there is a tradeoff, and as you said, we do just fine without those extra categories in many languages, we just sound a bit off. Living in a foreign country, completely immersed, ignoring your native language, you might find such changes happening. Arnold Schwarzenegger can't talk proper German/Austrian dialect anymore (just speaking about accent), but his English is a lot better than it was in his first movies.
This sounds like the old "we use only 10% of our potential" mantra from SciFi. We don't use 100% of the brain in parallel, but we do use 100%. The brain is no CPU you can use for anything all the time, it's more like a big cluster of FPGAs. While you can reprogram all of those, you usually don't ('cause you can't easily switch back to a backup) and only use the ones relevant to the task. 2 persons have voted this message useful | |||
| Ari Heptaglot Senior Member Norway Joined 6585 days ago 2314 posts - 5695 votes Speaks: Swedish*, English, French, Spanish, Portuguese, Mandarin, Cantonese Studies: Czech, Latin, German | Message 32 of 37 17 January 2015 at 6:33pm | IP Logged | ||
|
Oh, God no, I certainly didn't mean the 10% thing. But the human brain can store an estimated 2.5 petabytes of information, which is huge, and more than anyone will ever need. People like Jill Price, who can remember everything that's ever happened to her since she was a child, show clearly that the brain doesn't forget because of lack of space.
1 person has voted this message useful |
|  | 
Printable version
|
| Copyright 2024 FX Micheloud - All rights reserved No part of this website may be copied by any means without my written authorization. |
