Carpe Coffeam

Continue or start your personal language log here, including logs for challenge participants
BeaP
Green Belt
Posts: 405
Joined: Sun Oct 17, 2021 8:18 am
Languages: Hungarian (N), English, German, Spanish, French, Italian
x 1990

Re: Carpe Coffeam

Postby BeaP » Thu May 12, 2022 4:34 pm

I think putting a lot of pronunciation theory in coursebooks is a historical thing: we didn't have western tourists, cable TV, films, music. Also teachers were trained to emphasise pronunciation, because if you didn't learn it from them, you didn't learn it at all. This is a bit ironic, because it was the field they themselves were usually the worst at.

Monolingual books often don't include pronunciation exercises, and I think they have good reasons for this. Pronunciation needs to be taught according to the student's native language. Which sounds do you know? What can you hear? What can you produce? It's not the same for you, me or a native English speaker. I've found some completely idiotic exercises in coursebooks which were probably difficult and useful for a Korean learner.

I don't know how effectively those sounds can be trained that don't exist in your language. Intonation, stress and rhythm on the other hand can surely be imitated. I think Hungarian people don't do it because they think it's clownish, something ridiculous. They're shy to try to imitate someone well. But it's getting better with the internet.
2 x

garyb
Black Belt - 1st Dan
Posts: 1572
Joined: Mon Jul 20, 2015 12:35 pm
Location: Scotland
Languages: Native: English
Advanced: Italian, French
Intermediate: Spanish
Beginner: German, Japanese
Language Log: viewtopic.php?f=15&t=1855
x 5992
Contact:

Re: Carpe Coffeam

Postby garyb » Thu May 12, 2022 5:04 pm

Cavesa wrote:Yes, our experience differs, and I think the first part of it: language learning culture in each of our countries, is also shaping our perception of the more global online language learning world.

In the Czech resources, pronunciation and accent (self)shaming and focus is extremely strong.
Thanks for sharing your experience. Indeed I'm mostly familiar with resources for English speakers, and to a lesser extent for Romance language speakers e.g. Assimil Italian in French.

It seems to be one extreme or the other: almost no attention to pronunciation, or too much. Not much healthy middle ground of giving it the adequate and proper attention but also not encouraging perfectionism. And indeed, part of the problem is that it's often seen as something that can't be learned, at least not after a certain age, and people often use that to justify their inability to teach it or their own poor results.
2 x

User avatar
lusan
Green Belt
Posts: 463
Joined: Sat Aug 15, 2015 1:25 pm
Location: Greensboro, NC, USA
Languages: Spanish(Native)
English (Naïve)
French(Intermediate)
Italian(Intermediate)
Polish(In Alcatraz)
x 985

Re: Carpe Coffeam

Postby lusan » Thu May 12, 2022 10:53 pm

I am very happy for you. You work so hard. The future is just ahead! Congratulations!
3 x
Italian, polish, and French dance
FSI Basic French Lessons : 10 / 24 17 of 24 goal

User avatar
PeterMollenburg
Black Belt - 3rd Dan
Posts: 3229
Joined: Wed Jul 22, 2015 11:54 am
Location: Australia
Languages: English (N), French (B2-certified), Dutch (High A2?), Spanish (~A1), German (long-forgotten 99%), Norwegian (false starts in 2020 & 2021)
Language Log: https://forum.language-learners.org/vie ... 15&t=18080
x 8029

Re: Carpe Coffeam

Postby PeterMollenburg » Thu May 12, 2022 11:36 pm

BeaP wrote:Pronunciation needs to be taught according to the student's native language.


No, it doesn't. This was part of the purpose of IPA. It matters not which language is your native one - it's standardised, international. So an Asian, European or North American would understand Language XYZ depicted pronunciation regardless of background (language).

However, it's been under utilised and shied away from due to fear of over complicating the matter. It's no more complicated to roughly approximate pronunciation through an invented pronunciation system in a course book than to introduce IPA. But apparently learners are too dumb to learn something so accurate and apparently as complicated and standardised as IPA. Best not learn something you could use in another course or worse still for learning another foreign language later!
1 x

User avatar
tiia
Blue Belt
Posts: 735
Joined: Tue Mar 15, 2016 11:52 pm
Location: Finland
Languages: German (N), English (?), Finnish (C1), Spanish (B2??), Swedish (B2)
Language Log: viewtopic.php?t=2374
x 1997

Re: Carpe Coffeam

Postby tiia » Fri May 13, 2022 5:35 am

PeterMollenburg wrote:
BeaP wrote:Pronunciation needs to be taught according to the student's native language.


No, it doesn't. This was part of the purpose of IPA. It matters not which language is your native one - it's standardised, international. So an Asian, European or North American would understand Language XYZ depicted pronunciation regardless of background (language).

However, it's been under utilised and shied away from due to fear of over complicating the matter. It's no more complicated to roughly approximate pronunciation through an invented pronunciation system in a course book than to introduce IPA. But apparently learners are too dumb to learn something so accurate and apparently as complicated and standardised as IPA. Best not learn something you could use in another course or worse still for learning another foreign language later!


The thing ist just that not every language has all of the sounds. So you will still have to teach someone how to pronunce ü in German or th in English. But someone knowing e.g. Finnish doesn't need a lot of explanation on how to do the ü-sound. They already have it in their language (written as y). Not telling them directly ü=y would simply make things more complicated. But an English speaker may actually have to learn how to make the sound.
If the idea is to learn all the sounds possible to write with IPA, then one would need to learn to produce sounds that one never needs. People would forget how to produce them until they have to learn a language that actually uses them. So I don't think this would help too much.
I think one problem could be assuming or using English as a base language, because English is at least the least phonetic language that I know, so explanations not using IPA look often overly complicated.
5 x
Corrections for entries written in Finnish, Spanish or Swedish are welcome.
Project 30+X: 25 / 30

User avatar
PeterMollenburg
Black Belt - 3rd Dan
Posts: 3229
Joined: Wed Jul 22, 2015 11:54 am
Location: Australia
Languages: English (N), French (B2-certified), Dutch (High A2?), Spanish (~A1), German (long-forgotten 99%), Norwegian (false starts in 2020 & 2021)
Language Log: https://forum.language-learners.org/vie ... 15&t=18080
x 8029

Re: Carpe Coffeam

Postby PeterMollenburg » Fri May 13, 2022 6:06 am

tiia wrote:
PeterMollenburg wrote:
BeaP wrote:Pronunciation needs to be taught according to the student's native language.


No, it doesn't. This was part of the purpose of IPA. It matters not which language is your native one - it's standardised, international. So an Asian, European or North American would understand Language XYZ depicted pronunciation regardless of background (language).

However, it's been under utilised and shied away from due to fear of over complicating the matter. It's no more complicated to roughly approximate pronunciation through an invented pronunciation system in a course book than to introduce IPA. But apparently learners are too dumb to learn something so accurate and apparently as complicated and standardised as IPA. Best not learn something you could use in another course or worse still for learning another foreign language later!


The thing ist just that not every language has all of the sounds. So you will still have to teach someone how to pronunce ü in German or th in English.


While this is true and there might need to be some initial approximation of sounds even if using IPA as your base, at least then with it standardised it doesn't require a new discussion in a new course book. If you present a new phoneme (with audio) let's say in the pronunciation portion of a course book with sufficient examples alongside IPA and voilà, you've helped the student associate the new sound to the relevant IPA symbol.

Regardless of ones background however, you're right there still needs to be learning of how to produce certain sounds (like an English person producing the sound [y]). IPA like any pronunciation guide is representative, it's not going to do the work for us.

tiia wrote:If the idea is to learn all the sounds possible to write with IPA, then one would need to learn to produce sounds that one never needs.


No. This is unneccessary. When I started learning French, I learned the IPA symbols relevant to the French language. Why would I learn IPA for Spanish while learning French? Perhaps the way I worded my last post gave that impression, but it's not really what I meant.

However, with subsequent languages that I have learned, all I've needed to do is add a handful of new sounds, the remaining (predominantly composed of consonsant sounds) have been already learned.
1 x

User avatar
zenmonkey
Black Belt - 2nd Dan
Posts: 2528
Joined: Sun Jul 26, 2015 7:21 pm
Location: California, Germany and France
Languages: Spanish, English, French trilingual - German (B2/C1) on/off study: Persian, Hebrew, Tibetan, Setswana.
Some knowledge of Italian, Portuguese, Ladino, Yiddish ...
Want to tackle Tzotzil, Nahuatl
Language Log: viewtopic.php?f=15&t=859
x 7030
Contact:

Re: Carpe Coffeam

Postby zenmonkey » Fri May 13, 2022 8:37 am

PeterMollenburg wrote:
BeaP wrote:Pronunciation needs to be taught according to the student's native language.


No, it doesn't. This was part of the purpose of IPA. It matters not which language is your native one - it's standardised, international. So an Asian, European or North American would understand Language XYZ depicted pronunciation regardless of background (language).

However, it's been under utilised and shied away from due to fear of over complicating the matter. It's no more complicated to roughly approximate pronunciation through an invented pronunciation system in a course book than to introduce IPA. But apparently learners are too dumb to learn something so accurate and apparently as complicated and standardised as IPA. Best not learn something you could use in another course or worse still for learning another foreign language later!


The IPA is useful to describe differentiating sounds. It doesn't actually cover all productive sounds and is built around usability rather than accuracy. One of the issues is clearly that pronounced sounds as shown via spectral analysis demonstrate that dividing a continuous utterance into well-delimited segments is an approximation. And there are all sorts of complicated transition patterns, and co-articulations and some sounds only make sense in a certain phonetic context. IPA, like any writing system, remains an approximation.

The IPA is not some long-standing standard that is a universal reference. It continues to see important revisions and sounds are still added (latest about 12 years ago) and only really entered the public domain about 10 years ago. When I learned the IPA tones were not even consistently represented. And to this day, tonal representation for Chinese or Tibetan (non-Ladakh/Amdo) and other languages in IPA is just not practical - the Chao numbers are more practical and much more used. But again incomplete when applied to actual words and phrases. Tone sandi is poorly described by IPA or numerical, static representation.

Another concern is things like co-occurrence restrictions that are present in languages like Setswana and other Bantu languages where high low contrast matters and are generally not possible to describe with standard IPA. Extending the IPA helps but one definitely is not learning that while learning the most popular languages.

How about degrees of aspiration? IPA doesn't deal well with that either and in Tibetan (Choni) it's all represented with a single symbol. That's ok as far as differentiating representation goes but definitely aspirations have degrees.

These aren't examples pulled out of a hat - several people here have worked on learning Setswana - a material with IPA was pretty lacking and insufficient.

BeaP is correct, even with the IPA, you need to learn pronunciation based on the student's prior procedural knowledge of languages - from the issues with productive concerns to limitations in auditory perception - for example, a late learner of English from Mexico like my mother (who knew the IPA) will still have an issue with 'sheets' unless she actively trains - that work around pronunciation is dependent on her prior language experience, and it's an exercise that would be useless for French speakers.

For me, learning pronunciation is also a midterm process, it's part of learning to hear in order to also learn to produce speech. I'm definitely still struggling with final and medial ق which has often an IPA of /ɡ/ but somehow is different from گ /g/ and I don't hear it. The IPA isn't enough. It will come. Takes time.
8 x
I am a leaf on the wind, watch how I soar

BeaP
Green Belt
Posts: 405
Joined: Sun Oct 17, 2021 8:18 am
Languages: Hungarian (N), English, German, Spanish, French, Italian
x 1990

Re: Carpe Coffeam

Postby BeaP » Fri May 13, 2022 9:20 am

zenmonkey wrote:For me, learning pronunciation is also a midterm process, it's part of learning to hear in order to also learn to produce speech.

I've been thinking about this lately. So you think it's possible to train an adult (whose brain is already set for certain phonemes) to hear the difference between new sounds? I sometimes have the impression that people have bad pronunciation because they don't even hear what should be heard, and consequently they can't imitate it.
2 x

User avatar
zenmonkey
Black Belt - 2nd Dan
Posts: 2528
Joined: Sun Jul 26, 2015 7:21 pm
Location: California, Germany and France
Languages: Spanish, English, French trilingual - German (B2/C1) on/off study: Persian, Hebrew, Tibetan, Setswana.
Some knowledge of Italian, Portuguese, Ladino, Yiddish ...
Want to tackle Tzotzil, Nahuatl
Language Log: viewtopic.php?f=15&t=859
x 7030
Contact:

Re: Carpe Coffeam

Postby zenmonkey » Fri May 13, 2022 9:43 am

BeaP wrote:
zenmonkey wrote:For me, learning pronunciation is also a midterm process, it's part of learning to hear in order to also learn to produce speech.

I've been thinking about this lately. So you think it's possible to train an adult (whose brain is already set for certain phonemes) to hear the difference between new sounds? I sometimes have the impression that people have bad pronunciation because they don't even hear what should be heard, and consequently they can't imitate it.


I certainly have the same impression.

I do think an adult can learn corrective differentiation. With my German, I still have a French accent (which made my daughters laugh). Years of work with just shadowing really helped a lot. I think people don't hear and therefore don't correct. And a strategy of overtraining mechanical correct repetition does something. Now is there new plasticity that is built up? Probably - other research in using the auditory channels to process new information suggests it's trainable, to a degree. And there are effective accent reduction processes. But I also think there is a process of diminishing returns and at some point getting to 'ok' (whatever that means) is where we end up stopping. I'll always have some 'French' tones in my German - I'm happy with that - as long as my girls are no longer laughing and people aren't scrunching up their faces when I speak.

Before my hiatus from here, I was looking into speech analyzer software to work on accent reduction (Praat). I never really found something so useful for me that I was 'wow' this will do it. But I do know that some language therapists work with them as tools to develop new speaking patterns. Now is it actually new listening capabilities? I've read mixed things. I should spend more time researching that... :D
3 x
I am a leaf on the wind, watch how I soar

User avatar
PeterMollenburg
Black Belt - 3rd Dan
Posts: 3229
Joined: Wed Jul 22, 2015 11:54 am
Location: Australia
Languages: English (N), French (B2-certified), Dutch (High A2?), Spanish (~A1), German (long-forgotten 99%), Norwegian (false starts in 2020 & 2021)
Language Log: https://forum.language-learners.org/vie ... 15&t=18080
x 8029

Re: Carpe Coffeam

Postby PeterMollenburg » Fri May 13, 2022 10:08 am

zenmonkey wrote:
PeterMollenburg wrote:
BeaP wrote:Pronunciation needs to be taught according to the student's native language.


No, it doesn't. This was part of the purpose of IPA. It matters not which language is your native one - it's standardised, international. So an Asian, European or North American would understand Language XYZ depicted pronunciation regardless of background (language).

However, it's been under utilised and shied away from due to fear of over complicating the matter. It's no more complicated to roughly approximate pronunciation through an invented pronunciation system in a course book than to introduce IPA. But apparently learners are too dumb to learn something so accurate and apparently as complicated and standardised as IPA. Best not learn something you could use in another course or worse still for learning another foreign language later!


The IPA is useful to describe differentiating sounds. It doesn't actually cover all productive sounds and is built around usability rather than accuracy. One of the issues is clearly that pronounced sounds as shown via spectral analysis demonstrate that dividing a continuous utterance into well-delimited segments is an approximation. And there are all sorts of complicated transition patterns, and co-articulations and some sounds only make sense in a certain phonetic context. IPA, like any writing system, remains an approximation.

The IPA is not some long-standing standard that is a universal reference. It continues to see important revisions and sounds are still added (latest about 12 years ago) and only really entered the public domain about 10 years ago. When I learned the IPA tones were not even consistently represented. And to this day, tonal representation for Chinese or Tibetan (non-Ladakh/Amdo) and other languages in IPA is just not practical - the Chao numbers are more practical and much more used. But again incomplete when applied to actual words and phrases. Tone sandi is poorly described by IPA or numerical, static representation.

Another concern is things like co-occurrence restrictions that are present in languages like Setswana and other Bantu languages where high low contrast matters and are generally not possible to describe with standard IPA. Extending the IPA helps but one definitely is not learning that while learning the most popular languages.

How about degrees of aspiration? IPA doesn't deal well with that either and in Tibetan (Choni) it's all represented with a single symbol. That's ok as far as differentiating representation goes but definitely aspirations have degrees.

These aren't examples pulled out of a hat - several people here have worked on learning Setswana - a material with IPA was pretty lacking and insufficient.

BeaP is correct, even with the IPA, you need to learn pronunciation based on the student's prior procedural knowledge of languages - from the issues with productive concerns to limitations in auditory perception - for example, a late learner of English from Mexico like my mother (who knew the IPA) will still have an issue with 'sheets' unless she actively trains - that work around pronunciation is dependent on her prior language experience, and it's an exercise that would be useless for French speakers.

For me, learning pronunciation is also a midterm process, it's part of learning to hear in order to also learn to produce speech. I'm definitely still struggling with final and medial ق which has often an IPA of /ɡ/ but somehow is different from گ /g/ and I don't hear it. The IPA isn't enough. It will come. Takes time.


Okay, so it seems IPA has limitations I hadn't considered. I use tonal markings for Norwegian with IPA transcription without issue, but I don't know how I'd go with mandarin for example.

Nevertheless to me it is a better system to become accustomed to than TY's invented pronunciation systems or Assimil's. Both assume, they assume that the learner is either English or French. In IPA [y] is [y] even if it isn't 100% accurate it is more reliable than such examples like 'a' as in 'father'. Huh? 'A' in father for who? ie where's the speaker from? So I disagree, I feel BeaP is not entirely correct here as a standardised system such as IPA does elimate that problem (who's pronouncing sound xyz) by simply being standardised regardless of some limitations you have brought to light.
0 x


Return to “Language logs”

Who is online

Users browsing this forum: MapleLeaf and 2 guests