Correction... how useful...?

General discussion about learning languages
User avatar
Iversen
Black Belt - 4th Dan
Posts: 4759
Joined: Sun Jul 19, 2015 7:36 pm
Location: Denmark
Languages: Monolingual travels in Danish, English, German, Dutch, Swedish, French, Portuguese, Spanish, Catalan, Italian, Romanian and (part time) Esperanto
Ahem, not yet: Norwegian, Afrikaans, Platt, Scots, Russian, Serbian, Bulgarian, Albanian, Greek, Latin, Irish, Indonesian and a few more...
Language Log: viewtopic.php?f=15&t=1027
x 14924

Re: Correction... how useful...?

Postby Iversen » Fri Mar 24, 2023 9:09 am

luke wrote:I'd be inclined to think that minimal pair training that doesn't include production is incomplete.

Cainntear wrote:So do I, because I believe that the most effective way to build an accurate phoneme map is through producing the correct phonemes. How can a teacher judge the correctness of a student's phoneme map except by their production of phonemes, after all...?

luke wrote:I agree. Getting minimal pairs right is like being able to play note transitions smoothly and accurately on a musical instrument.


As a teacher Cainntear is obviously interested in knowing how well his students are doing, and then they of cause have to show their skills in practice - only then can the teacher judge whether they have organized their phonemes in the prescribed fashion. And the students of course have to learn to speak before that can demonstrate their skills and get the approving nod from the teacher. But that's how things are seen from a control point of view and doesn't in itself show how the students got there.

It's actually the old discussion about input and output once again, just at the phonological level. And my input to that discussion is that exact unbiased listening witout reference to the phoneme discussion has been underestimated. Minimal pairs is just the part of listening that concerns the borderlines between phonemes (that's how phonemes are defined, actually), but it's in my view just as important to be able to hear the differences between the allophones within a phoneme (how for instance a vowel phoneme differs with or without an r after it), and that must build on acribic listening to actual sounds - including sounds emitted by different native speakers.

As for the parallel between minimal pairs and note transitions in music - I simply don't understand that comparison :? . I spent several hours yesterday trying to revive my slumbering piano playing skills on my newly tuned piano, and there the phonemes were nicely separated - one key per note. I have also played violin and cello for many years, and there you have to hit one canonical spot for any given note - but at least there is a one dimensional continuum of notes on a given string, and there is one correct place to put your finger for each note (actually things are more complicated because of the notion of temperature, but let's leave the diabolus in musica aside for now). You may slide from one note to the next, but that movement will still be between different notes, each defined by one canonical finger position.

If I had to point to anything that gave me problems similar to those with the phoneme borders it would be tempo - what IS allegro? What IS adagio? On my metronome each name is associated with an interval (beats per minute), but in practice the speeds overlap. One conductor may play moderato faster than the allegretto of another, and a composer may write presto but the performer does it at a speed that would be called prestissimo on my metronome. But I still can't see how that can help me to learn languages EXCEPT that you have to learn to listen to the actual sounds in both cases - rather than just being content with the abstract units into which the actual sounds are organized. Only piano players are (partly) exempt from that requirement.. (afterthought: and writers that use a keyboard)
2 x

User avatar
luke
Brown Belt
Posts: 1243
Joined: Fri Aug 07, 2015 9:09 pm
Languages: English (N). Spanish (intermediate), Esperanto (B1), French (intermediate but rusting)
Language Log: https://forum.language-learners.org/vie ... 15&t=16948
x 3629

Re: Correction... how useful...?

Postby luke » Fri Mar 24, 2023 9:52 am

Iversen wrote:
luke wrote:Getting minimal pairs right is like being able to play note transitions smoothly and accurately on a musical instrument.
Only piano players are (partly) exempt from that requirement.. (afterthought: and writers that use a keyboard)

I'm probably missing the point. I had to go to the dictionary to improve my understanding of allophone.

With piano and keyboard (for typing), I was thinking in terms of hitting a neighboring note, or two notes when there should be one, or just not fully articulating the note. With typing, that might show up as a typographical error or a missing letter.
0 x
: 124 / 124 Cien años de soledad 20x
: 5479 / 5500 5500 pages - Reading
: 51 / 55 FSI Basic Spanish 3x
: 309 / 506 Camino a Macondo

Cainntear
Black Belt - 3rd Dan
Posts: 3449
Joined: Thu Jul 30, 2015 11:04 am
Location: Scotland
Languages: English(N)
Advanced: French,Spanish, Scottish Gaelic
Intermediate: Italian, Catalan, Corsican
Basic: Welsh
Dabbling: Polish, Russian etc
x 8601
Contact:

Re: Correction... how useful...?

Postby Cainntear » Fri Mar 24, 2023 9:53 am

jeff_lindqvist wrote:Speaking of music, I once heard a lecturer give examples of minimal pairs - they happened to use list intonation. :roll: I'm sure that even a tone deaf listener would be able to pass such a test.

That's triggered an old complaint of mine that was pretty common in computer language course audio for a while.

You could tell what order things were recorded in by listening to the intonation of the speaker.

"he likes" would often have an upwards list intonation, but not only would "she likes" have a downwards intonation, there would be contrastive intonation: she likes. I spotted the results in my own output -- I was learning male 3rd person pronouns as unstressed (or, depending on the language, weakly stressed), but I had the female third person as strongly stressed. I only managed to overcome this because I'd noticed I was doing it -- average learners wouldn't question it at all and would learn it wrong.
2 x

Cainntear
Black Belt - 3rd Dan
Posts: 3449
Joined: Thu Jul 30, 2015 11:04 am
Location: Scotland
Languages: English(N)
Advanced: French,Spanish, Scottish Gaelic
Intermediate: Italian, Catalan, Corsican
Basic: Welsh
Dabbling: Polish, Russian etc
x 8601
Contact:

Re: Correction... how useful...?

Postby Cainntear » Fri Mar 24, 2023 10:13 am

luke wrote:
Iversen wrote:
luke wrote:Getting minimal pairs right is like being able to play note transitions smoothly and accurately on a musical instrument.
Only piano players are (partly) exempt from that requirement.. (afterthought: and writers that use a keyboard)

I'm probably missing the point. I had to go to the dictionary to improve my understanding of allophone.

With piano and keyboard (for typing), I was thinking in terms of hitting a neighboring note, or two notes when there should be one, or just not fully articulating the note. With typing, that might show up as a typographical error or a missing letter.


I think computer keyboards are pretty hard to talk about allophones on.

Keyboard instruments perhaps less so, because the tune context determines which finger is best to strike the target key with, what angle the finger makes etc... but the produced sound is the same.

Stringed instruments are probably the best comparison to allophones, because you have pure tones and glides. A pure tone is only possible if you stop the string vibrating before and after the note and are *reeeally* careful in moving your finger not to squeak the string. If you switch from a lower note on the same string, you are likely to hit the new note slightly flat and slide upwards into it; if you switch from a higher note on the same string, you're likely to start slightly sharp and slide downwards. (And alternatively, you may overcompensate when going up and end up sharp have to correct downwards and similarly overcompensate when going down by ending up flat and having to correct upwards.) If you switch strings, there might be a brief overlap of vibration on the string you were playing and the string you are now playing.
All of these things are real and can be detected if you're listening carefully, but if you were asked to learn a tune from someone else's playing, you'd only really be likely to copy the notes and not do a close imitation of the player's style.

The exception, or course, is that when learning a new musical idiom, good musicians will actually consider the style of expression as important.

Nothing grinds my gears as much as a classically-trained violinist using a classical expression ("accent...?") when playing Scottish or Irish fiddle tunes, and thinking about it, I can kind of see that as like when a foreigner uses their own L1 phonology in the target language, so I'm probably changing my view on what musical "phonemes" really are.

Like the pure frequency and length of note is said to be the main thing, but there's far more expression that goes into a note being "natively expressed".

I'm a little lost as to what my own argument is here, actually...
2 x

User avatar
Iversen
Black Belt - 4th Dan
Posts: 4759
Joined: Sun Jul 19, 2015 7:36 pm
Location: Denmark
Languages: Monolingual travels in Danish, English, German, Dutch, Swedish, French, Portuguese, Spanish, Catalan, Italian, Romanian and (part time) Esperanto
Ahem, not yet: Norwegian, Afrikaans, Platt, Scots, Russian, Serbian, Bulgarian, Albanian, Greek, Latin, Irish, Indonesian and a few more...
Language Log: viewtopic.php?f=15&t=1027
x 14924

Re: Correction... how useful...?

Postby Iversen » Fri Mar 24, 2023 10:39 am

Cainntear just presented the best possible way to make me see the logic in Luke's comparison of slides in music to allophones in speech -now I'm happy, and I find my humble self vindicated in pointing to the importance of listening to allophones instead of just focusing on the phonemes :D
2 x

Cainntear
Black Belt - 3rd Dan
Posts: 3449
Joined: Thu Jul 30, 2015 11:04 am
Location: Scotland
Languages: English(N)
Advanced: French,Spanish, Scottish Gaelic
Intermediate: Italian, Catalan, Corsican
Basic: Welsh
Dabbling: Polish, Russian etc
x 8601
Contact:

Re: Correction... how useful...?

Postby Cainntear » Fri Mar 24, 2023 11:00 am

Iversen wrote:
luke wrote:I'd be inclined to think that minimal pair training that doesn't include production is incomplete.

Cainntear wrote:So do I, because I believe that the most effective way to build an accurate phoneme map is through producing the correct phonemes. How can a teacher judge the correctness of a student's phoneme map except by their production of phonemes, after all...?

luke wrote:I agree. Getting minimal pairs right is like being able to play note transitions smoothly and accurately on a musical instrument.


As a teacher Cainntear is obviously interested in knowing how well his students are doing, and then they of cause have to show their skills in practice - and then (and only then) can the teacher judge whether they have organized their phonemes in the prescribed fashion. And the students of course have to learn to speak before that can demonstrate their skills and get the approving nod from the teacher. But that's how things are seen from a control point of view and doesn't in itself show how the students got there.

No, but I'm definitely in favour of teachers attempting to understand how students get there, and improving their understanding over time, hence my big beef over people saying "Well student Y and student Z did well, so that proves my techniques work. Students A, B, C, ... ,W and X failed, but that's because they didn't try."

It's actually the old discussion about input and output once again, just at the phonological level. And my input to that discussion is that exact unbiased listening witout reference to the phoneme discussion has been underestimated.

I think I've made my view clear on that previously.

Comprehension is usually possible based on context.

There is therefore little motivation to perceive phonemes from listening.

Worse, the adult human is inclined to disregard phonemic differences as just a matter of accent, because it normally is, so won't pick them up from listening.

Worse again, the adult human is inclined to disregard differences in phonology as unimportant, and understand through the filter of their own phoneme map -- eg seseo and distinción speakers of Spanish: the seseo speaker will map the two phonemes for S and Z to a single one and can therefore understand but not mimic the distinción speaker, whereas the distinción speaker has two distinct phonemes but learns to treat the S sound as an acceptable allophone of a Z phoneme, so can both understand and mimic the seseo speaker.

I do believe that the wrinkle here is this phrase:
"exact unbiased listening"

What do you mean by "unbiased"? I suggest that you're talking about having a truly open mind and failing to acknowledge that you are in fact doing something that most people wouldn't and couldn't: whether as a result of intuition, through exposure to multiple languages early in life or because of your university studies, you have an internal understanding of how important phonemic differences are, and that means that you are able to listen for them.

My point is, as always, that if we have a special skill that allows us to do more than the average person, we have to be careful to recognise that and not assume that other people will automatically do what we do.

Minimal pairs is just the part of listening that concerns the borderlines between phonemes (that's how phonemes are defined, actually), but it's in my view just as important to be able to hear the differences between the allophones within a phoneme (how for instance a vowel phoneme differs with or without an r after it), and that must build on acribic listening to actual sounds - including sounds emitted by different native speakers.

Is it important to do from day 1, though? That's the big question.

The average uninformed learner doesn't fully appreciate the distinction between phonemic differences and non-phonemic ones, so there is a very real danger of a beginner hearing two allophones as distinct phonemes.

I cannot see how allophones can be comfortably learned without first learning phonemes, which means that the phoneme should really be said in its core form until such time as the learner has internalised it.

Hell most allophones are just natural results of making speaking more efficient anyway -- hence why languages typically maintain either their consonant or vowel phonemes in stronger form while letting the other weaken.
Eg English lets any old vowel drop to schwa directly before or after the tonic syllable and Russian lets "oh" sounds collapse to "ah" when unstressed. But both tend to be very particular about maintaining clear consonants. Whereas in Spanish, "pensado" with a full D phone would make the tongue move in a very sharp, pointy triangle, so you can round off the point by not actually touching your teeth at all, even though it's phonemically a dental consonant. You can only do this because Spanish is very particular about pronouncing vowels.
1 x

Cainntear
Black Belt - 3rd Dan
Posts: 3449
Joined: Thu Jul 30, 2015 11:04 am
Location: Scotland
Languages: English(N)
Advanced: French,Spanish, Scottish Gaelic
Intermediate: Italian, Catalan, Corsican
Basic: Welsh
Dabbling: Polish, Russian etc
x 8601
Contact:

Re: Correction... how useful...?

Postby Cainntear » Fri Mar 24, 2023 11:05 am

Iversen wrote:Cainntear just presented the best possible way to make me see the logic in Luke's comparison of slides in music to allophones in speech -now I'm happy, and I find my humble self vindicated in pointing to the importance of listening to allophones instead of just focusing on the phonemes :D

Ah yes, but here the only difference between musical styles is in allophone/expression -- there are no new "phonemes" to consider (well, ignoring microtonal traditions and old ways of teaching bagpipe music).

I would again argue that if new phonemes are required, the phonemes should probably come first (although it's a very nuanced matter and this discussion definitely has softened my stance so that I can genuinely think about the ramifications of everything).
1 x

tractor
Green Belt
Posts: 377
Joined: Sat Oct 29, 2016 10:58 am
Location: Norway
Languages: Norwegian (N), English, Spanish, Catalan, French, German, Italian, Latin
x 766

Re: Correction... how useful...?

Postby tractor » Fri Mar 24, 2023 8:47 pm

Cainntear wrote:The average uninformed learner doesn't fully appreciate the distinction between phonemic differences and non-phonemic ones, so there is a very real danger of a beginner hearing two allophones as distinct phonemes.

I don't think the average native spaker fully appreciates the distinction between phonemic and non-phonemic differences either. He or she has probably never heard about neither phonemes nor allophones, nor has the average learner. They most likely think in terms of sounds, not phonemens and allophones. Their concept of 'sound' doesn't necessarily correspond to a phonologist's defintion of 'phoneme'.

Does it matter if a learner of Spanish thinks of the two d's in un dedo as one sound or two different sounds, as long as they pronounce them correctly? I don't think it does.
Cainntear wrote:I cannot see how allophones can be comfortably learned without first learning phonemes, which means that the phoneme should really be said in its core form until such time as the learner has internalised it.

If you teach people the 'core form' first, you teach them to pronounce words incorrectly from the start. Then, after a while, they'll have to unlearn the wrong pronunciation in order to learn the proper one. Doesn't that approach go against the common wisdom of learning the right thing from the very beginning?

Having said that, I don't think that the teacher (or a language course) should necessarily teach all the details from the very beginning nor that every minor pronunciation error that a beginner makes should be pointed out. Nitpicking about non-phonemic differences can wait.

By the way, what is the 'core form' of the Spanish phoneme /d/ anyway? [d] or [ð]? If you go by frequency, I would gess it would be the latter, but personally I think of the former as the core form.
5 x

User avatar
tastyonions
Black Belt - 1st Dan
Posts: 1551
Joined: Sat Jul 18, 2015 5:39 pm
Location: Dallas, TX
Languages: EN (N), FR, ES, DE, IT, PT, NL, EL
x 3795

Re: Correction... how useful...?

Postby tastyonions » Fri Mar 24, 2023 9:04 pm

If I were to teach Spanish pronunciation to native English speakers, I'd do a demo of a big honking gringo accent, e.g. say "lado" with dark L and [d] rather than [ð] and drag out the "o" to a diphthong, and then do a demo of a decent accent without those features, pointing out all the changes as I went along. And repeat the same exercise across many different words.
1 x

Cainntear
Black Belt - 3rd Dan
Posts: 3449
Joined: Thu Jul 30, 2015 11:04 am
Location: Scotland
Languages: English(N)
Advanced: French,Spanish, Scottish Gaelic
Intermediate: Italian, Catalan, Corsican
Basic: Welsh
Dabbling: Polish, Russian etc
x 8601
Contact:

Re: Correction... how useful...?

Postby Cainntear » Fri Mar 24, 2023 9:51 pm

tractor wrote:
Cainntear wrote:The average uninformed learner doesn't fully appreciate the distinction between phonemic differences and non-phonemic ones, so there is a very real danger of a beginner hearing two allophones as distinct phonemes.

I don't think the average native spaker fully appreciates the distinction between phonemic and non-phonemic differences either. He or she has probably never heard about neither phonemes nor allophones, nor has the average learner. They most likely think in terms of sounds, not phonemens and allophones. Their concept of 'sound' doesn't necessarily correspond to a phonologist's defintion of 'phoneme'.

Just because they don't know the words phoneme and allophone doesn't mean they don't have an intuitive distinction between the two.

I'm pretty certain that neurolinguistics has confirmed the existence of phonemes and the fact that allophones are effectively recognised as the same meaningful unit of sound.

Does it matter if a learner of Spanish thinks of the two d's in un dedo as one sound or two different sounds, as long as they pronounce them correctly? I don't think it does.

So are you suggesting that having a different phoneme map from a native speaker doesn't mean you haven't learned the language wrong...?
Cainntear wrote:I cannot see how allophones can be comfortably learned without first learning phonemes, which means that the phoneme should really be said in its core form until such time as the learner has internalised it.

If you teach people the 'core form' first, you teach them to pronounce words incorrectly from the start. Then, after a while, they'll have to unlearn the wrong pronunciation in order to learn the proper one. Doesn't that approach go against the common wisdom of learning the right thing from the very beginning?

What is the right thing in the beginning?
I've always said I don't care about accent -- I learn phonology first, and then the accent takes care of itself. Aiming for a good accent early on before you've even built the phoneme map makes it worse in the long run.


By the way, what is the 'core form' of the Spanish phoneme /d/ anyway? [d] or [ð]? If you go by frequency, I would gess it would be the latter, but personally I think of the former as the core form.

[d] is the core form. The way I thought of [ð] while I was learning was that it was an attempt to say [d] that didn't quite make it because the corner was just rounded off.
[ð] doesn't make sense as a core form, because if that's what a /d/ phoneme means, why would you realise it as [d] when it means moving your tongue further...?
0 x


Return to “General Language Discussion”

Who is online

Users browsing this forum: D Anto, Google [Bot], ryanheise and 2 guests