Does shadowing improve listening?

[Iversen]

I personally have a construction in my head that downgrades listening to 'just being alert to interruptions' while I am speaking myself. So essentially I don't really understand how people can shadow in the version where they just reproduce what they hear (or think they hear) while they hear it. In contrast, I can see the point in echoing where you repeat after you have listened, and even more in the version where you also study the low level sound system of the target language so that you know what you can expect to hear.

That being said, I am fairly confident that the phoneme set of a language can be reliably determined - though with caveats such as the loss of nasalized oe in some French dialects. I do not trust the advice given about allophones in the books at all. Way back in the HTLAL days I let several different sampled voices read a passage from a Dutch homepage aloud, and then I wrote down exactly how the voices pronounced the diphtongs. And point one: they did not pronounce them the same way, 2) they didn't pronounce them as my books said they should. When we come down to the allophone level you just have to listen - but if you know something about phonology you may stand a better chance of analyzing the sounds correctly. And then my question is: how on Earth can you do that while you are babbling happily along yourself ??

So I would prefer echoing (one sentence max. in one go), and I would also want to know something about the speaker. The ideal thing would have to have each sentence spoken by different speakers with different backgrounds so that you can find out what the possible gamut is. The only other alternative would be to listen exclusively to one speaker, and then try to absorb the language as she or he renders it. I suppose that many teachers would favour this last strategy, but when it comes to support for listening the first strategy might be better because you become more diversity tolerant.

PS: it just occurs to me that there is one compromise that might work, namely the one where an authority figure says something and then 123 kids repeat in chorus what they heard. I think that is the method used in some schools abroad, but I would hate it.

[jackb]

Does repeating a sentence after a native speaker, making an effort to match all aspects of pronunciation (including prosody), improve your ability to understand said sentence, or similar structures, when you hear them? (Yes, there is an L1 definition involved – see technique below)

I changed my vote to yes because of this (above) combined with the OP's definition of shadowing.

The listen -> repeat -> listen loop of a small phrase/sentence is helpful for listening.

[Kraut]

In case some don't know: VLC-Player can create a passage of any length: one sentence .... ten sentences etc that can be played repeatedly until you stop it. You can find this function under "Ansicht- Erweiterte Steuerung".

[jeff_lindqvist]

No beginner nor intermediate learner can/should do this, the good practice for them is described above and better called "echoing".

That's what I think as well. Or echoing first and then shadowing the same recording when you master it. Argüelles recommends 'blind shadowing' first where you just parrot the sounds you hear. I don't see the point, to be honest, apart from 'becoming curious'.

Here's what Arguelles wrote more than fifteen years ago (Message 20 of 39):

Blind shadowing first and foremost when learning the phonetics of a language like Korean forces you listen more carefully and to reproduce the sounds you actually hear rather than attempting the approximate associations you will be given in any description of the sounds of the language.

When working through an Assimil or Linguaphone type method by shadowing, blind shadowing each lesson initially before listening to the text forces your mind to understand as much as it can on its own. You only need do this once or twice before looking at the text for clarification, but if you look at the text first and foremost, you will understand more when you listen/shadow, but you will not know how much you would have understood without doing this.

[Kraut]

Final word?

https://www.youtube.com/watch?v=GCFD96b ... KA&index=1

Shadowing and imitation
Cangaroo English

[leosmith]

I know this discussion has gone a certain direction, and I’m not trying to reel it in, but I’ll state once again that by “shadowing” I just meant “listening and repeating”. I know that Argüelles popularized the term to mean simultaneous listening and repeating, but since I’ve heard it so often as just meaning listening and repeating, I didn’t think to specify. I put an edit in the OP a few days ago.

I do not simultaneously listen and repeat. I tried it off and on for a couple years, following the advice of Olle Kjellin. At first I thought it was very effective for improving pronunciation, because it was easy to tell when my pronunciation was “outside the box” of the native pronunciation. But after a time I came to realize that there was a lot going on “inside the box” that I could not hear because I was talking at the same time that I was listening, so as Iversen pointed out, my listening accuracy was diminished. My M.O. is to do a lot of listening and repeating, with a fraction of a second between the end of native audio and the beginning of my voice. I do it very often in the beginning, and continue doing it, although less frequently, as I continue to learn.

https://www.youtube.com/watch?v=GCFD96bwDqk&list=PLmIoa_Hs27lM8QzzrDkkcA2gz_FS0ToKA&index=1

I actually enjoyed this video, and agree that simultaneously listening and repeating, especially with the additional obstacles that he demonstrates, is very difficult. Difficult, but learnable. The main thing I didn’t like about the video is that he says “You are never going to lose your accent, so don’t waste your time trying”. At first I thought what he really means is that we shouldn’t be trying to perfect our pronunciation, because it will never happen. But he goes beyond that a few minutes later, discouraging learners from imitating natives. In fact he says “I think that native speakers are not people that we should be trying to imitate. As a foreign learner, or as a learner of English as a second language, you should always try to be you…just speaking in English”. Obviously he is targeting English learners, but the advice is not good imo. Learners should make an effort to develop decent pronunciation in a foreign language – it makes a big difference in how you are perceived, and how well you are understood, by natives. Again, I’m not saying one needs to be perfect, but please just make an effort.

[lingohot]

Here's what Arguelles wrote more than fifteen years ago (Message 20 of 39):

Blind shadowing first and foremost when learning the phonetics of a language like Korean forces you listen more carefully and to reproduce the sounds you actually hear rather than attempting the approximate associations you will be given in any description of the sounds of the language.

When working through an Assimil or Linguaphone type method by shadowing, blind shadowing each lesson initially before listening to the text forces your mind to understand as much as it can on its own. You only need do this once or twice before looking at the text for clarification, but if you look at the text first and foremost, you will understand more when you listen/shadow, but you will not know how much you would have understood without doing this.

I don't really understand that. Is shadowing not listening (and then trying to repeat what you hear simultaneously)?
And: you can also find out how much you understand if you just listen to the recording (instead of parroting it at the same time), even better so, because you don't have your own parroting in the way and can fully concentrate on your listening comprehension.

But after a time I came to realize that there was a lot going on “inside the box” that I could not hear because I was talking at the same time that I was listening, so as Iversen pointed out, my listening accuracy was diminished.

This is also what I think. It's rather something that disturbs the initial 'discovering' of the language and its sounds etc., because you always have your own voice and what you think the language sounds like overlaying the actual recording that you are supposed to learn from. So I really think it's rather a good exercise for very advanced 'learners' who virtually understand anything they hear (which is one of the most difficult skills, actually!), not for beginners or intermediate learners (or even advanced learners, because you can already have an advanced level, but still have some difficulties with listening comprehension, especially when you're speaking at the same time).

[leosmith]

So I really think it's rather a good exercise for very advanced 'learners' who virtually understand anything they hear

In the video, it was demonstrated that it's even difficult to do in one's L1. Actually, the host speaks slowly, and as an experiment I shadowed him for a while. The first minute or so was rough, but after that I got used to it and I could do it very well. When he had the other guy try to shadow the Australian recording, I tried to shadow that too and couldn't, at least in part due to interference from the English guy shadowing at the same time. My guess is that I wouldn't do well with it even if I was the only one, since it was normal speed.

Anyway, I wonder though if this would be a good way to improve your pronunciation if you already know the script and have listened to the audio several times. I know I said "I could do it very well" for slow L1, but it was still quite taxing. Maybe knowing the script by heart would let you really concentrate on the pronunciation. For example, knowing a song really well allows me to sing it with pretty accurate pronunciation. Similar principal, imo, except songs often contain non-standard pronunciation, or pronunciation natives normally wouldn't use in conversation.

[Iversen]

Listening while reading the text (from a complete and faithful transcription, I hope) would for me be the perfect way to do listening training, but adding your own lamentable babbling on top of that would be overkill (or just 'kill ). Of course the final goal is to be able to dispense with the text and just listen, and beyond that the ability to understand lousy recordings without help, but knowing exactly what you are suppose to hear can't from my perspective be a bad thing.

Contrary to Arguelles I'm quite worried about the negative effect of speaking yourself at the same time, so I would prefer listening and reading followed by echoing (not chorusing, please) - and this should be repeated a couple of times for each short snippet to get the sounds right. And in principle it might be useful then to have a qualified native speaker to evaluate the result, but personally I would prefer working on my own as much as possible. Constant interference from a teacher is at best a nuisance, at worst hell on earth, but once you think you have reached your goal in your own opinion, a qualified opinion from a pronunciation expert may tell you exactly how and why you haven't - and THEN you can do the necessary corrections.

The main problem has hitherto been to get sufficiently loyal transcriptions of clearly spoken speech (or simple plain readings without the usual histrionics), but technology may solve that problem for us.

[Cainntear]

Final word?

https://www.youtube.com/watch?v=GCFD96b ... KA&index=1

Shadowing and imitation
Cangaroo English

If you're trying to put the cat among the pigeons, well done! Expert troll!

While the guy's opinions have some weight because he's a teacher, he says...

You are never going to lose your accent, so don't waste your time trying; and this is not just my personal opinion, this is scientific fact

He's talking as an authority... but on what grounds?

One thing about science is that it can't prove something impossible universally -- the best that can be said is usually something along the lines of "it may be possible, but if so, we don't know how to do it," or (when discussing a specific technique) "statistically, this technique hadn't proven capable of achieving it."

I personally am disproof of his assertion, so at the very best he could claim that it's only possible for a tiny minority of weirdos. I use that word "weirdo" because even if you failed to see the derision in his face after the Argüelles video, you can't help but hear his audibly laughing.

Argüelles is clearly not neurotypical -- autistic traits are very common among university academics, and his on-camera style shows him as being likely notably on the spectrum.

If he has commented about how the thing Argüelles sets out being something that a lot of people would just find an unacceptable practice, that would have been fine. I'm fact, I do wonder whether Argüelles over the years had developed a method unconsciously that allowed him to active blame to the student for their own failure: "they didn't do it right" is a great catchall write-off, and if you make something very specifically match what you (think you) do, and which people will feel so silly doing that they just don't replicate it, then when they fail you can simply call the line "they didn't do it right" and your ego is protected. Of course, you can ignore that your successful students didn't do it right either, so your technique claims credit for successes and denies responsibility for failures.

Then later on he demonstrated that it's difficult... OK, so what? Being difficult does nothing to prove it doesn't work. Having an English guy fail to replicate an Australian accent trying it out for several minutes as an entirely new technique... well, it proves nothing about the long -term effectiveness of it as a technique. Perhaps doing it every day for a month wild price successful as a way for an actor to acquire a particular target accent... or perhaps not, but that's not particularly relevant to language learners.